diff --git a/Documentation/RCU/checklist.rst b/Documentation/RCU/checklist.rst index 2efed9926c3f9..54a79d03438a0 100644 --- a/Documentation/RCU/checklist.rst +++ b/Documentation/RCU/checklist.rst @@ -214,8 +214,8 @@ over a rather long period of time, but improvements are always welcome! the rest of the system. 7. As of v4.20, a given kernel implements only one RCU flavor, - which is RCU-sched for PREEMPT=n and RCU-preempt for PREEMPT=y. - If the updater uses call_rcu() or synchronize_rcu(), + which is RCU-sched for PREEMPTION=n and RCU-preempt for + PREEMPTION=y. If the updater uses call_rcu() or synchronize_rcu(), then the corresponding readers my use rcu_read_lock() and rcu_read_unlock(), rcu_read_lock_bh() and rcu_read_unlock_bh(), or any pair of primitives that disables and re-enables preemption, diff --git a/Documentation/RCU/stallwarn.rst b/Documentation/RCU/stallwarn.rst index c9ab6af4d3be9..e97d1b4876ef4 100644 --- a/Documentation/RCU/stallwarn.rst +++ b/Documentation/RCU/stallwarn.rst @@ -25,7 +25,7 @@ So your kernel printed an RCU CPU stall warning. The next question is - A CPU looping with bottom halves disabled. -- For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the kernel +- For !CONFIG_PREEMPTION kernels, a CPU looping anywhere in the kernel without invoking schedule(). If the looping in the kernel is really expected and desirable behavior, you might need to add some calls to cond_resched(). @@ -44,7 +44,7 @@ So your kernel printed an RCU CPU stall warning. The next question is result in the ``rcu_.*kthread starved for`` console-log message, which will include additional debugging information. -- A CPU-bound real-time task in a CONFIG_PREEMPT kernel, which might +- A CPU-bound real-time task in a CONFIG_PREEMPTION kernel, which might happen to preempt a low-priority task in the middle of an RCU read-side critical section. This is especially damaging if that low-priority task is not permitted to run on any other CPU, diff --git a/Documentation/admin-guide/kdump/gdbmacros.txt b/Documentation/admin-guide/kdump/gdbmacros.txt index 220d0a80ca2c9..82aecdcae8a6c 100644 --- a/Documentation/admin-guide/kdump/gdbmacros.txt +++ b/Documentation/admin-guide/kdump/gdbmacros.txt @@ -170,57 +170,82 @@ document trapinfo address the kernel panicked. end -define dump_log_idx - set $idx = $arg0 - if ($argc > 1) - set $prev_flags = $arg1 +define dump_record + set var $desc = $arg0 + set var $info = $arg1 + if ($argc > 2) + set var $prev_flags = $arg2 else - set $prev_flags = 0 - end - set $msg = ((struct printk_log *) (log_buf + $idx)) - set $prefix = 1 - set $newline = 1 - set $log = log_buf + $idx + sizeof(*$msg) - - # prev & LOG_CONT && !(msg->flags & LOG_PREIX) - if (($prev_flags & 8) && !($msg->flags & 4)) - set $prefix = 0 + set var $prev_flags = 0 end - # msg->flags & LOG_CONT - if ($msg->flags & 8) + set var $prefix = 1 + set var $newline = 1 + + set var $begin = $desc->text_blk_lpos.begin % (1U << prb->text_data_ring.size_bits) + set var $next = $desc->text_blk_lpos.next % (1U << prb->text_data_ring.size_bits) + + # handle data-less record + if ($begin & 1) + set var $text_len = 0 + set var $log = "" + else + # handle wrapping data block + if ($begin > $next) + set var $begin = 0 + end + + # skip over descriptor id + set var $begin = $begin + sizeof(long) + + # handle truncated message + if ($next - $begin < $info->text_len) + set var $text_len = $next - $begin + else + set var $text_len = $info->text_len + end + + set var $log = &prb->text_data_ring.data[$begin] + end + + # prev & LOG_CONT && !(info->flags & LOG_PREIX) + if (($prev_flags & 8) && !($info->flags & 4)) + set var $prefix = 0 + end + + # info->flags & LOG_CONT + if ($info->flags & 8) # (prev & LOG_CONT && !(prev & LOG_NEWLINE)) if (($prev_flags & 8) && !($prev_flags & 2)) - set $prefix = 0 + set var $prefix = 0 end - # (!(msg->flags & LOG_NEWLINE)) - if (!($msg->flags & 2)) - set $newline = 0 + # (!(info->flags & LOG_NEWLINE)) + if (!($info->flags & 2)) + set var $newline = 0 end end if ($prefix) - printf "[%5lu.%06lu] ", $msg->ts_nsec / 1000000000, $msg->ts_nsec % 1000000000 + printf "[%5lu.%06lu] ", $info->ts_nsec / 1000000000, $info->ts_nsec % 1000000000 end - if ($msg->text_len != 0) - eval "printf \"%%%d.%ds\", $log", $msg->text_len, $msg->text_len + if ($text_len) + eval "printf \"%%%d.%ds\", $log", $text_len, $text_len end if ($newline) printf "\n" end - if ($msg->dict_len > 0) - set $dict = $log + $msg->text_len - set $idx = 0 - set $line = 1 - while ($idx < $msg->dict_len) - if ($line) - printf " " - set $line = 0 - end - set $c = $dict[$idx] + + # handle dictionary data + + set var $dict = &$info->dev_info.subsystem[0] + set var $dict_len = sizeof($info->dev_info.subsystem) + if ($dict[0] != '\0') + printf " SUBSYSTEM=" + set var $idx = 0 + while ($idx < $dict_len) + set var $c = $dict[$idx] if ($c == '\0') - printf "\n" - set $line = 1 + loop_break else if ($c < ' ' || $c >= 127 || $c == '\\') printf "\\x%02x", $c @@ -228,33 +253,67 @@ define dump_log_idx printf "%c", $c end end - set $idx = $idx + 1 + set var $idx = $idx + 1 + end + printf "\n" + end + + set var $dict = &$info->dev_info.device[0] + set var $dict_len = sizeof($info->dev_info.device) + if ($dict[0] != '\0') + printf " DEVICE=" + set var $idx = 0 + while ($idx < $dict_len) + set var $c = $dict[$idx] + if ($c == '\0') + loop_break + else + if ($c < ' ' || $c >= 127 || $c == '\\') + printf "\\x%02x", $c + else + printf "%c", $c + end + end + set var $idx = $idx + 1 end printf "\n" end end -document dump_log_idx - Dump a single log given its index in the log buffer. The first - parameter is the index into log_buf, the second is optional and - specified the previous log buffer's flags, used for properly - formatting continued lines. +document dump_record + Dump a single record. The first parameter is the descriptor, + the second parameter is the info, the third parameter is + optional and specifies the previous record's flags, used for + properly formatting continued lines. end define dmesg - set $i = log_first_idx - set $end_idx = log_first_idx - set $prev_flags = 0 + # definitions from kernel/printk/printk_ringbuffer.h + set var $desc_committed = 1 + set var $desc_finalized = 2 + set var $desc_sv_bits = sizeof(long) * 8 + set var $desc_flags_shift = $desc_sv_bits - 2 + set var $desc_flags_mask = 3 << $desc_flags_shift + set var $id_mask = ~$desc_flags_mask + + set var $desc_count = 1U << prb->desc_ring.count_bits + set var $prev_flags = 0 + + set var $id = prb->desc_ring.tail_id.counter + set var $end_id = prb->desc_ring.head_id.counter while (1) - set $msg = ((struct printk_log *) (log_buf + $i)) - if ($msg->len == 0) - set $i = 0 - else - dump_log_idx $i $prev_flags - set $i = $i + $msg->len - set $prev_flags = $msg->flags + set var $desc = &prb->desc_ring.descs[$id % $desc_count] + set var $info = &prb->desc_ring.infos[$id % $desc_count] + + # skip non-committed record + set var $state = 3 & ($desc->state_var.counter >> $desc_flags_shift) + if ($state == $desc_committed || $state == $desc_finalized) + dump_record $desc $info $prev_flags + set var $prev_flags = $info->flags end - if ($i == $end_idx) + + set var $id = ($id + 1) & $id_mask + if ($id == $end_id) loop_break end end diff --git a/Documentation/admin-guide/kdump/vmcoreinfo.rst b/Documentation/admin-guide/kdump/vmcoreinfo.rst index 2baad0bfb09d0..e44a6c01f3362 100644 --- a/Documentation/admin-guide/kdump/vmcoreinfo.rst +++ b/Documentation/admin-guide/kdump/vmcoreinfo.rst @@ -189,50 +189,123 @@ from this. Free areas descriptor. User-space tools use this value to iterate the free_area ranges. MAX_ORDER is used by the zone buddy allocator. -log_first_idx -------------- +prb +--- -Index of the first record stored in the buffer log_buf. Used by -user-space tools to read the strings in the log_buf. +A pointer to the printk ringbuffer (struct printk_ringbuffer). This +may be pointing to the static boot ringbuffer or the dynamically +allocated ringbuffer, depending on when the the core dump occurred. +Used by user-space tools to read the active kernel log buffer. -log_buf -------- +printk_rb_static +---------------- -Console output is written to the ring buffer log_buf at index -log_first_idx. Used to get the kernel log. +A pointer to the static boot printk ringbuffer. If @prb has a +different value, this is useful for viewing the initial boot messages, +which may have been overwritten in the dynamically allocated +ringbuffer. -log_buf_len ------------ - -log_buf's length. - -clear_idx +clear_seq --------- -The index that the next printk() record to read after the last clear -command. It indicates the first record after the last SYSLOG_ACTION -_CLEAR, like issued by 'dmesg -c'. Used by user-space tools to dump -the dmesg log. +The sequence number of the printk() record after the last clear +command. It indicates the first record after the last +SYSLOG_ACTION_CLEAR, like issued by 'dmesg -c'. Used by user-space +tools to dump a subset of the dmesg log. -log_next_idx ------------- +printk_ringbuffer +----------------- -The index of the next record to store in the buffer log_buf. Used to -compute the index of the current buffer position. +The size of a printk_ringbuffer structure. This structure contains all +information required for accessing the various components of the +kernel log buffer. -printk_log ----------- +(printk_ringbuffer, desc_ring|text_data_ring|dict_data_ring|fail) +----------------------------------------------------------------- -The size of a structure printk_log. Used to compute the size of -messages, and extract dmesg log. It encapsulates header information for -log_buf, such as timestamp, syslog level, etc. +Offsets for the various components of the printk ringbuffer. Used by +user-space tools to view the kernel log buffer without requiring the +declaration of the structure. -(printk_log, ts_nsec|len|text_len|dict_len) -------------------------------------------- +prb_desc_ring +------------- -It represents field offsets in struct printk_log. User space tools -parse it and check whether the values of printk_log's members have been -changed. +The size of the prb_desc_ring structure. This structure contains +information about the set of record descriptors. + +(prb_desc_ring, count_bits|descs|head_id|tail_id) +------------------------------------------------- + +Offsets for the fields describing the set of record descriptors. Used +by user-space tools to be able to traverse the descriptors without +requiring the declaration of the structure. + +prb_desc +-------- + +The size of the prb_desc structure. This structure contains +information about a single record descriptor. + +(prb_desc, info|state_var|text_blk_lpos|dict_blk_lpos) +------------------------------------------------------ + +Offsets for the fields describing a record descriptors. Used by +user-space tools to be able to read descriptors without requiring +the declaration of the structure. + +prb_data_blk_lpos +----------------- + +The size of the prb_data_blk_lpos structure. This structure contains +information about where the text or dictionary data (data block) is +located within the respective data ring. + +(prb_data_blk_lpos, begin|next) +------------------------------- + +Offsets for the fields describing the location of a data block. Used +by user-space tools to be able to locate data blocks without +requiring the declaration of the structure. + +printk_info +----------- + +The size of the printk_info structure. This structure contains all +the meta-data for a record. + +(printk_info, seq|ts_nsec|text_len|dict_len|caller_id) +------------------------------------------------------ + +Offsets for the fields providing the meta-data for a record. Used by +user-space tools to be able to read the information without requiring +the declaration of the structure. + +prb_data_ring +------------- + +The size of the prb_data_ring structure. This structure contains +information about a set of data blocks. + +(prb_data_ring, size_bits|data|head_lpos|tail_lpos) +--------------------------------------------------- + +Offsets for the fields describing a set of data blocks. Used by +user-space tools to be able to access the data blocks without +requiring the declaration of the structure. + +atomic_long_t +------------- + +The size of the atomic_long_t structure. Used by user-space tools to +be able to copy the full structure, regardless of its +architecture-specific implementation. + +(atomic_long_t, counter) +------------------------ + +Offset for the long value of an atomic_long_t variable. Used by +user-space tools to access the long value without requiring the +architecture-specific declaration. (free_area.free_list, MIGRATE_TYPES) ------------------------------------ diff --git a/Documentation/locking/seqlock.rst b/Documentation/locking/seqlock.rst index 62c5ad98c11ca..a334b584f2b34 100644 --- a/Documentation/locking/seqlock.rst +++ b/Documentation/locking/seqlock.rst @@ -139,6 +139,24 @@ with the associated LOCKTYPE lock acquired. Read path: same as in :ref:`seqcount_t`. + +.. _seqcount_latch_t: + +Latch sequence counters (``seqcount_latch_t``) +---------------------------------------------- + +Latch sequence counters are a multiversion concurrency control mechanism +where the embedded seqcount_t counter even/odd value is used to switch +between two copies of protected data. This allows the sequence counter +read path to safely interrupt its own write side critical section. + +Use seqcount_latch_t when the write side sections cannot be protected +from interruption by readers. This is typically the case when the read +side can be invoked from NMI handlers. + +Check `raw_write_seqcount_latch()` for more information. + + .. _seqlock_t: Sequential locks (``seqlock_t``) diff --git a/MAINTAINERS b/MAINTAINERS index 867157311dc8b..7ae63272d994c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -13960,6 +13960,7 @@ PRINTK M: Petr Mladek M: Sergey Senozhatsky R: Steven Rostedt +R: John Ogness S: Maintained F: include/linux/printk.h F: kernel/printk/ diff --git a/arch/Kconfig b/arch/Kconfig index af14a567b493f..5c8e173dc7c2b 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -34,6 +34,7 @@ config OPROFILE tristate "OProfile system profiling" depends on PROFILING depends on HAVE_OPROFILE + depends on !PREEMPT_RT select RING_BUFFER select RING_BUFFER_ALLOW_SWAP help @@ -414,6 +415,13 @@ config MMU_GATHER_NO_GATHER bool depends on MMU_GATHER_TABLE_FREE +config ARCH_WANT_IRQS_OFF_ACTIVATE_MM + bool + help + Temporary select until all architectures can be converted to have + irqs disabled over activate_mm. Architectures that do IPI based TLB + shootdowns should enable this. + config ARCH_HAVE_NMI_SAFE_CMPXCHG bool diff --git a/arch/alpha/include/asm/spinlock_types.h b/arch/alpha/include/asm/spinlock_types.h index 1d5716bc060be..6883bc952d224 100644 --- a/arch/alpha/include/asm/spinlock_types.h +++ b/arch/alpha/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef _ALPHA_SPINLOCK_TYPES_H #define _ALPHA_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - typedef struct { volatile unsigned int lock; } arch_spinlock_t; diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index e00d94b166587..b7c20565754da 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -31,6 +31,7 @@ config ARM select ARCH_OPTIONAL_KERNEL_RWX if ARCH_HAS_STRICT_KERNEL_RWX select ARCH_OPTIONAL_KERNEL_RWX_DEFAULT if CPU_V7 select ARCH_SUPPORTS_ATOMIC_RMW + select ARCH_SUPPORTS_RT if HAVE_POSIX_CPU_TIMERS_TASK_WORK select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU @@ -64,7 +65,7 @@ config ARM select HARDIRQS_SW_RESEND select HAVE_ARCH_AUDITSYSCALL if AEABI && !OABI_COMPAT select HAVE_ARCH_BITREVERSE if (CPU_32v7M || CPU_32v7) && !CPU_32v6 - select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32 && MMU + select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32 && MMU && !PREEMPT_RT select HAVE_ARCH_KGDB if !CPU_ENDIAN_BE32 && MMU select HAVE_ARCH_MMAP_RND_BITS if MMU select HAVE_ARCH_SECCOMP_FILTER if AEABI && !OABI_COMPAT @@ -102,6 +103,7 @@ config ARM select HAVE_PERF_EVENTS select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP + select HAVE_PREEMPT_LAZY select MMU_GATHER_RCU_TABLE_FREE if SMP && ARM_LPAE select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_RSEQ @@ -117,6 +119,7 @@ config ARM select OLD_SIGSUSPEND3 select PCI_SYSCALL if PCI select PERF_USE_VMALLOC + select HAVE_POSIX_CPU_TIMERS_TASK_WORK if !KVM select RTC_LIB select SYS_SUPPORTS_APM_EMULATION # Above selects are sorted alphabetically; please add new ones diff --git a/arch/arm/include/asm/spinlock_types.h b/arch/arm/include/asm/spinlock_types.h index 5976958647fe1..a37c0803954ba 100644 --- a/arch/arm/include/asm/spinlock_types.h +++ b/arch/arm/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef __ASM_SPINLOCK_TYPES_H #define __ASM_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - #define TICKET_SHIFT 16 typedef struct { diff --git a/arch/arm/include/asm/switch_to.h b/arch/arm/include/asm/switch_to.h index 007d8fea71572..285e6248454fc 100644 --- a/arch/arm/include/asm/switch_to.h +++ b/arch/arm/include/asm/switch_to.h @@ -4,6 +4,13 @@ #include +#if defined CONFIG_PREEMPT_RT && defined CONFIG_HIGHMEM +void switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p); +#else +static inline void +switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p) { } +#endif + /* * For v7 SMP cores running a preemptible kernel we may be pre-empted * during a TLB maintenance operation, so execute an inner-shareable dsb @@ -26,6 +33,7 @@ extern struct task_struct *__switch_to(struct task_struct *, struct thread_info #define switch_to(prev,next,last) \ do { \ __complete_pending_tlbi(); \ + switch_kmaps(prev, next); \ last = __switch_to(prev,task_thread_info(prev), task_thread_info(next)); \ } while (0) diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h index 536b6b979f634..875aaf9af9465 100644 --- a/arch/arm/include/asm/thread_info.h +++ b/arch/arm/include/asm/thread_info.h @@ -46,6 +46,7 @@ struct cpu_context_save { struct thread_info { unsigned long flags; /* low level flags */ int preempt_count; /* 0 => preemptable, <0 => bug */ + int preempt_lazy_count; /* 0 => preemptable, <0 => bug */ mm_segment_t addr_limit; /* address limit */ struct task_struct *task; /* main task structure */ __u32 cpu; /* cpu */ @@ -134,7 +135,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, #define TIF_SYSCALL_TRACE 4 /* syscall trace active */ #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */ #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */ -#define TIF_SECCOMP 7 /* seccomp syscall filtering active */ +#define TIF_NEED_RESCHED_LAZY 7 +#define TIF_SECCOMP 8 /* seccomp syscall filtering active */ #define TIF_USING_IWMMXT 17 #define TIF_MEMDIE 18 /* is terminating due to OOM killer */ @@ -143,6 +145,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME) +#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY) #define _TIF_UPROBE (1 << TIF_UPROBE) #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) @@ -158,7 +161,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp *, * Change these and you break ASM code in entry-common.S */ #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ - _TIF_NOTIFY_RESUME | _TIF_UPROBE) + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ + _TIF_NEED_RESCHED_LAZY) #endif /* __KERNEL__ */ #endif /* __ASM_ARM_THREAD_INFO_H */ diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c index a1570c8bab25a..88a0eb048ca1b 100644 --- a/arch/arm/kernel/asm-offsets.c +++ b/arch/arm/kernel/asm-offsets.c @@ -41,6 +41,7 @@ int main(void) BLANK(); DEFINE(TI_FLAGS, offsetof(struct thread_info, flags)); DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count)); + DEFINE(TI_PREEMPT_LAZY, offsetof(struct thread_info, preempt_lazy_count)); DEFINE(TI_ADDR_LIMIT, offsetof(struct thread_info, addr_limit)); DEFINE(TI_TASK, offsetof(struct thread_info, task)); DEFINE(TI_CPU, offsetof(struct thread_info, cpu)); diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 55a47df047738..1e689a727cb9b 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -206,11 +206,18 @@ ENDPROC(__dabt_svc) #ifdef CONFIG_PREEMPTION ldr r8, [tsk, #TI_PREEMPT] @ get preempt count - ldr r0, [tsk, #TI_FLAGS] @ get flags teq r8, #0 @ if preempt count != 0 + bne 1f @ return from exeption + ldr r0, [tsk, #TI_FLAGS] @ get flags + tst r0, #_TIF_NEED_RESCHED @ if NEED_RESCHED is set + blne svc_preempt @ preempt! + + ldr r8, [tsk, #TI_PREEMPT_LAZY] @ get preempt lazy count + teq r8, #0 @ if preempt lazy count != 0 movne r0, #0 @ force flags to 0 - tst r0, #_TIF_NEED_RESCHED + tst r0, #_TIF_NEED_RESCHED_LAZY blne svc_preempt +1: #endif svc_exit r5, irq = 1 @ return from exception @@ -225,8 +232,14 @@ ENDPROC(__irq_svc) 1: bl preempt_schedule_irq @ irq en/disable is done inside ldr r0, [tsk, #TI_FLAGS] @ get new tasks TI_FLAGS tst r0, #_TIF_NEED_RESCHED + bne 1b + tst r0, #_TIF_NEED_RESCHED_LAZY reteq r8 @ go again - b 1b + ldr r0, [tsk, #TI_PREEMPT_LAZY] @ get preempt lazy count + teq r0, #0 @ if preempt lazy count != 0 + beq 1b + ret r8 @ go again + #endif __und_fault: diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 271cb8a1eba1e..fd039b1b37319 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -53,7 +53,9 @@ saved_pc .req lr cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + tst r1, #((_TIF_SYSCALL_WORK | _TIF_WORK_MASK) & ~_TIF_SECCOMP) + bne fast_work_pending + tst r1, #_TIF_SECCOMP bne fast_work_pending @@ -90,8 +92,11 @@ ENDPROC(ret_fast_syscall) cmp r2, #TASK_SIZE blne addr_limit_check_failed ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK + tst r1, #((_TIF_SYSCALL_WORK | _TIF_WORK_MASK) & ~_TIF_SECCOMP) + bne do_slower_path + tst r1, #_TIF_SECCOMP beq no_work_pending +do_slower_path: UNWIND(.fnend ) ENDPROC(ret_fast_syscall) diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c index c1892f733f208..fb8c029549012 100644 --- a/arch/arm/kernel/signal.c +++ b/arch/arm/kernel/signal.c @@ -649,7 +649,8 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall) */ trace_hardirqs_off(); do { - if (likely(thread_flags & _TIF_NEED_RESCHED)) { + if (likely(thread_flags & (_TIF_NEED_RESCHED | + _TIF_NEED_RESCHED_LAZY))) { schedule(); } else { if (unlikely(!user_mode(regs))) diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index 5d9da61eff622..7f83a013dccca 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -680,11 +680,9 @@ void handle_IPI(int ipinr, struct pt_regs *regs) break; case IPI_CPU_BACKTRACE: - printk_nmi_enter(); irq_enter(); nmi_cpu_backtrace(regs); irq_exit(); - printk_nmi_exit(); break; default: diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c index efa4020250315..59487ee9fd612 100644 --- a/arch/arm/mm/fault.c +++ b/arch/arm/mm/fault.c @@ -400,6 +400,9 @@ do_translation_fault(unsigned long addr, unsigned int fsr, if (addr < TASK_SIZE) return do_page_fault(addr, fsr, regs); + if (interrupts_enabled(regs)) + local_irq_enable(); + if (user_mode(regs)) goto bad_area; @@ -470,6 +473,9 @@ do_translation_fault(unsigned long addr, unsigned int fsr, static int do_sect_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs) { + if (interrupts_enabled(regs)) + local_irq_enable(); + do_bad_area(addr, fsr, regs); return 0; } diff --git a/arch/arm/mm/highmem.c b/arch/arm/mm/highmem.c index 187fab227b509..954a115f62054 100644 --- a/arch/arm/mm/highmem.c +++ b/arch/arm/mm/highmem.c @@ -31,8 +31,14 @@ static inline pte_t get_fixmap_pte(unsigned long vaddr) return *ptep; } +static unsigned int fixmap_idx(int type) +{ + return FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id(); +} + void *kmap_atomic_high_prot(struct page *page, pgprot_t prot) { + pte_t pte = mk_pte(page, kmap_prot); unsigned int idx; unsigned long vaddr; void *kmap; @@ -53,7 +59,7 @@ void *kmap_atomic_high_prot(struct page *page, pgprot_t prot) type = kmap_atomic_idx_push(); - idx = FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id(); + idx = fixmap_idx(type); vaddr = __fix_to_virt(idx); #ifdef CONFIG_DEBUG_HIGHMEM /* @@ -61,13 +67,16 @@ void *kmap_atomic_high_prot(struct page *page, pgprot_t prot) * Make sure it was indeed properly unmapped. */ BUG_ON(!pte_none(get_fixmap_pte(vaddr))); +#endif +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = pte; #endif /* * When debugging is off, kunmap_atomic leaves the previous mapping * in place, so the contained TLB flush ensures the TLB is updated * with the new mapping. */ - set_fixmap_pte(idx, mk_pte(page, prot)); + set_fixmap_pte(idx, pte); return (void *)vaddr; } @@ -80,16 +89,19 @@ void kunmap_atomic_high(void *kvaddr) if (kvaddr >= (void *)FIXADDR_START) { type = kmap_atomic_idx(); - idx = FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id(); + idx = fixmap_idx(type); if (cache_is_vivt()) __cpuc_flush_dcache_area((void *)vaddr, PAGE_SIZE); +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = __pte(0); +#endif #ifdef CONFIG_DEBUG_HIGHMEM BUG_ON(vaddr != __fix_to_virt(idx)); - set_fixmap_pte(idx, __pte(0)); #else (void) idx; /* to kill a warning */ #endif + set_fixmap_pte(idx, __pte(0)); kmap_atomic_idx_pop(); } else if (vaddr >= PKMAP_ADDR(0) && vaddr < PKMAP_ADDR(LAST_PKMAP)) { /* this address was obtained through kmap_high_get() */ @@ -100,22 +112,51 @@ EXPORT_SYMBOL(kunmap_atomic_high); void *kmap_atomic_pfn(unsigned long pfn) { + pte_t pte = pfn_pte(pfn, kmap_prot); unsigned long vaddr; int idx, type; struct page *page = pfn_to_page(pfn); - preempt_disable(); + migrate_disable(); pagefault_disable(); if (!PageHighMem(page)) return page_address(page); type = kmap_atomic_idx_push(); - idx = FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id(); + idx = fixmap_idx(type); vaddr = __fix_to_virt(idx); #ifdef CONFIG_DEBUG_HIGHMEM BUG_ON(!pte_none(get_fixmap_pte(vaddr))); #endif - set_fixmap_pte(idx, pfn_pte(pfn, kmap_prot)); +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = pte; +#endif + set_fixmap_pte(idx, pte); return (void *)vaddr; } + +#if defined CONFIG_PREEMPT_RT +void switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p) +{ + int i; + + /* + * Clear @prev's kmap_atomic mappings + */ + for (i = 0; i < prev_p->kmap_idx; i++) { + int idx = fixmap_idx(i); + + set_fixmap_pte(idx, __pte(0)); + } + /* + * Restore @next_p's kmap_atomic mappings + */ + for (i = 0; i < next_p->kmap_idx; i++) { + int idx = fixmap_idx(i); + + if (!pte_none(next_p->kmap_pte[i])) + set_fixmap_pte(idx, next_p->kmap_pte[i]); + } +} +#endif diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 6d232837cbeee..4da217d5b84c2 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -75,6 +75,7 @@ config ARM64 select ARCH_SUPPORTS_ATOMIC_RMW select ARCH_SUPPORTS_INT128 if CC_HAS_INT128 && (GCC_VERSION >= 50000 || CC_IS_CLANG) select ARCH_SUPPORTS_NUMA_BALANCING + select ARCH_SUPPORTS_RT if HAVE_POSIX_CPU_TIMERS_TASK_WORK select ARCH_WANT_COMPAT_IPC_PARSE_VERSION if COMPAT select ARCH_WANT_DEFAULT_BPF_JIT select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT @@ -169,6 +170,7 @@ config ARM64 select HAVE_PERF_EVENTS select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP + select HAVE_PREEMPT_LAZY select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_FUNCTION_ARG_ACCESS_API select HAVE_FUTEX_CMPXCHG if FUTEX @@ -190,6 +192,7 @@ config ARM64 select PCI_DOMAINS_GENERIC if PCI select PCI_ECAM if (ACPI && PCI) select PCI_SYSCALL if PCI + select HAVE_POSIX_CPU_TIMERS_TASK_WORK if !KVM select POWER_RESET select POWER_SUPPLY select SPARSE_IRQ diff --git a/arch/arm64/include/asm/preempt.h b/arch/arm64/include/asm/preempt.h index 80e946b2abee2..994f997b15724 100644 --- a/arch/arm64/include/asm/preempt.h +++ b/arch/arm64/include/asm/preempt.h @@ -70,17 +70,43 @@ static inline bool __preempt_count_dec_and_test(void) * interrupt occurring between the non-atomic READ_ONCE/WRITE_ONCE * pair. */ - return !pc || !READ_ONCE(ti->preempt_count); + if (!pc || !READ_ONCE(ti->preempt_count)) + return true; +#ifdef CONFIG_PREEMPT_LAZY + if ((pc & ~PREEMPT_NEED_RESCHED)) + return false; + if (current_thread_info()->preempt_lazy_count) + return false; + return test_thread_flag(TIF_NEED_RESCHED_LAZY); +#else + return false; +#endif } static inline bool should_resched(int preempt_offset) { +#ifdef CONFIG_PREEMPT_LAZY + u64 pc = READ_ONCE(current_thread_info()->preempt_count); + if (pc == preempt_offset) + return true; + + if ((pc & ~PREEMPT_NEED_RESCHED) != preempt_offset) + return false; + + if (current_thread_info()->preempt_lazy_count) + return false; + return test_thread_flag(TIF_NEED_RESCHED_LAZY); +#else u64 pc = READ_ONCE(current_thread_info()->preempt_count); return pc == preempt_offset; +#endif } #ifdef CONFIG_PREEMPTION void preempt_schedule(void); +#ifdef CONFIG_PREEMPT_RT +void preempt_schedule_lock(void); +#endif #define __preempt_schedule() preempt_schedule() void preempt_schedule_notrace(void); #define __preempt_schedule_notrace() preempt_schedule_notrace() diff --git a/arch/arm64/include/asm/spinlock_types.h b/arch/arm64/include/asm/spinlock_types.h index 18782f0c47212..6672b05350b4e 100644 --- a/arch/arm64/include/asm/spinlock_types.h +++ b/arch/arm64/include/asm/spinlock_types.h @@ -5,10 +5,6 @@ #ifndef __ASM_SPINLOCK_TYPES_H #define __ASM_SPINLOCK_TYPES_H -#if !defined(__LINUX_SPINLOCK_TYPES_H) && !defined(__ASM_SPINLOCK_H) -# error "please don't include this file directly" -#endif - #include #include diff --git a/arch/arm64/include/asm/thread_info.h b/arch/arm64/include/asm/thread_info.h index 5e784e16ee895..5d9f16f92c11e 100644 --- a/arch/arm64/include/asm/thread_info.h +++ b/arch/arm64/include/asm/thread_info.h @@ -29,6 +29,7 @@ struct thread_info { #ifdef CONFIG_ARM64_SW_TTBR0_PAN u64 ttbr0; /* saved TTBR0_EL1 */ #endif + int preempt_lazy_count; /* 0 => preemptable, <0 => bug */ union { u64 preempt_count; /* 0 => preemptible, <0 => bug */ struct { @@ -67,6 +68,7 @@ void arch_release_task_struct(struct task_struct *tsk); #define TIF_FOREIGN_FPSTATE 3 /* CPU's FP state is not current's */ #define TIF_UPROBE 4 /* uprobe breakpoint or singlestep */ #define TIF_FSCHECK 5 /* Check FS is USER_DS on return */ +#define TIF_NEED_RESCHED_LAZY 6 #define TIF_SYSCALL_TRACE 8 /* syscall trace active */ #define TIF_SYSCALL_AUDIT 9 /* syscall auditing */ #define TIF_SYSCALL_TRACEPOINT 10 /* syscall tracepoint for ftrace */ @@ -93,14 +95,16 @@ void arch_release_task_struct(struct task_struct *tsk); #define _TIF_SYSCALL_EMU (1 << TIF_SYSCALL_EMU) #define _TIF_UPROBE (1 << TIF_UPROBE) #define _TIF_FSCHECK (1 << TIF_FSCHECK) +#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY) #define _TIF_SINGLESTEP (1 << TIF_SINGLESTEP) #define _TIF_32BIT (1 << TIF_32BIT) #define _TIF_SVE (1 << TIF_SVE) #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ _TIF_NOTIFY_RESUME | _TIF_FOREIGN_FPSTATE | \ - _TIF_UPROBE | _TIF_FSCHECK) + _TIF_UPROBE | _TIF_FSCHECK | _TIF_NEED_RESCHED_LAZY) +#define _TIF_NEED_RESCHED_MASK (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY) #define _TIF_SYSCALL_WORK (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | \ _TIF_SYSCALL_TRACEPOINT | _TIF_SECCOMP | \ _TIF_SYSCALL_EMU) diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c index 7d32fc959b1a0..b2f29bd2ae871 100644 --- a/arch/arm64/kernel/asm-offsets.c +++ b/arch/arm64/kernel/asm-offsets.c @@ -30,6 +30,7 @@ int main(void) BLANK(); DEFINE(TSK_TI_FLAGS, offsetof(struct task_struct, thread_info.flags)); DEFINE(TSK_TI_PREEMPT, offsetof(struct task_struct, thread_info.preempt_count)); + DEFINE(TSK_TI_PREEMPT_LAZY, offsetof(struct task_struct, thread_info.preempt_lazy_count)); DEFINE(TSK_TI_ADDR_LIMIT, offsetof(struct task_struct, thread_info.addr_limit)); #ifdef CONFIG_ARM64_SW_TTBR0_PAN DEFINE(TSK_TI_TTBR0, offsetof(struct task_struct, thread_info.ttbr0)); diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index 55af8b504b65a..de7cd345c633e 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -624,9 +624,18 @@ alternative_if ARM64_HAS_IRQ_PRIO_MASKING mrs x0, daif orr x24, x24, x0 alternative_else_nop_endif - cbnz x24, 1f // preempt count != 0 || NMI return path - bl arm64_preempt_schedule_irq // irq en/disable is done inside + + cbz x24, 1f // (need_resched + count) == 0 + cbnz w24, 2f // count != 0 + + ldr w24, [tsk, #TSK_TI_PREEMPT_LAZY] // get preempt lazy count + cbnz w24, 2f // preempt lazy count != 0 + + ldr x0, [tsk, #TSK_TI_FLAGS] // get flags + tbz x0, #TIF_NEED_RESCHED_LAZY, 2f // needs rescheduling? 1: + bl arm64_preempt_schedule_irq // irq en/disable is done inside +2: #endif #ifdef CONFIG_ARM64_PSEUDO_NMI diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c index 55c8f3ec6705d..638a41b27febb 100644 --- a/arch/arm64/kernel/fpsimd.c +++ b/arch/arm64/kernel/fpsimd.c @@ -224,6 +224,16 @@ static void sve_free(struct task_struct *task) __sve_free(task); } +static void *sve_free_atomic(struct task_struct *task) +{ + void *sve_state = task->thread.sve_state; + + WARN_ON(test_tsk_thread_flag(task, TIF_SVE)); + + task->thread.sve_state = NULL; + return sve_state; +} + /* * TIF_SVE controls whether a task can use SVE without trapping while * in userspace, and also the way a task's FPSIMD/SVE state is stored @@ -1020,6 +1030,7 @@ void fpsimd_thread_switch(struct task_struct *next) void fpsimd_flush_thread(void) { int vl, supported_vl; + void *mem = NULL; if (!system_supports_fpsimd()) return; @@ -1032,7 +1043,7 @@ void fpsimd_flush_thread(void) if (system_supports_sve()) { clear_thread_flag(TIF_SVE); - sve_free(current); + mem = sve_free_atomic(current); /* * Reset the task vector length as required. @@ -1066,6 +1077,7 @@ void fpsimd_flush_thread(void) } put_cpu_fpsimd_context(); + kfree(mem); } /* diff --git a/arch/arm64/kernel/signal.c b/arch/arm64/kernel/signal.c index 3b4f31f35e458..441a970fc7ceb 100644 --- a/arch/arm64/kernel/signal.c +++ b/arch/arm64/kernel/signal.c @@ -921,7 +921,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, /* Check valid user FS if needed */ addr_limit_user_check(); - if (thread_flags & _TIF_NEED_RESCHED) { + if (thread_flags & _TIF_NEED_RESCHED_MASK) { /* Unmask Debug and SError for the next task */ local_daif_restore(DAIF_PROCCTX_NOIRQ); diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index b588c3b5c2f07..19cb5b1012264 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -681,7 +681,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) * involves poking the GIC, which must be done in a * non-preemptible context. */ - preempt_disable(); + migrate_disable(); kvm_pmu_flush_hwstate(vcpu); @@ -730,7 +730,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) kvm_timer_sync_user(vcpu); kvm_vgic_sync_hwstate(vcpu); local_irq_enable(); - preempt_enable(); + migrate_enable(); continue; } @@ -802,7 +802,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) /* Exit types that need handling before we can be preempted */ handle_exit_early(vcpu, ret); - preempt_enable(); + migrate_enable(); ret = handle_exit(vcpu, ret); } diff --git a/arch/hexagon/include/asm/spinlock_types.h b/arch/hexagon/include/asm/spinlock_types.h index 19d233497ba52..de72fb23016dc 100644 --- a/arch/hexagon/include/asm/spinlock_types.h +++ b/arch/hexagon/include/asm/spinlock_types.h @@ -8,10 +8,6 @@ #ifndef _ASM_SPINLOCK_TYPES_H #define _ASM_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - typedef struct { volatile unsigned int lock; } arch_spinlock_t; diff --git a/arch/ia64/include/asm/spinlock_types.h b/arch/ia64/include/asm/spinlock_types.h index 6e345fefcdcab..681408d6816fb 100644 --- a/arch/ia64/include/asm/spinlock_types.h +++ b/arch/ia64/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef _ASM_IA64_SPINLOCK_TYPES_H #define _ASM_IA64_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - typedef struct { volatile unsigned int lock; } arch_spinlock_t; diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 8f328298f8cc8..7624e089b736c 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -2653,7 +2653,7 @@ config MIPS_CRC_SUPPORT # config HIGHMEM bool "High Memory Support" - depends on 32BIT && CPU_SUPPORTS_HIGHMEM && SYS_SUPPORTS_HIGHMEM && !CPU_MIPS32_3_5_EVA + depends on 32BIT && CPU_SUPPORTS_HIGHMEM && SYS_SUPPORTS_HIGHMEM && !CPU_MIPS32_3_5_EVA && !PREEMPT_RT config CPU_SUPPORTS_HIGHMEM bool diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 787e829b6f25c..7c4e90562f90b 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -143,6 +143,7 @@ config PPC select ARCH_MIGHT_HAVE_PC_SERIO select ARCH_OPTIONAL_KERNEL_RWX if ARCH_HAS_STRICT_KERNEL_RWX select ARCH_SUPPORTS_ATOMIC_RMW + select ARCH_SUPPORTS_RT if HAVE_POSIX_CPU_TIMERS_TASK_WORK select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if PPC64 select ARCH_USE_QUEUED_RWLOCKS if PPC_QUEUED_SPINLOCKS @@ -225,6 +226,7 @@ config PPC select HAVE_HARDLOCKUP_DETECTOR_PERF if PERF_EVENTS && HAVE_PERF_EVENTS_NMI && !HAVE_HARDLOCKUP_DETECTOR_ARCH select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP + select HAVE_PREEMPT_LAZY select MMU_GATHER_RCU_TABLE_FREE select MMU_GATHER_PAGE_SIZE select HAVE_REGS_AND_STACK_ACCESS_API @@ -246,6 +248,7 @@ config PPC select OLD_SIGSUSPEND select PCI_DOMAINS if PCI select PCI_SYSCALL if PCI + select HAVE_POSIX_CPU_TIMERS_TASK_WORK if !KVM select PPC_DAWR if PPC64 select RTC_LIB select SPARSE_IRQ @@ -403,7 +406,7 @@ menu "Kernel options" config HIGHMEM bool "High memory support" - depends on PPC32 + depends on PPC32 && !PREEMPT_RT source "kernel/Kconfig.hz" diff --git a/arch/powerpc/include/asm/spinlock_types.h b/arch/powerpc/include/asm/spinlock_types.h index c5d742f18021d..cc6922a011ba1 100644 --- a/arch/powerpc/include/asm/spinlock_types.h +++ b/arch/powerpc/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef _ASM_POWERPC_SPINLOCK_TYPES_H #define _ASM_POWERPC_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - #ifdef CONFIG_PPC_QUEUED_SPINLOCKS #include #include diff --git a/arch/powerpc/include/asm/stackprotector.h b/arch/powerpc/include/asm/stackprotector.h index 1c8460e235838..b1653c160bab9 100644 --- a/arch/powerpc/include/asm/stackprotector.h +++ b/arch/powerpc/include/asm/stackprotector.h @@ -24,7 +24,11 @@ static __always_inline void boot_init_stack_canary(void) unsigned long canary; /* Try to get a semi random initial value. */ +#ifdef CONFIG_PREEMPT_RT + canary = (unsigned long)&canary; +#else canary = get_random_canary(); +#endif canary ^= mftb(); canary ^= LINUX_VERSION_CODE; canary &= CANARY_MASK; diff --git a/arch/powerpc/include/asm/thread_info.h b/arch/powerpc/include/asm/thread_info.h index ca6c970257049..3eda5d38c4182 100644 --- a/arch/powerpc/include/asm/thread_info.h +++ b/arch/powerpc/include/asm/thread_info.h @@ -48,6 +48,8 @@ struct thread_info { int preempt_count; /* 0 => preemptable, <0 => BUG */ + int preempt_lazy_count; /* 0 => preemptable, + <0 => BUG */ unsigned long local_flags; /* private flags for thread */ #ifdef CONFIG_LIVEPATCH unsigned long *livepatch_sp; @@ -98,11 +100,12 @@ void arch_setup_new_exec(void); #define TIF_SINGLESTEP 8 /* singlestepping active */ #define TIF_NOHZ 9 /* in adaptive nohz mode */ #define TIF_SECCOMP 10 /* secure computing */ -#define TIF_RESTOREALL 11 /* Restore all regs (implies NOERROR) */ -#define TIF_NOERROR 12 /* Force successful syscall return */ + +#define TIF_NEED_RESCHED_LAZY 11 /* lazy rescheduling necessary */ +#define TIF_SYSCALL_TRACEPOINT 12 /* syscall tracepoint instrumentation */ + #define TIF_NOTIFY_RESUME 13 /* callback before returning to user */ #define TIF_UPROBE 14 /* breakpointed or single-stepping */ -#define TIF_SYSCALL_TRACEPOINT 15 /* syscall tracepoint instrumentation */ #define TIF_EMULATE_STACK_STORE 16 /* Is an instruction emulation for stack store? */ #define TIF_MEMDIE 17 /* is terminating due to OOM killer */ @@ -111,6 +114,9 @@ void arch_setup_new_exec(void); #endif #define TIF_POLLING_NRFLAG 19 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_32BIT 20 /* 32 bit binary */ +#define TIF_RESTOREALL 21 /* Restore all regs (implies NOERROR) */ +#define TIF_NOERROR 22 /* Force successful syscall return */ + /* as above, but as bit values */ #define _TIF_SYSCALL_TRACE (1<msr & MSR_EE)); again: - if (IS_ENABLED(CONFIG_PREEMPT)) { + if (IS_ENABLED(CONFIG_PREEMPTION)) { /* Return to preemptible kernel context */ if (unlikely(*ti_flagsp & _TIF_NEED_RESCHED)) { if (preempt_count() == 0) preempt_schedule_irq(); + } else if (unlikely(*ti_flagsp & _TIF_NEED_RESCHED_LAZY)) { + if (current_thread_info()->preempt_lazy_count == 0) + preempt_schedule_irq(); } } diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index d1ebe152f2107..f5a1468b8d5b9 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -170,7 +170,6 @@ extern void panic_flush_kmsg_start(void) extern void panic_flush_kmsg_end(void) { - printk_safe_flush_on_panic(); kmsg_dump(KMSG_DUMP_PANIC); bust_spinlocks(0); debug_locks_off(); @@ -260,12 +259,17 @@ static char *get_mmu_str(void) static int __die(const char *str, struct pt_regs *regs, long err) { + const char *pr = ""; + printk("Oops: %s, sig: %ld [#%d]\n", str, err, ++die_counter); + if (IS_ENABLED(CONFIG_PREEMPTION)) + pr = IS_ENABLED(CONFIG_PREEMPT_RT) ? " PREEMPT_RT" : " PREEMPT"; + printk("%s PAGE_SIZE=%luK%s%s%s%s%s%s %s\n", IS_ENABLED(CONFIG_CPU_LITTLE_ENDIAN) ? "LE" : "BE", PAGE_SIZE / 1024, get_mmu_str(), - IS_ENABLED(CONFIG_PREEMPT) ? " PREEMPT" : "", + pr, IS_ENABLED(CONFIG_SMP) ? " SMP" : "", IS_ENABLED(CONFIG_SMP) ? (" NR_CPUS=" __stringify(NR_CPUS)) : "", debug_pagealloc_enabled() ? " DEBUG_PAGEALLOC" : "", diff --git a/arch/powerpc/kernel/watchdog.c b/arch/powerpc/kernel/watchdog.c index af3c15a1d41eb..8ae46c5945d07 100644 --- a/arch/powerpc/kernel/watchdog.c +++ b/arch/powerpc/kernel/watchdog.c @@ -181,11 +181,6 @@ static void watchdog_smp_panic(int cpu, u64 tb) wd_smp_unlock(&flags); - printk_safe_flush(); - /* - * printk_safe_flush() seems to require another print - * before anything actually goes out to console. - */ if (sysctl_hardlockup_all_cpu_backtrace) trigger_allbutself_cpu_backtrace(); diff --git a/arch/powerpc/kexec/crash.c b/arch/powerpc/kexec/crash.c index c9a889880214e..d488311efab1f 100644 --- a/arch/powerpc/kexec/crash.c +++ b/arch/powerpc/kexec/crash.c @@ -311,9 +311,6 @@ void default_machine_crash_shutdown(struct pt_regs *regs) unsigned int i; int (*old_handler)(struct pt_regs *regs); - /* Avoid hardlocking with irresponsive CPU holding logbuf_lock */ - printk_nmi_enter(); - /* * This function is only called after the system * has panicked or is otherwise in a critical state. diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig index 549591d9aaa2c..efb5bfe93f705 100644 --- a/arch/powerpc/kvm/Kconfig +++ b/arch/powerpc/kvm/Kconfig @@ -178,6 +178,7 @@ config KVM_E500MC config KVM_MPIC bool "KVM in-kernel MPIC emulation" depends on KVM && E500 + depends on !PREEMPT_RT select HAVE_KVM_IRQCHIP select HAVE_KVM_IRQFD select HAVE_KVM_IRQ_ROUTING diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 6d47b4a3ce395..15eef4d607ed3 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -177,6 +178,7 @@ static int tce_build_pSeriesLP(unsigned long liobn, long tcenum, long tceshift, } static DEFINE_PER_CPU(__be64 *, tce_page); +static DEFINE_LOCAL_IRQ_LOCK(tcp_page_lock); static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages, unsigned long uaddr, @@ -198,7 +200,8 @@ static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, direction, attrs); } - local_irq_save(flags); /* to protect tcep and the page behind it */ + /* to protect tcep and the page behind it */ + local_lock_irqsave(tcp_page_lock, flags); tcep = __this_cpu_read(tce_page); @@ -209,7 +212,7 @@ static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, tcep = (__be64 *)__get_free_page(GFP_ATOMIC); /* If allocation fails, fall back to the loop implementation */ if (!tcep) { - local_irq_restore(flags); + local_unlock_irqrestore(tcp_page_lock, flags); return tce_build_pSeriesLP(tbl->it_index, tcenum, tbl->it_page_shift, npages, uaddr, direction, attrs); @@ -244,7 +247,7 @@ static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum, tcenum += limit; } while (npages > 0 && !rc); - local_irq_restore(flags); + local_unlock_irqrestore(tcp_page_lock, flags); if (unlikely(rc == H_NOT_ENOUGH_RESOURCES)) { ret = (int)rc; @@ -415,13 +418,14 @@ static int tce_setrange_multi_pSeriesLP(unsigned long start_pfn, DMA_BIDIRECTIONAL, 0); } - local_irq_disable(); /* to protect tcep and the page behind it */ + /* to protect tcep and the page behind it */ + local_lock_irq(tcp_page_lock); tcep = __this_cpu_read(tce_page); if (!tcep) { tcep = (__be64 *)__get_free_page(GFP_ATOMIC); if (!tcep) { - local_irq_enable(); + local_unlock_irq(tcp_page_lock); return -ENOMEM; } __this_cpu_write(tce_page, tcep); @@ -467,7 +471,7 @@ static int tce_setrange_multi_pSeriesLP(unsigned long start_pfn, /* error cleanup: caller will clear whole range */ - local_irq_enable(); + local_unlock_irq(tcp_page_lock); return rc; } diff --git a/arch/s390/include/asm/spinlock_types.h b/arch/s390/include/asm/spinlock_types.h index cfed272e4fd59..8e28e8176ec88 100644 --- a/arch/s390/include/asm/spinlock_types.h +++ b/arch/s390/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef __ASM_SPINLOCK_TYPES_H #define __ASM_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - typedef struct { int lock; } __attribute__ ((aligned (4))) arch_spinlock_t; diff --git a/arch/sh/include/asm/spinlock_types.h b/arch/sh/include/asm/spinlock_types.h index e82369f286a20..22ca9a98bbb8b 100644 --- a/arch/sh/include/asm/spinlock_types.h +++ b/arch/sh/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef __ASM_SH_SPINLOCK_TYPES_H #define __ASM_SH_SPINLOCK_TYPES_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - typedef struct { volatile unsigned int lock; } arch_spinlock_t; diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c index 5717c7cbdd97a..c4e46252377ed 100644 --- a/arch/sh/kernel/irq.c +++ b/arch/sh/kernel/irq.c @@ -148,6 +148,7 @@ void irq_ctx_exit(int cpu) hardirq_ctx[cpu] = NULL; } +#ifndef CONFIG_PREEMPT_RT void do_softirq_own_stack(void) { struct thread_info *curctx; @@ -175,6 +176,7 @@ void do_softirq_own_stack(void) "r5", "r6", "r7", "r8", "r9", "r15", "t", "pr" ); } +#endif #else static inline void handle_one_irq(unsigned int irq) { diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c index 3ec9f1402aad3..eb21682abfcb1 100644 --- a/arch/sparc/kernel/irq_64.c +++ b/arch/sparc/kernel/irq_64.c @@ -854,6 +854,7 @@ void __irq_entry handler_irq(int pil, struct pt_regs *regs) set_irq_regs(old_regs); } +#ifndef CONFIG_PREEMPT_RT void do_softirq_own_stack(void) { void *orig_sp, *sp = softirq_stack[smp_processor_id()]; @@ -868,6 +869,7 @@ void do_softirq_own_stack(void) __asm__ __volatile__("mov %0, %%sp" : : "r" (orig_sp)); } +#endif #ifdef CONFIG_HOTPLUG_CPU void fixup_irqs(void) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 7101ac64bb209..13a1f43f60347 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -92,6 +92,7 @@ config X86 select ARCH_SUPPORTS_ACPI select ARCH_SUPPORTS_ATOMIC_RMW select ARCH_SUPPORTS_NUMA_BALANCING if X86_64 + select ARCH_SUPPORTS_RT select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_QUEUED_RWLOCKS select ARCH_USE_QUEUED_SPINLOCKS @@ -208,6 +209,7 @@ config X86 select HAVE_PCI select HAVE_PERF_REGS select HAVE_PERF_USER_STACK_DUMP + select HAVE_PREEMPT_LAZY select MMU_GATHER_RCU_TABLE_FREE if PARAVIRT select HAVE_POSIX_CPU_TIMERS_TASK_WORK select HAVE_REGS_AND_STACK_ACCESS_API diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c index ad8a7188a2bf7..c9eb0ed6c284d 100644 --- a/arch/x86/crypto/aesni-intel_glue.c +++ b/arch/x86/crypto/aesni-intel_glue.c @@ -376,14 +376,14 @@ static int ecb_encrypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, true); - kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { + kernel_fpu_begin(); aesni_ecb_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr, nbytes & AES_BLOCK_MASK); + kernel_fpu_end(); nbytes &= AES_BLOCK_SIZE - 1; err = skcipher_walk_done(&walk, nbytes); } - kernel_fpu_end(); return err; } @@ -398,14 +398,14 @@ static int ecb_decrypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, true); - kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { + kernel_fpu_begin(); aesni_ecb_dec(ctx, walk.dst.virt.addr, walk.src.virt.addr, nbytes & AES_BLOCK_MASK); + kernel_fpu_end(); nbytes &= AES_BLOCK_SIZE - 1; err = skcipher_walk_done(&walk, nbytes); } - kernel_fpu_end(); return err; } @@ -420,14 +420,14 @@ static int cbc_encrypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, true); - kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { + kernel_fpu_begin(); aesni_cbc_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr, nbytes & AES_BLOCK_MASK, walk.iv); + kernel_fpu_end(); nbytes &= AES_BLOCK_SIZE - 1; err = skcipher_walk_done(&walk, nbytes); } - kernel_fpu_end(); return err; } @@ -442,14 +442,14 @@ static int cbc_decrypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, true); - kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { + kernel_fpu_begin(); aesni_cbc_dec(ctx, walk.dst.virt.addr, walk.src.virt.addr, nbytes & AES_BLOCK_MASK, walk.iv); + kernel_fpu_end(); nbytes &= AES_BLOCK_SIZE - 1; err = skcipher_walk_done(&walk, nbytes); } - kernel_fpu_end(); return err; } @@ -497,18 +497,20 @@ static int ctr_crypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, true); - kernel_fpu_begin(); while ((nbytes = walk.nbytes) >= AES_BLOCK_SIZE) { + kernel_fpu_begin(); aesni_ctr_enc_tfm(ctx, walk.dst.virt.addr, walk.src.virt.addr, nbytes & AES_BLOCK_MASK, walk.iv); + kernel_fpu_end(); nbytes &= AES_BLOCK_SIZE - 1; err = skcipher_walk_done(&walk, nbytes); } if (walk.nbytes) { + kernel_fpu_begin(); ctr_crypt_final(ctx, &walk); + kernel_fpu_end(); err = skcipher_walk_done(&walk, 0); } - kernel_fpu_end(); return err; } diff --git a/arch/x86/crypto/cast5_avx_glue.c b/arch/x86/crypto/cast5_avx_glue.c index 384ccb00f9e15..2f8df8ef8644e 100644 --- a/arch/x86/crypto/cast5_avx_glue.c +++ b/arch/x86/crypto/cast5_avx_glue.c @@ -46,7 +46,7 @@ static inline void cast5_fpu_end(bool fpu_enabled) static int ecb_crypt(struct skcipher_request *req, bool enc) { - bool fpu_enabled = false; + bool fpu_enabled; struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct cast5_ctx *ctx = crypto_skcipher_ctx(tfm); struct skcipher_walk walk; @@ -61,7 +61,7 @@ static int ecb_crypt(struct skcipher_request *req, bool enc) u8 *wsrc = walk.src.virt.addr; u8 *wdst = walk.dst.virt.addr; - fpu_enabled = cast5_fpu_begin(fpu_enabled, &walk, nbytes); + fpu_enabled = cast5_fpu_begin(false, &walk, nbytes); /* Process multi-block batch */ if (nbytes >= bsize * CAST5_PARALLEL_BLOCKS) { @@ -90,10 +90,9 @@ static int ecb_crypt(struct skcipher_request *req, bool enc) } while (nbytes >= bsize); done: + cast5_fpu_end(fpu_enabled); err = skcipher_walk_done(&walk, nbytes); } - - cast5_fpu_end(fpu_enabled); return err; } @@ -197,7 +196,7 @@ static int cbc_decrypt(struct skcipher_request *req) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct cast5_ctx *ctx = crypto_skcipher_ctx(tfm); - bool fpu_enabled = false; + bool fpu_enabled; struct skcipher_walk walk; unsigned int nbytes; int err; @@ -205,12 +204,11 @@ static int cbc_decrypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, false); while ((nbytes = walk.nbytes)) { - fpu_enabled = cast5_fpu_begin(fpu_enabled, &walk, nbytes); + fpu_enabled = cast5_fpu_begin(false, &walk, nbytes); nbytes = __cbc_decrypt(ctx, &walk); + cast5_fpu_end(fpu_enabled); err = skcipher_walk_done(&walk, nbytes); } - - cast5_fpu_end(fpu_enabled); return err; } @@ -277,7 +275,7 @@ static int ctr_crypt(struct skcipher_request *req) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); struct cast5_ctx *ctx = crypto_skcipher_ctx(tfm); - bool fpu_enabled = false; + bool fpu_enabled; struct skcipher_walk walk; unsigned int nbytes; int err; @@ -285,13 +283,12 @@ static int ctr_crypt(struct skcipher_request *req) err = skcipher_walk_virt(&walk, req, false); while ((nbytes = walk.nbytes) >= CAST5_BLOCK_SIZE) { - fpu_enabled = cast5_fpu_begin(fpu_enabled, &walk, nbytes); + fpu_enabled = cast5_fpu_begin(false, &walk, nbytes); nbytes = __ctr_crypt(&walk, ctx); + cast5_fpu_end(fpu_enabled); err = skcipher_walk_done(&walk, nbytes); } - cast5_fpu_end(fpu_enabled); - if (walk.nbytes) { ctr_crypt_final(&walk, ctx); err = skcipher_walk_done(&walk, 0); diff --git a/arch/x86/crypto/glue_helper.c b/arch/x86/crypto/glue_helper.c index d3d91a0abf88f..6d07747215147 100644 --- a/arch/x86/crypto/glue_helper.c +++ b/arch/x86/crypto/glue_helper.c @@ -24,7 +24,7 @@ int glue_ecb_req_128bit(const struct common_glue_ctx *gctx, void *ctx = crypto_skcipher_ctx(crypto_skcipher_reqtfm(req)); const unsigned int bsize = 128 / 8; struct skcipher_walk walk; - bool fpu_enabled = false; + bool fpu_enabled; unsigned int nbytes; int err; @@ -37,7 +37,7 @@ int glue_ecb_req_128bit(const struct common_glue_ctx *gctx, unsigned int i; fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, - &walk, fpu_enabled, nbytes); + &walk, false, nbytes); for (i = 0; i < gctx->num_funcs; i++) { func_bytes = bsize * gctx->funcs[i].num_blocks; @@ -55,10 +55,9 @@ int glue_ecb_req_128bit(const struct common_glue_ctx *gctx, if (nbytes < bsize) break; } + glue_fpu_end(fpu_enabled); err = skcipher_walk_done(&walk, nbytes); } - - glue_fpu_end(fpu_enabled); return err; } EXPORT_SYMBOL_GPL(glue_ecb_req_128bit); @@ -101,7 +100,7 @@ int glue_cbc_decrypt_req_128bit(const struct common_glue_ctx *gctx, void *ctx = crypto_skcipher_ctx(crypto_skcipher_reqtfm(req)); const unsigned int bsize = 128 / 8; struct skcipher_walk walk; - bool fpu_enabled = false; + bool fpu_enabled; unsigned int nbytes; int err; @@ -115,7 +114,7 @@ int glue_cbc_decrypt_req_128bit(const struct common_glue_ctx *gctx, u128 last_iv; fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, - &walk, fpu_enabled, nbytes); + &walk, false, nbytes); /* Start of the last block. */ src += nbytes / bsize - 1; dst += nbytes / bsize - 1; @@ -148,10 +147,10 @@ int glue_cbc_decrypt_req_128bit(const struct common_glue_ctx *gctx, done: u128_xor(dst, dst, (u128 *)walk.iv); *(u128 *)walk.iv = last_iv; + glue_fpu_end(fpu_enabled); err = skcipher_walk_done(&walk, nbytes); } - glue_fpu_end(fpu_enabled); return err; } EXPORT_SYMBOL_GPL(glue_cbc_decrypt_req_128bit); @@ -162,7 +161,7 @@ int glue_ctr_req_128bit(const struct common_glue_ctx *gctx, void *ctx = crypto_skcipher_ctx(crypto_skcipher_reqtfm(req)); const unsigned int bsize = 128 / 8; struct skcipher_walk walk; - bool fpu_enabled = false; + bool fpu_enabled; unsigned int nbytes; int err; @@ -176,7 +175,7 @@ int glue_ctr_req_128bit(const struct common_glue_ctx *gctx, le128 ctrblk; fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, - &walk, fpu_enabled, nbytes); + &walk, false, nbytes); be128_to_le128(&ctrblk, (be128 *)walk.iv); @@ -202,11 +201,10 @@ int glue_ctr_req_128bit(const struct common_glue_ctx *gctx, } le128_to_be128((be128 *)walk.iv, &ctrblk); + glue_fpu_end(fpu_enabled); err = skcipher_walk_done(&walk, nbytes); } - glue_fpu_end(fpu_enabled); - if (nbytes) { le128 ctrblk; u128 tmp; @@ -306,8 +304,14 @@ int glue_xts_req_128bit(const struct common_glue_ctx *gctx, tweak_fn(tweak_ctx, walk.iv, walk.iv); while (nbytes) { + fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit, + &walk, fpu_enabled, + nbytes < bsize ? bsize : nbytes); nbytes = __glue_xts_req_128bit(gctx, crypt_ctx, &walk); + glue_fpu_end(fpu_enabled); + fpu_enabled = false; + err = skcipher_walk_done(&walk, nbytes); nbytes = walk.nbytes; } diff --git a/arch/x86/include/asm/fpu/api.h b/arch/x86/include/asm/fpu/api.h index b774c52e5411f..7959a9c10e5a2 100644 --- a/arch/x86/include/asm/fpu/api.h +++ b/arch/x86/include/asm/fpu/api.h @@ -23,6 +23,7 @@ extern void kernel_fpu_begin(void); extern void kernel_fpu_end(void); extern bool irq_fpu_usable(void); extern void fpregs_mark_activate(void); +extern void kernel_fpu_resched(void); /* * Use fpregs_lock() while editing CPU's FPU registers or fpu->state. @@ -33,12 +34,18 @@ extern void fpregs_mark_activate(void); static inline void fpregs_lock(void) { preempt_disable(); - local_bh_disable(); + /* + * On RT disabling preemption is good enough because bottom halfs + * are always running in thread context. + */ + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_bh_disable(); } static inline void fpregs_unlock(void) { - local_bh_enable(); + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_bh_enable(); preempt_enable(); } diff --git a/arch/x86/include/asm/preempt.h b/arch/x86/include/asm/preempt.h index 69485ca13665f..471dec2d78e12 100644 --- a/arch/x86/include/asm/preempt.h +++ b/arch/x86/include/asm/preempt.h @@ -89,20 +89,54 @@ static __always_inline void __preempt_count_sub(int val) * a decrement which hits zero means we have no preempt_count and should * reschedule. */ -static __always_inline bool __preempt_count_dec_and_test(void) +static __always_inline bool ____preempt_count_dec_and_test(void) { return GEN_UNARY_RMWcc("decl", __preempt_count, e, __percpu_arg([var])); } +static __always_inline bool __preempt_count_dec_and_test(void) +{ + if (____preempt_count_dec_and_test()) + return true; +#ifdef CONFIG_PREEMPT_LAZY + if (preempt_count()) + return false; + if (current_thread_info()->preempt_lazy_count) + return false; + return test_thread_flag(TIF_NEED_RESCHED_LAZY); +#else + return false; +#endif +} + /* * Returns true when we need to resched and can (barring IRQ state). */ static __always_inline bool should_resched(int preempt_offset) { +#ifdef CONFIG_PREEMPT_LAZY + u32 tmp; + tmp = raw_cpu_read_4(__preempt_count); + if (tmp == preempt_offset) + return true; + + /* preempt count == 0 ? */ + tmp &= ~PREEMPT_NEED_RESCHED; + if (tmp != preempt_offset) + return false; + /* XXX PREEMPT_LOCK_OFFSET */ + if (current_thread_info()->preempt_lazy_count) + return false; + return test_thread_flag(TIF_NEED_RESCHED_LAZY); +#else return unlikely(raw_cpu_read_4(__preempt_count) == preempt_offset); +#endif } #ifdef CONFIG_PREEMPTION +#ifdef CONFIG_PREEMPT_RT + extern void preempt_schedule_lock(void); +#endif extern asmlinkage void preempt_schedule_thunk(void); # define __preempt_schedule() \ asm volatile ("call preempt_schedule_thunk" : ASM_CALL_CONSTRAINT) diff --git a/arch/x86/include/asm/signal.h b/arch/x86/include/asm/signal.h index 6fd8410a3910c..f3bf2f515edb3 100644 --- a/arch/x86/include/asm/signal.h +++ b/arch/x86/include/asm/signal.h @@ -28,6 +28,19 @@ typedef struct { #define SA_IA32_ABI 0x02000000u #define SA_X32_ABI 0x01000000u +/* + * Because some traps use the IST stack, we must keep preemption + * disabled while calling do_trap(), but do_trap() may call + * force_sig_info() which will grab the signal spin_locks for the + * task, which in PREEMPT_RT are mutexes. By defining + * ARCH_RT_DELAYS_SIGNAL_SEND the force_sig_info() will set + * TIF_NOTIFY_RESUME and set up the signal to be sent on exit of the + * trap. + */ +#if defined(CONFIG_PREEMPT_RT) +#define ARCH_RT_DELAYS_SIGNAL_SEND +#endif + #ifndef CONFIG_COMPAT typedef sigset_t compat_sigset_t; #endif diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h index 7fb482f0f25b0..3df0a95c9e134 100644 --- a/arch/x86/include/asm/stackprotector.h +++ b/arch/x86/include/asm/stackprotector.h @@ -65,7 +65,7 @@ */ static __always_inline void boot_init_stack_canary(void) { - u64 canary; + u64 canary = 0; u64 tsc; #ifdef CONFIG_X86_64 @@ -76,8 +76,14 @@ static __always_inline void boot_init_stack_canary(void) * of randomness. The TSC only matters for very early init, * there it already has some randomness on most systems. Later * on during the bootup the random pool has true entropy too. + * For preempt-rt we need to weaken the randomness a bit, as + * we can't call into the random generator from atomic context + * due to locking constraints. We just leave canary + * uninitialized and use the TSC based randomness on top of it. */ +#ifndef CONFIG_PREEMPT_RT get_random_bytes(&canary, sizeof(canary)); +#endif tsc = rdtsc(); canary += tsc + (tsc << 32UL); canary &= CANARY_MASK; diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index 267701ae3d86d..350e342d9f252 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -56,17 +56,24 @@ struct task_struct; struct thread_info { unsigned long flags; /* low level flags */ u32 status; /* thread synchronous flags */ + int preempt_lazy_count; /* 0 => lazy preemptable + <0 => BUG */ }; #define INIT_THREAD_INFO(tsk) \ { \ .flags = 0, \ + .preempt_lazy_count = 0, \ } #else /* !__ASSEMBLY__ */ #include +#define GET_THREAD_INFO(reg) \ + _ASM_MOV PER_CPU_VAR(cpu_current_top_of_stack),reg ; \ + _ASM_SUB $(THREAD_SIZE),reg ; + #endif /* @@ -93,6 +100,7 @@ struct thread_info { #define TIF_NOTSC 16 /* TSC is not accessible in userland */ #define TIF_IA32 17 /* IA32 compatibility process */ #define TIF_SLD 18 /* Restore split lock detection on context switch */ +#define TIF_NEED_RESCHED_LAZY 19 /* lazy rescheduling necessary */ #define TIF_MEMDIE 20 /* is terminating due to OOM killer */ #define TIF_POLLING_NRFLAG 21 /* idle is polling for TIF_NEED_RESCHED */ #define TIF_IO_BITMAP 22 /* uses I/O bitmap */ @@ -123,6 +131,7 @@ struct thread_info { #define _TIF_NOTSC (1 << TIF_NOTSC) #define _TIF_IA32 (1 << TIF_IA32) #define _TIF_SLD (1 << TIF_SLD) +#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY) #define _TIF_POLLING_NRFLAG (1 << TIF_POLLING_NRFLAG) #define _TIF_IO_BITMAP (1 << TIF_IO_BITMAP) #define _TIF_FORCED_TF (1 << TIF_FORCED_TF) @@ -156,6 +165,8 @@ struct thread_info { #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW) +#define _TIF_NEED_RESCHED_MASK (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY) + #define STACK_WARN (THREAD_SIZE/8) /* diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c index 31125448b174d..0a3270aadfce2 100644 --- a/arch/x86/kernel/cpu/mshyperv.c +++ b/arch/x86/kernel/cpu/mshyperv.c @@ -75,11 +75,12 @@ EXPORT_SYMBOL_GPL(hv_remove_vmbus_irq); DEFINE_IDTENTRY_SYSVEC(sysvec_hyperv_stimer0) { struct pt_regs *old_regs = set_irq_regs(regs); + u64 ip = regs ? instruction_pointer(regs) : 0; inc_irq_stat(hyperv_stimer0_count); if (hv_stimer0_handler) hv_stimer0_handler(); - add_interrupt_randomness(HYPERV_STIMER0_VECTOR, 0); + add_interrupt_randomness(HYPERV_STIMER0_VECTOR, 0, ip); ack_APIC_irq(); set_irq_regs(old_regs); diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c index eb86a2b831b15..23aaf9e132e9f 100644 --- a/arch/x86/kernel/fpu/core.c +++ b/arch/x86/kernel/fpu/core.c @@ -158,6 +158,18 @@ void kernel_fpu_end(void) } EXPORT_SYMBOL_GPL(kernel_fpu_end); +void kernel_fpu_resched(void) +{ + WARN_ON_FPU(!this_cpu_read(in_kernel_fpu)); + + if (should_resched(PREEMPT_OFFSET)) { + kernel_fpu_end(); + cond_resched(); + kernel_fpu_begin(); + } +} +EXPORT_SYMBOL_GPL(kernel_fpu_resched); + /* * Save the FPU state (mark it for reload if necessary): * diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c index 0b79efc87be52..93c6b88b382a0 100644 --- a/arch/x86/kernel/irq_32.c +++ b/arch/x86/kernel/irq_32.c @@ -131,6 +131,7 @@ int irq_init_percpu_irqstack(unsigned int cpu) return 0; } +#ifndef CONFIG_PREEMPT_RT void do_softirq_own_stack(void) { struct irq_stack *irqstk; @@ -147,6 +148,7 @@ void do_softirq_own_stack(void) call_on_stack(__do_softirq, isp); } +#endif void __handle_irq(struct irq_desc *desc, struct pt_regs *regs) { diff --git a/arch/x86/kernel/irq_64.c b/arch/x86/kernel/irq_64.c index 440eed558558d..7cfc4e6b7c94a 100644 --- a/arch/x86/kernel/irq_64.c +++ b/arch/x86/kernel/irq_64.c @@ -72,7 +72,9 @@ int irq_init_percpu_irqstack(unsigned int cpu) return map_irq_stack(cpu); } +#ifndef CONFIG_PREEMPT_RT void do_softirq_own_stack(void) { run_on_irqstack_cond(__do_softirq, NULL); } +#endif diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 4f2f54e1281c3..17189e47cee88 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include @@ -126,6 +127,35 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp) } EXPORT_SYMBOL_GPL(start_thread); +#ifdef CONFIG_PREEMPT_RT +static void switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p) +{ + int i; + + /* + * Clear @prev's kmap_atomic mappings + */ + for (i = 0; i < prev_p->kmap_idx; i++) { + int idx = i + KM_TYPE_NR * smp_processor_id(); + pte_t *ptep = kmap_pte - idx; + + kpte_clear_flush(ptep, __fix_to_virt(FIX_KMAP_BEGIN + idx)); + } + /* + * Restore @next_p's kmap_atomic mappings + */ + for (i = 0; i < next_p->kmap_idx; i++) { + int idx = i + KM_TYPE_NR * smp_processor_id(); + + if (!pte_none(next_p->kmap_pte[i])) + set_pte(kmap_pte - idx, next_p->kmap_pte[i]); + } +} +#else +static inline void +switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p) { } +#endif + /* * switch_to(x,y) should switch tasks from x to y. @@ -187,6 +217,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) switch_to_extra(prev_p, next_p); + switch_kmaps(prev_p, next_p); + /* * Leave lazy mode, flushing any hypercalls made here. * This must be done before restoring TLS segments so diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 49d925043171a..f70dffc2771f5 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -54,7 +54,7 @@ struct clocksource *art_related_clocksource; struct cyc2ns { struct cyc2ns_data data[2]; /* 0 + 2*16 = 32 */ - seqcount_t seq; /* 32 + 4 = 36 */ + seqcount_latch_t seq; /* 32 + 4 = 36 */ }; /* fits one cacheline */ @@ -73,14 +73,14 @@ __always_inline void cyc2ns_read_begin(struct cyc2ns_data *data) preempt_disable_notrace(); do { - seq = this_cpu_read(cyc2ns.seq.sequence); + seq = this_cpu_read(cyc2ns.seq.seqcount.sequence); idx = seq & 1; data->cyc2ns_offset = this_cpu_read(cyc2ns.data[idx].cyc2ns_offset); data->cyc2ns_mul = this_cpu_read(cyc2ns.data[idx].cyc2ns_mul); data->cyc2ns_shift = this_cpu_read(cyc2ns.data[idx].cyc2ns_shift); - } while (unlikely(seq != this_cpu_read(cyc2ns.seq.sequence))); + } while (unlikely(seq != this_cpu_read(cyc2ns.seq.seqcount.sequence))); } __always_inline void cyc2ns_read_end(void) @@ -186,7 +186,7 @@ static void __init cyc2ns_init_boot_cpu(void) { struct cyc2ns *c2n = this_cpu_ptr(&cyc2ns); - seqcount_init(&c2n->seq); + seqcount_latch_init(&c2n->seq); __set_cyc2ns_scale(tsc_khz, smp_processor_id(), rdtsc()); } @@ -203,7 +203,7 @@ static void __init cyc2ns_init_secondary_cpus(void) for_each_possible_cpu(cpu) { if (cpu != this_cpu) { - seqcount_init(&c2n->seq); + seqcount_latch_init(&c2n->seq); c2n = per_cpu_ptr(&cyc2ns, cpu); c2n->data[0] = data[0]; c2n->data[1] = data[1]; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ce856e0ece844..257dcd9ba1a0d 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -7513,6 +7513,14 @@ int kvm_arch_init(void *opaque) goto out; } +#ifdef CONFIG_PREEMPT_RT + if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) { + pr_err("RT requires X86_FEATURE_CONSTANT_TSC\n"); + r = -EOPNOTSUPP; + goto out; + } +#endif + r = -ENOMEM; x86_fpu_cache = kmem_cache_create("x86_fpu", sizeof(struct fpu), __alignof__(struct fpu), SLAB_ACCOUNT, diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c index 075fe51317b0f..95da91a7c7af5 100644 --- a/arch/x86/mm/highmem_32.c +++ b/arch/x86/mm/highmem_32.c @@ -8,12 +8,17 @@ void *kmap_atomic_high_prot(struct page *page, pgprot_t prot) { unsigned long vaddr; int idx, type; + pte_t pte; type = kmap_atomic_idx_push(); idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); BUG_ON(!pte_none(*(kmap_pte-idx))); - set_pte(kmap_pte-idx, mk_pte(page, prot)); + pte = mk_pte(page, prot); +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = pte; +#endif + set_pte(kmap_pte-idx, pte); arch_flush_lazy_mmu_mode(); return (void *)vaddr; @@ -50,6 +55,9 @@ void kunmap_atomic_high(void *kvaddr) * is a bad idea also, in case the page changes cacheability * attributes or becomes a protected page in a hypervisor. */ +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = __pte(0); +#endif kpte_clear_flush(kmap_pte-idx, vaddr); kmap_atomic_idx_pop(); arch_flush_lazy_mmu_mode(); diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c index f60398aeb6445..0ef360874c752 100644 --- a/arch/x86/mm/iomap_32.c +++ b/arch/x86/mm/iomap_32.c @@ -46,16 +46,22 @@ EXPORT_SYMBOL_GPL(iomap_free); void *kmap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot) { + pte_t pte = pfn_pte(pfn, prot); unsigned long vaddr; int idx, type; - preempt_disable(); + migrate_disable(); pagefault_disable(); type = kmap_atomic_idx_push(); idx = type + KM_TYPE_NR * smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); - set_pte(kmap_pte - idx, pfn_pte(pfn, prot)); + WARN_ON(!pte_none(*(kmap_pte - idx))); + +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = pte; +#endif + set_pte(kmap_pte - idx, pte); arch_flush_lazy_mmu_mode(); return (void *)vaddr; @@ -106,11 +112,14 @@ iounmap_atomic(void __iomem *kvaddr) * is a bad idea also, in case the page changes cacheability * attributes or becomes a protected page in a hypervisor. */ +#ifdef CONFIG_PREEMPT_RT + current->kmap_pte[type] = __pte(0); +#endif kpte_clear_flush(kmap_pte-idx, vaddr); kmap_atomic_idx_pop(); } pagefault_enable(); - preempt_enable(); + migrate_enable(); } EXPORT_SYMBOL_GPL(iounmap_atomic); diff --git a/arch/xtensa/include/asm/spinlock_types.h b/arch/xtensa/include/asm/spinlock_types.h index 64c9389254f13..dc846323b1cd7 100644 --- a/arch/xtensa/include/asm/spinlock_types.h +++ b/arch/xtensa/include/asm/spinlock_types.h @@ -2,10 +2,6 @@ #ifndef __ASM_SPINLOCK_TYPES_H #define __ASM_SPINLOCK_TYPES_H -#if !defined(__LINUX_SPINLOCK_TYPES_H) && !defined(__ASM_SPINLOCK_H) -# error "please don't include this file directly" -#endif - #include #include diff --git a/block/blk-mq.c b/block/blk-mq.c index cdced4aca2e81..1f9b312595a19 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -41,7 +41,7 @@ #include "blk-mq-sched.h" #include "blk-rq-qos.h" -static DEFINE_PER_CPU(struct list_head, blk_cpu_done); +static DEFINE_PER_CPU(struct llist_head, blk_cpu_done); static void blk_mq_poll_stats_start(struct request_queue *q); static void blk_mq_poll_stats_fn(struct blk_stat_callback *cb); @@ -565,80 +565,32 @@ void blk_mq_end_request(struct request *rq, blk_status_t error) } EXPORT_SYMBOL(blk_mq_end_request); -/* - * Softirq action handler - move entries to local list and loop over them - * while passing them to the queue registered handler. - */ -static __latent_entropy void blk_done_softirq(struct softirq_action *h) +static void blk_complete_reqs(struct llist_head *cpu_list) { - struct list_head *cpu_list, local_list; + struct llist_node *entry; + struct request *rq, *rq_next; - local_irq_disable(); - cpu_list = this_cpu_ptr(&blk_cpu_done); - list_replace_init(cpu_list, &local_list); - local_irq_enable(); + entry = llist_del_all(cpu_list); + entry = llist_reverse_order(entry); - while (!list_empty(&local_list)) { - struct request *rq; - - rq = list_entry(local_list.next, struct request, ipi_list); - list_del_init(&rq->ipi_list); + llist_for_each_entry_safe(rq, rq_next, entry, ipi_list) rq->q->mq_ops->complete(rq); - } } -static void blk_mq_trigger_softirq(struct request *rq) +static __latent_entropy void blk_done_softirq(struct softirq_action *h) { - struct list_head *list; - unsigned long flags; - - local_irq_save(flags); - list = this_cpu_ptr(&blk_cpu_done); - list_add_tail(&rq->ipi_list, list); - - /* - * If the list only contains our just added request, signal a raise of - * the softirq. If there are already entries there, someone already - * raised the irq but it hasn't run yet. - */ - if (list->next == &rq->ipi_list) - raise_softirq_irqoff(BLOCK_SOFTIRQ); - local_irq_restore(flags); + blk_complete_reqs(this_cpu_ptr(&blk_cpu_done)); } static int blk_softirq_cpu_dead(unsigned int cpu) { - /* - * If a CPU goes away, splice its entries to the current CPU - * and trigger a run of the softirq - */ - local_irq_disable(); - list_splice_init(&per_cpu(blk_cpu_done, cpu), - this_cpu_ptr(&blk_cpu_done)); - raise_softirq_irqoff(BLOCK_SOFTIRQ); - local_irq_enable(); - + blk_complete_reqs(&per_cpu(blk_cpu_done, cpu)); return 0; } - static void __blk_mq_complete_request_remote(void *data) { - struct request *rq = data; - - /* - * For most of single queue controllers, there is only one irq vector - * for handling I/O completion, and the only irq's affinity is set - * to all possible CPUs. On most of ARCHs, this affinity means the irq - * is handled on one specific CPU. - * - * So complete I/O requests in softirq context in case of single queue - * devices to avoid degrading I/O performance due to irqsoff latency. - */ - if (rq->q->nr_hw_queues == 1) - blk_mq_trigger_softirq(rq); - else - rq->q->mq_ops->complete(rq); + __raise_softirq_irqoff(BLOCK_SOFTIRQ); } static inline bool blk_mq_complete_need_ipi(struct request *rq) @@ -648,6 +600,14 @@ static inline bool blk_mq_complete_need_ipi(struct request *rq) if (!IS_ENABLED(CONFIG_SMP) || !test_bit(QUEUE_FLAG_SAME_COMP, &rq->q->queue_flags)) return false; + /* + * With force threaded interrupts enabled, raising softirq from an SMP + * function call will always result in waking the ksoftirqd thread. + * This is probably worse than completing the request on a different + * cache domain. + */ + if (force_irqthreads) + return false; /* same CPU or cache domain? Complete locally */ if (cpu == rq->mq_ctx->cpu || @@ -661,6 +621,7 @@ static inline bool blk_mq_complete_need_ipi(struct request *rq) bool blk_mq_complete_request_remote(struct request *rq) { + struct llist_head *cpu_list; WRITE_ONCE(rq->state, MQ_RQ_COMPLETE); /* @@ -671,14 +632,21 @@ bool blk_mq_complete_request_remote(struct request *rq) return false; if (blk_mq_complete_need_ipi(rq)) { - rq->csd.func = __blk_mq_complete_request_remote; - rq->csd.info = rq; - rq->csd.flags = 0; - smp_call_function_single_async(rq->mq_ctx->cpu, &rq->csd); + unsigned int cpu; + + cpu = rq->mq_ctx->cpu; + cpu_list = &per_cpu(blk_cpu_done, cpu); + if (llist_add(&rq->ipi_list, cpu_list)) { + rq->csd.func = __blk_mq_complete_request_remote; + rq->csd.flags = 0; + smp_call_function_single_async(cpu, &rq->csd); + } } else { if (rq->q->nr_hw_queues > 1) return false; - blk_mq_trigger_softirq(rq); + cpu_list = this_cpu_ptr(&blk_cpu_done); + if (llist_add(&rq->ipi_list, cpu_list)) + raise_softirq(BLOCK_SOFTIRQ); } return true; @@ -1603,14 +1571,14 @@ static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, return; if (!async && !(hctx->flags & BLK_MQ_F_BLOCKING)) { - int cpu = get_cpu(); + int cpu = get_cpu_light(); if (cpumask_test_cpu(cpu, hctx->cpumask)) { __blk_mq_run_hw_queue(hctx); - put_cpu(); + put_cpu_light(); return; } - put_cpu(); + put_cpu_light(); } kblockd_mod_delayed_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work, @@ -3881,7 +3849,7 @@ static int __init blk_mq_init(void) int i; for_each_possible_cpu(i) - INIT_LIST_HEAD(&per_cpu(blk_cpu_done, i)); + init_llist_head(&per_cpu(blk_cpu_done, i)); open_softirq(BLOCK_SOFTIRQ, blk_done_softirq); cpuhp_setup_state_nocalls(CPUHP_BLOCK_SOFTIRQ_DEAD, diff --git a/crypto/cryptd.c b/crypto/cryptd.c index a1bea0f4baa88..5f8ca8c1f59cd 100644 --- a/crypto/cryptd.c +++ b/crypto/cryptd.c @@ -36,6 +36,7 @@ static struct workqueue_struct *cryptd_wq; struct cryptd_cpu_queue { struct crypto_queue queue; struct work_struct work; + spinlock_t qlock; }; struct cryptd_queue { @@ -105,6 +106,7 @@ static int cryptd_init_queue(struct cryptd_queue *queue, cpu_queue = per_cpu_ptr(queue->cpu_queue, cpu); crypto_init_queue(&cpu_queue->queue, max_cpu_qlen); INIT_WORK(&cpu_queue->work, cryptd_queue_worker); + spin_lock_init(&cpu_queue->qlock); } pr_info("cryptd: max_cpu_qlen set to %d\n", max_cpu_qlen); return 0; @@ -129,8 +131,10 @@ static int cryptd_enqueue_request(struct cryptd_queue *queue, struct cryptd_cpu_queue *cpu_queue; refcount_t *refcnt; - cpu = get_cpu(); - cpu_queue = this_cpu_ptr(queue->cpu_queue); + cpu_queue = raw_cpu_ptr(queue->cpu_queue); + spin_lock_bh(&cpu_queue->qlock); + cpu = smp_processor_id(); + err = crypto_enqueue_request(&cpu_queue->queue, request); refcnt = crypto_tfm_ctx(request->tfm); @@ -146,7 +150,7 @@ static int cryptd_enqueue_request(struct cryptd_queue *queue, refcount_inc(refcnt); out_put_cpu: - put_cpu(); + spin_unlock_bh(&cpu_queue->qlock); return err; } @@ -162,16 +166,11 @@ static void cryptd_queue_worker(struct work_struct *work) cpu_queue = container_of(work, struct cryptd_cpu_queue, work); /* * Only handle one request at a time to avoid hogging crypto workqueue. - * preempt_disable/enable is used to prevent being preempted by - * cryptd_enqueue_request(). local_bh_disable/enable is used to prevent - * cryptd_enqueue_request() being accessed from software interrupts. */ - local_bh_disable(); - preempt_disable(); + spin_lock_bh(&cpu_queue->qlock); backlog = crypto_get_backlog(&cpu_queue->queue); req = crypto_dequeue_request(&cpu_queue->queue); - preempt_enable(); - local_bh_enable(); + spin_unlock_bh(&cpu_queue->qlock); if (!req) return; diff --git a/drivers/base/core.c b/drivers/base/core.c index bb5806a2bd4ca..f90e9f77bf8c2 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -4061,22 +4061,21 @@ void device_shutdown(void) */ #ifdef CONFIG_PRINTK -static int -create_syslog_header(const struct device *dev, char *hdr, size_t hdrlen) +static void +set_dev_info(const struct device *dev, struct dev_printk_info *dev_info) { const char *subsys; - size_t pos = 0; + + memset(dev_info, 0, sizeof(*dev_info)); if (dev->class) subsys = dev->class->name; else if (dev->bus) subsys = dev->bus->name; else - return 0; + return; - pos += snprintf(hdr + pos, hdrlen - pos, "SUBSYSTEM=%s", subsys); - if (pos >= hdrlen) - goto overflow; + strscpy(dev_info->subsystem, subsys, sizeof(dev_info->subsystem)); /* * Add device identifier DEVICE=: @@ -4092,41 +4091,28 @@ create_syslog_header(const struct device *dev, char *hdr, size_t hdrlen) c = 'b'; else c = 'c'; - pos++; - pos += snprintf(hdr + pos, hdrlen - pos, - "DEVICE=%c%u:%u", - c, MAJOR(dev->devt), MINOR(dev->devt)); + + snprintf(dev_info->device, sizeof(dev_info->device), + "%c%u:%u", c, MAJOR(dev->devt), MINOR(dev->devt)); } else if (strcmp(subsys, "net") == 0) { struct net_device *net = to_net_dev(dev); - pos++; - pos += snprintf(hdr + pos, hdrlen - pos, - "DEVICE=n%u", net->ifindex); + snprintf(dev_info->device, sizeof(dev_info->device), + "n%u", net->ifindex); } else { - pos++; - pos += snprintf(hdr + pos, hdrlen - pos, - "DEVICE=+%s:%s", subsys, dev_name(dev)); + snprintf(dev_info->device, sizeof(dev_info->device), + "+%s:%s", subsys, dev_name(dev)); } - - if (pos >= hdrlen) - goto overflow; - - return pos; - -overflow: - dev_WARN(dev, "device/subsystem name too long"); - return 0; } int dev_vprintk_emit(int level, const struct device *dev, const char *fmt, va_list args) { - char hdr[128]; - size_t hdrlen; + struct dev_printk_info dev_info; - hdrlen = create_syslog_header(dev, hdr, sizeof(hdr)); + set_dev_info(dev, &dev_info); - return vprintk_emit(0, level, hdrlen ? hdr : NULL, hdrlen, fmt, args); + return vprintk_emit(0, level, &dev_info, fmt, args); } EXPORT_SYMBOL(dev_vprintk_emit); diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index 9100ac36670af..4bd301dd22c3b 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -56,6 +56,40 @@ static void zram_free_page(struct zram *zram, size_t index); static int zram_bvec_read(struct zram *zram, struct bio_vec *bvec, u32 index, int offset, struct bio *bio); +#ifdef CONFIG_PREEMPT_RT +static void zram_meta_init_table_locks(struct zram *zram, size_t num_pages) +{ + size_t index; + + for (index = 0; index < num_pages; index++) + spin_lock_init(&zram->table[index].lock); +} + +static int zram_slot_trylock(struct zram *zram, u32 index) +{ + int ret; + + ret = spin_trylock(&zram->table[index].lock); + if (ret) + __set_bit(ZRAM_LOCK, &zram->table[index].flags); + return ret; +} + +static void zram_slot_lock(struct zram *zram, u32 index) +{ + spin_lock(&zram->table[index].lock); + __set_bit(ZRAM_LOCK, &zram->table[index].flags); +} + +static void zram_slot_unlock(struct zram *zram, u32 index) +{ + __clear_bit(ZRAM_LOCK, &zram->table[index].flags); + spin_unlock(&zram->table[index].lock); +} + +#else + +static void zram_meta_init_table_locks(struct zram *zram, size_t num_pages) { } static int zram_slot_trylock(struct zram *zram, u32 index) { @@ -71,6 +105,7 @@ static void zram_slot_unlock(struct zram *zram, u32 index) { bit_spin_unlock(ZRAM_LOCK, &zram->table[index].flags); } +#endif static inline bool init_done(struct zram *zram) { @@ -1158,6 +1193,7 @@ static bool zram_meta_alloc(struct zram *zram, u64 disksize) if (!huge_class_size) huge_class_size = zs_huge_class_size(zram->mem_pool); + zram_meta_init_table_locks(zram, num_pages); return true; } @@ -1220,6 +1256,7 @@ static int __zram_bvec_read(struct zram *zram, struct page *page, u32 index, unsigned long handle; unsigned int size; void *src, *dst; + struct zcomp_strm *zstrm; zram_slot_lock(zram, index); if (zram_test_flag(zram, index, ZRAM_WB)) { @@ -1250,6 +1287,7 @@ static int __zram_bvec_read(struct zram *zram, struct page *page, u32 index, size = zram_get_obj_size(zram, index); + zstrm = zcomp_stream_get(zram->comp); src = zs_map_object(zram->mem_pool, handle, ZS_MM_RO); if (size == PAGE_SIZE) { dst = kmap_atomic(page); @@ -1257,14 +1295,13 @@ static int __zram_bvec_read(struct zram *zram, struct page *page, u32 index, kunmap_atomic(dst); ret = 0; } else { - struct zcomp_strm *zstrm = zcomp_stream_get(zram->comp); dst = kmap_atomic(page); ret = zcomp_decompress(zstrm, src, size, dst); kunmap_atomic(dst); - zcomp_stream_put(zram->comp); } zs_unmap_object(zram->mem_pool, handle); + zcomp_stream_put(zram->comp); zram_slot_unlock(zram, index); /* Should NEVER happen. Return bio error if it does. */ diff --git a/drivers/block/zram/zram_drv.h b/drivers/block/zram/zram_drv.h index f2fd46daa7604..7e4dd447e1ddd 100644 --- a/drivers/block/zram/zram_drv.h +++ b/drivers/block/zram/zram_drv.h @@ -63,6 +63,7 @@ struct zram_table_entry { unsigned long element; }; unsigned long flags; + spinlock_t lock; #ifdef CONFIG_ZRAM_MEMORY_TRACKING ktime_t ac_time; #endif diff --git a/drivers/char/random.c b/drivers/char/random.c index d20ba1b104ca3..081a1a4b9d70f 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -1252,28 +1252,27 @@ static __u32 get_reg(struct fast_pool *f, struct pt_regs *regs) return *ptr; } -void add_interrupt_randomness(int irq, int irq_flags) +void add_interrupt_randomness(int irq, int irq_flags, __u64 ip) { struct entropy_store *r; struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness); - struct pt_regs *regs = get_irq_regs(); unsigned long now = jiffies; cycles_t cycles = random_get_entropy(); __u32 c_high, j_high; - __u64 ip; unsigned long seed; int credit = 0; if (cycles == 0) - cycles = get_reg(fast_pool, regs); + cycles = get_reg(fast_pool, NULL); c_high = (sizeof(cycles) > 4) ? cycles >> 32 : 0; j_high = (sizeof(now) > 4) ? now >> 32 : 0; fast_pool->pool[0] ^= cycles ^ j_high ^ irq; fast_pool->pool[1] ^= now ^ c_high; - ip = regs ? instruction_pointer(regs) : _RET_IP_; + if (!ip) + ip = _RET_IP_; fast_pool->pool[2] ^= ip; fast_pool->pool[3] ^= (sizeof(ip) > 4) ? ip >> 32 : - get_reg(fast_pool, regs); + get_reg(fast_pool, NULL); fast_mix(fast_pool); add_interrupt_bench(cycles); diff --git a/drivers/char/tpm/tpm-dev-common.c b/drivers/char/tpm/tpm-dev-common.c index 1784530b8387b..c08cbb306636b 100644 --- a/drivers/char/tpm/tpm-dev-common.c +++ b/drivers/char/tpm/tpm-dev-common.c @@ -20,7 +20,6 @@ #include "tpm-dev.h" static struct workqueue_struct *tpm_dev_wq; -static DEFINE_MUTEX(tpm_dev_wq_lock); static ssize_t tpm_dev_transmit(struct tpm_chip *chip, struct tpm_space *space, u8 *buf, size_t bufsiz) diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c index 0b214963539de..39cee10c59f88 100644 --- a/drivers/char/tpm/tpm_tis.c +++ b/drivers/char/tpm/tpm_tis.c @@ -49,6 +49,31 @@ static inline struct tpm_tis_tcg_phy *to_tpm_tis_tcg_phy(struct tpm_tis_data *da return container_of(data, struct tpm_tis_tcg_phy, priv); } +#ifdef CONFIG_PREEMPT_RT +/* + * Flushes previous write operations to chip so that a subsequent + * ioread*()s won't stall a cpu. + */ +static inline void tpm_tis_flush(void __iomem *iobase) +{ + ioread8(iobase + TPM_ACCESS(0)); +} +#else +#define tpm_tis_flush(iobase) do { } while (0) +#endif + +static inline void tpm_tis_iowrite8(u8 b, void __iomem *iobase, u32 addr) +{ + iowrite8(b, iobase + addr); + tpm_tis_flush(iobase); +} + +static inline void tpm_tis_iowrite32(u32 b, void __iomem *iobase, u32 addr) +{ + iowrite32(b, iobase + addr); + tpm_tis_flush(iobase); +} + static bool interrupts = true; module_param(interrupts, bool, 0444); MODULE_PARM_DESC(interrupts, "Enable interrupts"); @@ -146,7 +171,7 @@ static int tpm_tcg_write_bytes(struct tpm_tis_data *data, u32 addr, u16 len, struct tpm_tis_tcg_phy *phy = to_tpm_tis_tcg_phy(data); while (len--) - iowrite8(*value++, phy->iobase + addr); + tpm_tis_iowrite8(*value++, phy->iobase, addr); return 0; } @@ -173,7 +198,7 @@ static int tpm_tcg_write32(struct tpm_tis_data *data, u32 addr, u32 value) { struct tpm_tis_tcg_phy *phy = to_tpm_tis_tcg_phy(data); - iowrite32(value, phy->iobase + addr); + tpm_tis_iowrite32(value, phy->iobase, addr); return 0; } diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c index 3aa07c3b51369..25a8e81d17423 100644 --- a/drivers/firmware/efi/efi.c +++ b/drivers/firmware/efi/efi.c @@ -62,7 +62,7 @@ struct mm_struct efi_mm = { struct workqueue_struct *efi_rts_wq; -static bool disable_runtime; +static bool disable_runtime = IS_ENABLED(CONFIG_PREEMPT_RT); static int __init setup_noefi(char *arg) { disable_runtime = true; @@ -93,6 +93,9 @@ static int __init parse_efi_cmdline(char *str) if (parse_option_str(str, "noruntime")) disable_runtime = true; + if (parse_option_str(str, "runtime")) + disable_runtime = false; + if (parse_option_str(str, "nosoftreserve")) set_bit(EFI_MEM_NO_SOFT_RESERVE, &efi.flags); diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c index c89f5f7ccb06e..1b9d5e690a9f0 100644 --- a/drivers/gpu/drm/i915/display/intel_sprite.c +++ b/drivers/gpu/drm/i915/display/intel_sprite.c @@ -118,7 +118,8 @@ void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state) "PSR idle timed out 0x%x, atomic update may fail\n", psr_status); - local_irq_disable(); + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_irq_disable(); crtc->debug.min_vbl = min; crtc->debug.max_vbl = max; @@ -143,11 +144,13 @@ void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state) break; } - local_irq_enable(); + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_irq_enable(); timeout = schedule_timeout(timeout); - local_irq_disable(); + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_irq_disable(); } finish_wait(wq, &wait); @@ -180,7 +183,8 @@ void intel_pipe_update_start(const struct intel_crtc_state *new_crtc_state) return; irq_disable: - local_irq_disable(); + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_irq_disable(); } /** @@ -218,7 +222,8 @@ void intel_pipe_update_end(struct intel_crtc_state *new_crtc_state) new_crtc_state->uapi.event = NULL; } - local_irq_enable(); + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + local_irq_enable(); if (intel_vgpu_active(dev_priv)) return; diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c index 8ec3eecf3e397..31bb901530419 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c @@ -59,9 +59,10 @@ static int __engine_unpark(struct intel_wakeref *wf) static inline unsigned long __timeline_mark_lock(struct intel_context *ce) { - unsigned long flags; + unsigned long flags = 0; - local_irq_save(flags); + if (!force_irqthreads) + local_irq_save(flags); mutex_acquire(&ce->timeline->mutex.dep_map, 2, 0, _THIS_IP_); return flags; @@ -71,7 +72,8 @@ static inline void __timeline_mark_unlock(struct intel_context *ce, unsigned long flags) { mutex_release(&ce->timeline->mutex.dep_map, _THIS_IP_); - local_irq_restore(flags); + if (!force_irqthreads) + local_irq_restore(flags); } #else diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 1fa67700d8f49..4ee735b91e973 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -865,6 +865,7 @@ static bool i915_get_crtc_scanoutpos(struct drm_crtc *_crtc, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); /* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */ + preempt_disable_rt(); /* Get optional system timestamp before query. */ if (stime) @@ -916,6 +917,7 @@ static bool i915_get_crtc_scanoutpos(struct drm_crtc *_crtc, *etime = ktime_get(); /* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */ + preempt_enable_rt(); spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h index a4addcc649782..396b6598694d3 100644 --- a/drivers/gpu/drm/i915/i915_trace.h +++ b/drivers/gpu/drm/i915/i915_trace.h @@ -2,6 +2,10 @@ #if !defined(_I915_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) #define _I915_TRACE_H_ +#ifdef CONFIG_PREEMPT_RT +#define NOTRACE +#endif + #include #include #include @@ -778,7 +782,7 @@ DEFINE_EVENT(i915_request, i915_request_add, TP_ARGS(rq) ); -#if defined(CONFIG_DRM_I915_LOW_LEVEL_TRACEPOINTS) +#if defined(CONFIG_DRM_I915_LOW_LEVEL_TRACEPOINTS) && !defined(NOTRACE) DEFINE_EVENT(i915_request, i915_request_submit, TP_PROTO(struct i915_request *rq), TP_ARGS(rq) diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c index e0ae911ef427d..781edf5504360 100644 --- a/drivers/gpu/drm/radeon/radeon_display.c +++ b/drivers/gpu/drm/radeon/radeon_display.c @@ -1822,6 +1822,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, struct radeon_device *rdev = dev->dev_private; /* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */ + preempt_disable_rt(); /* Get optional system timestamp before query. */ if (stime) @@ -1914,6 +1915,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe, *etime = ktime_get(); /* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */ + preempt_enable_rt(); /* Decode into vertical and horizontal scanout position. */ *vpos = position & 0x1fff; diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h index 40e2b9f91163c..d9de4813ffac8 100644 --- a/drivers/hv/hyperv_vmbus.h +++ b/drivers/hv/hyperv_vmbus.h @@ -18,6 +18,7 @@ #include #include #include +#include #include "hv_trace.h" diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index 946d0aba101f4..9447315ccd7ce 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -1303,6 +1304,8 @@ static void vmbus_isr(void) void *page_addr = hv_cpu->synic_event_page; struct hv_message *msg; union hv_synic_event_flags *event; + struct pt_regs *regs = get_irq_regs(); + u64 ip = regs ? instruction_pointer(regs) : 0; bool handled = false; if (unlikely(page_addr == NULL)) @@ -1347,7 +1350,7 @@ static void vmbus_isr(void) tasklet_schedule(&hv_cpu->msg_dpc); } - add_interrupt_randomness(HYPERVISOR_CALLBACK_VECTOR, 0); + add_interrupt_randomness(HYPERVISOR_CALLBACK_VECTOR, 0, ip); } /* diff --git a/drivers/leds/trigger/Kconfig b/drivers/leds/trigger/Kconfig index ce9429ca6ddea..29ccbd6acf43f 100644 --- a/drivers/leds/trigger/Kconfig +++ b/drivers/leds/trigger/Kconfig @@ -64,6 +64,7 @@ config LEDS_TRIGGER_BACKLIGHT config LEDS_TRIGGER_CPU bool "LED CPU Trigger" + depends on !PREEMPT_RT help This allows LEDs to be controlled by active CPUs. This shows the active CPUs across an array of LEDs so you can see which diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 225380efd1e24..7ae33e2edd350 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2077,8 +2077,9 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request) struct raid5_percpu *percpu; unsigned long cpu; - cpu = get_cpu(); + cpu = get_cpu_light(); percpu = per_cpu_ptr(conf->percpu, cpu); + spin_lock(&percpu->lock); if (test_bit(STRIPE_OP_BIOFILL, &ops_request)) { ops_run_biofill(sh); overlap_clear++; @@ -2137,7 +2138,8 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request) if (test_and_clear_bit(R5_Overlap, &dev->flags)) wake_up(&sh->raid_conf->wait_for_overlap); } - put_cpu(); + spin_unlock(&percpu->lock); + put_cpu_light(); } static void free_stripe(struct kmem_cache *sc, struct stripe_head *sh) @@ -6902,6 +6904,7 @@ static int raid456_cpu_up_prepare(unsigned int cpu, struct hlist_node *node) __func__, cpu); return -ENOMEM; } + spin_lock_init(&per_cpu_ptr(conf->percpu, cpu)->lock); return 0; } diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h index 16fc29472f5c0..d16edbdcde6e6 100644 --- a/drivers/md/raid5.h +++ b/drivers/md/raid5.h @@ -627,6 +627,7 @@ struct r5conf { int recovery_disabled; /* per cpu variables */ struct raid5_percpu { + spinlock_t lock; /* Protection for -RT */ struct page *spare_page; /* Used when checking P/Q in raid6 */ void *scribble; /* space for constructing buffer * lists and performing address diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c index 0f9274960dc6b..dc97e4f1f4ad2 100644 --- a/drivers/scsi/fcoe/fcoe.c +++ b/drivers/scsi/fcoe/fcoe.c @@ -1452,11 +1452,11 @@ static int fcoe_rcv(struct sk_buff *skb, struct net_device *netdev, static int fcoe_alloc_paged_crc_eof(struct sk_buff *skb, int tlen) { struct fcoe_percpu_s *fps; - int rc; + int rc, cpu = get_cpu_light(); - fps = &get_cpu_var(fcoe_percpu); + fps = &per_cpu(fcoe_percpu, cpu); rc = fcoe_get_paged_crc_eof(skb, tlen, fps); - put_cpu_var(fcoe_percpu); + put_cpu_light(); return rc; } @@ -1641,11 +1641,11 @@ static inline int fcoe_filter_frames(struct fc_lport *lport, return 0; } - stats = per_cpu_ptr(lport->stats, get_cpu()); + stats = per_cpu_ptr(lport->stats, get_cpu_light()); stats->InvalidCRCCount++; if (stats->InvalidCRCCount < 5) printk(KERN_WARNING "fcoe: dropping frame with CRC error\n"); - put_cpu(); + put_cpu_light(); return -EINVAL; } @@ -1686,7 +1686,7 @@ static void fcoe_recv_frame(struct sk_buff *skb) */ hp = (struct fcoe_hdr *) skb_network_header(skb); - stats = per_cpu_ptr(lport->stats, get_cpu()); + stats = per_cpu_ptr(lport->stats, get_cpu_light()); if (unlikely(FC_FCOE_DECAPS_VER(hp) != FC_FCOE_VER)) { if (stats->ErrorFrames < 5) printk(KERN_WARNING "fcoe: FCoE version " @@ -1718,13 +1718,13 @@ static void fcoe_recv_frame(struct sk_buff *skb) goto drop; if (!fcoe_filter_frames(lport, fp)) { - put_cpu(); + put_cpu_light(); fc_exch_recv(lport, fp); return; } drop: stats->ErrorFrames++; - put_cpu(); + put_cpu_light(); kfree_skb(skb); } diff --git a/drivers/scsi/fcoe/fcoe_ctlr.c b/drivers/scsi/fcoe/fcoe_ctlr.c index 5ea426effa609..0d6b9acc7cf81 100644 --- a/drivers/scsi/fcoe/fcoe_ctlr.c +++ b/drivers/scsi/fcoe/fcoe_ctlr.c @@ -828,7 +828,7 @@ static unsigned long fcoe_ctlr_age_fcfs(struct fcoe_ctlr *fip) INIT_LIST_HEAD(&del_list); - stats = per_cpu_ptr(fip->lp->stats, get_cpu()); + stats = per_cpu_ptr(fip->lp->stats, get_cpu_light()); list_for_each_entry_safe(fcf, next, &fip->fcfs, list) { deadline = fcf->time + fcf->fka_period + fcf->fka_period / 2; @@ -864,7 +864,7 @@ static unsigned long fcoe_ctlr_age_fcfs(struct fcoe_ctlr *fip) sel_time = fcf->time; } } - put_cpu(); + put_cpu_light(); list_for_each_entry_safe(fcf, next, &del_list, list) { /* Removes fcf from current list */ diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c index 96a2952cf626b..1cb773a4a02e7 100644 --- a/drivers/scsi/libfc/fc_exch.c +++ b/drivers/scsi/libfc/fc_exch.c @@ -826,10 +826,10 @@ static struct fc_exch *fc_exch_em_alloc(struct fc_lport *lport, } memset(ep, 0, sizeof(*ep)); - cpu = get_cpu(); + cpu = get_cpu_light(); pool = per_cpu_ptr(mp->pool, cpu); spin_lock_bh(&pool->lock); - put_cpu(); + put_cpu_light(); /* peek cache of free slot */ if (pool->left != FC_XID_UNKNOWN) { diff --git a/drivers/tty/serial/8250/8250.h b/drivers/tty/serial/8250/8250.h index 52bb21205bb68..5cbcaafbb4aa2 100644 --- a/drivers/tty/serial/8250/8250.h +++ b/drivers/tty/serial/8250/8250.h @@ -130,12 +130,55 @@ static inline void serial_dl_write(struct uart_8250_port *up, int value) up->dl_write(up, value); } +static inline void serial8250_set_IER(struct uart_8250_port *up, + unsigned char ier) +{ + struct uart_port *port = &up->port; + unsigned int flags; + bool is_console; + + is_console = uart_console(port); + + if (is_console) + console_atomic_lock(&flags); + + serial_out(up, UART_IER, ier); + + if (is_console) + console_atomic_unlock(flags); +} + +static inline unsigned char serial8250_clear_IER(struct uart_8250_port *up) +{ + struct uart_port *port = &up->port; + unsigned int clearval = 0; + unsigned int prior; + unsigned int flags; + bool is_console; + + is_console = uart_console(port); + + if (up->capabilities & UART_CAP_UUE) + clearval = UART_IER_UUE; + + if (is_console) + console_atomic_lock(&flags); + + prior = serial_port_in(port, UART_IER); + serial_port_out(port, UART_IER, clearval); + + if (is_console) + console_atomic_unlock(flags); + + return prior; +} + static inline bool serial8250_set_THRI(struct uart_8250_port *up) { if (up->ier & UART_IER_THRI) return false; up->ier |= UART_IER_THRI; - serial_out(up, UART_IER, up->ier); + serial8250_set_IER(up, up->ier); return true; } @@ -144,7 +187,7 @@ static inline bool serial8250_clear_THRI(struct uart_8250_port *up) if (!(up->ier & UART_IER_THRI)) return false; up->ier &= ~UART_IER_THRI; - serial_out(up, UART_IER, up->ier); + serial8250_set_IER(up, up->ier); return true; } diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c index cae61d1ebec5a..47dd230562719 100644 --- a/drivers/tty/serial/8250/8250_core.c +++ b/drivers/tty/serial/8250/8250_core.c @@ -274,10 +274,8 @@ static void serial8250_backup_timeout(struct timer_list *t) * Must disable interrupts or else we risk racing with the interrupt * based handler. */ - if (up->port.irq) { - ier = serial_in(up, UART_IER); - serial_out(up, UART_IER, 0); - } + if (up->port.irq) + ier = serial8250_clear_IER(up); iir = serial_in(up, UART_IIR); @@ -300,7 +298,7 @@ static void serial8250_backup_timeout(struct timer_list *t) serial8250_tx_chars(up); if (up->port.irq) - serial_out(up, UART_IER, ier); + serial8250_set_IER(up, ier); spin_unlock_irqrestore(&up->port.lock, flags); @@ -578,6 +576,14 @@ serial8250_register_ports(struct uart_driver *drv, struct device *dev) #ifdef CONFIG_SERIAL_8250_CONSOLE +static void univ8250_console_write_atomic(struct console *co, const char *s, + unsigned int count) +{ + struct uart_8250_port *up = &serial8250_ports[co->index]; + + serial8250_console_write_atomic(up, s, count); +} + static void univ8250_console_write(struct console *co, const char *s, unsigned int count) { @@ -671,6 +677,7 @@ static int univ8250_console_match(struct console *co, char *name, int idx, static struct console univ8250_console = { .name = "ttyS", + .write_atomic = univ8250_console_write_atomic, .write = univ8250_console_write, .device = uart_console_device, .setup = univ8250_console_setup, diff --git a/drivers/tty/serial/8250/8250_fsl.c b/drivers/tty/serial/8250/8250_fsl.c index 0d0c80905c581..f8116a8839d8a 100644 --- a/drivers/tty/serial/8250/8250_fsl.c +++ b/drivers/tty/serial/8250/8250_fsl.c @@ -53,9 +53,18 @@ int fsl8250_handle_irq(struct uart_port *port) /* Stop processing interrupts on input overrun */ if ((orig_lsr & UART_LSR_OE) && (up->overrun_backoff_time_ms > 0)) { + unsigned int ca_flags; unsigned long delay; + bool is_console; + is_console = uart_console(port); + + if (is_console) + console_atomic_lock(&ca_flags); up->ier = port->serial_in(port, UART_IER); + if (is_console) + console_atomic_unlock(ca_flags); + if (up->ier & (UART_IER_RLSI | UART_IER_RDI)) { port->ops->stop_rx(port); } else { diff --git a/drivers/tty/serial/8250/8250_ingenic.c b/drivers/tty/serial/8250/8250_ingenic.c index dde766fa465ff..f4cceca82748e 100644 --- a/drivers/tty/serial/8250/8250_ingenic.c +++ b/drivers/tty/serial/8250/8250_ingenic.c @@ -146,6 +146,8 @@ OF_EARLYCON_DECLARE(x1000_uart, "ingenic,x1000-uart", static void ingenic_uart_serial_out(struct uart_port *p, int offset, int value) { + unsigned int flags; + bool is_console; int ier; switch (offset) { @@ -167,7 +169,12 @@ static void ingenic_uart_serial_out(struct uart_port *p, int offset, int value) * If we have enabled modem status IRQs we should enable * modem mode. */ + is_console = uart_console(p); + if (is_console) + console_atomic_lock(&flags); ier = p->serial_in(p, UART_IER); + if (is_console) + console_atomic_unlock(flags); if (ier & UART_IER_MSI) value |= UART_MCR_MDCE | UART_MCR_FCM; diff --git a/drivers/tty/serial/8250/8250_mtk.c b/drivers/tty/serial/8250/8250_mtk.c index 7b0dec14c8b80..d323ff7051f69 100644 --- a/drivers/tty/serial/8250/8250_mtk.c +++ b/drivers/tty/serial/8250/8250_mtk.c @@ -213,12 +213,37 @@ static void mtk8250_shutdown(struct uart_port *port) static void mtk8250_disable_intrs(struct uart_8250_port *up, int mask) { - serial_out(up, UART_IER, serial_in(up, UART_IER) & (~mask)); + struct uart_port *port = &up->port; + unsigned int flags; + unsigned int ier; + bool is_console; + + is_console = uart_console(port); + + if (is_console) + console_atomic_lock(&flags); + + ier = serial_in(up, UART_IER); + serial_out(up, UART_IER, ier & (~mask)); + + if (is_console) + console_atomic_unlock(flags); } static void mtk8250_enable_intrs(struct uart_8250_port *up, int mask) { - serial_out(up, UART_IER, serial_in(up, UART_IER) | mask); + struct uart_port *port = &up->port; + unsigned int flags; + unsigned int ier; + + if (uart_console(port)) + console_atomic_lock(&flags); + + ier = serial_in(up, UART_IER); + serial_out(up, UART_IER, ier | mask); + + if (uart_console(port)) + console_atomic_unlock(flags); } static void mtk8250_set_flow_ctrl(struct uart_8250_port *up, int mode) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index c71d647eb87a0..4325a15be5731 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -757,7 +757,7 @@ static void serial8250_set_sleep(struct uart_8250_port *p, int sleep) serial_out(p, UART_EFR, UART_EFR_ECB); serial_out(p, UART_LCR, 0); } - serial_out(p, UART_IER, sleep ? UART_IERX_SLEEP : 0); + serial8250_set_IER(p, sleep ? UART_IERX_SLEEP : 0); if (p->capabilities & UART_CAP_EFR) { serial_out(p, UART_LCR, UART_LCR_CONF_MODE_B); serial_out(p, UART_EFR, efr); @@ -1429,7 +1429,7 @@ static void serial8250_stop_rx(struct uart_port *port) up->ier &= ~(UART_IER_RLSI | UART_IER_RDI); up->port.read_status_mask &= ~UART_LSR_DR; - serial_port_out(port, UART_IER, up->ier); + serial8250_set_IER(up, up->ier); serial8250_rpm_put(up); } @@ -1459,7 +1459,7 @@ void serial8250_em485_stop_tx(struct uart_8250_port *p) serial8250_clear_and_reinit_fifos(p); p->ier |= UART_IER_RLSI | UART_IER_RDI; - serial_port_out(&p->port, UART_IER, p->ier); + serial8250_set_IER(p, p->ier); } } EXPORT_SYMBOL_GPL(serial8250_em485_stop_tx); @@ -1687,7 +1687,7 @@ static void serial8250_disable_ms(struct uart_port *port) mctrl_gpio_disable_ms(up->gpios); up->ier &= ~UART_IER_MSI; - serial_port_out(port, UART_IER, up->ier); + serial8250_set_IER(up, up->ier); } static void serial8250_enable_ms(struct uart_port *port) @@ -1703,7 +1703,7 @@ static void serial8250_enable_ms(struct uart_port *port) up->ier |= UART_IER_MSI; serial8250_rpm_get(up); - serial_port_out(port, UART_IER, up->ier); + serial8250_set_IER(up, up->ier); serial8250_rpm_put(up); } @@ -2118,14 +2118,7 @@ static void serial8250_put_poll_char(struct uart_port *port, struct uart_8250_port *up = up_to_u8250p(port); serial8250_rpm_get(up); - /* - * First save the IER then disable the interrupts - */ - ier = serial_port_in(port, UART_IER); - if (up->capabilities & UART_CAP_UUE) - serial_port_out(port, UART_IER, UART_IER_UUE); - else - serial_port_out(port, UART_IER, 0); + ier = serial8250_clear_IER(up); wait_for_xmitr(up, BOTH_EMPTY); /* @@ -2138,7 +2131,7 @@ static void serial8250_put_poll_char(struct uart_port *port, * and restore the IER */ wait_for_xmitr(up, BOTH_EMPTY); - serial_port_out(port, UART_IER, ier); + serial8250_set_IER(up, ier); serial8250_rpm_put(up); } @@ -2441,7 +2434,7 @@ void serial8250_do_shutdown(struct uart_port *port) */ spin_lock_irqsave(&port->lock, flags); up->ier = 0; - serial_port_out(port, UART_IER, 0); + serial8250_set_IER(up, 0); spin_unlock_irqrestore(&port->lock, flags); synchronize_irq(port->irq); @@ -2768,7 +2761,7 @@ serial8250_do_set_termios(struct uart_port *port, struct ktermios *termios, if (up->capabilities & UART_CAP_RTOIE) up->ier |= UART_IER_RTOIE; - serial_port_out(port, UART_IER, up->ier); + serial8250_set_IER(up, up->ier); if (up->capabilities & UART_CAP_EFR) { unsigned char efr = 0; @@ -3234,7 +3227,7 @@ EXPORT_SYMBOL_GPL(serial8250_set_defaults); #ifdef CONFIG_SERIAL_8250_CONSOLE -static void serial8250_console_putchar(struct uart_port *port, int ch) +static void serial8250_console_putchar_locked(struct uart_port *port, int ch) { struct uart_8250_port *up = up_to_u8250p(port); @@ -3242,6 +3235,18 @@ static void serial8250_console_putchar(struct uart_port *port, int ch) serial_port_out(port, UART_TX, ch); } +static void serial8250_console_putchar(struct uart_port *port, int ch) +{ + struct uart_8250_port *up = up_to_u8250p(port); + unsigned int flags; + + wait_for_xmitr(up, UART_LSR_THRE); + + console_atomic_lock(&flags); + serial8250_console_putchar_locked(port, ch); + console_atomic_unlock(flags); +} + /* * Restore serial console when h/w power-off detected */ @@ -3263,6 +3268,32 @@ static void serial8250_console_restore(struct uart_8250_port *up) serial8250_out_MCR(up, UART_MCR_DTR | UART_MCR_RTS); } +void serial8250_console_write_atomic(struct uart_8250_port *up, + const char *s, unsigned int count) +{ + struct uart_port *port = &up->port; + unsigned int flags; + unsigned int ier; + + console_atomic_lock(&flags); + + touch_nmi_watchdog(); + + ier = serial8250_clear_IER(up); + + if (atomic_fetch_inc(&up->console_printing)) { + uart_console_write(port, "\n", 1, + serial8250_console_putchar_locked); + } + uart_console_write(port, s, count, serial8250_console_putchar_locked); + atomic_dec(&up->console_printing); + + wait_for_xmitr(up, BOTH_EMPTY); + serial8250_set_IER(up, ier); + + console_atomic_unlock(flags); +} + /* * Print a string to the serial port trying not to disturb * any possible real use of the port... @@ -3279,24 +3310,12 @@ void serial8250_console_write(struct uart_8250_port *up, const char *s, struct uart_port *port = &up->port; unsigned long flags; unsigned int ier; - int locked = 1; touch_nmi_watchdog(); - if (oops_in_progress) - locked = spin_trylock_irqsave(&port->lock, flags); - else - spin_lock_irqsave(&port->lock, flags); + spin_lock_irqsave(&port->lock, flags); - /* - * First save the IER then disable the interrupts - */ - ier = serial_port_in(port, UART_IER); - - if (up->capabilities & UART_CAP_UUE) - serial_port_out(port, UART_IER, UART_IER_UUE); - else - serial_port_out(port, UART_IER, 0); + ier = serial8250_clear_IER(up); /* check scratch reg to see if port powered off during system sleep */ if (up->canary && (up->canary != serial_port_in(port, UART_SCR))) { @@ -3310,7 +3329,9 @@ void serial8250_console_write(struct uart_8250_port *up, const char *s, mdelay(port->rs485.delay_rts_before_send); } + atomic_inc(&up->console_printing); uart_console_write(port, s, count, serial8250_console_putchar); + atomic_dec(&up->console_printing); /* * Finally, wait for transmitter to become empty @@ -3323,8 +3344,7 @@ void serial8250_console_write(struct uart_8250_port *up, const char *s, if (em485->tx_stopped) up->rs485_stop_tx(up); } - - serial_port_out(port, UART_IER, ier); + serial8250_set_IER(up, ier); /* * The receive handling will happen properly because the @@ -3336,8 +3356,7 @@ void serial8250_console_write(struct uart_8250_port *up, const char *s, if (up->msr_saved_flags) serial8250_modem_status(up); - if (locked) - spin_unlock_irqrestore(&port->lock, flags); + spin_unlock_irqrestore(&port->lock, flags); } static unsigned int probe_baud(struct uart_port *port) @@ -3357,6 +3376,7 @@ static unsigned int probe_baud(struct uart_port *port) int serial8250_console_setup(struct uart_port *port, char *options, bool probe) { + struct uart_8250_port *up = up_to_u8250p(port); int baud = 9600; int bits = 8; int parity = 'n'; @@ -3366,6 +3386,8 @@ int serial8250_console_setup(struct uart_port *port, char *options, bool probe) if (!port->iobase && !port->membase) return -ENODEV; + atomic_set(&up->console_printing, 0); + if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); else if (probe) diff --git a/drivers/tty/serial/amba-pl011.c b/drivers/tty/serial/amba-pl011.c index 67498594d7d7e..e5cf42bbabf35 100644 --- a/drivers/tty/serial/amba-pl011.c +++ b/drivers/tty/serial/amba-pl011.c @@ -2198,18 +2198,24 @@ pl011_console_write(struct console *co, const char *s, unsigned int count) { struct uart_amba_port *uap = amba_ports[co->index]; unsigned int old_cr = 0, new_cr; - unsigned long flags; + unsigned long flags = 0; int locked = 1; clk_enable(uap->clk); - local_irq_save(flags); + /* + * local_irq_save(flags); + * + * This local_irq_save() is nonsense. If we come in via sysrq + * handling then interrupts are already disabled. Aside of + * that the port.sysrq check is racy on SMP regardless. + */ if (uap->port.sysrq) locked = 0; else if (oops_in_progress) - locked = spin_trylock(&uap->port.lock); + locked = spin_trylock_irqsave(&uap->port.lock, flags); else - spin_lock(&uap->port.lock); + spin_lock_irqsave(&uap->port.lock, flags); /* * First save the CR then disable the interrupts @@ -2235,8 +2241,7 @@ pl011_console_write(struct console *co, const char *s, unsigned int count) pl011_write(old_cr, uap, REG_CR); if (locked) - spin_unlock(&uap->port.lock); - local_irq_restore(flags); + spin_unlock_irqrestore(&uap->port.lock, flags); clk_disable(uap->clk); } diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c index 76b94d0ff5865..80371598efea4 100644 --- a/drivers/tty/serial/omap-serial.c +++ b/drivers/tty/serial/omap-serial.c @@ -1301,13 +1301,10 @@ serial_omap_console_write(struct console *co, const char *s, pm_runtime_get_sync(up->dev); - local_irq_save(flags); - if (up->port.sysrq) - locked = 0; - else if (oops_in_progress) - locked = spin_trylock(&up->port.lock); + if (up->port.sysrq || oops_in_progress) + locked = spin_trylock_irqsave(&up->port.lock, flags); else - spin_lock(&up->port.lock); + spin_lock_irqsave(&up->port.lock, flags); /* * First save the IER then disable the interrupts @@ -1336,8 +1333,7 @@ serial_omap_console_write(struct console *co, const char *s, pm_runtime_mark_last_busy(up->dev); pm_runtime_put_autosuspend(up->dev); if (locked) - spin_unlock(&up->port.lock); - local_irq_restore(flags); + spin_unlock_irqrestore(&up->port.lock, flags); } static int __init diff --git a/fs/afs/dir_silly.c b/fs/afs/dir_silly.c index 04f75a44f2432..60cbce1995a59 100644 --- a/fs/afs/dir_silly.c +++ b/fs/afs/dir_silly.c @@ -236,7 +236,7 @@ int afs_silly_iput(struct dentry *dentry, struct inode *inode) struct dentry *alias; int ret; - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); _enter("%p{%pd},%llx", dentry, dentry, vnode->fid.vnode); diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 6df0922e7e304..5b2371b26bf85 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -81,7 +81,7 @@ cifs_prime_dcache(struct dentry *parent, struct qstr *name, struct inode *inode; struct super_block *sb = parent->d_sb; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); cifs_dbg(FYI, "%s: for %s\n", __func__, name->name); diff --git a/fs/dcache.c b/fs/dcache.c index ea0485861d937..26a187abf13a6 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2503,9 +2503,10 @@ EXPORT_SYMBOL(d_rehash); static inline unsigned start_dir_add(struct inode *dir) { + preempt_disable_rt(); for (;;) { - unsigned n = dir->i_dir_seq; - if (!(n & 1) && cmpxchg(&dir->i_dir_seq, n, n + 1) == n) + unsigned n = dir->__i_dir_seq; + if (!(n & 1) && cmpxchg(&dir->__i_dir_seq, n, n + 1) == n) return n; cpu_relax(); } @@ -2513,26 +2514,30 @@ static inline unsigned start_dir_add(struct inode *dir) static inline void end_dir_add(struct inode *dir, unsigned n) { - smp_store_release(&dir->i_dir_seq, n + 2); + smp_store_release(&dir->__i_dir_seq, n + 2); + preempt_enable_rt(); } static void d_wait_lookup(struct dentry *dentry) { - if (d_in_lookup(dentry)) { - DECLARE_WAITQUEUE(wait, current); - add_wait_queue(dentry->d_wait, &wait); - do { - set_current_state(TASK_UNINTERRUPTIBLE); - spin_unlock(&dentry->d_lock); - schedule(); - spin_lock(&dentry->d_lock); - } while (d_in_lookup(dentry)); - } + struct swait_queue __wait; + + if (!d_in_lookup(dentry)) + return; + + INIT_LIST_HEAD(&__wait.task_list); + do { + prepare_to_swait_exclusive(dentry->d_wait, &__wait, TASK_UNINTERRUPTIBLE); + spin_unlock(&dentry->d_lock); + schedule(); + spin_lock(&dentry->d_lock); + } while (d_in_lookup(dentry)); + finish_swait(dentry->d_wait, &__wait); } struct dentry *d_alloc_parallel(struct dentry *parent, const struct qstr *name, - wait_queue_head_t *wq) + struct swait_queue_head *wq) { unsigned int hash = name->hash; struct hlist_bl_head *b = in_lookup_hash(parent, hash); @@ -2546,7 +2551,7 @@ struct dentry *d_alloc_parallel(struct dentry *parent, retry: rcu_read_lock(); - seq = smp_load_acquire(&parent->d_inode->i_dir_seq); + seq = smp_load_acquire(&parent->d_inode->__i_dir_seq); r_seq = read_seqbegin(&rename_lock); dentry = __d_lookup_rcu(parent, name, &d_seq); if (unlikely(dentry)) { @@ -2574,7 +2579,7 @@ struct dentry *d_alloc_parallel(struct dentry *parent, } hlist_bl_lock(b); - if (unlikely(READ_ONCE(parent->d_inode->i_dir_seq) != seq)) { + if (unlikely(READ_ONCE(parent->d_inode->__i_dir_seq) != seq)) { hlist_bl_unlock(b); rcu_read_unlock(); goto retry; @@ -2647,7 +2652,7 @@ void __d_lookup_done(struct dentry *dentry) hlist_bl_lock(b); dentry->d_flags &= ~DCACHE_PAR_LOOKUP; __hlist_bl_del(&dentry->d_u.d_in_lookup_hash); - wake_up_all(dentry->d_wait); + swake_up_all(dentry->d_wait); dentry->d_wait = NULL; hlist_bl_unlock(b); INIT_HLIST_NODE(&dentry->d_u.d_alias); diff --git a/fs/exec.c b/fs/exec.c index a91003e28eaae..d4fb18baf1fb1 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1130,11 +1130,24 @@ static int exec_mmap(struct mm_struct *mm) } task_lock(tsk); - active_mm = tsk->active_mm; membarrier_exec_mmap(mm); - tsk->mm = mm; + + local_irq_disable(); + active_mm = tsk->active_mm; tsk->active_mm = mm; + tsk->mm = mm; + /* + * This prevents preemption while active_mm is being loaded and + * it and mm are being updated, which could cause problems for + * lazy tlb mm refcounting when these are updated by context + * switches. Not all architectures can handle irqs off over + * activate_mm yet. + */ + if (!IS_ENABLED(CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM)) + local_irq_enable(); activate_mm(active_mm, mm); + if (IS_ENABLED(CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM)) + local_irq_enable(); tsk->mm->vmacache_seqnum = 0; vmacache_flush(tsk); task_unlock(tsk); diff --git a/fs/fuse/readdir.c b/fs/fuse/readdir.c index 90e3f01bd796b..ec8fd82117f4c 100644 --- a/fs/fuse/readdir.c +++ b/fs/fuse/readdir.c @@ -158,7 +158,7 @@ static int fuse_direntplus_link(struct file *file, struct inode *dir = d_inode(parent); struct fuse_conn *fc; struct inode *inode; - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); if (!o->nodeid) { /* diff --git a/fs/inode.c b/fs/inode.c index 72c4c347afb70..1b27d29265cfa 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -158,7 +158,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) inode->i_bdev = NULL; inode->i_cdev = NULL; inode->i_link = NULL; - inode->i_dir_seq = 0; + inode->__i_dir_seq = 0; inode->i_rdev = 0; inode->dirtied_when = 0; diff --git a/fs/io-wq.c b/fs/io-wq.c index 414beb5438836..f1cf739d63c21 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -87,7 +87,7 @@ enum { */ struct io_wqe { struct { - spinlock_t lock; + raw_spinlock_t lock; struct io_wq_work_list work_list; unsigned long hash_map; unsigned flags; @@ -148,7 +148,7 @@ static bool __io_worker_unuse(struct io_wqe *wqe, struct io_worker *worker) if (current->files != worker->restore_files) { __acquire(&wqe->lock); - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); dropped_lock = true; task_lock(current); @@ -166,7 +166,7 @@ static bool __io_worker_unuse(struct io_wqe *wqe, struct io_worker *worker) if (worker->mm) { if (!dropped_lock) { __acquire(&wqe->lock); - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); dropped_lock = true; } __set_current_state(TASK_RUNNING); @@ -220,17 +220,17 @@ static void io_worker_exit(struct io_worker *worker) worker->flags = 0; preempt_enable(); - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); hlist_nulls_del_rcu(&worker->nulls_node); list_del_rcu(&worker->all_list); if (__io_worker_unuse(wqe, worker)) { __release(&wqe->lock); - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); } acct->nr_workers--; nr_workers = wqe->acct[IO_WQ_ACCT_BOUND].nr_workers + wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers; - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); /* all workers gone, wq exit can proceed */ if (!nr_workers && refcount_dec_and_test(&wqe->wq->refs)) @@ -504,7 +504,7 @@ static void io_worker_handle_work(struct io_worker *worker) else if (!wq_list_empty(&wqe->work_list)) wqe->flags |= IO_WQE_FLAG_STALLED; - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); if (!work) break; io_assign_current_work(worker, work); @@ -538,17 +538,17 @@ static void io_worker_handle_work(struct io_worker *worker) io_wqe_enqueue(wqe, linked); if (hash != -1U && !next_hashed) { - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); wqe->hash_map &= ~BIT_ULL(hash); wqe->flags &= ~IO_WQE_FLAG_STALLED; /* skip unnecessary unlock-lock wqe->lock */ if (!work) goto get_next; - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); } } while (work); - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); } while (1); } @@ -563,7 +563,7 @@ static int io_wqe_worker(void *data) while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)) { set_current_state(TASK_INTERRUPTIBLE); loop: - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); if (io_wqe_run_queue(wqe)) { __set_current_state(TASK_RUNNING); io_worker_handle_work(worker); @@ -574,7 +574,7 @@ static int io_wqe_worker(void *data) __release(&wqe->lock); goto loop; } - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); if (signal_pending(current)) flush_signals(current); if (schedule_timeout(WORKER_IDLE_TIMEOUT)) @@ -586,11 +586,11 @@ static int io_wqe_worker(void *data) } if (test_bit(IO_WQ_BIT_EXIT, &wq->state)) { - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); if (!wq_list_empty(&wqe->work_list)) io_worker_handle_work(worker); else - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); } io_worker_exit(worker); @@ -630,9 +630,9 @@ void io_wq_worker_sleeping(struct task_struct *tsk) worker->flags &= ~IO_WORKER_F_RUNNING; - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); io_wqe_dec_running(wqe, worker); - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); } static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) @@ -656,7 +656,7 @@ static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) return false; } - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list); list_add_tail_rcu(&worker->all_list, &wqe->all_list); worker->flags |= IO_WORKER_F_FREE; @@ -665,7 +665,7 @@ static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) if (!acct->nr_workers && (worker->flags & IO_WORKER_F_BOUND)) worker->flags |= IO_WORKER_F_FIXED; acct->nr_workers++; - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); if (index == IO_WQ_ACCT_UNBOUND) atomic_inc(&wq->user->processes); @@ -720,12 +720,12 @@ static int io_wq_manager(void *data) if (!node_online(node)) continue; - spin_lock_irq(&wqe->lock); + raw_spin_lock_irq(&wqe->lock); if (io_wqe_need_worker(wqe, IO_WQ_ACCT_BOUND)) fork_worker[IO_WQ_ACCT_BOUND] = true; if (io_wqe_need_worker(wqe, IO_WQ_ACCT_UNBOUND)) fork_worker[IO_WQ_ACCT_UNBOUND] = true; - spin_unlock_irq(&wqe->lock); + raw_spin_unlock_irq(&wqe->lock); if (fork_worker[IO_WQ_ACCT_BOUND]) create_io_worker(wq, wqe, IO_WQ_ACCT_BOUND); if (fork_worker[IO_WQ_ACCT_UNBOUND]) @@ -821,10 +821,10 @@ static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work) } work_flags = work->flags; - spin_lock_irqsave(&wqe->lock, flags); + raw_spin_lock_irqsave(&wqe->lock, flags); io_wqe_insert_work(wqe, work); wqe->flags &= ~IO_WQE_FLAG_STALLED; - spin_unlock_irqrestore(&wqe->lock, flags); + raw_spin_unlock_irqrestore(&wqe->lock, flags); if ((work_flags & IO_WQ_WORK_CONCURRENT) || !atomic_read(&acct->nr_running)) @@ -951,13 +951,13 @@ static void io_wqe_cancel_pending_work(struct io_wqe *wqe, unsigned long flags; retry: - spin_lock_irqsave(&wqe->lock, flags); + raw_spin_lock_irqsave(&wqe->lock, flags); wq_list_for_each(node, prev, &wqe->work_list) { work = container_of(node, struct io_wq_work, list); if (!match->fn(work, match->data)) continue; io_wqe_remove_pending(wqe, work, prev); - spin_unlock_irqrestore(&wqe->lock, flags); + raw_spin_unlock_irqrestore(&wqe->lock, flags); io_run_cancel(work, wqe); match->nr_pending++; if (!match->cancel_all) @@ -966,7 +966,7 @@ static void io_wqe_cancel_pending_work(struct io_wqe *wqe, /* not safe to continue after unlock */ goto retry; } - spin_unlock_irqrestore(&wqe->lock, flags); + raw_spin_unlock_irqrestore(&wqe->lock, flags); } static void io_wqe_cancel_running_work(struct io_wqe *wqe, @@ -1074,7 +1074,7 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) } atomic_set(&wqe->acct[IO_WQ_ACCT_UNBOUND].nr_running, 0); wqe->wq = wq; - spin_lock_init(&wqe->lock); + raw_spin_lock_init(&wqe->lock); INIT_WQ_LIST(&wqe->work_list); INIT_HLIST_NULLS_HEAD(&wqe->free_list, 0); INIT_LIST_HEAD(&wqe->all_list); diff --git a/fs/namei.c b/fs/namei.c index e99e2a9da0f7d..8b7cbca327ded 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1520,7 +1520,7 @@ static struct dentry *__lookup_slow(const struct qstr *name, { struct dentry *dentry, *old; struct inode *inode = dir->d_inode; - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); /* Don't go there if it's already dead */ if (unlikely(IS_DEADDIR(inode))) @@ -3018,7 +3018,7 @@ static struct dentry *lookup_open(struct nameidata *nd, struct file *file, struct dentry *dentry; int error, create_error = 0; umode_t mode = op->mode; - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); if (unlikely(IS_DEADDIR(dir_inode))) return ERR_PTR(-ENOENT); diff --git a/fs/namespace.c b/fs/namespace.c index bae0e95b3713a..c7e53f6ba7df8 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -321,8 +322,11 @@ int __mnt_want_write(struct vfsmount *m) * incremented count after it has set MNT_WRITE_HOLD. */ smp_mb(); - while (READ_ONCE(mnt->mnt.mnt_flags) & MNT_WRITE_HOLD) - cpu_relax(); + while (READ_ONCE(mnt->mnt.mnt_flags) & MNT_WRITE_HOLD) { + preempt_enable(); + cpu_chill(); + preempt_disable(); + } /* * After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will * be set to match its requirements. So we must not load that until diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index cb52db9a0cfb7..b5703ca9492cd 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -484,7 +484,7 @@ void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry, unsigned long dir_verifier) { struct qstr filename = QSTR_INIT(entry->name, entry->len); - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); struct dentry *dentry; struct dentry *alias; struct inode *inode; @@ -1665,7 +1665,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, struct file *file, unsigned open_flags, umode_t mode) { - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); struct nfs_open_context *ctx; struct dentry *res; struct iattr attr = { .ia_valid = ATTR_OPEN }; diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index b27ebdccef703..f86c98a7ed047 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include #include @@ -180,7 +180,7 @@ nfs_async_unlink(struct dentry *dentry, const struct qstr *name) data->cred = get_current_cred(); data->res.dir_attr = &data->dir_attr; - init_waitqueue_head(&data->wq); + init_swait_queue_head(&data->wq); status = -EBUSY; spin_lock(&dentry->d_lock); diff --git a/fs/proc/array.c b/fs/proc/array.c index 65ec2029fa802..7052441be9679 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -382,9 +382,9 @@ static inline void task_context_switch_counts(struct seq_file *m, static void task_cpus_allowed(struct seq_file *m, struct task_struct *task) { seq_printf(m, "Cpus_allowed:\t%*pb\n", - cpumask_pr_args(task->cpus_ptr)); + cpumask_pr_args(&task->cpus_mask)); seq_printf(m, "Cpus_allowed_list:\t%*pbl\n", - cpumask_pr_args(task->cpus_ptr)); + cpumask_pr_args(&task->cpus_mask)); } static inline void task_core_dumping(struct seq_file *m, struct mm_struct *mm) diff --git a/fs/proc/base.c b/fs/proc/base.c index 617db4e0faa09..9e4520113c022 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -96,6 +96,7 @@ #include #include #include +#include #include #include "internal.h" #include "fd.h" @@ -2033,7 +2034,7 @@ bool proc_fill_cache(struct file *file, struct dir_context *ctx, child = d_hash_and_lookup(dir, &qname); if (!child) { - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); child = d_alloc_parallel(dir, &qname, &wq); if (IS_ERR(child)) goto end_instantiate; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 6c1166ccdaea5..dab5f1749ef1d 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -685,7 +685,7 @@ static bool proc_sys_fill_cache(struct file *file, child = d_lookup(dir, &qname); if (!child) { - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq); child = d_alloc_parallel(dir, &qname, &wq); if (IS_ERR(child)) return false; diff --git a/include/asm-generic/preempt.h b/include/asm-generic/preempt.h index d683f5e6d7913..71c1535db56a2 100644 --- a/include/asm-generic/preempt.h +++ b/include/asm-generic/preempt.h @@ -79,6 +79,9 @@ static __always_inline bool should_resched(int preempt_offset) } #ifdef CONFIG_PREEMPTION +#ifdef CONFIG_PREEMPT_RT +extern void preempt_schedule_lock(void); +#endif extern asmlinkage void preempt_schedule(void); #define __preempt_schedule() preempt_schedule() extern asmlinkage void preempt_schedule_notrace(void); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 868e11face00e..28231798f1613 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -154,7 +154,7 @@ struct request { */ union { struct hlist_node hash; /* merge hash */ - struct list_head ipi_list; + struct llist_node ipi_list; }; /* diff --git a/include/linux/bottom_half.h b/include/linux/bottom_half.h index a19519f4241dc..eed86eb0a1dea 100644 --- a/include/linux/bottom_half.h +++ b/include/linux/bottom_half.h @@ -4,7 +4,7 @@ #include -#ifdef CONFIG_TRACE_IRQFLAGS +#if defined(CONFIG_PREEMPT_RT) || defined(CONFIG_TRACE_IRQFLAGS) extern void __local_bh_disable_ip(unsigned long ip, unsigned int cnt); #else static __always_inline void __local_bh_disable_ip(unsigned long ip, unsigned int cnt) @@ -32,4 +32,10 @@ static inline void local_bh_enable(void) __local_bh_enable_ip(_THIS_IP_, SOFTIRQ_DISABLE_OFFSET); } +#ifdef CONFIG_PREEMPT_RT +extern bool local_bh_blocked(void); +#else +static inline bool local_bh_blocked(void) { return false; } +#endif + #endif /* _LINUX_BH_H */ diff --git a/include/linux/console.h b/include/linux/console.h index 0670d3491e0ea..00d7437a92e11 100644 --- a/include/linux/console.h +++ b/include/linux/console.h @@ -137,10 +137,12 @@ static inline int con_debug_leave(void) #define CON_ANYTIME (16) /* Safe to call when cpu is offline */ #define CON_BRL (32) /* Used for a braille device */ #define CON_EXTENDED (64) /* Use the extended output format a la /dev/kmsg */ +#define CON_HANDOVER (128) /* Device was previously a boot console. */ struct console { char name[16]; void (*write)(struct console *, const char *, unsigned); + void (*write_atomic)(struct console *, const char *, unsigned); int (*read)(struct console *, char *, unsigned); struct tty_driver *(*device)(struct console *, int *); void (*unblank)(void); @@ -150,6 +152,8 @@ struct console { short flags; short index; int cflag; + atomic64_t printk_seq; + struct task_struct *thread; void *data; struct console *next; }; @@ -230,4 +234,7 @@ extern void console_init(void); void dummycon_register_output_notifier(struct notifier_block *nb); void dummycon_unregister_output_notifier(struct notifier_block *nb); +extern void console_atomic_lock(unsigned int *flags); +extern void console_atomic_unlock(unsigned int flags); + #endif /* _LINUX_CONSOLE_H */ diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index bf9181cef444d..26b29e78dbc64 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -151,6 +151,7 @@ enum cpuhp_state { CPUHP_AP_ONLINE, CPUHP_TEARDOWN_CPU, CPUHP_AP_ONLINE_IDLE, + CPUHP_AP_SCHED_WAIT_EMPTY, CPUHP_AP_SMPBOOT_THREADS, CPUHP_AP_X86_VDSO_VMA_ONLINE, CPUHP_AP_IRQ_AFFINITY_ONLINE, diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index f0d895d6ac39f..383684e30f123 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -199,6 +199,11 @@ static inline int cpumask_any_and_distribute(const struct cpumask *src1p, return cpumask_next_and(-1, src1p, src2p); } +static inline int cpumask_any_distribute(const struct cpumask *srcp) +{ + return cpumask_first(srcp); +} + #define for_each_cpu(cpu, mask) \ for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask) #define for_each_cpu_not(cpu, mask) \ @@ -252,6 +257,7 @@ int cpumask_any_but(const struct cpumask *mask, unsigned int cpu); unsigned int cpumask_local_spread(unsigned int i, int node); int cpumask_any_and_distribute(const struct cpumask *src1p, const struct cpumask *src2p); +int cpumask_any_distribute(const struct cpumask *srcp); /** * for_each_cpu - iterate over every cpu in a mask diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h index 6594dbc34a374..206bde8308b2d 100644 --- a/include/linux/crash_core.h +++ b/include/linux/crash_core.h @@ -55,6 +55,9 @@ phys_addr_t paddr_vmcoreinfo_note(void); #define VMCOREINFO_OFFSET(name, field) \ vmcoreinfo_append_str("OFFSET(%s.%s)=%lu\n", #name, #field, \ (unsigned long)offsetof(struct name, field)) +#define VMCOREINFO_TYPE_OFFSET(name, field) \ + vmcoreinfo_append_str("OFFSET(%s.%s)=%lu\n", #name, #field, \ + (unsigned long)offsetof(name, field)) #define VMCOREINFO_LENGTH(name, value) \ vmcoreinfo_append_str("LENGTH(%s)=%lu\n", #name, (unsigned long)value) #define VMCOREINFO_NUMBER(name) \ diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 65d975bf93904..1e23dd02ac4e8 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -106,7 +106,7 @@ struct dentry { union { struct list_head d_lru; /* LRU list */ - wait_queue_head_t *d_wait; /* in-lookup ones only */ + struct swait_queue_head *d_wait; /* in-lookup ones only */ }; struct list_head d_child; /* child of parent list */ struct list_head d_subdirs; /* our children */ @@ -238,7 +238,7 @@ extern void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op extern struct dentry * d_alloc(struct dentry *, const struct qstr *); extern struct dentry * d_alloc_anon(struct super_block *); extern struct dentry * d_alloc_parallel(struct dentry *, const struct qstr *, - wait_queue_head_t *); + struct swait_queue_head *); extern struct dentry * d_splice_alias(struct inode *, struct dentry *); extern struct dentry * d_add_ci(struct dentry *, struct inode *, struct qstr *); extern struct dentry * d_exact_alias(struct dentry *, struct inode *); diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h index e7e45f0cc7da5..5a9e3e3769ce0 100644 --- a/include/linux/debug_locks.h +++ b/include/linux/debug_locks.h @@ -2,9 +2,8 @@ #ifndef __LINUX_DEBUG_LOCKING_H #define __LINUX_DEBUG_LOCKING_H -#include #include -#include +#include struct task_struct; diff --git a/include/linux/delay.h b/include/linux/delay.h index 1d0e2ce6b6d9f..02b37178b54f4 100644 --- a/include/linux/delay.h +++ b/include/linux/delay.h @@ -76,4 +76,10 @@ static inline void fsleep(unsigned long usecs) msleep(DIV_ROUND_UP(usecs, 1000)); } +#ifdef CONFIG_PREEMPT_RT +extern void cpu_chill(void); +#else +# define cpu_chill() cpu_relax() +#endif + #endif /* defined(_LINUX_DELAY_H) */ diff --git a/include/linux/dev_printk.h b/include/linux/dev_printk.h index 3028b644b4fbd..6f009559ee540 100644 --- a/include/linux/dev_printk.h +++ b/include/linux/dev_printk.h @@ -21,6 +21,14 @@ struct device; +#define PRINTK_INFO_SUBSYSTEM_LEN 16 +#define PRINTK_INFO_DEVICE_LEN 48 + +struct dev_printk_info { + char subsystem[PRINTK_INFO_SUBSYSTEM_LEN]; + char device[PRINTK_INFO_DEVICE_LEN]; +}; + #ifdef CONFIG_PRINTK __printf(3, 0) __cold diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h index 159c7476b11b4..7f5b22e7c3542 100644 --- a/include/linux/entry-common.h +++ b/include/linux/entry-common.h @@ -69,7 +69,7 @@ #define EXIT_TO_USER_MODE_WORK \ (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_UPROBE | \ - _TIF_NEED_RESCHED | _TIF_PATCH_PENDING | \ + _TIF_NEED_RESCHED_MASK | _TIF_PATCH_PENDING | \ ARCH_EXIT_TO_USER_MODE_WORK) /** diff --git a/include/linux/fs.h b/include/linux/fs.h index 7519ae003a082..946b42e7c72da 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -704,7 +704,7 @@ struct inode { struct block_device *i_bdev; struct cdev *i_cdev; char *i_link; - unsigned i_dir_seq; + unsigned __i_dir_seq; }; __u32 i_generation; diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index 754f67ac4326a..41408d5e40149 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -8,6 +8,7 @@ #include #include #include +#include extern void synchronize_irq(unsigned int irq); extern bool synchronize_hardirq(unsigned int irq); @@ -115,7 +116,6 @@ extern void rcu_nmi_exit(void); do { \ lockdep_off(); \ arch_nmi_enter(); \ - printk_nmi_enter(); \ BUG_ON(in_nmi() == NMI_MASK); \ __preempt_count_add(NMI_OFFSET + HARDIRQ_OFFSET); \ } while (0) @@ -134,7 +134,6 @@ extern void rcu_nmi_exit(void); do { \ BUG_ON(!in_nmi()); \ __preempt_count_sub(NMI_OFFSET + HARDIRQ_OFFSET); \ - printk_nmi_exit(); \ arch_nmi_exit(); \ lockdep_on(); \ } while (0) diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 14e6202ce47f1..aa41a6b3d4829 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -8,6 +8,7 @@ #include #include #include +#include #include @@ -83,7 +84,7 @@ static inline void kunmap(struct page *page) */ static inline void *kmap_atomic_prot(struct page *page, pgprot_t prot) { - preempt_disable(); + migrate_disable(); pagefault_disable(); if (!PageHighMem(page)) return page_address(page); @@ -153,7 +154,7 @@ static inline void kunmap(struct page *page) static inline void *kmap_atomic(struct page *page) { - preempt_disable(); + migrate_disable(); pagefault_disable(); return page_address(page); } @@ -178,32 +179,51 @@ static inline void kunmap_atomic_high(void *addr) #if defined(CONFIG_HIGHMEM) || defined(CONFIG_X86_32) +#ifndef CONFIG_PREEMPT_RT DECLARE_PER_CPU(int, __kmap_atomic_idx); +#endif static inline int kmap_atomic_idx_push(void) { +#ifndef CONFIG_PREEMPT_RT int idx = __this_cpu_inc_return(__kmap_atomic_idx) - 1; -#ifdef CONFIG_DEBUG_HIGHMEM +# ifdef CONFIG_DEBUG_HIGHMEM WARN_ON_ONCE(in_irq() && !irqs_disabled()); BUG_ON(idx >= KM_TYPE_NR); -#endif +# endif return idx; +#else + current->kmap_idx++; + BUG_ON(current->kmap_idx > KM_TYPE_NR); + return current->kmap_idx - 1; +#endif } static inline int kmap_atomic_idx(void) { +#ifndef CONFIG_PREEMPT_RT return __this_cpu_read(__kmap_atomic_idx) - 1; +#else + return current->kmap_idx - 1; +#endif } static inline void kmap_atomic_idx_pop(void) { -#ifdef CONFIG_DEBUG_HIGHMEM +#ifndef CONFIG_PREEMPT_RT +# ifdef CONFIG_DEBUG_HIGHMEM int idx = __this_cpu_dec_return(__kmap_atomic_idx); BUG_ON(idx < 0); -#else +# else __this_cpu_dec(__kmap_atomic_idx); +# endif +#else + current->kmap_idx--; +# ifdef CONFIG_DEBUG_HIGHMEM + BUG_ON(current->kmap_idx < 0); +# endif #endif } @@ -218,7 +238,7 @@ do { \ BUILD_BUG_ON(__same_type((addr), struct page *)); \ kunmap_atomic_high(addr); \ pagefault_enable(); \ - preempt_enable(); \ + migrate_enable(); \ } while (0) diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index f9aee35384610..a490d9f801aac 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -560,7 +560,7 @@ struct softirq_action asmlinkage void do_softirq(void); asmlinkage void __do_softirq(void); -#ifdef __ARCH_HAS_DO_SOFTIRQ +#if defined(__ARCH_HAS_DO_SOFTIRQ) && !defined(CONFIG_PREEMPT_RT) void do_softirq_own_stack(void); #else static inline void do_softirq_own_stack(void) @@ -654,7 +654,7 @@ enum TASKLET_STATE_RUN /* Tasklet is running (SMP only) */ }; -#ifdef CONFIG_SMP +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT) static inline int tasklet_trylock(struct tasklet_struct *t) { return !test_and_set_bit(TASKLET_STATE_RUN, &(t)->state); @@ -665,15 +665,11 @@ static inline void tasklet_unlock(struct tasklet_struct *t) smp_mb__before_atomic(); clear_bit(TASKLET_STATE_RUN, &(t)->state); } - -static inline void tasklet_unlock_wait(struct tasklet_struct *t) -{ - while (test_bit(TASKLET_STATE_RUN, &(t)->state)) { barrier(); } -} +void tasklet_unlock_wait(struct tasklet_struct *t); #else -#define tasklet_trylock(t) 1 -#define tasklet_unlock_wait(t) do { } while (0) -#define tasklet_unlock(t) do { } while (0) +static inline int tasklet_trylock(struct tasklet_struct *t) { return 1; } +static inline void tasklet_unlock(struct tasklet_struct *t) { } +static inline void tasklet_unlock_wait(struct tasklet_struct *t) { } #endif extern void __tasklet_schedule(struct tasklet_struct *t); diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h index 30823780c1928..f941f2d7d71ce 100644 --- a/include/linux/irq_work.h +++ b/include/linux/irq_work.h @@ -55,4 +55,10 @@ static inline void irq_work_run(void) { } static inline void irq_work_single(void *arg) { } #endif +#if defined(CONFIG_IRQ_WORK) && defined(CONFIG_PREEMPT_RT) +void irq_work_tick_soft(void); +#else +static inline void irq_work_tick_soft(void) { } +#endif + #endif /* _LINUX_IRQ_WORK_H */ diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index 5745491303e03..2b9caf39fb075 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -68,6 +68,7 @@ struct irq_desc { unsigned int irqs_unhandled; atomic_t threads_handled; int threads_handled_last; + u64 random_ip; raw_spinlock_t lock; struct cpumask *percpu_enabled; const struct cpumask *percpu_affinity; diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h index 3ed4e8771b64e..a437b2e70d379 100644 --- a/include/linux/irqflags.h +++ b/include/linux/irqflags.h @@ -71,14 +71,6 @@ do { \ do { \ __this_cpu_dec(hardirq_context); \ } while (0) -# define lockdep_softirq_enter() \ -do { \ - current->softirq_context++; \ -} while (0) -# define lockdep_softirq_exit() \ -do { \ - current->softirq_context--; \ -} while (0) # define lockdep_hrtimer_enter(__hrtimer) \ ({ \ @@ -140,6 +132,21 @@ do { \ # define lockdep_irq_work_exit(__work) do { } while (0) #endif +#if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PREEMPT_RT) +# define lockdep_softirq_enter() \ +do { \ + current->softirq_context++; \ +} while (0) +# define lockdep_softirq_exit() \ +do { \ + current->softirq_context--; \ +} while (0) + +#else +# define lockdep_softirq_enter() do { } while (0) +# define lockdep_softirq_exit() do { } while (0) +#endif + #if defined(CONFIG_IRQSOFF_TRACER) || \ defined(CONFIG_PREEMPT_TRACER) extern void stop_critical_timings(void); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index c25b8e41c0ea3..b3b514a1f12fc 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -218,6 +218,10 @@ extern void __cant_sleep(const char *file, int line, int preempt_offset); */ # define might_sleep() \ do { __might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0) + +# define might_sleep_no_state_check() \ + do { ___might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0) + /** * cant_sleep - annotation for functions that cannot sleep * @@ -249,6 +253,7 @@ extern void __cant_sleep(const char *file, int line, int preempt_offset); static inline void __might_sleep(const char *file, int line, int preempt_offset) { } # define might_sleep() do { might_resched(); } while (0) +# define might_sleep_no_state_check() do { might_resched(); } while (0) # define cant_sleep() do { } while (0) # define sched_annotate_sleep() do { } while (0) # define non_block_start() do { } while (0) diff --git a/include/linux/local_lock_internal.h b/include/linux/local_lock_internal.h index 4a8795b21d774..998d2c34cf0c2 100644 --- a/include/linux/local_lock_internal.h +++ b/include/linux/local_lock_internal.h @@ -7,33 +7,90 @@ #include typedef struct { -#ifdef CONFIG_DEBUG_LOCK_ALLOC +#ifdef CONFIG_PREEMPT_RT + spinlock_t lock; + struct task_struct *owner; + int nestcnt; + +#elif defined(CONFIG_DEBUG_LOCK_ALLOC) struct lockdep_map dep_map; struct task_struct *owner; #endif } local_lock_t; -#ifdef CONFIG_DEBUG_LOCK_ALLOC -# define LL_DEP_MAP_INIT(lockname) \ +#ifdef CONFIG_PREEMPT_RT + +#define INIT_LOCAL_LOCK(lockname) { \ + __SPIN_LOCK_UNLOCKED((lockname).lock), \ + .owner = NULL, \ + .nestcnt = 0, \ + } +#else + +# ifdef CONFIG_DEBUG_LOCK_ALLOC +# define LL_DEP_MAP_INIT(lockname) \ .dep_map = { \ .name = #lockname, \ .wait_type_inner = LD_WAIT_CONFIG, \ } -#else -# define LL_DEP_MAP_INIT(lockname) -#endif +# else +# define LL_DEP_MAP_INIT(lockname) +# endif #define INIT_LOCAL_LOCK(lockname) { LL_DEP_MAP_INIT(lockname) } -#define __local_lock_init(lock) \ +#endif + +#ifdef CONFIG_PREEMPT_RT + +static inline void ___local_lock_init(local_lock_t *l) +{ + l->owner = NULL; + l->nestcnt = 0; +} + +#define __local_lock_init(l) \ +do { \ + spin_lock_init(&(l)->lock); \ + ___local_lock_init(l); \ +} while (0) + +#else + +#define __local_lock_init(l) \ do { \ static struct lock_class_key __key; \ \ - debug_check_no_locks_freed((void *)lock, sizeof(*lock));\ - lockdep_init_map_wait(&(lock)->dep_map, #lock, &__key, 0, LD_WAIT_CONFIG);\ + debug_check_no_locks_freed((void *)l, sizeof(*l)); \ + lockdep_init_map_wait(&(l)->dep_map, #l, &__key, 0, LD_WAIT_CONFIG);\ } while (0) +#endif -#ifdef CONFIG_DEBUG_LOCK_ALLOC +#ifdef CONFIG_PREEMPT_RT + +static inline void local_lock_acquire(local_lock_t *l) +{ + if (l->owner != current) { + spin_lock(&l->lock); + DEBUG_LOCKS_WARN_ON(l->owner); + DEBUG_LOCKS_WARN_ON(l->nestcnt); + l->owner = current; + } + l->nestcnt++; +} + +static inline void local_lock_release(local_lock_t *l) +{ + DEBUG_LOCKS_WARN_ON(l->nestcnt == 0); + DEBUG_LOCKS_WARN_ON(l->owner != current); + if (--l->nestcnt) + return; + + l->owner = NULL; + spin_unlock(&l->lock); +} + +#elif defined(CONFIG_DEBUG_LOCK_ALLOC) static inline void local_lock_acquire(local_lock_t *l) { lock_map_acquire(&l->dep_map); @@ -55,10 +112,45 @@ static inline void local_lock_release(local_lock_t *l) { } #define __local_lock(lock) \ do { \ - preempt_disable(); \ + migrate_disable(); \ local_lock_acquire(this_cpu_ptr(lock)); \ } while (0) +#define __local_unlock(lock) \ + do { \ + local_lock_release(this_cpu_ptr(lock)); \ + migrate_enable(); \ + } while (0) + +#ifdef CONFIG_PREEMPT_RT + +#define __local_lock_irq(lock) \ + do { \ + migrate_disable(); \ + local_lock_acquire(this_cpu_ptr(lock)); \ + } while (0) + +#define __local_lock_irqsave(lock, flags) \ + do { \ + migrate_disable(); \ + flags = 0; \ + local_lock_acquire(this_cpu_ptr(lock)); \ + } while (0) + +#define __local_unlock_irq(lock) \ + do { \ + local_lock_release(this_cpu_ptr(lock)); \ + migrate_enable(); \ + } while (0) + +#define __local_unlock_irqrestore(lock, flags) \ + do { \ + local_lock_release(this_cpu_ptr(lock)); \ + migrate_enable(); \ + } while (0) + +#else + #define __local_lock_irq(lock) \ do { \ local_irq_disable(); \ @@ -71,12 +163,6 @@ static inline void local_lock_release(local_lock_t *l) { } local_lock_acquire(this_cpu_ptr(lock)); \ } while (0) -#define __local_unlock(lock) \ - do { \ - local_lock_release(this_cpu_ptr(lock)); \ - preempt_enable(); \ - } while (0) - #define __local_unlock_irq(lock) \ do { \ local_lock_release(this_cpu_ptr(lock)); \ @@ -88,3 +174,5 @@ static inline void local_lock_release(local_lock_t *l) { } local_lock_release(this_cpu_ptr(lock)); \ local_irq_restore(flags); \ } while (0) + +#endif diff --git a/include/linux/mhi.h b/include/linux/mhi.h index c4a940d989121..2b4ed30d47294 100644 --- a/include/linux/mhi.h +++ b/include/linux/mhi.h @@ -9,10 +9,9 @@ #include #include #include -#include #include #include -#include +#include #include #include diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index ed028af3cb19b..cbdc39fea2ff0 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -548,6 +549,9 @@ struct mm_struct { bool tlb_flush_batched; #endif struct uprobes_state uprobes_state; +#ifdef CONFIG_PREEMPT_RT + struct rcu_head delayed_drop; +#endif #ifdef CONFIG_HUGETLB_PAGE atomic_long_t hugetlb_usage; #endif diff --git a/include/linux/mutex.h b/include/linux/mutex.h index dcd185cbfe793..90f090efcb58d 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -22,6 +22,20 @@ struct ww_acquire_ctx; +#ifdef CONFIG_DEBUG_LOCK_ALLOC +# define __DEP_MAP_MUTEX_INITIALIZER(lockname) \ + , .dep_map = { \ + .name = #lockname, \ + .wait_type_inner = LD_WAIT_SLEEP, \ + } +#else +# define __DEP_MAP_MUTEX_INITIALIZER(lockname) +#endif + +#ifdef CONFIG_PREEMPT_RT +# include +#else + /* * Simple, straightforward mutexes with strict semantics: * @@ -68,14 +82,6 @@ struct mutex { struct ww_class; struct ww_acquire_ctx; -struct ww_mutex { - struct mutex base; - struct ww_acquire_ctx *ctx; -#ifdef CONFIG_DEBUG_MUTEXES - struct ww_class *ww_class; -#endif -}; - /* * This is the control structure for tasks blocked on mutex, * which resides on the blocked task's kernel stack: @@ -119,16 +125,6 @@ do { \ __mutex_init((mutex), #mutex, &__key); \ } while (0) -#ifdef CONFIG_DEBUG_LOCK_ALLOC -# define __DEP_MAP_MUTEX_INITIALIZER(lockname) \ - , .dep_map = { \ - .name = #lockname, \ - .wait_type_inner = LD_WAIT_SLEEP, \ - } -#else -# define __DEP_MAP_MUTEX_INITIALIZER(lockname) -#endif - #define __MUTEX_INITIALIZER(lockname) \ { .owner = ATOMIC_LONG_INIT(0) \ , .wait_lock = __SPIN_LOCK_UNLOCKED(lockname.wait_lock) \ @@ -224,4 +220,6 @@ enum mutex_trylock_recursive_enum { extern /* __deprecated */ __must_check enum mutex_trylock_recursive_enum mutex_trylock_recursive(struct mutex *lock); +#endif /* !PREEMPT_RT */ + #endif /* __LINUX_MUTEX_H */ diff --git a/include/linux/mutex_rt.h b/include/linux/mutex_rt.h new file mode 100644 index 0000000000000..f0b2e07cd5c57 --- /dev/null +++ b/include/linux/mutex_rt.h @@ -0,0 +1,130 @@ +// SPDX-License-Identifier: GPL-2.0-only +#ifndef __LINUX_MUTEX_RT_H +#define __LINUX_MUTEX_RT_H + +#ifndef __LINUX_MUTEX_H +#error "Please include mutex.h" +#endif + +#include + +/* FIXME: Just for __lockfunc */ +#include + +struct mutex { + struct rt_mutex lock; +#ifdef CONFIG_DEBUG_LOCK_ALLOC + struct lockdep_map dep_map; +#endif +}; + +#define __MUTEX_INITIALIZER(mutexname) \ + { \ + .lock = __RT_MUTEX_INITIALIZER(mutexname.lock) \ + __DEP_MAP_MUTEX_INITIALIZER(mutexname) \ + } + +#define DEFINE_MUTEX(mutexname) \ + struct mutex mutexname = __MUTEX_INITIALIZER(mutexname) + +extern void __mutex_do_init(struct mutex *lock, const char *name, struct lock_class_key *key); +extern void __lockfunc _mutex_lock(struct mutex *lock); +extern void __lockfunc _mutex_lock_io_nested(struct mutex *lock, int subclass); +extern int __lockfunc _mutex_lock_interruptible(struct mutex *lock); +extern int __lockfunc _mutex_lock_killable(struct mutex *lock); +extern void __lockfunc _mutex_lock_nested(struct mutex *lock, int subclass); +extern void __lockfunc _mutex_lock_nest_lock(struct mutex *lock, struct lockdep_map *nest_lock); +extern int __lockfunc _mutex_lock_interruptible_nested(struct mutex *lock, int subclass); +extern int __lockfunc _mutex_lock_killable_nested(struct mutex *lock, int subclass); +extern int __lockfunc _mutex_trylock(struct mutex *lock); +extern void __lockfunc _mutex_unlock(struct mutex *lock); + +#define mutex_is_locked(l) rt_mutex_is_locked(&(l)->lock) +#define mutex_lock(l) _mutex_lock(l) +#define mutex_lock_interruptible(l) _mutex_lock_interruptible(l) +#define mutex_lock_killable(l) _mutex_lock_killable(l) +#define mutex_trylock(l) _mutex_trylock(l) +#define mutex_unlock(l) _mutex_unlock(l) +#define mutex_lock_io(l) _mutex_lock_io_nested(l, 0); + +#define __mutex_owner(l) ((l)->lock.owner) + +#ifdef CONFIG_DEBUG_MUTEXES +#define mutex_destroy(l) rt_mutex_destroy(&(l)->lock) +#else +static inline void mutex_destroy(struct mutex *lock) {} +#endif + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +# define mutex_lock_nested(l, s) _mutex_lock_nested(l, s) +# define mutex_lock_interruptible_nested(l, s) \ + _mutex_lock_interruptible_nested(l, s) +# define mutex_lock_killable_nested(l, s) \ + _mutex_lock_killable_nested(l, s) +# define mutex_lock_io_nested(l, s) _mutex_lock_io_nested(l, s) + +# define mutex_lock_nest_lock(lock, nest_lock) \ +do { \ + typecheck(struct lockdep_map *, &(nest_lock)->dep_map); \ + _mutex_lock_nest_lock(lock, &(nest_lock)->dep_map); \ +} while (0) + +#else +# define mutex_lock_nested(l, s) _mutex_lock(l) +# define mutex_lock_interruptible_nested(l, s) \ + _mutex_lock_interruptible(l) +# define mutex_lock_killable_nested(l, s) \ + _mutex_lock_killable(l) +# define mutex_lock_nest_lock(lock, nest_lock) mutex_lock(lock) +# define mutex_lock_io_nested(l, s) _mutex_lock_io_nested(l, s) +#endif + +# define mutex_init(mutex) \ +do { \ + static struct lock_class_key __key; \ + \ + rt_mutex_init(&(mutex)->lock); \ + __mutex_do_init((mutex), #mutex, &__key); \ +} while (0) + +# define __mutex_init(mutex, name, key) \ +do { \ + rt_mutex_init(&(mutex)->lock); \ + __mutex_do_init((mutex), name, key); \ +} while (0) + +/** + * These values are chosen such that FAIL and SUCCESS match the + * values of the regular mutex_trylock(). + */ +enum mutex_trylock_recursive_enum { + MUTEX_TRYLOCK_FAILED = 0, + MUTEX_TRYLOCK_SUCCESS = 1, + MUTEX_TRYLOCK_RECURSIVE, +}; +/** + * mutex_trylock_recursive - trylock variant that allows recursive locking + * @lock: mutex to be locked + * + * This function should not be used, _ever_. It is purely for hysterical GEM + * raisins, and once those are gone this will be removed. + * + * Returns: + * MUTEX_TRYLOCK_FAILED - trylock failed, + * MUTEX_TRYLOCK_SUCCESS - lock acquired, + * MUTEX_TRYLOCK_RECURSIVE - we already owned the lock. + */ +int __rt_mutex_owner_current(struct rt_mutex *lock); + +static inline /* __deprecated */ __must_check enum mutex_trylock_recursive_enum +mutex_trylock_recursive(struct mutex *lock) +{ + if (unlikely(__rt_mutex_owner_current(&lock->lock))) + return MUTEX_TRYLOCK_RECURSIVE; + + return mutex_trylock(lock); +} + +extern int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock); + +#endif diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 69cb46f7b8d2c..fc682adab66a5 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1670,7 +1670,7 @@ struct nfs_unlinkdata { struct nfs_removeargs args; struct nfs_removeres res; struct dentry *dentry; - wait_queue_head_t wq; + struct swait_queue_head wq; const struct cred *cred; struct nfs_fattr dir_attr; long timeout; diff --git a/include/linux/pid.h b/include/linux/pid.h index 176d6cf80e7c1..4daecc34c097f 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -3,6 +3,7 @@ #define _LINUX_PID_H #include +#include #include #include diff --git a/include/linux/preempt.h b/include/linux/preempt.h index 7d9c1c0e149c0..8a47b9b1bade1 100644 --- a/include/linux/preempt.h +++ b/include/linux/preempt.h @@ -77,10 +77,14 @@ /* preempt_count() and related functions, depends on PREEMPT_NEED_RESCHED */ #include -#define hardirq_count() (preempt_count() & HARDIRQ_MASK) -#define softirq_count() (preempt_count() & SOFTIRQ_MASK) -#define irq_count() (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_MASK \ - | NMI_MASK)) +#define pc_nmi_count() (preempt_count() & NMI_MASK) +#define hardirq_count() (preempt_count() & HARDIRQ_MASK) +#ifdef CONFIG_PREEMPT_RT +# define softirq_count() (current->softirq_disable_cnt & SOFTIRQ_MASK) +#else +# define softirq_count() (preempt_count() & SOFTIRQ_MASK) +#endif +#define irq_count() (pc_nmi_count() | hardirq_count() | softirq_count()) /* * Are we doing bottom half or hardware interrupt processing? @@ -95,13 +99,12 @@ * Note: due to the BH disabled confusion: in_softirq(),in_interrupt() really * should not be used in new code. */ +#define in_nmi() (pc_nmi_count()) #define in_irq() (hardirq_count()) -#define in_softirq() (softirq_count()) #define in_interrupt() (irq_count()) +#define in_softirq() (softirq_count()) #define in_serving_softirq() (softirq_count() & SOFTIRQ_OFFSET) -#define in_nmi() (preempt_count() & NMI_MASK) -#define in_task() (!(preempt_count() & \ - (NMI_MASK | HARDIRQ_MASK | SOFTIRQ_OFFSET))) +#define in_task() (!(irq_count() & (NMI_MASK | HARDIRQ_MASK | SOFTIRQ_OFFSET))) /* * The preempt_count offset after preempt_disable(); @@ -115,7 +118,11 @@ /* * The preempt_count offset after spin_lock() */ +#if !defined(CONFIG_PREEMPT_RT) #define PREEMPT_LOCK_OFFSET PREEMPT_DISABLE_OFFSET +#else +#define PREEMPT_LOCK_OFFSET 0 +#endif /* * The preempt_count offset needed for things like: @@ -164,6 +171,20 @@ extern void preempt_count_sub(int val); #define preempt_count_inc() preempt_count_add(1) #define preempt_count_dec() preempt_count_sub(1) +#ifdef CONFIG_PREEMPT_LAZY +#define add_preempt_lazy_count(val) do { preempt_lazy_count() += (val); } while (0) +#define sub_preempt_lazy_count(val) do { preempt_lazy_count() -= (val); } while (0) +#define inc_preempt_lazy_count() add_preempt_lazy_count(1) +#define dec_preempt_lazy_count() sub_preempt_lazy_count(1) +#define preempt_lazy_count() (current_thread_info()->preempt_lazy_count) +#else +#define add_preempt_lazy_count(val) do { } while (0) +#define sub_preempt_lazy_count(val) do { } while (0) +#define inc_preempt_lazy_count() do { } while (0) +#define dec_preempt_lazy_count() do { } while (0) +#define preempt_lazy_count() (0) +#endif + #ifdef CONFIG_PREEMPT_COUNT #define preempt_disable() \ @@ -172,13 +193,25 @@ do { \ barrier(); \ } while (0) +#define preempt_lazy_disable() \ +do { \ + inc_preempt_lazy_count(); \ + barrier(); \ +} while (0) + #define sched_preempt_enable_no_resched() \ do { \ barrier(); \ preempt_count_dec(); \ } while (0) -#define preempt_enable_no_resched() sched_preempt_enable_no_resched() +#ifdef CONFIG_PREEMPT_RT +# define preempt_enable_no_resched() sched_preempt_enable_no_resched() +# define preempt_check_resched_rt() preempt_check_resched() +#else +# define preempt_enable_no_resched() preempt_enable() +# define preempt_check_resched_rt() barrier(); +#endif #define preemptible() (preempt_count() == 0 && !irqs_disabled()) @@ -203,6 +236,18 @@ do { \ __preempt_schedule(); \ } while (0) +/* + * open code preempt_check_resched() because it is not exported to modules and + * used by local_unlock() or bpf_enable_instrumentation(). + */ +#define preempt_lazy_enable() \ +do { \ + dec_preempt_lazy_count(); \ + barrier(); \ + if (should_resched(0)) \ + __preempt_schedule(); \ +} while (0) + #else /* !CONFIG_PREEMPTION */ #define preempt_enable() \ do { \ @@ -210,6 +255,12 @@ do { \ preempt_count_dec(); \ } while (0) +#define preempt_lazy_enable() \ +do { \ + dec_preempt_lazy_count(); \ + barrier(); \ +} while (0) + #define preempt_enable_notrace() \ do { \ barrier(); \ @@ -248,6 +299,7 @@ do { \ #define preempt_disable_notrace() barrier() #define preempt_enable_no_resched_notrace() barrier() #define preempt_enable_notrace() barrier() +#define preempt_check_resched_rt() barrier() #define preemptible() 0 #endif /* CONFIG_PREEMPT_COUNT */ @@ -268,10 +320,22 @@ do { \ } while (0) #define preempt_fold_need_resched() \ do { \ - if (tif_need_resched()) \ + if (tif_need_resched_now()) \ set_preempt_need_resched(); \ } while (0) +#ifdef CONFIG_PREEMPT_RT +# define preempt_disable_rt() preempt_disable() +# define preempt_enable_rt() preempt_enable() +# define preempt_disable_nort() barrier() +# define preempt_enable_nort() barrier() +#else +# define preempt_disable_rt() barrier() +# define preempt_enable_rt() barrier() +# define preempt_disable_nort() preempt_disable() +# define preempt_enable_nort() preempt_enable() +#endif + #ifdef CONFIG_PREEMPT_NOTIFIERS struct preempt_notifier; @@ -322,6 +386,80 @@ static inline void preempt_notifier_init(struct preempt_notifier *notifier, #endif +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + +/* + * Migrate-Disable and why it is undesired. + * + * When a preempted task becomes elegible to run under the ideal model (IOW it + * becomes one of the M highest priority tasks), it might still have to wait + * for the preemptee's migrate_disable() section to complete. Thereby suffering + * a reduction in bandwidth in the exact duration of the migrate_disable() + * section. + * + * Per this argument, the change from preempt_disable() to migrate_disable() + * gets us: + * + * - a higher priority tasks gains reduced wake-up latency; with preempt_disable() + * it would have had to wait for the lower priority task. + * + * - a lower priority tasks; which under preempt_disable() could've instantly + * migrated away when another CPU becomes available, is now constrained + * by the ability to push the higher priority task away, which might itself be + * in a migrate_disable() section, reducing it's available bandwidth. + * + * IOW it trades latency / moves the interference term, but it stays in the + * system, and as long as it remains unbounded, the system is not fully + * deterministic. + * + * + * The reason we have it anyway. + * + * PREEMPT_RT breaks a number of assumptions traditionally held. By forcing a + * number of primitives into becoming preemptible, they would also allow + * migration. This turns out to break a bunch of per-cpu usage. To this end, + * all these primitives employ migirate_disable() to restore this implicit + * assumption. + * + * This is a 'temporary' work-around at best. The correct solution is getting + * rid of the above assumptions and reworking the code to employ explicit + * per-cpu locking or short preempt-disable regions. + * + * The end goal must be to get rid of migrate_disable(), alternatively we need + * a schedulability theory that does not depend on abritrary migration. + * + * + * Notes on the implementation. + * + * The implementation is particularly tricky since existing code patterns + * dictate neither migrate_disable() nor migrate_enable() is allowed to block. + * This means that it cannot use cpus_read_lock() to serialize against hotplug, + * nor can it easily migrate itself into a pending affinity mask change on + * migrate_enable(). + * + * + * Note: even non-work-conserving schedulers like semi-partitioned depends on + * migration, so migrate_disable() is not only a problem for + * work-conserving schedulers. + * + */ +extern void migrate_disable(void); +extern void migrate_enable(void); + +#elif defined(CONFIG_PREEMPT_RT) + +static inline void migrate_disable(void) +{ + preempt_lazy_disable(); +} + +static inline void migrate_enable(void) +{ + preempt_lazy_enable(); +} + +#else /* !CONFIG_PREEMPT_RT */ + /** * migrate_disable - Prevent migration of the current task * @@ -352,4 +490,6 @@ static __always_inline void migrate_enable(void) preempt_enable(); } +#endif /* CONFIG_SMP && CONFIG_PREEMPT_RT */ + #endif /* __LINUX_PREEMPT_H */ diff --git a/include/linux/printk.h b/include/linux/printk.h index 34c1a7be3e014..c49d5bb3f8ffa 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -147,22 +147,12 @@ static inline __printf(1, 2) __cold void early_printk(const char *s, ...) { } #endif -#ifdef CONFIG_PRINTK_NMI -extern void printk_nmi_enter(void); -extern void printk_nmi_exit(void); -extern void printk_nmi_direct_enter(void); -extern void printk_nmi_direct_exit(void); -#else -static inline void printk_nmi_enter(void) { } -static inline void printk_nmi_exit(void) { } -static inline void printk_nmi_direct_enter(void) { } -static inline void printk_nmi_direct_exit(void) { } -#endif /* PRINTK_NMI */ +struct dev_printk_info; #ifdef CONFIG_PRINTK -asmlinkage __printf(5, 0) +asmlinkage __printf(4, 0) int vprintk_emit(int facility, int level, - const char *dict, size_t dictlen, + const struct dev_printk_info *dev_info, const char *fmt, va_list args); asmlinkage __printf(1, 0) @@ -203,8 +193,6 @@ __printf(1, 2) void dump_stack_set_arch_desc(const char *fmt, ...); void dump_stack_print_info(const char *log_lvl); void show_regs_print_info(const char *log_lvl); extern asmlinkage void dump_stack(void) __cold; -extern void printk_safe_flush(void); -extern void printk_safe_flush_on_panic(void); #else static inline __printf(1, 0) int vprintk(const char *s, va_list args) @@ -268,14 +256,6 @@ static inline void show_regs_print_info(const char *log_lvl) static inline void dump_stack(void) { } - -static inline void printk_safe_flush(void) -{ -} - -static inline void printk_safe_flush_on_panic(void) -{ -} #endif extern int kptr_restrict; diff --git a/include/linux/random.h b/include/linux/random.h index f45b8be3e3c4e..0e41d05278091 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -35,7 +35,7 @@ static inline void add_latent_entropy(void) {} extern void add_input_randomness(unsigned int type, unsigned int code, unsigned int value) __latent_entropy; -extern void add_interrupt_randomness(int irq, int irq_flags) __latent_entropy; +extern void add_interrupt_randomness(int irq, int irq_flags, __u64 ip) __latent_entropy; extern void get_random_bytes(void *buf, int nbytes); extern int wait_for_random_bytes(void); diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h index d7db179963221..c33b0e16d04b4 100644 --- a/include/linux/rbtree.h +++ b/include/linux/rbtree.h @@ -19,19 +19,9 @@ #include #include +#include #include -struct rb_node { - unsigned long __rb_parent_color; - struct rb_node *rb_right; - struct rb_node *rb_left; -} __attribute__((aligned(sizeof(long)))); - /* The alignment might seem pointless, but allegedly CRIS needs it */ - -struct rb_root { - struct rb_node *rb_node; -}; - #define rb_parent(r) ((struct rb_node *)((r)->__rb_parent_color & ~3)) #define RB_ROOT (struct rb_root) { NULL, } @@ -112,21 +102,6 @@ static inline void rb_link_node_rcu(struct rb_node *node, struct rb_node *parent typeof(*pos), field); 1; }); \ pos = n) -/* - * Leftmost-cached rbtrees. - * - * We do not cache the rightmost node based on footprint - * size vs number of potential users that could benefit - * from O(1) rb_last(). Just not worth it, users that want - * this feature can always implement the logic explicitly. - * Furthermore, users that want to cache both pointers may - * find it a bit asymmetric, but that's ok. - */ -struct rb_root_cached { - struct rb_root rb_root; - struct rb_node *rb_leftmost; -}; - #define RB_ROOT_CACHED (struct rb_root_cached) { {NULL, }, NULL } /* Same as rb_first(), but O(1) */ diff --git a/include/linux/rbtree_latch.h b/include/linux/rbtree_latch.h index 7d012faa509a4..3d1a9e716b803 100644 --- a/include/linux/rbtree_latch.h +++ b/include/linux/rbtree_latch.h @@ -42,8 +42,8 @@ struct latch_tree_node { }; struct latch_tree_root { - seqcount_t seq; - struct rb_root tree[2]; + seqcount_latch_t seq; + struct rb_root tree[2]; }; /** @@ -206,7 +206,7 @@ latch_tree_find(void *key, struct latch_tree_root *root, do { seq = raw_read_seqcount_latch(&root->seq); node = __lt_find(key, root, seq & 1, ops->comp); - } while (read_seqcount_retry(&root->seq, seq)); + } while (read_seqcount_latch_retry(&root->seq, seq)); return node; } diff --git a/include/linux/rbtree_type.h b/include/linux/rbtree_type.h new file mode 100644 index 0000000000000..77a89dd2c7c66 --- /dev/null +++ b/include/linux/rbtree_type.h @@ -0,0 +1,31 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +#ifndef _LINUX_RBTREE_TYPE_H +#define _LINUX_RBTREE_TYPE_H + +struct rb_node { + unsigned long __rb_parent_color; + struct rb_node *rb_right; + struct rb_node *rb_left; +} __attribute__((aligned(sizeof(long)))); +/* The alignment might seem pointless, but allegedly CRIS needs it */ + +struct rb_root { + struct rb_node *rb_node; +}; + +/* + * Leftmost-cached rbtrees. + * + * We do not cache the rightmost node based on footprint + * size vs number of potential users that could benefit + * from O(1) rb_last(). Just not worth it, users that want + * this feature can always implement the logic explicitly. + * Furthermore, users that want to cache both pointers may + * find it a bit asymmetric, but that's ok. + */ +struct rb_root_cached { + struct rb_root rb_root; + struct rb_node *rb_leftmost; +}; + +#endif diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index d15d46db61f75..76d19f339419c 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -52,6 +52,11 @@ void __rcu_read_unlock(void); * types of kernel builds, the rcu_read_lock() nesting depth is unknowable. */ #define rcu_preempt_depth() (current->rcu_read_lock_nesting) +#ifndef CONFIG_PREEMPT_RT +#define sched_rcu_preempt_depth() rcu_preempt_depth() +#else +static inline int sched_rcu_preempt_depth(void) { return 0; } +#endif #else /* #ifdef CONFIG_PREEMPT_RCU */ @@ -70,6 +75,8 @@ static inline int rcu_preempt_depth(void) return 0; } +#define sched_rcu_preempt_depth() rcu_preempt_depth() + #endif /* #else #ifdef CONFIG_PREEMPT_RCU */ /* Internal to kernel */ @@ -312,7 +319,8 @@ static inline void rcu_preempt_sleep_check(void) { } #define rcu_sleep_check() \ do { \ rcu_preempt_sleep_check(); \ - RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map), \ + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) \ + RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map), \ "Illegal context switch in RCU-bh read-side critical section"); \ RCU_LOCKDEP_WARN(lock_is_held(&rcu_sched_lock_map), \ "Illegal context switch in RCU-sched read-side critical section"); \ diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h index 6fd615a0eea94..b02009f530263 100644 --- a/include/linux/rtmutex.h +++ b/include/linux/rtmutex.h @@ -14,11 +14,15 @@ #define __LINUX_RT_MUTEX_H #include -#include -#include +#include +#include extern int max_lock_depth; /* for sysctl */ +#ifdef CONFIG_DEBUG_MUTEXES +#include +#endif + /** * The rt_mutex structure * @@ -31,12 +35,7 @@ struct rt_mutex { raw_spinlock_t wait_lock; struct rb_root_cached waiters; struct task_struct *owner; -#ifdef CONFIG_DEBUG_RT_MUTEXES int save_state; - const char *name, *file; - int line; - void *magic; -#endif #ifdef CONFIG_DEBUG_LOCK_ALLOC struct lockdep_map dep_map; #endif @@ -49,6 +48,7 @@ struct hrtimer_sleeper; extern int rt_mutex_debug_check_no_locks_freed(const void *from, unsigned long len); extern void rt_mutex_debug_check_no_locks_held(struct task_struct *task); + extern void rt_mutex_debug_task_free(struct task_struct *tsk); #else static inline int rt_mutex_debug_check_no_locks_freed(const void *from, unsigned long len) @@ -56,25 +56,15 @@ struct hrtimer_sleeper; return 0; } # define rt_mutex_debug_check_no_locks_held(task) do { } while (0) +# define rt_mutex_debug_task_free(t) do { } while (0) #endif -#ifdef CONFIG_DEBUG_RT_MUTEXES -# define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) \ - , .name = #mutexname, .file = __FILE__, .line = __LINE__ - -# define rt_mutex_init(mutex) \ +#define rt_mutex_init(mutex) \ do { \ static struct lock_class_key __key; \ __rt_mutex_init(mutex, __func__, &__key); \ } while (0) - extern void rt_mutex_debug_task_free(struct task_struct *tsk); -#else -# define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) -# define rt_mutex_init(mutex) __rt_mutex_init(mutex, NULL, NULL) -# define rt_mutex_debug_task_free(t) do { } while (0) -#endif - #ifdef CONFIG_DEBUG_LOCK_ALLOC #define __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname) \ , .dep_map = { .name = #mutexname } @@ -82,12 +72,19 @@ do { \ #define __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname) #endif -#define __RT_MUTEX_INITIALIZER(mutexname) \ - { .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \ +#define __RT_MUTEX_INITIALIZER_PLAIN(mutexname) \ + .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \ , .waiters = RB_ROOT_CACHED \ , .owner = NULL \ - __DEBUG_RT_MUTEX_INITIALIZER(mutexname) \ - __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname)} + __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname) + +#define __RT_MUTEX_INITIALIZER(mutexname) \ + { __RT_MUTEX_INITIALIZER_PLAIN(mutexname) \ + , .save_state = 0 } + +#define __RT_MUTEX_INITIALIZER_SAVE_STATE(mutexname) \ + { __RT_MUTEX_INITIALIZER_PLAIN(mutexname) \ + , .save_state = 1 } #define DEFINE_RT_MUTEX(mutexname) \ struct rt_mutex mutexname = __RT_MUTEX_INITIALIZER(mutexname) @@ -115,9 +112,6 @@ extern void rt_mutex_lock(struct rt_mutex *lock); #endif extern int rt_mutex_lock_interruptible(struct rt_mutex *lock); -extern int rt_mutex_timed_lock(struct rt_mutex *lock, - struct hrtimer_sleeper *timeout); - extern int rt_mutex_trylock(struct rt_mutex *lock); extern void rt_mutex_unlock(struct rt_mutex *lock); diff --git a/include/linux/rwlock_rt.h b/include/linux/rwlock_rt.h new file mode 100644 index 0000000000000..aafdb0a685d54 --- /dev/null +++ b/include/linux/rwlock_rt.h @@ -0,0 +1,109 @@ +// SPDX-License-Identifier: GPL-2.0-only +#ifndef __LINUX_RWLOCK_RT_H +#define __LINUX_RWLOCK_RT_H + +#ifndef __LINUX_SPINLOCK_H +#error Do not include directly. Use spinlock.h +#endif + +extern void __lockfunc rt_write_lock(rwlock_t *rwlock); +extern void __lockfunc rt_read_lock(rwlock_t *rwlock); +extern int __lockfunc rt_write_trylock(rwlock_t *rwlock); +extern int __lockfunc rt_read_trylock(rwlock_t *rwlock); +extern void __lockfunc rt_write_unlock(rwlock_t *rwlock); +extern void __lockfunc rt_read_unlock(rwlock_t *rwlock); +extern int __lockfunc rt_read_can_lock(rwlock_t *rwlock); +extern int __lockfunc rt_write_can_lock(rwlock_t *rwlock); +extern void __rt_rwlock_init(rwlock_t *rwlock, char *name, struct lock_class_key *key); + +#define read_can_lock(rwlock) rt_read_can_lock(rwlock) +#define write_can_lock(rwlock) rt_write_can_lock(rwlock) + +#define read_trylock(lock) __cond_lock(lock, rt_read_trylock(lock)) +#define write_trylock(lock) __cond_lock(lock, rt_write_trylock(lock)) + +static inline int __write_trylock_rt_irqsave(rwlock_t *lock, unsigned long *flags) +{ + *flags = 0; + return rt_write_trylock(lock); +} + +#define write_trylock_irqsave(lock, flags) \ + __cond_lock(lock, __write_trylock_rt_irqsave(lock, &(flags))) + +#define read_lock_irqsave(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + rt_read_lock(lock); \ + flags = 0; \ + } while (0) + +#define write_lock_irqsave(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + rt_write_lock(lock); \ + flags = 0; \ + } while (0) + +#define read_lock(lock) rt_read_lock(lock) + +#define read_lock_bh(lock) \ + do { \ + local_bh_disable(); \ + rt_read_lock(lock); \ + } while (0) + +#define read_lock_irq(lock) read_lock(lock) + +#define write_lock(lock) rt_write_lock(lock) + +#define write_lock_bh(lock) \ + do { \ + local_bh_disable(); \ + rt_write_lock(lock); \ + } while (0) + +#define write_lock_irq(lock) write_lock(lock) + +#define read_unlock(lock) rt_read_unlock(lock) + +#define read_unlock_bh(lock) \ + do { \ + rt_read_unlock(lock); \ + local_bh_enable(); \ + } while (0) + +#define read_unlock_irq(lock) read_unlock(lock) + +#define write_unlock(lock) rt_write_unlock(lock) + +#define write_unlock_bh(lock) \ + do { \ + rt_write_unlock(lock); \ + local_bh_enable(); \ + } while (0) + +#define write_unlock_irq(lock) write_unlock(lock) + +#define read_unlock_irqrestore(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + (void) flags; \ + rt_read_unlock(lock); \ + } while (0) + +#define write_unlock_irqrestore(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + (void) flags; \ + rt_write_unlock(lock); \ + } while (0) + +#define rwlock_init(rwl) \ +do { \ + static struct lock_class_key __key; \ + \ + __rt_rwlock_init(rwl, #rwl, &__key); \ +} while (0) + +#endif diff --git a/include/linux/rwlock_types.h b/include/linux/rwlock_types.h index 3bd03e18061c1..0ad226b5d8fd5 100644 --- a/include/linux/rwlock_types.h +++ b/include/linux/rwlock_types.h @@ -1,6 +1,10 @@ #ifndef __LINUX_RWLOCK_TYPES_H #define __LINUX_RWLOCK_TYPES_H +#if !defined(__LINUX_SPINLOCK_TYPES_H) +# error "Do not include directly, include spinlock_types.h" +#endif + /* * include/linux/rwlock_types.h - generic rwlock type definitions * and initializers diff --git a/include/linux/rwlock_types_rt.h b/include/linux/rwlock_types_rt.h new file mode 100644 index 0000000000000..4762391d659bf --- /dev/null +++ b/include/linux/rwlock_types_rt.h @@ -0,0 +1,56 @@ +// SPDX-License-Identifier: GPL-2.0-only +#ifndef __LINUX_RWLOCK_TYPES_RT_H +#define __LINUX_RWLOCK_TYPES_RT_H + +#ifndef __LINUX_SPINLOCK_TYPES_H +#error "Do not include directly. Include spinlock_types.h instead" +#endif + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +# define RW_DEP_MAP_INIT(lockname) .dep_map = { .name = #lockname } +#else +# define RW_DEP_MAP_INIT(lockname) +#endif + +typedef struct rt_rw_lock rwlock_t; + +#define __RW_LOCK_UNLOCKED(name) __RWLOCK_RT_INITIALIZER(name) + +#define DEFINE_RWLOCK(name) \ + rwlock_t name = __RW_LOCK_UNLOCKED(name) + +/* + * A reader biased implementation primarily for CPU pinning. + * + * Can be selected as general replacement for the single reader RT rwlock + * variant + */ +struct rt_rw_lock { + struct rt_mutex rtmutex; + atomic_t readers; +#ifdef CONFIG_DEBUG_LOCK_ALLOC + struct lockdep_map dep_map; +#endif +}; + +#define READER_BIAS (1U << 31) +#define WRITER_BIAS (1U << 30) + +#define __RWLOCK_RT_INITIALIZER(name) \ +{ \ + .readers = ATOMIC_INIT(READER_BIAS), \ + .rtmutex = __RT_MUTEX_INITIALIZER_SAVE_STATE(name.rtmutex), \ + RW_DEP_MAP_INIT(name) \ +} + +void __rwlock_biased_rt_init(struct rt_rw_lock *lock, const char *name, + struct lock_class_key *key); + +#define rwlock_biased_rt_init(rwlock) \ + do { \ + static struct lock_class_key __key; \ + \ + __rwlock_biased_rt_init((rwlock), #rwlock, &__key); \ + } while (0) + +#endif diff --git a/include/linux/rwsem-rt.h b/include/linux/rwsem-rt.h new file mode 100644 index 0000000000000..7f7e748ef5222 --- /dev/null +++ b/include/linux/rwsem-rt.h @@ -0,0 +1,69 @@ +// SPDX-License-Identifier: GPL-2.0-only +#ifndef _LINUX_RWSEM_RT_H +#define _LINUX_RWSEM_RT_H + +#ifndef _LINUX_RWSEM_H +#error "Include rwsem.h" +#endif + +#include +#include + +#define READER_BIAS (1U << 31) +#define WRITER_BIAS (1U << 30) + +struct rw_semaphore { + atomic_t readers; + struct rt_mutex rtmutex; +#ifdef CONFIG_DEBUG_LOCK_ALLOC + struct lockdep_map dep_map; +#endif +}; + +#define __RWSEM_INITIALIZER(name) \ +{ \ + .readers = ATOMIC_INIT(READER_BIAS), \ + .rtmutex = __RT_MUTEX_INITIALIZER(name.rtmutex), \ + RW_DEP_MAP_INIT(name) \ +} + +#define DECLARE_RWSEM(lockname) \ + struct rw_semaphore lockname = __RWSEM_INITIALIZER(lockname) + +extern void __rwsem_init(struct rw_semaphore *rwsem, const char *name, + struct lock_class_key *key); + +#define __init_rwsem(sem, name, key) \ +do { \ + rt_mutex_init(&(sem)->rtmutex); \ + __rwsem_init((sem), (name), (key)); \ +} while (0) + +#define init_rwsem(sem) \ +do { \ + static struct lock_class_key __key; \ + \ + __init_rwsem((sem), #sem, &__key); \ +} while (0) + +static inline int rwsem_is_locked(struct rw_semaphore *sem) +{ + return atomic_read(&sem->readers) != READER_BIAS; +} + +static inline int rwsem_is_contended(struct rw_semaphore *sem) +{ + return atomic_read(&sem->readers) > 0; +} + +extern void __down_read(struct rw_semaphore *sem); +extern int __down_read_killable(struct rw_semaphore *sem); +extern int __down_read_trylock(struct rw_semaphore *sem); +extern void __down_write(struct rw_semaphore *sem); +extern int __must_check __down_write_killable(struct rw_semaphore *sem); +extern int __down_write_trylock(struct rw_semaphore *sem); +extern void __up_read(struct rw_semaphore *sem); +extern void __up_write(struct rw_semaphore *sem); +extern void __downgrade_write(struct rw_semaphore *sem); + +#endif diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h index 25e3fde856178..9101367852bc9 100644 --- a/include/linux/rwsem.h +++ b/include/linux/rwsem.h @@ -16,6 +16,11 @@ #include #include #include + +#ifdef CONFIG_PREEMPT_RT +#include +#else /* PREEMPT_RT */ + #ifdef CONFIG_RWSEM_SPIN_ON_OWNER #include #endif @@ -119,6 +124,13 @@ static inline int rwsem_is_contended(struct rw_semaphore *sem) return !list_empty(&sem->wait_list); } +#endif /* !PREEMPT_RT */ + +/* + * The functions below are the same for all rwsem implementations including + * the RT specific variant. + */ + /* * lock for reading */ diff --git a/include/linux/sched.h b/include/linux/sched.h index afe01e232935f..c72ae6627e967 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -34,6 +34,7 @@ #include #include #include +#include /* task_struct member predeclarations (sorted alphabetically): */ struct audit_context; @@ -110,12 +111,8 @@ struct task_group; __TASK_TRACED | EXIT_DEAD | EXIT_ZOMBIE | \ TASK_PARKED) -#define task_is_traced(task) ((task->state & __TASK_TRACED) != 0) - #define task_is_stopped(task) ((task->state & __TASK_STOPPED) != 0) -#define task_is_stopped_or_traced(task) ((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0) - #ifdef CONFIG_DEBUG_ATOMIC_SLEEP /* @@ -139,6 +136,9 @@ struct task_group; smp_store_mb(current->state, (state_value)); \ } while (0) +#define __set_current_state_no_track(state_value) \ + current->state = (state_value); + #define set_special_state(state_value) \ do { \ unsigned long flags; /* may shadow */ \ @@ -192,6 +192,9 @@ struct task_group; #define set_current_state(state_value) \ smp_store_mb(current->state, (state_value)) +#define __set_current_state_no_track(state_value) \ + __set_current_state(state_value) + /* * set_special_state() should be used for those states when the blocking task * can not use the regular condition based wait-loop. In that case we must @@ -638,6 +641,8 @@ struct task_struct { #endif /* -1 unrunnable, 0 runnable, >0 stopped: */ volatile long state; + /* saved state for "spinlock sleepers" */ + volatile long saved_state; /* * This begins the randomizable portion of task_struct. Only @@ -713,6 +718,11 @@ struct task_struct { int nr_cpus_allowed; const cpumask_t *cpus_ptr; cpumask_t cpus_mask; + void *migration_pending; +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + unsigned short migration_disabled; +#endif + unsigned short migration_flags; #ifdef CONFIG_PREEMPT_RCU int rcu_read_lock_nesting; @@ -941,11 +951,16 @@ struct task_struct { /* Signal handlers: */ struct signal_struct *signal; struct sighand_struct __rcu *sighand; + struct sigqueue *sigqueue_cache; sigset_t blocked; sigset_t real_blocked; /* Restored if set_restore_sigmask() was used: */ sigset_t saved_sigmask; struct sigpending pending; +#ifdef CONFIG_PREEMPT_RT + /* TODO: move me into ->restart_block ? */ + struct kernel_siginfo forced_info; +#endif unsigned long sas_ss_sp; size_t sas_ss_size; unsigned int sas_ss_flags; @@ -972,6 +987,7 @@ struct task_struct { raw_spinlock_t pi_lock; struct wake_q_node wake_q; + struct wake_q_node wake_q_sleeper; #ifdef CONFIG_RT_MUTEXES /* PI waiters blocked on a rt_mutex held by this task: */ @@ -999,6 +1015,9 @@ struct task_struct { int softirq_context; int irq_config; #endif +#ifdef CONFIG_PREEMPT_RT + int softirq_disable_cnt; +#endif #ifdef CONFIG_LOCKDEP # define MAX_LOCK_DEPTH 48UL @@ -1280,6 +1299,12 @@ struct task_struct { unsigned int sequential_io; unsigned int sequential_io_avg; #endif +#ifdef CONFIG_PREEMPT_RT +# if defined CONFIG_HIGHMEM || defined CONFIG_X86_32 + int kmap_idx; + pte_t kmap_pte[KM_TYPE_NR]; +# endif +#endif #ifdef CONFIG_DEBUG_ATOMIC_SLEEP unsigned long task_state_change; #endif @@ -1722,6 +1747,7 @@ extern struct task_struct *find_get_task_by_vpid(pid_t nr); extern int wake_up_state(struct task_struct *tsk, unsigned int state); extern int wake_up_process(struct task_struct *tsk); +extern int wake_up_lock_sleeper(struct task_struct *tsk); extern void wake_up_new_task(struct task_struct *tsk); #ifdef CONFIG_SMP @@ -1812,6 +1838,89 @@ static inline int test_tsk_need_resched(struct task_struct *tsk) return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED)); } +#ifdef CONFIG_PREEMPT_LAZY +static inline void set_tsk_need_resched_lazy(struct task_struct *tsk) +{ + set_tsk_thread_flag(tsk,TIF_NEED_RESCHED_LAZY); +} + +static inline void clear_tsk_need_resched_lazy(struct task_struct *tsk) +{ + clear_tsk_thread_flag(tsk,TIF_NEED_RESCHED_LAZY); +} + +static inline int test_tsk_need_resched_lazy(struct task_struct *tsk) +{ + return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED_LAZY)); +} + +static inline int need_resched_lazy(void) +{ + return test_thread_flag(TIF_NEED_RESCHED_LAZY); +} + +static inline int need_resched_now(void) +{ + return test_thread_flag(TIF_NEED_RESCHED); +} + +#else +static inline void clear_tsk_need_resched_lazy(struct task_struct *tsk) { } +static inline int need_resched_lazy(void) { return 0; } + +static inline int need_resched_now(void) +{ + return test_thread_flag(TIF_NEED_RESCHED); +} + +#endif + + +static inline bool __task_is_stopped_or_traced(struct task_struct *task) +{ + if (task->state & (__TASK_STOPPED | __TASK_TRACED)) + return true; +#ifdef CONFIG_PREEMPT_RT + if (task->saved_state & (__TASK_STOPPED | __TASK_TRACED)) + return true; +#endif + return false; +} + +static inline bool task_is_stopped_or_traced(struct task_struct *task) +{ + bool traced_stopped; + +#ifdef CONFIG_PREEMPT_RT + unsigned long flags; + + raw_spin_lock_irqsave(&task->pi_lock, flags); + traced_stopped = __task_is_stopped_or_traced(task); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); +#else + traced_stopped = __task_is_stopped_or_traced(task); +#endif + return traced_stopped; +} + +static inline bool task_is_traced(struct task_struct *task) +{ + bool traced = false; + + if (task->state & __TASK_TRACED) + return true; +#ifdef CONFIG_PREEMPT_RT + /* in case the task is sleeping on tasklist_lock */ + raw_spin_lock_irq(&task->pi_lock); + if (task->state & __TASK_TRACED) + traced = true; + else if (task->saved_state & __TASK_TRACED) + traced = true; + raw_spin_unlock_irq(&task->pi_lock); +#endif + return traced; +} + /* * cond_resched() and cond_resched_lock(): latency reduction via * explicit rescheduling in places that are safe. The return diff --git a/include/linux/sched/hotplug.h b/include/linux/sched/hotplug.h index 9a62ffdd296f0..412cdaba33eb4 100644 --- a/include/linux/sched/hotplug.h +++ b/include/linux/sched/hotplug.h @@ -11,8 +11,10 @@ extern int sched_cpu_activate(unsigned int cpu); extern int sched_cpu_deactivate(unsigned int cpu); #ifdef CONFIG_HOTPLUG_CPU +extern int sched_cpu_wait_empty(unsigned int cpu); extern int sched_cpu_dying(unsigned int cpu); #else +# define sched_cpu_wait_empty NULL # define sched_cpu_dying NULL #endif diff --git a/include/linux/sched/mm.h b/include/linux/sched/mm.h index f889e332912fb..1a08d3d41805c 100644 --- a/include/linux/sched/mm.h +++ b/include/linux/sched/mm.h @@ -49,6 +49,17 @@ static inline void mmdrop(struct mm_struct *mm) __mmdrop(mm); } +#ifdef CONFIG_PREEMPT_RT +extern void __mmdrop_delayed(struct rcu_head *rhp); +static inline void mmdrop_delayed(struct mm_struct *mm) +{ + if (atomic_dec_and_test(&mm->mm_count)) + call_rcu(&mm->delayed_drop, __mmdrop_delayed); +} +#else +# define mmdrop_delayed(mm) mmdrop(mm) +#endif + /* * This has to be called after a get_task_mm()/mmget_not_zero() * followed by taking the mmap_lock for writing before modifying the diff --git a/include/linux/sched/rt.h b/include/linux/sched/rt.h index e5af028c08b49..994c25640e156 100644 --- a/include/linux/sched/rt.h +++ b/include/linux/sched/rt.h @@ -39,20 +39,12 @@ static inline struct task_struct *rt_mutex_get_top_task(struct task_struct *p) } extern void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task); extern void rt_mutex_adjust_pi(struct task_struct *p); -static inline bool tsk_is_pi_blocked(struct task_struct *tsk) -{ - return tsk->pi_blocked_on != NULL; -} #else static inline struct task_struct *rt_mutex_get_top_task(struct task_struct *task) { return NULL; } # define rt_mutex_adjust_pi(p) do { } while (0) -static inline bool tsk_is_pi_blocked(struct task_struct *tsk) -{ - return false; -} #endif extern void normalize_rt_tasks(void); diff --git a/include/linux/sched/wake_q.h b/include/linux/sched/wake_q.h index 26a2013ac39c4..6e2dff721547c 100644 --- a/include/linux/sched/wake_q.h +++ b/include/linux/sched/wake_q.h @@ -58,6 +58,17 @@ static inline bool wake_q_empty(struct wake_q_head *head) extern void wake_q_add(struct wake_q_head *head, struct task_struct *task); extern void wake_q_add_safe(struct wake_q_head *head, struct task_struct *task); -extern void wake_up_q(struct wake_q_head *head); +extern void wake_q_add_sleeper(struct wake_q_head *head, struct task_struct *task); +extern void __wake_up_q(struct wake_q_head *head, bool sleeper); + +static inline void wake_up_q(struct wake_q_head *head) +{ + __wake_up_q(head, false); +} + +static inline void wake_up_q_sleeper(struct wake_q_head *head) +{ + __wake_up_q(head, true); +} #endif /* _LINUX_SCHED_WAKE_Q_H */ diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h index 962d9768945f0..76e44e6c01004 100644 --- a/include/linux/seqlock.h +++ b/include/linux/seqlock.h @@ -17,6 +17,7 @@ #include #include #include +#include #include #include @@ -53,7 +54,7 @@ * * If the write serialization mechanism is one of the common kernel * locking primitives, use a sequence counter with associated lock - * (seqcount_LOCKTYPE_t) instead. + * (seqcount_LOCKNAME_t) instead. * * If it's desired to automatically handle the sequence counter writer * serialization and non-preemptibility requirements, use a sequential @@ -117,7 +118,7 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s) #define SEQCNT_ZERO(name) { .sequence = 0, SEQCOUNT_DEP_MAP_INIT(name) } /* - * Sequence counters with associated locks (seqcount_LOCKTYPE_t) + * Sequence counters with associated locks (seqcount_LOCKNAME_t) * * A sequence counter which associates the lock used for writer * serialization at initialization time. This enables lockdep to validate @@ -131,63 +132,117 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s) * See Documentation/locking/seqlock.rst */ -#ifdef CONFIG_LOCKDEP +/* + * For PREEMPT_RT, seqcount_LOCKNAME_t write side critical sections cannot + * disable preemption. It can lead to higher latencies, and the write side + * sections will not be able to acquire locks which become sleeping locks + * (e.g. spinlock_t). + * + * To remain preemptible while avoiding a possible livelock caused by the + * reader preempting the writer, use a different technique: let the reader + * detect if a seqcount_LOCKNAME_t writer is in progress. If that is the + * case, acquire then release the associated LOCKNAME writer serialization + * lock. This will allow any possibly-preempted writer to make progress + * until the end of its writer serialization lock critical section. + * + * This lock-unlock technique must be implemented for all of PREEMPT_RT + * sleeping locks. See Documentation/locking/locktypes.rst + */ +#if defined(CONFIG_LOCKDEP) || defined(CONFIG_PREEMPT_RT) #define __SEQ_LOCK(expr) expr #else #define __SEQ_LOCK(expr) #endif /** - * typedef seqcount_LOCKNAME_t - sequence counter with LOCKTYPR associated + * typedef seqcount_LOCKNAME_t - sequence counter with LOCKNAME associated * @seqcount: The real sequence counter - * @lock: Pointer to the associated spinlock + * @lock: Pointer to the associated lock * - * A plain sequence counter with external writer synchronization by a - * spinlock. The spinlock is associated to the sequence count in the + * A plain sequence counter with external writer synchronization by + * LOCKNAME @lock. The lock is associated to the sequence counter in the * static initializer or init function. This enables lockdep to validate * that the write side critical section is properly serialized. - */ - -/** - * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t - * @s: Pointer to the seqcount_LOCKNAME_t instance - * @lock: Pointer to the associated LOCKTYPE + * + * LOCKNAME: raw_spinlock, spinlock, rwlock, mutex, or ww_mutex. */ /* - * SEQCOUNT_LOCKTYPE() - Instantiate seqcount_LOCKNAME_t and helpers - * @locktype: actual typename - * @lockname: name + * seqcount_LOCKNAME_init() - runtime initializer for seqcount_LOCKNAME_t + * @s: Pointer to the seqcount_LOCKNAME_t instance + * @lock: Pointer to the associated lock + */ + +#define seqcount_LOCKNAME_init(s, _lock, lockname) \ + do { \ + seqcount_##lockname##_t *____s = (s); \ + seqcount_init(&____s->seqcount); \ + __SEQ_LOCK(____s->lock = (_lock)); \ + } while (0) + +#define seqcount_raw_spinlock_init(s, lock) seqcount_LOCKNAME_init(s, lock, raw_spinlock) +#define seqcount_spinlock_init(s, lock) seqcount_LOCKNAME_init(s, lock, spinlock) +#define seqcount_rwlock_init(s, lock) seqcount_LOCKNAME_init(s, lock, rwlock); +#define seqcount_mutex_init(s, lock) seqcount_LOCKNAME_init(s, lock, mutex); +#define seqcount_ww_mutex_init(s, lock) seqcount_LOCKNAME_init(s, lock, ww_mutex); + +/* + * SEQCOUNT_LOCKNAME() - Instantiate seqcount_LOCKNAME_t and helpers + * seqprop_LOCKNAME_*() - Property accessors for seqcount_LOCKNAME_t + * + * @lockname: "LOCKNAME" part of seqcount_LOCKNAME_t + * @locktype: LOCKNAME canonical C data type * @preemptible: preemptibility of above locktype * @lockmember: argument for lockdep_assert_held() + * @lockbase: associated lock release function (prefix only) + * @lock_acquire: associated lock acquisition function (full call) */ -#define SEQCOUNT_LOCKTYPE(locktype, lockname, preemptible, lockmember) \ +#define SEQCOUNT_LOCKNAME(lockname, locktype, preemptible, lockmember, lockbase, lock_acquire) \ typedef struct seqcount_##lockname { \ seqcount_t seqcount; \ __SEQ_LOCK(locktype *lock); \ } seqcount_##lockname##_t; \ \ -static __always_inline void \ -seqcount_##lockname##_init(seqcount_##lockname##_t *s, locktype *lock) \ -{ \ - seqcount_init(&s->seqcount); \ - __SEQ_LOCK(s->lock = lock); \ -} \ - \ static __always_inline seqcount_t * \ -__seqcount_##lockname##_ptr(seqcount_##lockname##_t *s) \ +__seqprop_##lockname##_ptr(seqcount_##lockname##_t *s) \ { \ return &s->seqcount; \ } \ \ -static __always_inline bool \ -__seqcount_##lockname##_preemptible(seqcount_##lockname##_t *s) \ +static __always_inline unsigned \ +__seqprop_##lockname##_sequence(const seqcount_##lockname##_t *s) \ { \ - return preemptible; \ + unsigned seq = READ_ONCE(s->seqcount.sequence); \ + \ + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) \ + return seq; \ + \ + if (preemptible && unlikely(seq & 1)) { \ + __SEQ_LOCK(lock_acquire); \ + __SEQ_LOCK(lockbase##_unlock(s->lock)); \ + \ + /* \ + * Re-read the sequence counter since the (possibly \ + * preempted) writer made progress. \ + */ \ + seq = READ_ONCE(s->seqcount.sequence); \ + } \ + \ + return seq; \ +} \ + \ +static __always_inline bool \ +__seqprop_##lockname##_preemptible(const seqcount_##lockname##_t *s) \ +{ \ + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) \ + return preemptible; \ + \ + /* PREEMPT_RT relies on the above LOCK+UNLOCK */ \ + return false; \ } \ \ static __always_inline void \ -__seqcount_##lockname##_assert(seqcount_##lockname##_t *s) \ +__seqprop_##lockname##_assert(const seqcount_##lockname##_t *s) \ { \ __SEQ_LOCK(lockdep_assert_held(lockmember)); \ } @@ -196,50 +251,56 @@ __seqcount_##lockname##_assert(seqcount_##lockname##_t *s) \ * __seqprop() for seqcount_t */ -static inline seqcount_t *__seqcount_ptr(seqcount_t *s) +static inline seqcount_t *__seqprop_ptr(seqcount_t *s) { return s; } -static inline bool __seqcount_preemptible(seqcount_t *s) +static inline unsigned __seqprop_sequence(const seqcount_t *s) +{ + return READ_ONCE(s->sequence); +} + +static inline bool __seqprop_preemptible(const seqcount_t *s) { return false; } -static inline void __seqcount_assert(seqcount_t *s) +static inline void __seqprop_assert(const seqcount_t *s) { lockdep_assert_preemption_disabled(); } -SEQCOUNT_LOCKTYPE(raw_spinlock_t, raw_spinlock, false, s->lock) -SEQCOUNT_LOCKTYPE(spinlock_t, spinlock, false, s->lock) -SEQCOUNT_LOCKTYPE(rwlock_t, rwlock, false, s->lock) -SEQCOUNT_LOCKTYPE(struct mutex, mutex, true, s->lock) -SEQCOUNT_LOCKTYPE(struct ww_mutex, ww_mutex, true, &s->lock->base) +#define __SEQ_RT IS_ENABLED(CONFIG_PREEMPT_RT) -/** +SEQCOUNT_LOCKNAME(raw_spinlock, raw_spinlock_t, false, s->lock, raw_spin, raw_spin_lock(s->lock)) +SEQCOUNT_LOCKNAME(spinlock, spinlock_t, __SEQ_RT, s->lock, spin, spin_lock(s->lock)) +SEQCOUNT_LOCKNAME(rwlock, rwlock_t, __SEQ_RT, s->lock, read, read_lock(s->lock)) +SEQCOUNT_LOCKNAME(mutex, struct mutex, true, s->lock, mutex, mutex_lock(s->lock)) +SEQCOUNT_LOCKNAME(ww_mutex, struct ww_mutex, true, &s->lock->base, ww_mutex, ww_mutex_lock(s->lock, NULL)) + +/* * SEQCNT_LOCKNAME_ZERO - static initializer for seqcount_LOCKNAME_t * @name: Name of the seqcount_LOCKNAME_t instance - * @lock: Pointer to the associated LOCKTYPE + * @lock: Pointer to the associated LOCKNAME */ -#define SEQCOUNT_LOCKTYPE_ZERO(seq_name, assoc_lock) { \ +#define SEQCOUNT_LOCKNAME_ZERO(seq_name, assoc_lock) { \ .seqcount = SEQCNT_ZERO(seq_name.seqcount), \ __SEQ_LOCK(.lock = (assoc_lock)) \ } -#define SEQCNT_SPINLOCK_ZERO(name, lock) SEQCOUNT_LOCKTYPE_ZERO(name, lock) -#define SEQCNT_RAW_SPINLOCK_ZERO(name, lock) SEQCOUNT_LOCKTYPE_ZERO(name, lock) -#define SEQCNT_RWLOCK_ZERO(name, lock) SEQCOUNT_LOCKTYPE_ZERO(name, lock) -#define SEQCNT_MUTEX_ZERO(name, lock) SEQCOUNT_LOCKTYPE_ZERO(name, lock) -#define SEQCNT_WW_MUTEX_ZERO(name, lock) SEQCOUNT_LOCKTYPE_ZERO(name, lock) - +#define SEQCNT_RAW_SPINLOCK_ZERO(name, lock) SEQCOUNT_LOCKNAME_ZERO(name, lock) +#define SEQCNT_SPINLOCK_ZERO(name, lock) SEQCOUNT_LOCKNAME_ZERO(name, lock) +#define SEQCNT_RWLOCK_ZERO(name, lock) SEQCOUNT_LOCKNAME_ZERO(name, lock) +#define SEQCNT_MUTEX_ZERO(name, lock) SEQCOUNT_LOCKNAME_ZERO(name, lock) +#define SEQCNT_WW_MUTEX_ZERO(name, lock) SEQCOUNT_LOCKNAME_ZERO(name, lock) #define __seqprop_case(s, lockname, prop) \ - seqcount_##lockname##_t: __seqcount_##lockname##_##prop((void *)(s)) + seqcount_##lockname##_t: __seqprop_##lockname##_##prop((void *)(s)) #define __seqprop(s, prop) _Generic(*(s), \ - seqcount_t: __seqcount_##prop((void *)(s)), \ + seqcount_t: __seqprop_##prop((void *)(s)), \ __seqprop_case((s), raw_spinlock, prop), \ __seqprop_case((s), spinlock, prop), \ __seqprop_case((s), rwlock, prop), \ @@ -247,12 +308,13 @@ SEQCOUNT_LOCKTYPE(struct ww_mutex, ww_mutex, true, &s->lock->base) __seqprop_case((s), ww_mutex, prop)) #define __seqcount_ptr(s) __seqprop(s, ptr) +#define __seqcount_sequence(s) __seqprop(s, sequence) #define __seqcount_lock_preemptible(s) __seqprop(s, preemptible) #define __seqcount_assert_lock_held(s) __seqprop(s, assert) /** * __read_seqcount_begin() - begin a seqcount_t read section w/o barrier - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * __read_seqcount_begin is like read_seqcount_begin, but has no smp_rmb() * barrier. Callers should ensure that smp_rmb() or equivalent ordering is @@ -265,56 +327,45 @@ SEQCOUNT_LOCKTYPE(struct ww_mutex, ww_mutex, true, &s->lock->base) * Return: count to be passed to read_seqcount_retry() */ #define __read_seqcount_begin(s) \ - __read_seqcount_t_begin(__seqcount_ptr(s)) - -static inline unsigned __read_seqcount_t_begin(const seqcount_t *s) -{ - unsigned ret; - -repeat: - ret = READ_ONCE(s->sequence); - if (unlikely(ret & 1)) { - cpu_relax(); - goto repeat; - } - kcsan_atomic_next(KCSAN_SEQLOCK_REGION_MAX); - return ret; -} +({ \ + unsigned seq; \ + \ + while ((seq = __seqcount_sequence(s)) & 1) \ + cpu_relax(); \ + \ + kcsan_atomic_next(KCSAN_SEQLOCK_REGION_MAX); \ + seq; \ +}) /** * raw_read_seqcount_begin() - begin a seqcount_t read section w/o lockdep - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * Return: count to be passed to read_seqcount_retry() */ #define raw_read_seqcount_begin(s) \ - raw_read_seqcount_t_begin(__seqcount_ptr(s)) - -static inline unsigned raw_read_seqcount_t_begin(const seqcount_t *s) -{ - unsigned ret = __read_seqcount_t_begin(s); - smp_rmb(); - return ret; -} +({ \ + unsigned seq = __read_seqcount_begin(s); \ + \ + smp_rmb(); \ + seq; \ +}) /** * read_seqcount_begin() - begin a seqcount_t read critical section - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * Return: count to be passed to read_seqcount_retry() */ #define read_seqcount_begin(s) \ - read_seqcount_t_begin(__seqcount_ptr(s)) - -static inline unsigned read_seqcount_t_begin(const seqcount_t *s) -{ - seqcount_lockdep_reader_access(s); - return raw_read_seqcount_t_begin(s); -} +({ \ + seqcount_lockdep_reader_access(__seqcount_ptr(s)); \ + raw_read_seqcount_begin(s); \ +}) /** * raw_read_seqcount() - read the raw seqcount_t counter value - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * raw_read_seqcount opens a read critical section of the given * seqcount_t, without any lockdep checking, and without checking or @@ -324,20 +375,18 @@ static inline unsigned read_seqcount_t_begin(const seqcount_t *s) * Return: count to be passed to read_seqcount_retry() */ #define raw_read_seqcount(s) \ - raw_read_seqcount_t(__seqcount_ptr(s)) - -static inline unsigned raw_read_seqcount_t(const seqcount_t *s) -{ - unsigned ret = READ_ONCE(s->sequence); - smp_rmb(); - kcsan_atomic_next(KCSAN_SEQLOCK_REGION_MAX); - return ret; -} +({ \ + unsigned seq = __seqcount_sequence(s); \ + \ + smp_rmb(); \ + kcsan_atomic_next(KCSAN_SEQLOCK_REGION_MAX); \ + seq; \ +}) /** * raw_seqcount_begin() - begin a seqcount_t read critical section w/o * lockdep and w/o counter stabilization - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * raw_seqcount_begin opens a read critical section of the given * seqcount_t. Unlike read_seqcount_begin(), this function will not wait @@ -352,20 +401,17 @@ static inline unsigned raw_read_seqcount_t(const seqcount_t *s) * Return: count to be passed to read_seqcount_retry() */ #define raw_seqcount_begin(s) \ - raw_seqcount_t_begin(__seqcount_ptr(s)) - -static inline unsigned raw_seqcount_t_begin(const seqcount_t *s) -{ - /* - * If the counter is odd, let read_seqcount_retry() fail - * by decrementing the counter. - */ - return raw_read_seqcount_t(s) & ~1; -} +({ \ + /* \ + * If the counter is odd, let read_seqcount_retry() fail \ + * by decrementing the counter. \ + */ \ + raw_read_seqcount(s) & ~1; \ +}) /** * __read_seqcount_retry() - end a seqcount_t read section w/o barrier - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * @start: count, from read_seqcount_begin() * * __read_seqcount_retry is like read_seqcount_retry, but has no smp_rmb() @@ -389,7 +435,7 @@ static inline int __read_seqcount_t_retry(const seqcount_t *s, unsigned start) /** * read_seqcount_retry() - end a seqcount_t read critical section - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * @start: count, from read_seqcount_begin() * * read_seqcount_retry closes the read critical section of given @@ -409,7 +455,7 @@ static inline int read_seqcount_t_retry(const seqcount_t *s, unsigned start) /** * raw_write_seqcount_begin() - start a seqcount_t write section w/o lockdep - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants */ #define raw_write_seqcount_begin(s) \ do { \ @@ -428,7 +474,7 @@ static inline void raw_write_seqcount_t_begin(seqcount_t *s) /** * raw_write_seqcount_end() - end a seqcount_t write section w/o lockdep - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants */ #define raw_write_seqcount_end(s) \ do { \ @@ -448,7 +494,7 @@ static inline void raw_write_seqcount_t_end(seqcount_t *s) /** * write_seqcount_begin_nested() - start a seqcount_t write section with * custom lockdep nesting level - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * @subclass: lockdep nesting level * * See Documentation/locking/lockdep-design.rst @@ -471,7 +517,7 @@ static inline void write_seqcount_t_begin_nested(seqcount_t *s, int subclass) /** * write_seqcount_begin() - start a seqcount_t write side critical section - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * write_seqcount_begin opens a write side critical section of the given * seqcount_t. @@ -497,7 +543,7 @@ static inline void write_seqcount_t_begin(seqcount_t *s) /** * write_seqcount_end() - end a seqcount_t write side critical section - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * The write section must've been opened with write_seqcount_begin(). */ @@ -517,7 +563,7 @@ static inline void write_seqcount_t_end(seqcount_t *s) /** * raw_write_seqcount_barrier() - do a seqcount_t write barrier - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * This can be used to provide an ordering guarantee instead of the usual * consistency guarantee. It is one wmb cheaper, because it can collapse @@ -571,7 +617,7 @@ static inline void raw_write_seqcount_t_barrier(seqcount_t *s) /** * write_seqcount_invalidate() - invalidate in-progress seqcount_t read * side operations - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * @s: Pointer to seqcount_t or any of the seqcount_LOCKNAME_t variants * * After write_seqcount_invalidate, no seqcount_t read side operations * will complete successfully and see data older than this. @@ -587,34 +633,73 @@ static inline void write_seqcount_t_invalidate(seqcount_t *s) kcsan_nestable_atomic_end(); } -/** - * raw_read_seqcount_latch() - pick even/odd seqcount_t latch data copy - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants +/* + * Latch sequence counters (seqcount_latch_t) * - * Use seqcount_t latching to switch between two storage places protected - * by a sequence counter. Doing so allows having interruptible, preemptible, - * seqcount_t write side critical sections. + * A sequence counter variant where the counter even/odd value is used to + * switch between two copies of protected data. This allows the read path, + * typically NMIs, to safely interrupt the write side critical section. * - * Check raw_write_seqcount_latch() for more details and a full reader and - * writer usage example. - * - * Return: sequence counter raw value. Use the lowest bit as an index for - * picking which data copy to read. The full counter value must then be - * checked with read_seqcount_retry(). + * As the write sections are fully preemptible, no special handling for + * PREEMPT_RT is needed. */ -#define raw_read_seqcount_latch(s) \ - raw_read_seqcount_t_latch(__seqcount_ptr(s)) +typedef struct { + seqcount_t seqcount; +} seqcount_latch_t; -static inline int raw_read_seqcount_t_latch(seqcount_t *s) -{ - /* Pairs with the first smp_wmb() in raw_write_seqcount_latch() */ - int seq = READ_ONCE(s->sequence); /* ^^^ */ - return seq; +/** + * SEQCNT_LATCH_ZERO() - static initializer for seqcount_latch_t + * @seq_name: Name of the seqcount_latch_t instance + */ +#define SEQCNT_LATCH_ZERO(seq_name) { \ + .seqcount = SEQCNT_ZERO(seq_name.seqcount), \ } /** - * raw_write_seqcount_latch() - redirect readers to even/odd copy - * @s: Pointer to seqcount_t or any of the seqcount_locktype_t variants + * seqcount_latch_init() - runtime initializer for seqcount_latch_t + * @s: Pointer to the seqcount_latch_t instance + */ +static inline void seqcount_latch_init(seqcount_latch_t *s) +{ + seqcount_init(&s->seqcount); +} + +/** + * raw_read_seqcount_latch() - pick even/odd latch data copy + * @s: Pointer to seqcount_latch_t + * + * See raw_write_seqcount_latch() for details and a full reader/writer + * usage example. + * + * Return: sequence counter raw value. Use the lowest bit as an index for + * picking which data copy to read. The full counter must then be checked + * with read_seqcount_latch_retry(). + */ +static inline unsigned raw_read_seqcount_latch(const seqcount_latch_t *s) +{ + /* + * Pairs with the first smp_wmb() in raw_write_seqcount_latch(). + * Due to the dependent load, a full smp_rmb() is not needed. + */ + return READ_ONCE(s->seqcount.sequence); +} + +/** + * read_seqcount_latch_retry() - end a seqcount_latch_t read section + * @s: Pointer to seqcount_latch_t + * @start: count, from raw_read_seqcount_latch() + * + * Return: true if a read section retry is required, else false + */ +static inline int +read_seqcount_latch_retry(const seqcount_latch_t *s, unsigned start) +{ + return read_seqcount_retry(&s->seqcount, start); +} + +/** + * raw_write_seqcount_latch() - redirect latch readers to even/odd copy + * @s: Pointer to seqcount_latch_t * * The latch technique is a multiversion concurrency control method that allows * queries during non-atomic modifications. If you can guarantee queries never @@ -633,7 +718,7 @@ static inline int raw_read_seqcount_t_latch(seqcount_t *s) * The basic form is a data structure like:: * * struct latch_struct { - * seqcount_t seq; + * seqcount_latch_t seq; * struct data_struct data[2]; * }; * @@ -643,13 +728,13 @@ static inline int raw_read_seqcount_t_latch(seqcount_t *s) * void latch_modify(struct latch_struct *latch, ...) * { * smp_wmb(); // Ensure that the last data[1] update is visible - * latch->seq++; + * latch->seq.sequence++; * smp_wmb(); // Ensure that the seqcount update is visible * * modify(latch->data[0], ...); * * smp_wmb(); // Ensure that the data[0] update is visible - * latch->seq++; + * latch->seq.sequence++; * smp_wmb(); // Ensure that the seqcount update is visible * * modify(latch->data[1], ...); @@ -668,8 +753,8 @@ static inline int raw_read_seqcount_t_latch(seqcount_t *s) * idx = seq & 0x01; * entry = data_query(latch->data[idx], ...); * - * // read_seqcount_retry() includes needed smp_rmb() - * } while (read_seqcount_retry(&latch->seq, seq)); + * // This includes needed smp_rmb() + * } while (read_seqcount_latch_retry(&latch->seq, seq)); * * return entry; * } @@ -688,19 +773,16 @@ static inline int raw_read_seqcount_t_latch(seqcount_t *s) * to miss an entire modification sequence, once it resumes it might * observe the new entry. * - * NOTE: + * NOTE2: * * When data is a dynamic data structure; one should use regular RCU * patterns to manage the lifetimes of the objects within. */ -#define raw_write_seqcount_latch(s) \ - raw_write_seqcount_t_latch(__seqcount_ptr(s)) - -static inline void raw_write_seqcount_t_latch(seqcount_t *s) +static inline void raw_write_seqcount_latch(seqcount_latch_t *s) { - smp_wmb(); /* prior stores before incrementing "sequence" */ - s->sequence++; - smp_wmb(); /* increment "sequence" before following stores */ + smp_wmb(); /* prior stores before incrementing "sequence" */ + s->seqcount.sequence++; + smp_wmb(); /* increment "sequence" before following stores */ } /* @@ -714,13 +796,17 @@ static inline void raw_write_seqcount_t_latch(seqcount_t *s) * - Documentation/locking/seqlock.rst */ typedef struct { - struct seqcount seqcount; + /* + * Make sure that readers don't starve writers on PREEMPT_RT: use + * seqcount_spinlock_t instead of seqcount_t. Check __SEQ_LOCK(). + */ + seqcount_spinlock_t seqcount; spinlock_t lock; } seqlock_t; #define __SEQLOCK_UNLOCKED(lockname) \ { \ - .seqcount = SEQCNT_ZERO(lockname), \ + .seqcount = SEQCNT_SPINLOCK_ZERO(lockname, &(lockname).lock), \ .lock = __SPIN_LOCK_UNLOCKED(lockname) \ } @@ -730,8 +816,8 @@ typedef struct { */ #define seqlock_init(sl) \ do { \ - seqcount_init(&(sl)->seqcount); \ spin_lock_init(&(sl)->lock); \ + seqcount_spinlock_init(&(sl)->seqcount, &(sl)->lock); \ } while (0) /** @@ -778,6 +864,12 @@ static inline unsigned read_seqretry(const seqlock_t *sl, unsigned start) return read_seqcount_retry(&sl->seqcount, start); } +/* + * For all seqlock_t write side functions, use write_seqcount_*t*_begin() + * instead of the generic write_seqcount_begin(). This way, no redundant + * lockdep_assert_held() checks are added. + */ + /** * write_seqlock() - start a seqlock_t write side critical section * @sl: Pointer to seqlock_t @@ -794,7 +886,7 @@ static inline unsigned read_seqretry(const seqlock_t *sl, unsigned start) static inline void write_seqlock(seqlock_t *sl) { spin_lock(&sl->lock); - write_seqcount_t_begin(&sl->seqcount); + write_seqcount_t_begin(&sl->seqcount.seqcount); } /** @@ -806,7 +898,7 @@ static inline void write_seqlock(seqlock_t *sl) */ static inline void write_sequnlock(seqlock_t *sl) { - write_seqcount_t_end(&sl->seqcount); + write_seqcount_t_end(&sl->seqcount.seqcount); spin_unlock(&sl->lock); } @@ -820,7 +912,7 @@ static inline void write_sequnlock(seqlock_t *sl) static inline void write_seqlock_bh(seqlock_t *sl) { spin_lock_bh(&sl->lock); - write_seqcount_t_begin(&sl->seqcount); + write_seqcount_t_begin(&sl->seqcount.seqcount); } /** @@ -833,7 +925,7 @@ static inline void write_seqlock_bh(seqlock_t *sl) */ static inline void write_sequnlock_bh(seqlock_t *sl) { - write_seqcount_t_end(&sl->seqcount); + write_seqcount_t_end(&sl->seqcount.seqcount); spin_unlock_bh(&sl->lock); } @@ -847,7 +939,7 @@ static inline void write_sequnlock_bh(seqlock_t *sl) static inline void write_seqlock_irq(seqlock_t *sl) { spin_lock_irq(&sl->lock); - write_seqcount_t_begin(&sl->seqcount); + write_seqcount_t_begin(&sl->seqcount.seqcount); } /** @@ -859,7 +951,7 @@ static inline void write_seqlock_irq(seqlock_t *sl) */ static inline void write_sequnlock_irq(seqlock_t *sl) { - write_seqcount_t_end(&sl->seqcount); + write_seqcount_t_end(&sl->seqcount.seqcount); spin_unlock_irq(&sl->lock); } @@ -868,7 +960,7 @@ static inline unsigned long __write_seqlock_irqsave(seqlock_t *sl) unsigned long flags; spin_lock_irqsave(&sl->lock, flags); - write_seqcount_t_begin(&sl->seqcount); + write_seqcount_t_begin(&sl->seqcount.seqcount); return flags; } @@ -897,7 +989,7 @@ static inline unsigned long __write_seqlock_irqsave(seqlock_t *sl) static inline void write_sequnlock_irqrestore(seqlock_t *sl, unsigned long flags) { - write_seqcount_t_end(&sl->seqcount); + write_seqcount_t_end(&sl->seqcount.seqcount); spin_unlock_irqrestore(&sl->lock, flags); } diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h index 2b70f736b091d..68d756373b53b 100644 --- a/include/linux/serial_8250.h +++ b/include/linux/serial_8250.h @@ -7,6 +7,7 @@ #ifndef _LINUX_SERIAL_8250_H #define _LINUX_SERIAL_8250_H +#include #include #include #include @@ -125,6 +126,8 @@ struct uart_8250_port { #define MSR_SAVE_FLAGS UART_MSR_ANY_DELTA unsigned char msr_saved_flags; + atomic_t console_printing; + struct uart_8250_dma *dma; const struct uart_8250_ops *ops; @@ -180,6 +183,8 @@ void serial8250_init_port(struct uart_8250_port *up); void serial8250_set_defaults(struct uart_8250_port *up); void serial8250_console_write(struct uart_8250_port *up, const char *s, unsigned int count); +void serial8250_console_write_atomic(struct uart_8250_port *up, const char *s, + unsigned int count); int serial8250_console_setup(struct uart_port *port, char *options, bool probe); int serial8250_console_exit(struct uart_port *port); diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h index a5a5d1d4d7b1d..0470d1582b09a 100644 --- a/include/linux/shmem_fs.h +++ b/include/linux/shmem_fs.h @@ -31,7 +31,7 @@ struct shmem_sb_info { struct percpu_counter used_blocks; /* How many are allocated */ unsigned long max_inodes; /* How many inodes are allowed */ unsigned long free_inodes; /* How many are left for allocation */ - spinlock_t stat_lock; /* Serialize shmem_sb_info changes */ + raw_spinlock_t stat_lock; /* Serialize shmem_sb_info changes */ umode_t mode; /* Mount mode for root directory */ unsigned char huge; /* Whether to try for hugepages */ kuid_t uid; /* Mount uid for root directory */ diff --git a/include/linux/signal.h b/include/linux/signal.h index 7bbc0e9cf0847..3030d984a1442 100644 --- a/include/linux/signal.h +++ b/include/linux/signal.h @@ -263,6 +263,7 @@ static inline void init_sigpending(struct sigpending *sig) } extern void flush_sigqueue(struct sigpending *queue); +extern void flush_task_sigqueue(struct task_struct *tsk); /* Test if 'sig' is valid signal. Use this instead of testing _NSIG directly */ static inline int valid_signal(unsigned long sig) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 04a18e01b362e..187ea2e1854e3 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -295,6 +295,7 @@ struct sk_buff_head { __u32 qlen; spinlock_t lock; + raw_spinlock_t raw_lock; }; struct sk_buff; @@ -1884,6 +1885,12 @@ static inline void skb_queue_head_init(struct sk_buff_head *list) __skb_queue_head_init(list); } +static inline void skb_queue_head_init_raw(struct sk_buff_head *list) +{ + raw_spin_lock_init(&list->raw_lock); + __skb_queue_head_init(list); +} + static inline void skb_queue_head_init_class(struct sk_buff_head *list, struct lock_class_key *class) { diff --git a/include/linux/smp.h b/include/linux/smp.h index 80d557ef8a11c..47d666fa4fbae 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -236,6 +236,9 @@ static inline int get_boot_cpu_id(void) #define get_cpu() ({ preempt_disable(); __smp_processor_id(); }) #define put_cpu() preempt_enable() +#define get_cpu_light() ({ migrate_disable(); __smp_processor_id(); }) +#define put_cpu_light() migrate_enable() + /* * Callback to arch code if there's nosmp or maxcpus=0 on the * boot command line: diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index f2f12d746dbdb..38ff58065dfb1 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -309,7 +309,11 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) __releases(lock) }) /* Include rwlock functions */ -#include +#ifdef CONFIG_PREEMPT_RT +# include +#else +# include +#endif /* * Pull the _spin_*()/_read_*()/_write_*() functions/declarations: @@ -320,6 +324,10 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) __releases(lock) # include #endif +#ifdef CONFIG_PREEMPT_RT +# include +#else /* PREEMPT_RT */ + /* * Map the spin_lock functions to the raw variants for PREEMPT_RT=n */ @@ -454,6 +462,8 @@ static __always_inline int spin_is_contended(spinlock_t *lock) #define assert_spin_locked(lock) assert_raw_spin_locked(&(lock)->rlock) +#endif /* !PREEMPT_RT */ + /* * Pull the atomic_t declaration: * (asm-mips/atomic.h needs above definitions) diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h index 19a9be9d97ee5..da38149f28431 100644 --- a/include/linux/spinlock_api_smp.h +++ b/include/linux/spinlock_api_smp.h @@ -187,6 +187,8 @@ static inline int __raw_spin_trylock_bh(raw_spinlock_t *lock) return 0; } -#include +#ifndef CONFIG_PREEMPT_RT +# include +#endif #endif /* __LINUX_SPINLOCK_API_SMP_H */ diff --git a/include/linux/spinlock_rt.h b/include/linux/spinlock_rt.h new file mode 100644 index 0000000000000..3085132eae383 --- /dev/null +++ b/include/linux/spinlock_rt.h @@ -0,0 +1,155 @@ +// SPDX-License-Identifier: GPL-2.0-only +#ifndef __LINUX_SPINLOCK_RT_H +#define __LINUX_SPINLOCK_RT_H + +#ifndef __LINUX_SPINLOCK_H +#error Do not include directly. Use spinlock.h +#endif + +#include + +extern void +__rt_spin_lock_init(spinlock_t *lock, const char *name, struct lock_class_key *key); + +#define spin_lock_init(slock) \ +do { \ + static struct lock_class_key __key; \ + \ + rt_mutex_init(&(slock)->lock); \ + __rt_spin_lock_init(slock, #slock, &__key); \ +} while (0) + +extern void __lockfunc rt_spin_lock(spinlock_t *lock); +extern void __lockfunc rt_spin_lock_nested(spinlock_t *lock, int subclass); +extern void __lockfunc rt_spin_lock_nest_lock(spinlock_t *lock, struct lockdep_map *nest_lock); +extern void __lockfunc rt_spin_unlock(spinlock_t *lock); +extern void __lockfunc rt_spin_lock_unlock(spinlock_t *lock); +extern int __lockfunc rt_spin_trylock_irqsave(spinlock_t *lock, unsigned long *flags); +extern int __lockfunc rt_spin_trylock_bh(spinlock_t *lock); +extern int __lockfunc rt_spin_trylock(spinlock_t *lock); +extern int atomic_dec_and_spin_lock(atomic_t *atomic, spinlock_t *lock); + +/* + * lockdep-less calls, for derived types like rwlock: + * (for trylock they can use rt_mutex_trylock() directly. + * Migrate disable handling must be done at the call site. + */ +extern void __lockfunc __rt_spin_lock(struct rt_mutex *lock); +extern void __lockfunc __rt_spin_trylock(struct rt_mutex *lock); +extern void __lockfunc __rt_spin_unlock(struct rt_mutex *lock); + +#define spin_lock(lock) rt_spin_lock(lock) + +#define spin_lock_bh(lock) \ + do { \ + local_bh_disable(); \ + rt_spin_lock(lock); \ + } while (0) + +#define spin_lock_irq(lock) spin_lock(lock) + +#define spin_do_trylock(lock) __cond_lock(lock, rt_spin_trylock(lock)) + +#define spin_trylock(lock) \ +({ \ + int __locked; \ + __locked = spin_do_trylock(lock); \ + __locked; \ +}) + +#ifdef CONFIG_LOCKDEP +# define spin_lock_nested(lock, subclass) \ + do { \ + rt_spin_lock_nested(lock, subclass); \ + } while (0) + +#define spin_lock_bh_nested(lock, subclass) \ + do { \ + local_bh_disable(); \ + rt_spin_lock_nested(lock, subclass); \ + } while (0) + +# define spin_lock_nest_lock(lock, subclass) \ + do { \ + typecheck(struct lockdep_map *, &(subclass)->dep_map); \ + rt_spin_lock_nest_lock(lock, &(subclass)->dep_map); \ + } while (0) + +# define spin_lock_irqsave_nested(lock, flags, subclass) \ + do { \ + typecheck(unsigned long, flags); \ + flags = 0; \ + rt_spin_lock_nested(lock, subclass); \ + } while (0) +#else +# define spin_lock_nested(lock, subclass) spin_lock(((void)(subclass), (lock))) +# define spin_lock_nest_lock(lock, subclass) spin_lock(((void)(subclass), (lock))) +# define spin_lock_bh_nested(lock, subclass) spin_lock_bh(((void)(subclass), (lock))) + +# define spin_lock_irqsave_nested(lock, flags, subclass) \ + do { \ + typecheck(unsigned long, flags); \ + flags = 0; \ + spin_lock(((void)(subclass), (lock))); \ + } while (0) +#endif + +#define spin_lock_irqsave(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + flags = 0; \ + spin_lock(lock); \ + } while (0) + +#define spin_unlock(lock) rt_spin_unlock(lock) + +#define spin_unlock_bh(lock) \ + do { \ + rt_spin_unlock(lock); \ + local_bh_enable(); \ + } while (0) + +#define spin_unlock_irq(lock) spin_unlock(lock) + +#define spin_unlock_irqrestore(lock, flags) \ + do { \ + typecheck(unsigned long, flags); \ + (void) flags; \ + spin_unlock(lock); \ + } while (0) + +#define spin_trylock_bh(lock) __cond_lock(lock, rt_spin_trylock_bh(lock)) +#define spin_trylock_irq(lock) spin_trylock(lock) + +#define spin_trylock_irqsave(lock, flags) \ +({ \ + int __locked; \ + \ + typecheck(unsigned long, flags); \ + flags = 0; \ + __locked = spin_trylock(lock); \ + __locked; \ +}) + +#ifdef CONFIG_GENERIC_LOCKBREAK +# define spin_is_contended(lock) ((lock)->break_lock) +#else +# define spin_is_contended(lock) (((void)(lock), 0)) +#endif + +static inline int spin_can_lock(spinlock_t *lock) +{ + return !rt_mutex_is_locked(&lock->lock); +} + +static inline int spin_is_locked(spinlock_t *lock) +{ + return rt_mutex_is_locked(&lock->lock); +} + +static inline void assert_spin_locked(spinlock_t *lock) +{ + BUG_ON(!spin_is_locked(lock)); +} + +#endif diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h index b981caafe8bf1..8d896d3e1a017 100644 --- a/include/linux/spinlock_types.h +++ b/include/linux/spinlock_types.h @@ -9,93 +9,15 @@ * Released under the General Public License (GPL). */ -#if defined(CONFIG_SMP) -# include +#include + +#ifndef CONFIG_PREEMPT_RT +# include +# include #else -# include +# include +# include +# include #endif -#include - -typedef struct raw_spinlock { - arch_spinlock_t raw_lock; -#ifdef CONFIG_DEBUG_SPINLOCK - unsigned int magic, owner_cpu; - void *owner; -#endif -#ifdef CONFIG_DEBUG_LOCK_ALLOC - struct lockdep_map dep_map; -#endif -} raw_spinlock_t; - -#define SPINLOCK_MAGIC 0xdead4ead - -#define SPINLOCK_OWNER_INIT ((void *)-1L) - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -# define RAW_SPIN_DEP_MAP_INIT(lockname) \ - .dep_map = { \ - .name = #lockname, \ - .wait_type_inner = LD_WAIT_SPIN, \ - } -# define SPIN_DEP_MAP_INIT(lockname) \ - .dep_map = { \ - .name = #lockname, \ - .wait_type_inner = LD_WAIT_CONFIG, \ - } -#else -# define RAW_SPIN_DEP_MAP_INIT(lockname) -# define SPIN_DEP_MAP_INIT(lockname) -#endif - -#ifdef CONFIG_DEBUG_SPINLOCK -# define SPIN_DEBUG_INIT(lockname) \ - .magic = SPINLOCK_MAGIC, \ - .owner_cpu = -1, \ - .owner = SPINLOCK_OWNER_INIT, -#else -# define SPIN_DEBUG_INIT(lockname) -#endif - -#define __RAW_SPIN_LOCK_INITIALIZER(lockname) \ - { \ - .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ - SPIN_DEBUG_INIT(lockname) \ - RAW_SPIN_DEP_MAP_INIT(lockname) } - -#define __RAW_SPIN_LOCK_UNLOCKED(lockname) \ - (raw_spinlock_t) __RAW_SPIN_LOCK_INITIALIZER(lockname) - -#define DEFINE_RAW_SPINLOCK(x) raw_spinlock_t x = __RAW_SPIN_LOCK_UNLOCKED(x) - -typedef struct spinlock { - union { - struct raw_spinlock rlock; - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -# define LOCK_PADSIZE (offsetof(struct raw_spinlock, dep_map)) - struct { - u8 __padding[LOCK_PADSIZE]; - struct lockdep_map dep_map; - }; -#endif - }; -} spinlock_t; - -#define ___SPIN_LOCK_INITIALIZER(lockname) \ - { \ - .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ - SPIN_DEBUG_INIT(lockname) \ - SPIN_DEP_MAP_INIT(lockname) } - -#define __SPIN_LOCK_INITIALIZER(lockname) \ - { { .rlock = ___SPIN_LOCK_INITIALIZER(lockname) } } - -#define __SPIN_LOCK_UNLOCKED(lockname) \ - (spinlock_t) __SPIN_LOCK_INITIALIZER(lockname) - -#define DEFINE_SPINLOCK(x) spinlock_t x = __SPIN_LOCK_UNLOCKED(x) - -#include - #endif /* __LINUX_SPINLOCK_TYPES_H */ diff --git a/include/linux/spinlock_types_nort.h b/include/linux/spinlock_types_nort.h new file mode 100644 index 0000000000000..e4549f0dd197b --- /dev/null +++ b/include/linux/spinlock_types_nort.h @@ -0,0 +1,39 @@ +#ifndef __LINUX_SPINLOCK_TYPES_NORT_H +#define __LINUX_SPINLOCK_TYPES_NORT_H + +#ifndef __LINUX_SPINLOCK_TYPES_H +#error "Do not include directly. Include spinlock_types.h instead" +#endif + +/* + * The non RT version maps spinlocks to raw_spinlocks + */ +typedef struct spinlock { + union { + struct raw_spinlock rlock; + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +# define LOCK_PADSIZE (offsetof(struct raw_spinlock, dep_map)) + struct { + u8 __padding[LOCK_PADSIZE]; + struct lockdep_map dep_map; + }; +#endif + }; +} spinlock_t; + +#define ___SPIN_LOCK_INITIALIZER(lockname) \ +{ \ + .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ + SPIN_DEBUG_INIT(lockname) \ + SPIN_DEP_MAP_INIT(lockname) } + +#define __SPIN_LOCK_INITIALIZER(lockname) \ + { { .rlock = ___SPIN_LOCK_INITIALIZER(lockname) } } + +#define __SPIN_LOCK_UNLOCKED(lockname) \ + (spinlock_t) __SPIN_LOCK_INITIALIZER(lockname) + +#define DEFINE_SPINLOCK(x) spinlock_t x = __SPIN_LOCK_UNLOCKED(x) + +#endif diff --git a/include/linux/spinlock_types_raw.h b/include/linux/spinlock_types_raw.h new file mode 100644 index 0000000000000..1d4a180e983d7 --- /dev/null +++ b/include/linux/spinlock_types_raw.h @@ -0,0 +1,65 @@ +#ifndef __LINUX_SPINLOCK_TYPES_RAW_H +#define __LINUX_SPINLOCK_TYPES_RAW_H + +#include + +#if defined(CONFIG_SMP) +# include +#else +# include +#endif + +#include + +typedef struct raw_spinlock { + arch_spinlock_t raw_lock; +#ifdef CONFIG_DEBUG_SPINLOCK + unsigned int magic, owner_cpu; + void *owner; +#endif +#ifdef CONFIG_DEBUG_LOCK_ALLOC + struct lockdep_map dep_map; +#endif +} raw_spinlock_t; + +#define SPINLOCK_MAGIC 0xdead4ead + +#define SPINLOCK_OWNER_INIT ((void *)-1L) + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +# define RAW_SPIN_DEP_MAP_INIT(lockname) \ + .dep_map = { \ + .name = #lockname, \ + .wait_type_inner = LD_WAIT_SPIN, \ + } +# define SPIN_DEP_MAP_INIT(lockname) \ + .dep_map = { \ + .name = #lockname, \ + .wait_type_inner = LD_WAIT_CONFIG, \ + } +#else +# define RAW_SPIN_DEP_MAP_INIT(lockname) +# define SPIN_DEP_MAP_INIT(lockname) +#endif + +#ifdef CONFIG_DEBUG_SPINLOCK +# define SPIN_DEBUG_INIT(lockname) \ + .magic = SPINLOCK_MAGIC, \ + .owner_cpu = -1, \ + .owner = SPINLOCK_OWNER_INIT, +#else +# define SPIN_DEBUG_INIT(lockname) +#endif + +#define __RAW_SPIN_LOCK_INITIALIZER(lockname) \ +{ \ + .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \ + SPIN_DEBUG_INIT(lockname) \ + RAW_SPIN_DEP_MAP_INIT(lockname) } + +#define __RAW_SPIN_LOCK_UNLOCKED(lockname) \ + (raw_spinlock_t) __RAW_SPIN_LOCK_INITIALIZER(lockname) + +#define DEFINE_RAW_SPINLOCK(x) raw_spinlock_t x = __RAW_SPIN_LOCK_UNLOCKED(x) + +#endif diff --git a/include/linux/spinlock_types_rt.h b/include/linux/spinlock_types_rt.h new file mode 100644 index 0000000000000..446da786e5d5c --- /dev/null +++ b/include/linux/spinlock_types_rt.h @@ -0,0 +1,38 @@ +// SPDX-License-Identifier: GPL-2.0-only +#ifndef __LINUX_SPINLOCK_TYPES_RT_H +#define __LINUX_SPINLOCK_TYPES_RT_H + +#ifndef __LINUX_SPINLOCK_TYPES_H +#error "Do not include directly. Include spinlock_types.h instead" +#endif + +#include + +/* + * PREEMPT_RT: spinlocks - an RT mutex plus lock-break field: + */ +typedef struct spinlock { + struct rt_mutex lock; + unsigned int break_lock; +#ifdef CONFIG_DEBUG_LOCK_ALLOC + struct lockdep_map dep_map; +#endif +} spinlock_t; + +#define __RT_SPIN_INITIALIZER(name) \ + { \ + .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(name.wait_lock), \ + .save_state = 1, \ + } +/* +.wait_list = PLIST_HEAD_INIT_RAW((name).lock.wait_list, (name).lock.wait_lock) +*/ + +#define __SPIN_LOCK_UNLOCKED(name) \ + { .lock = __RT_SPIN_INITIALIZER(name.lock), \ + SPIN_DEP_MAP_INIT(name) } + +#define DEFINE_SPINLOCK(name) \ + spinlock_t name = __SPIN_LOCK_UNLOCKED(name) + +#endif diff --git a/include/linux/spinlock_types_up.h b/include/linux/spinlock_types_up.h index c09b6407ae1b3..b0243ba07fb78 100644 --- a/include/linux/spinlock_types_up.h +++ b/include/linux/spinlock_types_up.h @@ -1,10 +1,6 @@ #ifndef __LINUX_SPINLOCK_TYPES_UP_H #define __LINUX_SPINLOCK_TYPES_UP_H -#ifndef __LINUX_SPINLOCK_TYPES_H -# error "please don't include this file directly" -#endif - /* * include/linux/spinlock_types_up.h - spinlock type definitions for UP * diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h index 76d8b09384a7a..30577c3aecf81 100644 --- a/include/linux/stop_machine.h +++ b/include/linux/stop_machine.h @@ -24,6 +24,7 @@ typedef int (*cpu_stop_fn_t)(void *arg); struct cpu_stop_work { struct list_head list; /* cpu_stopper->works */ cpu_stop_fn_t fn; + unsigned long caller; void *arg; struct cpu_stop_done *done; }; @@ -36,6 +37,8 @@ void stop_machine_park(int cpu); void stop_machine_unpark(int cpu); void stop_machine_yield(const struct cpumask *cpumask); +extern void print_stop_info(const char *log_lvl, struct task_struct *task); + #else /* CONFIG_SMP */ #include @@ -80,6 +83,8 @@ static inline bool stop_one_cpu_nowait(unsigned int cpu, return false; } +static inline void print_stop_info(const char *log_lvl, struct task_struct *task) { } + #endif /* CONFIG_SMP */ /* diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index e93e249a4e9bf..c88b9cecc78aa 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -97,7 +97,17 @@ static inline int test_ti_thread_flag(struct thread_info *ti, int flag) #define test_thread_flag(flag) \ test_ti_thread_flag(current_thread_info(), flag) -#define tif_need_resched() test_thread_flag(TIF_NEED_RESCHED) +#ifdef CONFIG_PREEMPT_LAZY +#define tif_need_resched() (test_thread_flag(TIF_NEED_RESCHED) || \ + test_thread_flag(TIF_NEED_RESCHED_LAZY)) +#define tif_need_resched_now() (test_thread_flag(TIF_NEED_RESCHED)) +#define tif_need_resched_lazy() test_thread_flag(TIF_NEED_RESCHED_LAZY)) + +#else +#define tif_need_resched() test_thread_flag(TIF_NEED_RESCHED) +#define tif_need_resched_now() test_thread_flag(TIF_NEED_RESCHED) +#define tif_need_resched_lazy() 0 +#endif #ifndef CONFIG_HAVE_ARCH_WITHIN_STACK_FRAMES static inline int arch_within_stack_frames(const void * const stack, diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h index 5c69433540494..75e3acd596358 100644 --- a/include/linux/trace_events.h +++ b/include/linux/trace_events.h @@ -67,6 +67,8 @@ struct trace_entry { unsigned char flags; unsigned char preempt_count; int pid; + unsigned char migrate_disable; + unsigned char preempt_lazy_count; }; #define TRACE_EVENT_TYPE_MAX \ diff --git a/include/linux/u64_stats_sync.h b/include/linux/u64_stats_sync.h index c6abb79501b33..72bf38b97df88 100644 --- a/include/linux/u64_stats_sync.h +++ b/include/linux/u64_stats_sync.h @@ -66,7 +66,7 @@ #include struct u64_stats_sync { -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) +#if BITS_PER_LONG==32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) seqcount_t seq; #endif }; @@ -117,22 +117,26 @@ static inline void u64_stats_inc(u64_stats_t *p) static inline void u64_stats_init(struct u64_stats_sync *syncp) { -#if BITS_PER_LONG == 32 && defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) seqcount_init(&syncp->seq); #endif } static inline void u64_stats_update_begin(struct u64_stats_sync *syncp) { -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + preempt_disable(); write_seqcount_begin(&syncp->seq); #endif } static inline void u64_stats_update_end(struct u64_stats_sync *syncp) { -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) write_seqcount_end(&syncp->seq); + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + preempt_enable(); #endif } @@ -141,8 +145,11 @@ u64_stats_update_begin_irqsave(struct u64_stats_sync *syncp) { unsigned long flags = 0; -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) - local_irq_save(flags); +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + preempt_disable(); + else + local_irq_save(flags); write_seqcount_begin(&syncp->seq); #endif return flags; @@ -152,15 +159,18 @@ static inline void u64_stats_update_end_irqrestore(struct u64_stats_sync *syncp, unsigned long flags) { -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) write_seqcount_end(&syncp->seq); - local_irq_restore(flags); + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + preempt_enable(); + else + local_irq_restore(flags); #endif } static inline unsigned int __u64_stats_fetch_begin(const struct u64_stats_sync *syncp) { -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) return read_seqcount_begin(&syncp->seq); #else return 0; @@ -169,7 +179,7 @@ static inline unsigned int __u64_stats_fetch_begin(const struct u64_stats_sync * static inline unsigned int u64_stats_fetch_begin(const struct u64_stats_sync *syncp) { -#if BITS_PER_LONG==32 && !defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (!defined(CONFIG_SMP) && !defined(CONFIG_PREEMPT_RT)) preempt_disable(); #endif return __u64_stats_fetch_begin(syncp); @@ -178,7 +188,7 @@ static inline unsigned int u64_stats_fetch_begin(const struct u64_stats_sync *sy static inline bool __u64_stats_fetch_retry(const struct u64_stats_sync *syncp, unsigned int start) { -#if BITS_PER_LONG==32 && defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT)) return read_seqcount_retry(&syncp->seq, start); #else return false; @@ -188,7 +198,7 @@ static inline bool __u64_stats_fetch_retry(const struct u64_stats_sync *syncp, static inline bool u64_stats_fetch_retry(const struct u64_stats_sync *syncp, unsigned int start) { -#if BITS_PER_LONG==32 && !defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && (!defined(CONFIG_SMP) && !defined(CONFIG_PREEMPT_RT)) preempt_enable(); #endif return __u64_stats_fetch_retry(syncp, start); @@ -202,7 +212,9 @@ static inline bool u64_stats_fetch_retry(const struct u64_stats_sync *syncp, */ static inline unsigned int u64_stats_fetch_begin_irq(const struct u64_stats_sync *syncp) { -#if BITS_PER_LONG==32 && !defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && defined(CONFIG_PREEMPT_RT) + preempt_disable(); +#elif BITS_PER_LONG == 32 && !defined(CONFIG_SMP) local_irq_disable(); #endif return __u64_stats_fetch_begin(syncp); @@ -211,7 +223,9 @@ static inline unsigned int u64_stats_fetch_begin_irq(const struct u64_stats_sync static inline bool u64_stats_fetch_retry_irq(const struct u64_stats_sync *syncp, unsigned int start) { -#if BITS_PER_LONG==32 && !defined(CONFIG_SMP) +#if BITS_PER_LONG == 32 && defined(CONFIG_PREEMPT_RT) + preempt_enable(); +#elif BITS_PER_LONG == 32 && !defined(CONFIG_SMP) local_irq_enable(); #endif return __u64_stats_fetch_retry(syncp, start); diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index 7557c1070fd7f..ed4b631843276 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -63,7 +63,9 @@ DECLARE_PER_CPU(struct vm_event_state, vm_event_states); */ static inline void __count_vm_event(enum vm_event_item item) { + preempt_disable_rt(); raw_cpu_inc(vm_event_states.event[item]); + preempt_enable_rt(); } static inline void count_vm_event(enum vm_event_item item) @@ -73,7 +75,9 @@ static inline void count_vm_event(enum vm_event_item item) static inline void __count_vm_events(enum vm_event_item item, long delta) { + preempt_disable_rt(); raw_cpu_add(vm_event_states.event[item], delta); + preempt_enable_rt(); } static inline void count_vm_events(enum vm_event_item item, long delta) diff --git a/include/linux/wait.h b/include/linux/wait.h index 27fb99cfeb026..93b42387b4c63 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -10,6 +10,7 @@ #include #include +#include typedef struct wait_queue_entry wait_queue_entry_t; diff --git a/include/linux/ww_mutex.h b/include/linux/ww_mutex.h index 850424e5d0306..8ef2feb0d8fef 100644 --- a/include/linux/ww_mutex.h +++ b/include/linux/ww_mutex.h @@ -28,6 +28,14 @@ struct ww_class { unsigned int is_wait_die; }; +struct ww_mutex { + struct mutex base; + struct ww_acquire_ctx *ctx; +#ifdef CONFIG_DEBUG_MUTEXES + struct ww_class *ww_class; +#endif +}; + struct ww_acquire_ctx { struct task_struct *task; unsigned long stamp; diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h index 1424e02cef90c..163f8415e5db5 100644 --- a/include/net/gen_stats.h +++ b/include/net/gen_stats.h @@ -6,6 +6,7 @@ #include #include #include +#include /* Note: this used to be in include/uapi/linux/gen_stats.h */ struct gnet_stats_basic_packed { @@ -42,15 +43,15 @@ int gnet_stats_start_copy_compat(struct sk_buff *skb, int type, spinlock_t *lock, struct gnet_dump *d, int padattr); -int gnet_stats_copy_basic(const seqcount_t *running, +int gnet_stats_copy_basic(net_seqlock_t *running, struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b); -void __gnet_stats_copy_basic(const seqcount_t *running, +void __gnet_stats_copy_basic(net_seqlock_t *running, struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b); -int gnet_stats_copy_basic_hw(const seqcount_t *running, +int gnet_stats_copy_basic_hw(net_seqlock_t *running, struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b); @@ -70,13 +71,13 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct net_rate_estimator __rcu **rate_est, spinlock_t *lock, - seqcount_t *running, struct nlattr *opt); + net_seqlock_t *running, struct nlattr *opt); void gen_kill_estimator(struct net_rate_estimator __rcu **ptr); int gen_replace_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct net_rate_estimator __rcu **ptr, spinlock_t *lock, - seqcount_t *running, struct nlattr *opt); + net_seqlock_t *running, struct nlattr *opt); bool gen_estimator_active(struct net_rate_estimator __rcu **ptr); bool gen_estimator_read(struct net_rate_estimator __rcu **ptr, struct gnet_stats_rate_est64 *sample); diff --git a/include/net/net_seq_lock.h b/include/net/net_seq_lock.h new file mode 100644 index 0000000000000..67710bace7418 --- /dev/null +++ b/include/net/net_seq_lock.h @@ -0,0 +1,15 @@ +#ifndef __NET_NET_SEQ_LOCK_H__ +#define __NET_NET_SEQ_LOCK_H__ + +#ifdef CONFIG_PREEMPT_RT +# define net_seqlock_t seqlock_t +# define net_seq_begin(__r) read_seqbegin(__r) +# define net_seq_retry(__r, __s) read_seqretry(__r, __s) + +#else +# define net_seqlock_t seqcount_t +# define net_seq_begin(__r) read_seqcount_begin(__r) +# define net_seq_retry(__r, __s) read_seqcount_retry(__r, __s) +#endif + +#endif diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index d60e7c39d60c7..40be4443b6bdb 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -100,7 +101,7 @@ struct Qdisc { struct sk_buff_head gso_skb ____cacheline_aligned_in_smp; struct qdisc_skb_head q; struct gnet_stats_basic_packed bstats; - seqcount_t running; + net_seqlock_t running; struct gnet_stats_queue qstats; unsigned long state; struct Qdisc *next_sched; @@ -138,7 +139,11 @@ static inline bool qdisc_is_running(struct Qdisc *qdisc) { if (qdisc->flags & TCQ_F_NOLOCK) return spin_is_locked(&qdisc->seqlock); +#ifdef CONFIG_PREEMPT_RT + return spin_is_locked(&qdisc->running.lock) ? true : false; +#else return (raw_read_seqcount(&qdisc->running) & 1) ? true : false; +#endif } static inline bool qdisc_is_percpu_stats(const struct Qdisc *q) @@ -162,17 +167,35 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc) } else if (qdisc_is_running(qdisc)) { return false; } +#ifdef CONFIG_PREEMPT_RT + if (spin_trylock(&qdisc->running.lock)) { + seqcount_t *s = &qdisc->running.seqcount.seqcount; + /* + * Variant of write_seqcount_t_begin() telling lockdep that a + * trylock was attempted. + */ + raw_write_seqcount_t_begin(s); + seqcount_acquire(&s->dep_map, 0, 1, _RET_IP_); + return true; + } + return false; +#else /* Variant of write_seqcount_begin() telling lockdep a trylock * was attempted. */ raw_write_seqcount_begin(&qdisc->running); seqcount_acquire(&qdisc->running.dep_map, 0, 1, _RET_IP_); return true; +#endif } static inline void qdisc_run_end(struct Qdisc *qdisc) { +#ifdef CONFIG_PREEMPT_RT + write_sequnlock(&qdisc->running); +#else write_seqcount_end(&qdisc->running); +#endif if (qdisc->flags & TCQ_F_NOLOCK) spin_unlock(&qdisc->seqlock); } @@ -547,7 +570,7 @@ static inline spinlock_t *qdisc_root_sleeping_lock(const struct Qdisc *qdisc) return qdisc_lock(root); } -static inline seqcount_t *qdisc_root_sleeping_running(const struct Qdisc *qdisc) +static inline net_seqlock_t *qdisc_root_sleeping_running(const struct Qdisc *qdisc) { struct Qdisc *root = qdisc_root_sleeping(qdisc); diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index fec25b9cfbafa..f9bf9d8f30b89 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -646,6 +646,18 @@ DECLARE_TRACE(sched_update_nr_running_tp, TP_PROTO(struct rq *rq, int change), TP_ARGS(rq, change)); +DECLARE_TRACE(sched_migrate_disable_tp, + TP_PROTO(struct task_struct *p), + TP_ARGS(p)); + +DECLARE_TRACE(sched_migrate_enable_tp, + TP_PROTO(struct task_struct *p), + TP_ARGS(p)); + +DECLARE_TRACE(sched_migrate_pull_tp, + TP_PROTO(struct task_struct *p), + TP_ARGS(p)); + #endif /* _TRACE_SCHED_H */ /* This part must be outside protection */ diff --git a/init/Kconfig b/init/Kconfig index d6a0b31b13dc9..c48887283f88a 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -682,7 +682,8 @@ config IKHEADERS config LOG_BUF_SHIFT int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" - range 12 25 + range 12 25 if !H8300 + range 12 19 if H8300 default 17 depends on PRINTK help @@ -964,6 +965,7 @@ config CFS_BANDWIDTH config RT_GROUP_SCHED bool "Group scheduling for SCHED_RR/FIFO" depends on CGROUP_SCHED + depends on !PREEMPT_RT default n help This feature lets you explicitly allocate real CPU bandwidth @@ -1871,6 +1873,7 @@ choice config SLAB bool "SLAB" + depends on !PREEMPT_RT select HAVE_HARDENED_USERCOPY_ALLOCATOR help The regular slab allocator that is established and known to work @@ -1891,6 +1894,7 @@ config SLUB config SLOB depends on EXPERT bool "SLOB (Simple Allocator)" + depends on !PREEMPT_RT help SLOB replaces the stock allocator with a drastically simpler allocator. SLOB is generally more space efficient but @@ -1957,7 +1961,7 @@ config SHUFFLE_PAGE_ALLOCATOR config SLUB_CPU_PARTIAL default y - depends on SLUB && SMP + depends on SLUB && SMP && !PREEMPT_RT bool "SLUB per cpu partial cache" help Per cpu partial caches accelerate objects allocation and freeing diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks index 3de8fd11873b4..4198f0273ecdc 100644 --- a/kernel/Kconfig.locks +++ b/kernel/Kconfig.locks @@ -251,7 +251,7 @@ config ARCH_USE_QUEUED_RWLOCKS config QUEUED_RWLOCKS def_bool y if ARCH_USE_QUEUED_RWLOCKS - depends on SMP + depends on SMP && !PREEMPT_RT config ARCH_HAS_MMIOWB bool diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt index bf82259cff965..c1203c14fee92 100644 --- a/kernel/Kconfig.preempt +++ b/kernel/Kconfig.preempt @@ -1,5 +1,11 @@ # SPDX-License-Identifier: GPL-2.0-only +config HAVE_PREEMPT_LAZY + bool + +config PREEMPT_LAZY + def_bool y if HAVE_PREEMPT_LAZY && PREEMPT_RT + choice prompt "Preemption Model" default PREEMPT_NONE diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 642415b8c3c9e..daca5e74d75ed 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -345,7 +345,7 @@ void cpuset_read_unlock(void) percpu_up_read(&cpuset_rwsem); } -static DEFINE_SPINLOCK(callback_lock); +static DEFINE_RAW_SPINLOCK(callback_lock); static struct workqueue_struct *cpuset_migrate_mm_wq; @@ -1257,7 +1257,7 @@ static int update_parent_subparts_cpumask(struct cpuset *cpuset, int cmd, * Newly added CPUs will be removed from effective_cpus and * newly deleted ones will be added back to effective_cpus. */ - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); if (adding) { cpumask_or(parent->subparts_cpus, parent->subparts_cpus, tmp->addmask); @@ -1276,7 +1276,7 @@ static int update_parent_subparts_cpumask(struct cpuset *cpuset, int cmd, } parent->nr_subparts_cpus = cpumask_weight(parent->subparts_cpus); - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); return cmd == partcmd_update; } @@ -1381,7 +1381,7 @@ static void update_cpumasks_hier(struct cpuset *cs, struct tmpmasks *tmp) continue; rcu_read_unlock(); - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cpumask_copy(cp->effective_cpus, tmp->new_cpus); if (cp->nr_subparts_cpus && @@ -1412,7 +1412,7 @@ static void update_cpumasks_hier(struct cpuset *cs, struct tmpmasks *tmp) = cpumask_weight(cp->subparts_cpus); } } - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); WARN_ON(!is_in_v2_mode() && !cpumask_equal(cp->cpus_allowed, cp->effective_cpus)); @@ -1530,7 +1530,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, return -EINVAL; } - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed); /* @@ -1541,7 +1541,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, cs->cpus_allowed); cs->nr_subparts_cpus = cpumask_weight(cs->subparts_cpus); } - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); update_cpumasks_hier(cs, &tmp); @@ -1735,9 +1735,9 @@ static void update_nodemasks_hier(struct cpuset *cs, nodemask_t *new_mems) continue; rcu_read_unlock(); - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cp->effective_mems = *new_mems; - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); WARN_ON(!is_in_v2_mode() && !nodes_equal(cp->mems_allowed, cp->effective_mems)); @@ -1805,9 +1805,9 @@ static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs, if (retval < 0) goto done; - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cs->mems_allowed = trialcs->mems_allowed; - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); /* use trialcs->mems_allowed as a temp variable */ update_nodemasks_hier(cs, &trialcs->mems_allowed); @@ -1898,9 +1898,9 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, spread_flag_changed = ((is_spread_slab(cs) != is_spread_slab(trialcs)) || (is_spread_page(cs) != is_spread_page(trialcs))); - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cs->flags = trialcs->flags; - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed) rebuild_sched_domains_locked(); @@ -2409,7 +2409,7 @@ static int cpuset_common_seq_show(struct seq_file *sf, void *v) cpuset_filetype_t type = seq_cft(sf)->private; int ret = 0; - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); switch (type) { case FILE_CPULIST: @@ -2431,7 +2431,7 @@ static int cpuset_common_seq_show(struct seq_file *sf, void *v) ret = -EINVAL; } - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); return ret; } @@ -2744,14 +2744,14 @@ static int cpuset_css_online(struct cgroup_subsys_state *css) cpuset_inc(); - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); if (is_in_v2_mode()) { cpumask_copy(cs->effective_cpus, parent->effective_cpus); cs->effective_mems = parent->effective_mems; cs->use_parent_ecpus = true; parent->child_ecpus_count++; } - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); if (!test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags)) goto out_unlock; @@ -2778,12 +2778,12 @@ static int cpuset_css_online(struct cgroup_subsys_state *css) } rcu_read_unlock(); - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cs->mems_allowed = parent->mems_allowed; cs->effective_mems = parent->mems_allowed; cpumask_copy(cs->cpus_allowed, parent->cpus_allowed); cpumask_copy(cs->effective_cpus, parent->cpus_allowed); - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); out_unlock: percpu_up_write(&cpuset_rwsem); put_online_cpus(); @@ -2839,7 +2839,7 @@ static void cpuset_css_free(struct cgroup_subsys_state *css) static void cpuset_bind(struct cgroup_subsys_state *root_css) { percpu_down_write(&cpuset_rwsem); - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); if (is_in_v2_mode()) { cpumask_copy(top_cpuset.cpus_allowed, cpu_possible_mask); @@ -2850,7 +2850,7 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css) top_cpuset.mems_allowed = top_cpuset.effective_mems; } - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); percpu_up_write(&cpuset_rwsem); } @@ -2947,12 +2947,12 @@ hotplug_update_tasks_legacy(struct cpuset *cs, { bool is_empty; - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cpumask_copy(cs->cpus_allowed, new_cpus); cpumask_copy(cs->effective_cpus, new_cpus); cs->mems_allowed = *new_mems; cs->effective_mems = *new_mems; - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); /* * Don't call update_tasks_cpumask() if the cpuset becomes empty, @@ -2989,10 +2989,10 @@ hotplug_update_tasks(struct cpuset *cs, if (nodes_empty(*new_mems)) *new_mems = parent_cs(cs)->effective_mems; - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); cpumask_copy(cs->effective_cpus, new_cpus); cs->effective_mems = *new_mems; - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); if (cpus_updated) update_tasks_cpumask(cs); @@ -3147,7 +3147,7 @@ static void cpuset_hotplug_workfn(struct work_struct *work) /* synchronize cpus_allowed to cpu_active_mask */ if (cpus_updated) { - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); if (!on_dfl) cpumask_copy(top_cpuset.cpus_allowed, &new_cpus); /* @@ -3167,17 +3167,17 @@ static void cpuset_hotplug_workfn(struct work_struct *work) } } cpumask_copy(top_cpuset.effective_cpus, &new_cpus); - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); /* we don't mess with cpumasks of tasks in top_cpuset */ } /* synchronize mems_allowed to N_MEMORY */ if (mems_updated) { - spin_lock_irq(&callback_lock); + raw_spin_lock_irq(&callback_lock); if (!on_dfl) top_cpuset.mems_allowed = new_mems; top_cpuset.effective_mems = new_mems; - spin_unlock_irq(&callback_lock); + raw_spin_unlock_irq(&callback_lock); update_tasks_nodemask(&top_cpuset); } @@ -3278,11 +3278,11 @@ void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask) { unsigned long flags; - spin_lock_irqsave(&callback_lock, flags); + raw_spin_lock_irqsave(&callback_lock, flags); rcu_read_lock(); guarantee_online_cpus(task_cs(tsk), pmask); rcu_read_unlock(); - spin_unlock_irqrestore(&callback_lock, flags); + raw_spin_unlock_irqrestore(&callback_lock, flags); } /** @@ -3343,11 +3343,11 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk) nodemask_t mask; unsigned long flags; - spin_lock_irqsave(&callback_lock, flags); + raw_spin_lock_irqsave(&callback_lock, flags); rcu_read_lock(); guarantee_online_mems(task_cs(tsk), &mask); rcu_read_unlock(); - spin_unlock_irqrestore(&callback_lock, flags); + raw_spin_unlock_irqrestore(&callback_lock, flags); return mask; } @@ -3439,14 +3439,14 @@ bool __cpuset_node_allowed(int node, gfp_t gfp_mask) return true; /* Not hardwall and node outside mems_allowed: scan up cpusets */ - spin_lock_irqsave(&callback_lock, flags); + raw_spin_lock_irqsave(&callback_lock, flags); rcu_read_lock(); cs = nearest_hardwall_ancestor(task_cs(current)); allowed = node_isset(node, cs->mems_allowed); rcu_read_unlock(); - spin_unlock_irqrestore(&callback_lock, flags); + raw_spin_unlock_irqrestore(&callback_lock, flags); return allowed; } diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c index d51175cedfca4..b424f3157b345 100644 --- a/kernel/cgroup/rstat.c +++ b/kernel/cgroup/rstat.c @@ -149,8 +149,9 @@ static void cgroup_rstat_flush_locked(struct cgroup *cgrp, bool may_sleep) raw_spinlock_t *cpu_lock = per_cpu_ptr(&cgroup_rstat_cpu_lock, cpu); struct cgroup *pos = NULL; + unsigned long flags; - raw_spin_lock(cpu_lock); + raw_spin_lock_irqsave(cpu_lock, flags); while ((pos = cgroup_rstat_cpu_pop_updated(pos, cgrp, cpu))) { struct cgroup_subsys_state *css; @@ -162,7 +163,7 @@ static void cgroup_rstat_flush_locked(struct cgroup *cgrp, bool may_sleep) css->ss->css_rstat_flush(css, cpu); rcu_read_unlock(); } - raw_spin_unlock(cpu_lock); + raw_spin_unlock_irqrestore(cpu_lock, flags); /* if @may_sleep, play nice and yield if necessary */ if (may_sleep && (need_resched() || diff --git a/kernel/cpu.c b/kernel/cpu.c index 6ff2578ecf17d..fa535eaa48265 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -1602,7 +1602,7 @@ static struct cpuhp_step cpuhp_hp_states[] = { .name = "ap:online", }, /* - * Handled on controll processor until the plugged processor manages + * Handled on control processor until the plugged processor manages * this itself. */ [CPUHP_TEARDOWN_CPU] = { @@ -1611,6 +1611,13 @@ static struct cpuhp_step cpuhp_hp_states[] = { .teardown.single = takedown_cpu, .cant_stop = true, }, + + [CPUHP_AP_SCHED_WAIT_EMPTY] = { + .name = "sched:waitempty", + .startup.single = NULL, + .teardown.single = sched_cpu_wait_empty, + }, + /* Handle smpboot threads park/unpark */ [CPUHP_AP_SMPBOOT_THREADS] = { .name = "smpboot/threads:online", diff --git a/kernel/entry/common.c b/kernel/entry/common.c index 6fdb6105e6d61..adbfcef76991f 100644 --- a/kernel/entry/common.c +++ b/kernel/entry/common.c @@ -148,9 +148,17 @@ static unsigned long exit_to_user_mode_loop(struct pt_regs *regs, local_irq_enable_exit_to_user(ti_work); - if (ti_work & _TIF_NEED_RESCHED) + if (ti_work & _TIF_NEED_RESCHED_MASK) schedule(); +#ifdef ARCH_RT_DELAYS_SIGNAL_SEND + if (unlikely(current->forced_info.si_signo)) { + struct task_struct *t = current; + force_sig_info(&t->forced_info); + t->forced_info.si_signo = 0; + } +#endif + if (ti_work & _TIF_UPROBE) uprobe_notify_resume(regs); @@ -354,7 +362,7 @@ void irqentry_exit_cond_resched(void) rcu_irq_exit_check_preempt(); if (IS_ENABLED(CONFIG_DEBUG_ENTRY)) WARN_ON_ONCE(!on_thread_stack()); - if (need_resched()) + if (should_resched(0)) preempt_schedule_irq(); } } diff --git a/kernel/exit.c b/kernel/exit.c index 733e80f334e71..a1756dcc17f2b 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -151,7 +151,7 @@ static void __exit_signal(struct task_struct *tsk) * Do this under ->siglock, we can race with another thread * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals. */ - flush_sigqueue(&tsk->pending); + flush_task_sigqueue(tsk); tsk->sighand = NULL; spin_unlock(&sighand->siglock); diff --git a/kernel/fork.c b/kernel/fork.c index da8d360fb0326..2cf99526192ea 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -42,6 +42,7 @@ #include #include #include +#include #include #include #include @@ -287,7 +288,7 @@ static inline void free_thread_stack(struct task_struct *tsk) return; } - vfree_atomic(tsk->stack); + vfree(tsk->stack); return; } #endif @@ -687,6 +688,19 @@ void __mmdrop(struct mm_struct *mm) } EXPORT_SYMBOL_GPL(__mmdrop); +#ifdef CONFIG_PREEMPT_RT +/* + * RCU callback for delayed mm drop. Not strictly rcu, but we don't + * want another facility to make this work. + */ +void __mmdrop_delayed(struct rcu_head *rhp) +{ + struct mm_struct *mm = container_of(rhp, struct mm_struct, delayed_drop); + + __mmdrop(mm); +} +#endif + static void mmdrop_async_fn(struct work_struct *work) { struct mm_struct *mm; @@ -728,6 +742,15 @@ void __put_task_struct(struct task_struct *tsk) WARN_ON(refcount_read(&tsk->usage)); WARN_ON(tsk == current); + /* + * Remove function-return probe instances associated with this + * task and put them back on the free list. + */ + kprobe_flush_task(tsk); + + /* Task is done with its stack. */ + put_task_stack(tsk); + cgroup_free(tsk); task_numa_free(tsk, true); security_task_free(tsk); @@ -924,6 +947,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node) tsk->splice_pipe = NULL; tsk->task_frag.page = NULL; tsk->wake_q.next = NULL; + tsk->wake_q_sleeper.next = NULL; account_kernel_stack(tsk, 1); @@ -1970,6 +1994,7 @@ static __latent_entropy struct task_struct *copy_process( spin_lock_init(&p->alloc_lock); init_sigpending(&p->pending); + p->sigqueue_cache = NULL; p->utime = p->stime = p->gtime = 0; #ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME diff --git a/kernel/futex.c b/kernel/futex.c index a5876694a60eb..1ca3581043fea 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -1479,6 +1479,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_ struct task_struct *new_owner; bool postunlock = false; DEFINE_WAKE_Q(wake_q); + DEFINE_WAKE_Q(wake_sleeper_q); int ret = 0; new_owner = rt_mutex_next_owner(&pi_state->pi_mutex); @@ -1538,13 +1539,13 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_pi_state *pi_ pi_state->owner = new_owner; raw_spin_unlock(&new_owner->pi_lock); - postunlock = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q); - + postunlock = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q, + &wake_sleeper_q); out_unlock: raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock); if (postunlock) - rt_mutex_postunlock(&wake_q); + rt_mutex_postunlock(&wake_q, &wake_sleeper_q); return ret; } @@ -2145,6 +2146,16 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags, */ requeue_pi_wake_futex(this, &key2, hb2); continue; + } else if (ret == -EAGAIN) { + /* + * Waiter was woken by timeout or + * signal and has set pi_blocked_on to + * PI_WAKEUP_INPROGRESS before we + * tried to enqueue it on the rtmutex. + */ + this->pi_state = NULL; + put_pi_state(pi_state); + continue; } else if (ret) { /* * rt_mutex_start_proxy_lock() detected a @@ -2830,7 +2841,7 @@ static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, goto no_block; } - rt_mutex_init_waiter(&rt_waiter); + rt_mutex_init_waiter(&rt_waiter, false); /* * On PREEMPT_RT_FULL, when hb->lock becomes an rt_mutex, we must not @@ -3171,7 +3182,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, struct hrtimer_sleeper timeout, *to; struct futex_pi_state *pi_state = NULL; struct rt_mutex_waiter rt_waiter; - struct futex_hash_bucket *hb; + struct futex_hash_bucket *hb, *hb2; union futex_key key2 = FUTEX_KEY_INIT; struct futex_q q = futex_q_init; int res, ret; @@ -3192,7 +3203,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, * The waiter is allocated on our stack, manipulated by the requeue * code while we sleep on uaddr. */ - rt_mutex_init_waiter(&rt_waiter); + rt_mutex_init_waiter(&rt_waiter, false); ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2, FUTEX_WRITE); if (unlikely(ret != 0)) @@ -3223,20 +3234,55 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, /* Queue the futex_q, drop the hb lock, wait for wakeup. */ futex_wait_queue_me(hb, &q, to); - spin_lock(&hb->lock); - ret = handle_early_requeue_pi_wakeup(hb, &q, &key2, to); - spin_unlock(&hb->lock); - if (ret) - goto out; + /* + * On RT we must avoid races with requeue and trying to block + * on two mutexes (hb->lock and uaddr2's rtmutex) by + * serializing access to pi_blocked_on with pi_lock. + */ + raw_spin_lock_irq(¤t->pi_lock); + if (current->pi_blocked_on) { + /* + * We have been requeued or are in the process of + * being requeued. + */ + raw_spin_unlock_irq(¤t->pi_lock); + } else { + /* + * Setting pi_blocked_on to PI_WAKEUP_INPROGRESS + * prevents a concurrent requeue from moving us to the + * uaddr2 rtmutex. After that we can safely acquire + * (and possibly block on) hb->lock. + */ + current->pi_blocked_on = PI_WAKEUP_INPROGRESS; + raw_spin_unlock_irq(¤t->pi_lock); + + spin_lock(&hb->lock); + + /* + * Clean up pi_blocked_on. We might leak it otherwise + * when we succeeded with the hb->lock in the fast + * path. + */ + raw_spin_lock_irq(¤t->pi_lock); + current->pi_blocked_on = NULL; + raw_spin_unlock_irq(¤t->pi_lock); + + ret = handle_early_requeue_pi_wakeup(hb, &q, &key2, to); + spin_unlock(&hb->lock); + if (ret) + goto out; + } /* - * In order for us to be here, we know our q.key == key2, and since - * we took the hb->lock above, we also know that futex_requeue() has - * completed and we no longer have to concern ourselves with a wakeup - * race with the atomic proxy lock acquisition by the requeue code. The - * futex_requeue dropped our key1 reference and incremented our key2 - * reference count. + * In order to be here, we have either been requeued, are in + * the process of being requeued, or requeue successfully + * acquired uaddr2 on our behalf. If pi_blocked_on was + * non-null above, we may be racing with a requeue. Do not + * rely on q->lock_ptr to be hb2->lock until after blocking on + * hb->lock or hb2->lock. The futex_requeue dropped our key1 + * reference and incremented our key2 reference count. */ + hb2 = hash_futex(&key2); /* Check if the requeue code acquired the second futex for us. */ if (!q.rt_waiter) { @@ -3245,7 +3291,8 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, * did a lock-steal - fix up the PI-state in that case. */ if (q.pi_state && (q.pi_state->owner != current)) { - spin_lock(q.lock_ptr); + spin_lock(&hb2->lock); + BUG_ON(&hb2->lock != q.lock_ptr); ret = fixup_pi_state_owner(uaddr2, &q, current); if (ret && rt_mutex_owner(&q.pi_state->pi_mutex) == current) { pi_state = q.pi_state; @@ -3256,7 +3303,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, * the requeue_pi() code acquired for us. */ put_pi_state(q.pi_state); - spin_unlock(q.lock_ptr); + spin_unlock(&hb2->lock); } } else { struct rt_mutex *pi_mutex; @@ -3270,7 +3317,8 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, pi_mutex = &q.pi_state->pi_mutex; ret = rt_mutex_wait_proxy_lock(pi_mutex, to, &rt_waiter); - spin_lock(q.lock_ptr); + spin_lock(&hb2->lock); + BUG_ON(&hb2->lock != q.lock_ptr); if (ret && !rt_mutex_cleanup_proxy_lock(pi_mutex, &rt_waiter)) ret = 0; diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 762a928e18f92..7929fcdb78174 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -192,10 +192,16 @@ irqreturn_t handle_irq_event_percpu(struct irq_desc *desc) { irqreturn_t retval; unsigned int flags = 0; + struct pt_regs *regs = get_irq_regs(); + u64 ip = regs ? instruction_pointer(regs) : 0; retval = __handle_irq_event_percpu(desc, &flags); - add_interrupt_randomness(desc->irq_data.irq, flags); +#ifdef CONFIG_PREEMPT_RT + desc->random_ip = ip; +#else + add_interrupt_randomness(desc->irq_data.irq, flags, ip); +#endif if (!noirqdebug) note_interrupt(desc, retval); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 5df903fccb609..881e13ec97093 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -1175,6 +1175,12 @@ static int irq_thread(void *data) if (action_ret == IRQ_WAKE_THREAD) irq_wake_secondary(desc, action); + if (IS_ENABLED(CONFIG_PREEMPT_RT)) { + migrate_disable(); + add_interrupt_randomness(action->irq, 0, + desc->random_ip ^ (unsigned long) action); + migrate_enable(); + } wake_threads_waitq(desc); } @@ -2711,7 +2717,7 @@ EXPORT_SYMBOL_GPL(irq_get_irqchip_state); * This call sets the internal irqchip state of an interrupt, * depending on the value of @which. * - * This function should be called with preemption disabled if the + * This function should be called with migration disabled if the * interrupt controller has per-cpu registers. */ int irq_set_irqchip_state(unsigned int irq, enum irqchip_irq_state which, diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index f865e5f4d3825..dc7311dd74b11 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -443,6 +443,10 @@ MODULE_PARM_DESC(noirqdebug, "Disable irq lockup detection when true"); static int __init irqfixup_setup(char *str) { +#ifdef CONFIG_PREEMPT_RT + pr_warn("irqfixup boot option not supported w/ CONFIG_PREEMPT_RT\n"); + return 1; +#endif irqfixup = 1; printk(KERN_WARNING "Misrouted IRQ fixup support enabled.\n"); printk(KERN_WARNING "This may impact system performance.\n"); @@ -455,6 +459,10 @@ module_param(irqfixup, int, 0644); static int __init irqpoll_setup(char *str) { +#ifdef CONFIG_PREEMPT_RT + pr_warn("irqpoll boot option not supported w/ CONFIG_PREEMPT_RT\n"); + return 1; +#endif irqfixup = 2; printk(KERN_WARNING "Misrouted IRQ fixup and polling support " "enabled\n"); diff --git a/kernel/irq_work.c b/kernel/irq_work.c index eca83965b631a..8183d30e1bb1c 100644 --- a/kernel/irq_work.c +++ b/kernel/irq_work.c @@ -18,6 +18,7 @@ #include #include #include +#include #include @@ -52,13 +53,19 @@ void __weak arch_irq_work_raise(void) /* Enqueue on current CPU, work must already be claimed and preempt disabled */ static void __irq_work_queue_local(struct irq_work *work) { + struct llist_head *list; + bool lazy_work, realtime = IS_ENABLED(CONFIG_PREEMPT_RT); + + lazy_work = atomic_read(&work->flags) & IRQ_WORK_LAZY; + /* If the work is "lazy", handle it from next tick if any */ - if (atomic_read(&work->flags) & IRQ_WORK_LAZY) { - if (llist_add(&work->llnode, this_cpu_ptr(&lazy_list)) && - tick_nohz_tick_stopped()) - arch_irq_work_raise(); - } else { - if (llist_add(&work->llnode, this_cpu_ptr(&raised_list))) + if (lazy_work || (realtime && !(atomic_read(&work->flags) & IRQ_WORK_HARD_IRQ))) + list = this_cpu_ptr(&lazy_list); + else + list = this_cpu_ptr(&raised_list); + + if (llist_add(&work->llnode, list)) { + if (!lazy_work || tick_nohz_tick_stopped()) arch_irq_work_raise(); } } @@ -102,7 +109,13 @@ bool irq_work_queue_on(struct irq_work *work, int cpu) if (cpu != smp_processor_id()) { /* Arch remote IPI send/receive backend aren't NMI safe */ WARN_ON_ONCE(in_nmi()); - __smp_call_single_queue(cpu, &work->llnode); + + if (IS_ENABLED(CONFIG_PREEMPT_RT) && !(atomic_read(&work->flags) & IRQ_WORK_HARD_IRQ)) { + if (llist_add(&work->llnode, &per_cpu(lazy_list, cpu))) + arch_send_call_function_single_ipi(cpu); + } else { + __smp_call_single_queue(cpu, &work->llnode); + } } else { __irq_work_queue_local(work); } @@ -120,9 +133,8 @@ bool irq_work_needs_cpu(void) raised = this_cpu_ptr(&raised_list); lazy = this_cpu_ptr(&lazy_list); - if (llist_empty(raised) || arch_irq_work_has_interrupt()) - if (llist_empty(lazy)) - return false; + if (llist_empty(raised) && llist_empty(lazy)) + return false; /* All work should have been flushed before going offline */ WARN_ON_ONCE(cpu_is_offline(smp_processor_id())); @@ -160,8 +172,12 @@ static void irq_work_run_list(struct llist_head *list) struct irq_work *work, *tmp; struct llist_node *llnode; +#ifndef CONFIG_PREEMPT_RT + /* + * nort: On RT IRQ-work may run in SOFTIRQ context. + */ BUG_ON(!irqs_disabled()); - +#endif if (llist_empty(list)) return; @@ -177,7 +193,16 @@ static void irq_work_run_list(struct llist_head *list) void irq_work_run(void) { irq_work_run_list(this_cpu_ptr(&raised_list)); - irq_work_run_list(this_cpu_ptr(&lazy_list)); + if (IS_ENABLED(CONFIG_PREEMPT_RT)) { + /* + * NOTE: we raise softirq via IPI for safety, + * and execute in irq_work_tick() to move the + * overhead from hard to soft irq context. + */ + if (!llist_empty(this_cpu_ptr(&lazy_list))) + raise_softirq(TIMER_SOFTIRQ); + } else + irq_work_run_list(this_cpu_ptr(&lazy_list)); } EXPORT_SYMBOL_GPL(irq_work_run); @@ -187,8 +212,17 @@ void irq_work_tick(void) if (!llist_empty(raised) && !arch_irq_work_has_interrupt()) irq_work_run_list(raised); + + if (!IS_ENABLED(CONFIG_PREEMPT_RT)) + irq_work_run_list(this_cpu_ptr(&lazy_list)); +} + +#if defined(CONFIG_IRQ_WORK) && defined(CONFIG_PREEMPT_RT) +void irq_work_tick_soft(void) +{ irq_work_run_list(this_cpu_ptr(&lazy_list)); } +#endif /* * Synchronize against the irq_work @entry, ensures the entry is not diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c index c19c0dad1ebef..c85754463de6d 100644 --- a/kernel/kexec_core.c +++ b/kernel/kexec_core.c @@ -978,7 +978,6 @@ void crash_kexec(struct pt_regs *regs) old_cpu = atomic_cmpxchg(&panic_cpu, PANIC_CPU_INVALID, this_cpu); if (old_cpu == PANIC_CPU_INVALID) { /* This is the 1st CPU which comes here, so go ahead. */ - printk_safe_flush_on_panic(); __crash_kexec(regs); /* diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c index 35859da8bd4f7..dfff31ed644a6 100644 --- a/kernel/ksysfs.c +++ b/kernel/ksysfs.c @@ -138,6 +138,15 @@ KERNEL_ATTR_RO(vmcoreinfo); #endif /* CONFIG_CRASH_CORE */ +#if defined(CONFIG_PREEMPT_RT) +static ssize_t realtime_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", 1); +} +KERNEL_ATTR_RO(realtime); +#endif + /* whether file capabilities are enabled */ static ssize_t fscaps_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) @@ -228,6 +237,9 @@ static struct attribute * kernel_attrs[] = { #ifndef CONFIG_TINY_RCU &rcu_expedited_attr.attr, &rcu_normal_attr.attr, +#endif +#ifdef CONFIG_PREEMPT_RT + &realtime_attr.attr, #endif NULL }; diff --git a/kernel/locking/Makefile b/kernel/locking/Makefile index 6d11cfb9b41f2..c7fbf737e16ea 100644 --- a/kernel/locking/Makefile +++ b/kernel/locking/Makefile @@ -3,7 +3,7 @@ # and is generally not a function of system call inputs. KCOV_INSTRUMENT := n -obj-y += mutex.o semaphore.o rwsem.o percpu-rwsem.o +obj-y += semaphore.o rwsem.o percpu-rwsem.o # Avoid recursion lockdep -> KCSAN -> ... -> lockdep. KCSAN_SANITIZE_lockdep.o := n @@ -15,19 +15,23 @@ CFLAGS_REMOVE_mutex-debug.o = $(CC_FLAGS_FTRACE) CFLAGS_REMOVE_rtmutex-debug.o = $(CC_FLAGS_FTRACE) endif -obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o obj-$(CONFIG_LOCKDEP) += lockdep.o ifeq ($(CONFIG_PROC_FS),y) obj-$(CONFIG_LOCKDEP) += lockdep_proc.o endif obj-$(CONFIG_SMP) += spinlock.o -obj-$(CONFIG_LOCK_SPIN_ON_OWNER) += osq_lock.o obj-$(CONFIG_PROVE_LOCKING) += spinlock.o obj-$(CONFIG_QUEUED_SPINLOCKS) += qspinlock.o obj-$(CONFIG_RT_MUTEXES) += rtmutex.o obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock_debug.o +ifneq ($(CONFIG_PREEMPT_RT),y) +obj-y += mutex.o +obj-$(CONFIG_LOCK_SPIN_ON_OWNER) += osq_lock.o +obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o +endif +obj-$(CONFIG_PREEMPT_RT) += mutex-rt.o rwsem-rt.o rwlock-rt.o obj-$(CONFIG_QUEUED_RWLOCKS) += qrwlock.o obj-$(CONFIG_LOCK_TORTURE_TEST) += locktorture.o obj-$(CONFIG_WW_MUTEX_SELFTEST) += test-ww_mutex.o diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 2facbbd146ec2..b870708af5810 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -4896,6 +4896,7 @@ static void check_flags(unsigned long flags) } } +#ifndef CONFIG_PREEMPT_RT /* * We dont accurately track softirq state in e.g. * hardirq contexts (such as on 4KSTACKS), so only @@ -4910,6 +4911,7 @@ static void check_flags(unsigned long flags) DEBUG_LOCKS_WARN_ON(!current->softirqs_enabled); } } +#endif if (!debug_locks) print_irqtrace_events(current); diff --git a/kernel/locking/mutex-rt.c b/kernel/locking/mutex-rt.c new file mode 100644 index 0000000000000..2b849e6b9b4ae --- /dev/null +++ b/kernel/locking/mutex-rt.c @@ -0,0 +1,224 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Real-Time Preemption Support + * + * started by Ingo Molnar: + * + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * + * historic credit for proving that Linux spinlocks can be implemented via + * RT-aware mutexes goes to many people: The Pmutex project (Dirk Grambow + * and others) who prototyped it on 2.4 and did lots of comparative + * research and analysis; TimeSys, for proving that you can implement a + * fully preemptible kernel via the use of IRQ threading and mutexes; + * Bill Huey for persuasively arguing on lkml that the mutex model is the + * right one; and to MontaVista, who ported pmutexes to 2.6. + * + * This code is a from-scratch implementation and is not based on pmutexes, + * but the idea of converting spinlocks to mutexes is used here too. + * + * lock debugging, locking tree, deadlock detection: + * + * Copyright (C) 2004, LynuxWorks, Inc., Igor Manyilov, Bill Huey + * Released under the General Public License (GPL). + * + * Includes portions of the generic R/W semaphore implementation from: + * + * Copyright (c) 2001 David Howells (dhowells@redhat.com). + * - Derived partially from idea by Andrea Arcangeli + * - Derived also from comments by Linus + * + * Pending ownership of locks and ownership stealing: + * + * Copyright (C) 2005, Kihon Technologies Inc., Steven Rostedt + * + * (also by Steven Rostedt) + * - Converted single pi_lock to individual task locks. + * + * By Esben Nielsen: + * Doing priority inheritance with help of the scheduler. + * + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner + * - major rework based on Esben Nielsens initial patch + * - replaced thread_info references by task_struct refs + * - removed task->pending_owner dependency + * - BKL drop/reacquire for semaphore style locks to avoid deadlocks + * in the scheduler return path as discussed with Steven Rostedt + * + * Copyright (C) 2006, Kihon Technologies Inc. + * Steven Rostedt + * - debugged and patched Thomas Gleixner's rework. + * - added back the cmpxchg to the rework. + * - turned atomic require back on for SMP. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "rtmutex_common.h" + +/* + * struct mutex functions + */ +void __mutex_do_init(struct mutex *mutex, const char *name, + struct lock_class_key *key) +{ +#ifdef CONFIG_DEBUG_LOCK_ALLOC + /* + * Make sure we are not reinitializing a held lock: + */ + debug_check_no_locks_freed((void *)mutex, sizeof(*mutex)); + lockdep_init_map(&mutex->dep_map, name, key, 0); +#endif + mutex->lock.save_state = 0; +} +EXPORT_SYMBOL(__mutex_do_init); + +static int _mutex_lock_blk_flush(struct mutex *lock, int state) +{ + /* + * Flush blk before ->pi_blocked_on is set. At schedule() time it is too + * late if one of the callbacks needs to acquire a sleeping lock. + */ + if (blk_needs_flush_plug(current)) + blk_schedule_flush_plug(current); + return __rt_mutex_lock_state(&lock->lock, state); +} + +void __lockfunc _mutex_lock(struct mutex *lock) +{ + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); + _mutex_lock_blk_flush(lock, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(_mutex_lock); + +void __lockfunc _mutex_lock_io_nested(struct mutex *lock, int subclass) +{ + int token; + + token = io_schedule_prepare(); + + mutex_acquire_nest(&lock->dep_map, subclass, 0, NULL, _RET_IP_); + __rt_mutex_lock_state(&lock->lock, TASK_UNINTERRUPTIBLE); + + io_schedule_finish(token); +} +EXPORT_SYMBOL_GPL(_mutex_lock_io_nested); + +int __lockfunc _mutex_lock_interruptible(struct mutex *lock) +{ + int ret; + + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); + ret = _mutex_lock_blk_flush(lock, TASK_INTERRUPTIBLE); + if (ret) + mutex_release(&lock->dep_map, _RET_IP_); + return ret; +} +EXPORT_SYMBOL(_mutex_lock_interruptible); + +int __lockfunc _mutex_lock_killable(struct mutex *lock) +{ + int ret; + + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); + ret = _mutex_lock_blk_flush(lock, TASK_KILLABLE); + if (ret) + mutex_release(&lock->dep_map, _RET_IP_); + return ret; +} +EXPORT_SYMBOL(_mutex_lock_killable); + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +void __lockfunc _mutex_lock_nested(struct mutex *lock, int subclass) +{ + mutex_acquire_nest(&lock->dep_map, subclass, 0, NULL, _RET_IP_); + _mutex_lock_blk_flush(lock, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(_mutex_lock_nested); + +void __lockfunc _mutex_lock_nest_lock(struct mutex *lock, struct lockdep_map *nest) +{ + mutex_acquire_nest(&lock->dep_map, 0, 0, nest, _RET_IP_); + _mutex_lock_blk_flush(lock, TASK_UNINTERRUPTIBLE); +} +EXPORT_SYMBOL(_mutex_lock_nest_lock); + +int __lockfunc _mutex_lock_interruptible_nested(struct mutex *lock, int subclass) +{ + int ret; + + mutex_acquire_nest(&lock->dep_map, subclass, 0, NULL, _RET_IP_); + ret = _mutex_lock_blk_flush(lock, TASK_INTERRUPTIBLE); + if (ret) + mutex_release(&lock->dep_map, _RET_IP_); + return ret; +} +EXPORT_SYMBOL(_mutex_lock_interruptible_nested); + +int __lockfunc _mutex_lock_killable_nested(struct mutex *lock, int subclass) +{ + int ret; + + mutex_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + ret = _mutex_lock_blk_flush(lock, TASK_KILLABLE); + if (ret) + mutex_release(&lock->dep_map, _RET_IP_); + return ret; +} +EXPORT_SYMBOL(_mutex_lock_killable_nested); +#endif + +int __lockfunc _mutex_trylock(struct mutex *lock) +{ + int ret = __rt_mutex_trylock(&lock->lock); + + if (ret) + mutex_acquire(&lock->dep_map, 0, 1, _RET_IP_); + + return ret; +} +EXPORT_SYMBOL(_mutex_trylock); + +void __lockfunc _mutex_unlock(struct mutex *lock) +{ + mutex_release(&lock->dep_map, _RET_IP_); + __rt_mutex_unlock(&lock->lock); +} +EXPORT_SYMBOL(_mutex_unlock); + +/** + * atomic_dec_and_mutex_lock - return holding mutex if we dec to 0 + * @cnt: the atomic which we are to dec + * @lock: the mutex to return holding if we dec to 0 + * + * return true and hold lock if we dec to 0, return false otherwise + */ +int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock) +{ + /* dec if we can't possibly hit 0 */ + if (atomic_add_unless(cnt, -1, 1)) + return 0; + /* we might hit 0, so take the lock */ + mutex_lock(lock); + if (!atomic_dec_and_test(cnt)) { + /* when we actually did the dec, we didn't hit 0 */ + mutex_unlock(lock); + return 0; + } + /* we hit 0, and we hold the lock */ + return 1; +} +EXPORT_SYMBOL(atomic_dec_and_mutex_lock); diff --git a/kernel/locking/rtmutex-debug.c b/kernel/locking/rtmutex-debug.c index 36e69100e8e06..fb150100335fc 100644 --- a/kernel/locking/rtmutex-debug.c +++ b/kernel/locking/rtmutex-debug.c @@ -32,110 +32,12 @@ #include "rtmutex_common.h" -static void printk_task(struct task_struct *p) -{ - if (p) - printk("%16s:%5d [%p, %3d]", p->comm, task_pid_nr(p), p, p->prio); - else - printk(""); -} - -static void printk_lock(struct rt_mutex *lock, int print_owner) -{ - if (lock->name) - printk(" [%p] {%s}\n", - lock, lock->name); - else - printk(" [%p] {%s:%d}\n", - lock, lock->file, lock->line); - - if (print_owner && rt_mutex_owner(lock)) { - printk(".. ->owner: %p\n", lock->owner); - printk(".. held by: "); - printk_task(rt_mutex_owner(lock)); - printk("\n"); - } -} - void rt_mutex_debug_task_free(struct task_struct *task) { DEBUG_LOCKS_WARN_ON(!RB_EMPTY_ROOT(&task->pi_waiters.rb_root)); DEBUG_LOCKS_WARN_ON(task->pi_blocked_on); } -/* - * We fill out the fields in the waiter to store the information about - * the deadlock. We print when we return. act_waiter can be NULL in - * case of a remove waiter operation. - */ -void debug_rt_mutex_deadlock(enum rtmutex_chainwalk chwalk, - struct rt_mutex_waiter *act_waiter, - struct rt_mutex *lock) -{ - struct task_struct *task; - - if (!debug_locks || chwalk == RT_MUTEX_FULL_CHAINWALK || !act_waiter) - return; - - task = rt_mutex_owner(act_waiter->lock); - if (task && task != current) { - act_waiter->deadlock_task_pid = get_pid(task_pid(task)); - act_waiter->deadlock_lock = lock; - } -} - -void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter) -{ - struct task_struct *task; - - if (!waiter->deadlock_lock || !debug_locks) - return; - - rcu_read_lock(); - task = pid_task(waiter->deadlock_task_pid, PIDTYPE_PID); - if (!task) { - rcu_read_unlock(); - return; - } - - if (!debug_locks_off()) { - rcu_read_unlock(); - return; - } - - pr_warn("\n"); - pr_warn("============================================\n"); - pr_warn("WARNING: circular locking deadlock detected!\n"); - pr_warn("%s\n", print_tainted()); - pr_warn("--------------------------------------------\n"); - printk("%s/%d is deadlocking current task %s/%d\n\n", - task->comm, task_pid_nr(task), - current->comm, task_pid_nr(current)); - - printk("\n1) %s/%d is trying to acquire this lock:\n", - current->comm, task_pid_nr(current)); - printk_lock(waiter->lock, 1); - - printk("\n2) %s/%d is blocked on this lock:\n", - task->comm, task_pid_nr(task)); - printk_lock(waiter->deadlock_lock, 1); - - debug_show_held_locks(current); - debug_show_held_locks(task); - - printk("\n%s/%d's [blocked] stackdump:\n\n", - task->comm, task_pid_nr(task)); - show_stack(task, NULL, KERN_DEFAULT); - printk("\n%s/%d's [current] stackdump:\n\n", - current->comm, task_pid_nr(current)); - dump_stack(); - debug_show_all_locks(); - rcu_read_unlock(); - - printk("[ turning off deadlock detection." - "Please report this trace. ]\n\n"); -} - void debug_rt_mutex_lock(struct rt_mutex *lock) { } @@ -158,12 +60,10 @@ void debug_rt_mutex_proxy_unlock(struct rt_mutex *lock) void debug_rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) { memset(waiter, 0x11, sizeof(*waiter)); - waiter->deadlock_task_pid = NULL; } void debug_rt_mutex_free_waiter(struct rt_mutex_waiter *waiter) { - put_pid(waiter->deadlock_task_pid); memset(waiter, 0x22, sizeof(*waiter)); } @@ -173,10 +73,8 @@ void debug_rt_mutex_init(struct rt_mutex *lock, const char *name, struct lock_cl * Make sure we are not reinitializing a held lock: */ debug_check_no_locks_freed((void *)lock, sizeof(*lock)); - lock->name = name; #ifdef CONFIG_DEBUG_LOCK_ALLOC lockdep_init_map(&lock->dep_map, name, key, 0); #endif } - diff --git a/kernel/locking/rtmutex-debug.h b/kernel/locking/rtmutex-debug.h index fc549713bba3f..659e93e256c67 100644 --- a/kernel/locking/rtmutex-debug.h +++ b/kernel/locking/rtmutex-debug.h @@ -18,20 +18,9 @@ extern void debug_rt_mutex_unlock(struct rt_mutex *lock); extern void debug_rt_mutex_proxy_lock(struct rt_mutex *lock, struct task_struct *powner); extern void debug_rt_mutex_proxy_unlock(struct rt_mutex *lock); -extern void debug_rt_mutex_deadlock(enum rtmutex_chainwalk chwalk, - struct rt_mutex_waiter *waiter, - struct rt_mutex *lock); -extern void debug_rt_mutex_print_deadlock(struct rt_mutex_waiter *waiter); -# define debug_rt_mutex_reset_waiter(w) \ - do { (w)->deadlock_lock = NULL; } while (0) static inline bool debug_rt_mutex_detect_deadlock(struct rt_mutex_waiter *waiter, enum rtmutex_chainwalk walk) { return (waiter != NULL); } - -static inline void rt_mutex_print_deadlock(struct rt_mutex_waiter *w) -{ - debug_rt_mutex_print_deadlock(w); -} diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index cfdd5b93264d7..cc0d7e99be00a 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -8,6 +8,11 @@ * Copyright (C) 2005-2006 Timesys Corp., Thomas Gleixner * Copyright (C) 2005 Kihon Technologies Inc., Steven Rostedt * Copyright (C) 2006 Esben Nielsen + * Adaptive Spinlocks: + * Copyright (C) 2008 Novell, Inc., Gregory Haskins, Sven Dietrich, + * and Peter Morreale, + * Adaptive Spinlocks simplification: + * Copyright (C) 2008 Red Hat, Inc., Steven Rostedt * * See Documentation/locking/rt-mutex-design.rst for details. */ @@ -19,6 +24,7 @@ #include #include #include +#include #include "rtmutex_common.h" @@ -136,6 +142,12 @@ static void fixup_rt_mutex_waiters(struct rt_mutex *lock) WRITE_ONCE(*p, owner & ~RT_MUTEX_HAS_WAITERS); } +static int rt_mutex_real_waiter(struct rt_mutex_waiter *waiter) +{ + return waiter && waiter != PI_WAKEUP_INPROGRESS && + waiter != PI_REQUEUE_INPROGRESS; +} + /* * We can speed up the acquire/release, if there's no debugging state to be * set up. @@ -227,7 +239,7 @@ static inline bool unlock_rt_mutex_safe(struct rt_mutex *lock, * Only use with rt_mutex_waiter_{less,equal}() */ #define task_to_waiter(p) \ - &(struct rt_mutex_waiter){ .prio = (p)->prio, .deadline = (p)->dl.deadline } + &(struct rt_mutex_waiter){ .prio = (p)->prio, .deadline = (p)->dl.deadline, .task = (p) } static inline int rt_mutex_waiter_less(struct rt_mutex_waiter *left, @@ -267,6 +279,27 @@ rt_mutex_waiter_equal(struct rt_mutex_waiter *left, return 1; } +#define STEAL_NORMAL 0 +#define STEAL_LATERAL 1 + +static inline int +rt_mutex_steal(struct rt_mutex *lock, struct rt_mutex_waiter *waiter, int mode) +{ + struct rt_mutex_waiter *top_waiter = rt_mutex_top_waiter(lock); + + if (waiter == top_waiter || rt_mutex_waiter_less(waiter, top_waiter)) + return 1; + + /* + * Note that RT tasks are excluded from lateral-steals + * to prevent the introduction of an unbounded latency. + */ + if (mode == STEAL_NORMAL || rt_task(waiter->task)) + return 0; + + return rt_mutex_waiter_equal(waiter, top_waiter); +} + static void rt_mutex_enqueue(struct rt_mutex *lock, struct rt_mutex_waiter *waiter) { @@ -371,6 +404,14 @@ static bool rt_mutex_cond_detect_deadlock(struct rt_mutex_waiter *waiter, return debug_rt_mutex_detect_deadlock(waiter, chwalk); } +static void rt_mutex_wake_waiter(struct rt_mutex_waiter *waiter) +{ + if (waiter->savestate) + wake_up_lock_sleeper(waiter->task); + else + wake_up_process(waiter->task); +} + /* * Max number of times we'll walk the boosting chain: */ @@ -378,7 +419,8 @@ int max_lock_depth = 1024; static inline struct rt_mutex *task_blocked_on_lock(struct task_struct *p) { - return p->pi_blocked_on ? p->pi_blocked_on->lock : NULL; + return rt_mutex_real_waiter(p->pi_blocked_on) ? + p->pi_blocked_on->lock : NULL; } /* @@ -514,7 +556,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, * reached or the state of the chain has changed while we * dropped the locks. */ - if (!waiter) + if (!rt_mutex_real_waiter(waiter)) goto out_unlock_pi; /* @@ -597,7 +639,6 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, * walk, we detected a deadlock. */ if (lock == orig_lock || rt_mutex_owner(lock) == top_task) { - debug_rt_mutex_deadlock(chwalk, orig_waiter, lock); raw_spin_unlock(&lock->wait_lock); ret = -EDEADLK; goto out_unlock_pi; @@ -694,13 +735,16 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, * follow here. This is the end of the chain we are walking. */ if (!rt_mutex_owner(lock)) { + struct rt_mutex_waiter *lock_top_waiter; + /* * If the requeue [7] above changed the top waiter, * then we need to wake the new top waiter up to try * to get the lock. */ - if (prerequeue_top_waiter != rt_mutex_top_waiter(lock)) - wake_up_process(rt_mutex_top_waiter(lock)->task); + lock_top_waiter = rt_mutex_top_waiter(lock); + if (prerequeue_top_waiter != lock_top_waiter) + rt_mutex_wake_waiter(lock_top_waiter); raw_spin_unlock_irq(&lock->wait_lock); return 0; } @@ -801,9 +845,11 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task, * @task: The task which wants to acquire the lock * @waiter: The waiter that is queued to the lock's wait tree if the * callsite called task_blocked_on_lock(), otherwise NULL + * @mode: Lock steal mode (STEAL_NORMAL, STEAL_LATERAL) */ -static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, - struct rt_mutex_waiter *waiter) +static int __try_to_take_rt_mutex(struct rt_mutex *lock, + struct task_struct *task, + struct rt_mutex_waiter *waiter, int mode) { lockdep_assert_held(&lock->wait_lock); @@ -839,12 +885,11 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, */ if (waiter) { /* - * If waiter is not the highest priority waiter of - * @lock, give up. + * If waiter is not the highest priority waiter of @lock, + * or its peer when lateral steal is allowed, give up. */ - if (waiter != rt_mutex_top_waiter(lock)) + if (!rt_mutex_steal(lock, waiter, mode)) return 0; - /* * We can acquire the lock. Remove the waiter from the * lock waiters tree. @@ -862,14 +907,12 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, */ if (rt_mutex_has_waiters(lock)) { /* - * If @task->prio is greater than or equal to - * the top waiter priority (kernel view), - * @task lost. + * If @task->prio is greater than the top waiter + * priority (kernel view), or equal to it when a + * lateral steal is forbidden, @task lost. */ - if (!rt_mutex_waiter_less(task_to_waiter(task), - rt_mutex_top_waiter(lock))) + if (!rt_mutex_steal(lock, task_to_waiter(task), mode)) return 0; - /* * The current top waiter stays enqueued. We * don't have to change anything in the lock @@ -916,6 +959,329 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, return 1; } +#ifdef CONFIG_PREEMPT_RT +/* + * preemptible spin_lock functions: + */ +static inline void rt_spin_lock_fastlock(struct rt_mutex *lock, + void (*slowfn)(struct rt_mutex *lock)) +{ + might_sleep_no_state_check(); + + if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) + return; + else + slowfn(lock); +} + +static inline void rt_spin_lock_fastunlock(struct rt_mutex *lock, + void (*slowfn)(struct rt_mutex *lock)) +{ + if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) + return; + else + slowfn(lock); +} +#ifdef CONFIG_SMP +/* + * Note that owner is a speculative pointer and dereferencing relies + * on rcu_read_lock() and the check against the lock owner. + */ +static int adaptive_wait(struct rt_mutex *lock, + struct task_struct *owner) +{ + int res = 0; + + rcu_read_lock(); + for (;;) { + if (owner != rt_mutex_owner(lock)) + break; + /* + * Ensure that owner->on_cpu is dereferenced _after_ + * checking the above to be valid. + */ + barrier(); + if (!owner->on_cpu) { + res = 1; + break; + } + cpu_relax(); + } + rcu_read_unlock(); + return res; +} +#else +static int adaptive_wait(struct rt_mutex *lock, + struct task_struct *orig_owner) +{ + return 1; +} +#endif + +static int task_blocks_on_rt_mutex(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + struct task_struct *task, + enum rtmutex_chainwalk chwalk); +/* + * Slow path lock function spin_lock style: this variant is very + * careful not to miss any non-lock wakeups. + * + * We store the current state under p->pi_lock in p->saved_state and + * the try_to_wake_up() code handles this accordingly. + */ +void __sched rt_spin_lock_slowlock_locked(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + unsigned long flags) +{ + struct task_struct *lock_owner, *self = current; + struct rt_mutex_waiter *top_waiter; + int ret; + + if (__try_to_take_rt_mutex(lock, self, NULL, STEAL_LATERAL)) + return; + + BUG_ON(rt_mutex_owner(lock) == self); + + /* + * We save whatever state the task is in and we'll restore it + * after acquiring the lock taking real wakeups into account + * as well. We are serialized via pi_lock against wakeups. See + * try_to_wake_up(). + */ + raw_spin_lock(&self->pi_lock); + self->saved_state = self->state; + __set_current_state_no_track(TASK_UNINTERRUPTIBLE); + raw_spin_unlock(&self->pi_lock); + + ret = task_blocks_on_rt_mutex(lock, waiter, self, RT_MUTEX_MIN_CHAINWALK); + BUG_ON(ret); + + for (;;) { + /* Try to acquire the lock again. */ + if (__try_to_take_rt_mutex(lock, self, waiter, STEAL_LATERAL)) + break; + + top_waiter = rt_mutex_top_waiter(lock); + lock_owner = rt_mutex_owner(lock); + + raw_spin_unlock_irqrestore(&lock->wait_lock, flags); + + if (top_waiter != waiter || adaptive_wait(lock, lock_owner)) + preempt_schedule_lock(); + + raw_spin_lock_irqsave(&lock->wait_lock, flags); + + raw_spin_lock(&self->pi_lock); + __set_current_state_no_track(TASK_UNINTERRUPTIBLE); + raw_spin_unlock(&self->pi_lock); + } + + /* + * Restore the task state to current->saved_state. We set it + * to the original state above and the try_to_wake_up() code + * has possibly updated it when a real (non-rtmutex) wakeup + * happened while we were blocked. Clear saved_state so + * try_to_wakeup() does not get confused. + */ + raw_spin_lock(&self->pi_lock); + __set_current_state_no_track(self->saved_state); + self->saved_state = TASK_RUNNING; + raw_spin_unlock(&self->pi_lock); + + /* + * try_to_take_rt_mutex() sets the waiter bit + * unconditionally. We might have to fix that up: + */ + fixup_rt_mutex_waiters(lock); + + BUG_ON(rt_mutex_has_waiters(lock) && waiter == rt_mutex_top_waiter(lock)); + BUG_ON(!RB_EMPTY_NODE(&waiter->tree_entry)); +} + +static void noinline __sched rt_spin_lock_slowlock(struct rt_mutex *lock) +{ + struct rt_mutex_waiter waiter; + unsigned long flags; + + rt_mutex_init_waiter(&waiter, true); + + raw_spin_lock_irqsave(&lock->wait_lock, flags); + rt_spin_lock_slowlock_locked(lock, &waiter, flags); + raw_spin_unlock_irqrestore(&lock->wait_lock, flags); + debug_rt_mutex_free_waiter(&waiter); +} + +static bool __sched __rt_mutex_unlock_common(struct rt_mutex *lock, + struct wake_q_head *wake_q, + struct wake_q_head *wq_sleeper); +/* + * Slow path to release a rt_mutex spin_lock style + */ +void __sched rt_spin_lock_slowunlock(struct rt_mutex *lock) +{ + unsigned long flags; + DEFINE_WAKE_Q(wake_q); + DEFINE_WAKE_Q(wake_sleeper_q); + bool postunlock; + + raw_spin_lock_irqsave(&lock->wait_lock, flags); + postunlock = __rt_mutex_unlock_common(lock, &wake_q, &wake_sleeper_q); + raw_spin_unlock_irqrestore(&lock->wait_lock, flags); + + if (postunlock) + rt_mutex_postunlock(&wake_q, &wake_sleeper_q); +} + +void __lockfunc rt_spin_lock(spinlock_t *lock) +{ + spin_acquire(&lock->dep_map, 0, 0, _RET_IP_); + rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock); + rcu_read_lock(); + migrate_disable(); +} +EXPORT_SYMBOL(rt_spin_lock); + +void __lockfunc __rt_spin_lock(struct rt_mutex *lock) +{ + rt_spin_lock_fastlock(lock, rt_spin_lock_slowlock); +} + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +void __lockfunc rt_spin_lock_nested(spinlock_t *lock, int subclass) +{ + spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock); + rcu_read_lock(); + migrate_disable(); +} +EXPORT_SYMBOL(rt_spin_lock_nested); + +void __lockfunc rt_spin_lock_nest_lock(spinlock_t *lock, + struct lockdep_map *nest_lock) +{ + spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_); + rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock); + rcu_read_lock(); + migrate_disable(); +} +EXPORT_SYMBOL(rt_spin_lock_nest_lock); +#endif + +void __lockfunc rt_spin_unlock(spinlock_t *lock) +{ + /* NOTE: we always pass in '1' for nested, for simplicity */ + spin_release(&lock->dep_map, _RET_IP_); + migrate_enable(); + rcu_read_unlock(); + rt_spin_lock_fastunlock(&lock->lock, rt_spin_lock_slowunlock); +} +EXPORT_SYMBOL(rt_spin_unlock); + +void __lockfunc __rt_spin_unlock(struct rt_mutex *lock) +{ + rt_spin_lock_fastunlock(lock, rt_spin_lock_slowunlock); +} +EXPORT_SYMBOL(__rt_spin_unlock); + +/* + * Wait for the lock to get unlocked: instead of polling for an unlock + * (like raw spinlocks do), we lock and unlock, to force the kernel to + * schedule if there's contention: + */ +void __lockfunc rt_spin_lock_unlock(spinlock_t *lock) +{ + spin_lock(lock); + spin_unlock(lock); +} +EXPORT_SYMBOL(rt_spin_lock_unlock); + +int __lockfunc rt_spin_trylock(spinlock_t *lock) +{ + int ret; + + ret = __rt_mutex_trylock(&lock->lock); + if (ret) { + spin_acquire(&lock->dep_map, 0, 1, _RET_IP_); + rcu_read_lock(); + migrate_disable(); + } + return ret; +} +EXPORT_SYMBOL(rt_spin_trylock); + +int __lockfunc rt_spin_trylock_bh(spinlock_t *lock) +{ + int ret; + + local_bh_disable(); + ret = __rt_mutex_trylock(&lock->lock); + if (ret) { + spin_acquire(&lock->dep_map, 0, 1, _RET_IP_); + rcu_read_lock(); + migrate_disable(); + } else { + local_bh_enable(); + } + return ret; +} +EXPORT_SYMBOL(rt_spin_trylock_bh); + +void +__rt_spin_lock_init(spinlock_t *lock, const char *name, struct lock_class_key *key) +{ +#ifdef CONFIG_DEBUG_LOCK_ALLOC + /* + * Make sure we are not reinitializing a held lock: + */ + debug_check_no_locks_freed((void *)lock, sizeof(*lock)); + lockdep_init_map(&lock->dep_map, name, key, 0); +#endif +} +EXPORT_SYMBOL(__rt_spin_lock_init); + +#endif /* PREEMPT_RT */ + +#ifdef CONFIG_PREEMPT_RT + static inline int __sched +__mutex_lock_check_stamp(struct rt_mutex *lock, struct ww_acquire_ctx *ctx) +{ + struct ww_mutex *ww = container_of(lock, struct ww_mutex, base.lock); + struct ww_acquire_ctx *hold_ctx = READ_ONCE(ww->ctx); + + if (!hold_ctx) + return 0; + + if (unlikely(ctx == hold_ctx)) + return -EALREADY; + + if (ctx->stamp - hold_ctx->stamp <= LONG_MAX && + (ctx->stamp != hold_ctx->stamp || ctx > hold_ctx)) { +#ifdef CONFIG_DEBUG_MUTEXES + DEBUG_LOCKS_WARN_ON(ctx->contending_lock); + ctx->contending_lock = ww; +#endif + return -EDEADLK; + } + + return 0; +} +#else + static inline int __sched +__mutex_lock_check_stamp(struct rt_mutex *lock, struct ww_acquire_ctx *ctx) +{ + BUG(); + return 0; +} + +#endif + +static inline int +try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task, + struct rt_mutex_waiter *waiter) +{ + return __try_to_take_rt_mutex(lock, task, waiter, STEAL_NORMAL); +} + /* * Task blocks on lock. * @@ -948,6 +1314,22 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, return -EDEADLK; raw_spin_lock(&task->pi_lock); + /* + * In the case of futex requeue PI, this will be a proxy + * lock. The task will wake unaware that it is enqueueed on + * this lock. Avoid blocking on two locks and corrupting + * pi_blocked_on via the PI_WAKEUP_INPROGRESS + * flag. futex_wait_requeue_pi() sets this when it wakes up + * before requeue (due to a signal or timeout). Do not enqueue + * the task if PI_WAKEUP_INPROGRESS is set. + */ + if (task != current && task->pi_blocked_on == PI_WAKEUP_INPROGRESS) { + raw_spin_unlock(&task->pi_lock); + return -EAGAIN; + } + + BUG_ON(rt_mutex_real_waiter(task->pi_blocked_on)); + waiter->task = task; waiter->lock = lock; waiter->prio = task->prio; @@ -971,7 +1353,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, rt_mutex_enqueue_pi(owner, waiter); rt_mutex_adjust_prio(owner); - if (owner->pi_blocked_on) + if (rt_mutex_real_waiter(owner->pi_blocked_on)) chain_walk = 1; } else if (rt_mutex_cond_detect_deadlock(waiter, chwalk)) { chain_walk = 1; @@ -1013,6 +1395,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock, * Called with lock->wait_lock held and interrupts disabled. */ static void mark_wakeup_next_waiter(struct wake_q_head *wake_q, + struct wake_q_head *wake_sleeper_q, struct rt_mutex *lock) { struct rt_mutex_waiter *waiter; @@ -1052,7 +1435,10 @@ static void mark_wakeup_next_waiter(struct wake_q_head *wake_q, * Pairs with preempt_enable() in rt_mutex_postunlock(); */ preempt_disable(); - wake_q_add(wake_q, waiter->task); + if (waiter->savestate) + wake_q_add_sleeper(wake_sleeper_q, waiter->task); + else + wake_q_add(wake_q, waiter->task); raw_spin_unlock(¤t->pi_lock); } @@ -1067,7 +1453,7 @@ static void remove_waiter(struct rt_mutex *lock, { bool is_top_waiter = (waiter == rt_mutex_top_waiter(lock)); struct task_struct *owner = rt_mutex_owner(lock); - struct rt_mutex *next_lock; + struct rt_mutex *next_lock = NULL; lockdep_assert_held(&lock->wait_lock); @@ -1093,7 +1479,8 @@ static void remove_waiter(struct rt_mutex *lock, rt_mutex_adjust_prio(owner); /* Store the lock on which owner is blocked or NULL */ - next_lock = task_blocked_on_lock(owner); + if (rt_mutex_real_waiter(owner->pi_blocked_on)) + next_lock = task_blocked_on_lock(owner); raw_spin_unlock(&owner->pi_lock); @@ -1129,26 +1516,28 @@ void rt_mutex_adjust_pi(struct task_struct *task) raw_spin_lock_irqsave(&task->pi_lock, flags); waiter = task->pi_blocked_on; - if (!waiter || rt_mutex_waiter_equal(waiter, task_to_waiter(task))) { + if (!rt_mutex_real_waiter(waiter) || + rt_mutex_waiter_equal(waiter, task_to_waiter(task))) { raw_spin_unlock_irqrestore(&task->pi_lock, flags); return; } next_lock = waiter->lock; - raw_spin_unlock_irqrestore(&task->pi_lock, flags); /* gets dropped in rt_mutex_adjust_prio_chain()! */ get_task_struct(task); + raw_spin_unlock_irqrestore(&task->pi_lock, flags); rt_mutex_adjust_prio_chain(task, RT_MUTEX_MIN_CHAINWALK, NULL, next_lock, NULL, task); } -void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) +void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter, bool savestate) { debug_rt_mutex_init_waiter(waiter); RB_CLEAR_NODE(&waiter->pi_tree_entry); RB_CLEAR_NODE(&waiter->tree_entry); waiter->task = NULL; + waiter->savestate = savestate; } /** @@ -1164,7 +1553,8 @@ void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter) static int __sched __rt_mutex_slowlock(struct rt_mutex *lock, int state, struct hrtimer_sleeper *timeout, - struct rt_mutex_waiter *waiter) + struct rt_mutex_waiter *waiter, + struct ww_acquire_ctx *ww_ctx) { int ret = 0; @@ -1173,24 +1563,23 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state, if (try_to_take_rt_mutex(lock, current, waiter)) break; - /* - * TASK_INTERRUPTIBLE checks for signals and - * timeout. Ignored otherwise. - */ - if (likely(state == TASK_INTERRUPTIBLE)) { - /* Signal pending? */ - if (signal_pending(current)) - ret = -EINTR; - if (timeout && !timeout->task) - ret = -ETIMEDOUT; + if (timeout && !timeout->task) { + ret = -ETIMEDOUT; + break; + } + if (signal_pending_state(state, current)) { + ret = -EINTR; + break; + } + + if (ww_ctx && ww_ctx->acquired > 0) { + ret = __mutex_lock_check_stamp(lock, ww_ctx); if (ret) break; } raw_spin_unlock_irq(&lock->wait_lock); - debug_rt_mutex_print_deadlock(waiter); - schedule(); raw_spin_lock_irq(&lock->wait_lock); @@ -1211,29 +1600,163 @@ static void rt_mutex_handle_deadlock(int res, int detect_deadlock, if (res != -EDEADLOCK || detect_deadlock) return; - /* - * Yell lowdly and stop the task right here. - */ - rt_mutex_print_deadlock(w); while (1) { set_current_state(TASK_INTERRUPTIBLE); schedule(); } } +static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww, + struct ww_acquire_ctx *ww_ctx) +{ +#ifdef CONFIG_DEBUG_MUTEXES + /* + * If this WARN_ON triggers, you used ww_mutex_lock to acquire, + * but released with a normal mutex_unlock in this call. + * + * This should never happen, always use ww_mutex_unlock. + */ + DEBUG_LOCKS_WARN_ON(ww->ctx); + + /* + * Not quite done after calling ww_acquire_done() ? + */ + DEBUG_LOCKS_WARN_ON(ww_ctx->done_acquire); + + if (ww_ctx->contending_lock) { + /* + * After -EDEADLK you tried to + * acquire a different ww_mutex? Bad! + */ + DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock != ww); + + /* + * You called ww_mutex_lock after receiving -EDEADLK, + * but 'forgot' to unlock everything else first? + */ + DEBUG_LOCKS_WARN_ON(ww_ctx->acquired > 0); + ww_ctx->contending_lock = NULL; + } + + /* + * Naughty, using a different class will lead to undefined behavior! + */ + DEBUG_LOCKS_WARN_ON(ww_ctx->ww_class != ww->ww_class); +#endif + ww_ctx->acquired++; +} + +#ifdef CONFIG_PREEMPT_RT +static void ww_mutex_account_lock(struct rt_mutex *lock, + struct ww_acquire_ctx *ww_ctx) +{ + struct ww_mutex *ww = container_of(lock, struct ww_mutex, base.lock); + struct rt_mutex_waiter *waiter, *n; + + /* + * This branch gets optimized out for the common case, + * and is only important for ww_mutex_lock. + */ + ww_mutex_lock_acquired(ww, ww_ctx); + ww->ctx = ww_ctx; + + /* + * Give any possible sleeping processes the chance to wake up, + * so they can recheck if they have to back off. + */ + rbtree_postorder_for_each_entry_safe(waiter, n, &lock->waiters.rb_root, + tree_entry) { + /* XXX debug rt mutex waiter wakeup */ + + BUG_ON(waiter->lock != lock); + rt_mutex_wake_waiter(waiter); + } +} + +#else + +static void ww_mutex_account_lock(struct rt_mutex *lock, + struct ww_acquire_ctx *ww_ctx) +{ + BUG(); +} +#endif + +int __sched rt_mutex_slowlock_locked(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + enum rtmutex_chainwalk chwalk, + struct ww_acquire_ctx *ww_ctx, + struct rt_mutex_waiter *waiter) +{ + int ret; + +#ifdef CONFIG_PREEMPT_RT + if (ww_ctx) { + struct ww_mutex *ww; + + ww = container_of(lock, struct ww_mutex, base.lock); + if (unlikely(ww_ctx == READ_ONCE(ww->ctx))) + return -EALREADY; + } +#endif + + /* Try to acquire the lock again: */ + if (try_to_take_rt_mutex(lock, current, NULL)) { + if (ww_ctx) + ww_mutex_account_lock(lock, ww_ctx); + return 0; + } + + set_current_state(state); + + /* Setup the timer, when timeout != NULL */ + if (unlikely(timeout)) + hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); + + ret = task_blocks_on_rt_mutex(lock, waiter, current, chwalk); + + if (likely(!ret)) { + /* sleep on the mutex */ + ret = __rt_mutex_slowlock(lock, state, timeout, waiter, + ww_ctx); + } else if (ww_ctx) { + /* ww_mutex received EDEADLK, let it become EALREADY */ + ret = __mutex_lock_check_stamp(lock, ww_ctx); + BUG_ON(!ret); + } + + if (unlikely(ret)) { + __set_current_state(TASK_RUNNING); + remove_waiter(lock, waiter); + /* ww_mutex wants to report EDEADLK/EALREADY, let it */ + if (!ww_ctx) + rt_mutex_handle_deadlock(ret, chwalk, waiter); + } else if (ww_ctx) { + ww_mutex_account_lock(lock, ww_ctx); + } + + /* + * try_to_take_rt_mutex() sets the waiter bit + * unconditionally. We might have to fix that up. + */ + fixup_rt_mutex_waiters(lock); + return ret; +} + /* * Slow path lock function: */ static int __sched rt_mutex_slowlock(struct rt_mutex *lock, int state, struct hrtimer_sleeper *timeout, - enum rtmutex_chainwalk chwalk) + enum rtmutex_chainwalk chwalk, + struct ww_acquire_ctx *ww_ctx) { struct rt_mutex_waiter waiter; unsigned long flags; int ret = 0; - rt_mutex_init_waiter(&waiter); + rt_mutex_init_waiter(&waiter, false); /* * Technically we could use raw_spin_[un]lock_irq() here, but this can @@ -1245,35 +1768,8 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state, */ raw_spin_lock_irqsave(&lock->wait_lock, flags); - /* Try to acquire the lock again: */ - if (try_to_take_rt_mutex(lock, current, NULL)) { - raw_spin_unlock_irqrestore(&lock->wait_lock, flags); - return 0; - } - - set_current_state(state); - - /* Setup the timer, when timeout != NULL */ - if (unlikely(timeout)) - hrtimer_start_expires(&timeout->timer, HRTIMER_MODE_ABS); - - ret = task_blocks_on_rt_mutex(lock, &waiter, current, chwalk); - - if (likely(!ret)) - /* sleep on the mutex */ - ret = __rt_mutex_slowlock(lock, state, timeout, &waiter); - - if (unlikely(ret)) { - __set_current_state(TASK_RUNNING); - remove_waiter(lock, &waiter); - rt_mutex_handle_deadlock(ret, chwalk, &waiter); - } - - /* - * try_to_take_rt_mutex() sets the waiter bit - * unconditionally. We might have to fix that up. - */ - fixup_rt_mutex_waiters(lock); + ret = rt_mutex_slowlock_locked(lock, state, timeout, chwalk, ww_ctx, + &waiter); raw_spin_unlock_irqrestore(&lock->wait_lock, flags); @@ -1334,7 +1830,8 @@ static inline int rt_mutex_slowtrylock(struct rt_mutex *lock) * Return whether the current task needs to call rt_mutex_postunlock(). */ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock, - struct wake_q_head *wake_q) + struct wake_q_head *wake_q, + struct wake_q_head *wake_sleeper_q) { unsigned long flags; @@ -1388,7 +1885,7 @@ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock, * * Queue the next waiter for wakeup once we release the wait_lock. */ - mark_wakeup_next_waiter(wake_q, lock); + mark_wakeup_next_waiter(wake_q, wake_sleeper_q, lock); raw_spin_unlock_irqrestore(&lock->wait_lock, flags); return true; /* call rt_mutex_postunlock() */ @@ -1402,29 +1899,16 @@ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock, */ static inline int rt_mutex_fastlock(struct rt_mutex *lock, int state, + struct ww_acquire_ctx *ww_ctx, int (*slowfn)(struct rt_mutex *lock, int state, struct hrtimer_sleeper *timeout, - enum rtmutex_chainwalk chwalk)) + enum rtmutex_chainwalk chwalk, + struct ww_acquire_ctx *ww_ctx)) { if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) return 0; - return slowfn(lock, state, NULL, RT_MUTEX_MIN_CHAINWALK); -} - -static inline int -rt_mutex_timed_fastlock(struct rt_mutex *lock, int state, - struct hrtimer_sleeper *timeout, - enum rtmutex_chainwalk chwalk, - int (*slowfn)(struct rt_mutex *lock, int state, - struct hrtimer_sleeper *timeout, - enum rtmutex_chainwalk chwalk)) -{ - if (chwalk == RT_MUTEX_MIN_CHAINWALK && - likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) - return 0; - - return slowfn(lock, state, timeout, chwalk); + return slowfn(lock, state, NULL, RT_MUTEX_MIN_CHAINWALK, ww_ctx); } static inline int @@ -1440,9 +1924,11 @@ rt_mutex_fasttrylock(struct rt_mutex *lock, /* * Performs the wakeup of the the top-waiter and re-enables preemption. */ -void rt_mutex_postunlock(struct wake_q_head *wake_q) +void rt_mutex_postunlock(struct wake_q_head *wake_q, + struct wake_q_head *wake_sleeper_q) { wake_up_q(wake_q); + wake_up_q_sleeper(wake_sleeper_q); /* Pairs with preempt_disable() in rt_mutex_slowunlock() */ preempt_enable(); @@ -1451,23 +1937,46 @@ void rt_mutex_postunlock(struct wake_q_head *wake_q) static inline void rt_mutex_fastunlock(struct rt_mutex *lock, bool (*slowfn)(struct rt_mutex *lock, - struct wake_q_head *wqh)) + struct wake_q_head *wqh, + struct wake_q_head *wq_sleeper)) { DEFINE_WAKE_Q(wake_q); + DEFINE_WAKE_Q(wake_sleeper_q); if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) return; - if (slowfn(lock, &wake_q)) - rt_mutex_postunlock(&wake_q); + if (slowfn(lock, &wake_q, &wake_sleeper_q)) + rt_mutex_postunlock(&wake_q, &wake_sleeper_q); +} + +int __sched __rt_mutex_lock_state(struct rt_mutex *lock, int state) +{ + might_sleep(); + return rt_mutex_fastlock(lock, state, NULL, rt_mutex_slowlock); +} + +/** + * rt_mutex_lock_state - lock a rt_mutex with a given state + * + * @lock: The rt_mutex to be locked + * @state: The state to set when blocking on the rt_mutex + */ +static inline int __sched rt_mutex_lock_state(struct rt_mutex *lock, + unsigned int subclass, int state) +{ + int ret; + + mutex_acquire(&lock->dep_map, subclass, 0, _RET_IP_); + ret = __rt_mutex_lock_state(lock, state); + if (ret) + mutex_release(&lock->dep_map, _RET_IP_); + return ret; } static inline void __rt_mutex_lock(struct rt_mutex *lock, unsigned int subclass) { - might_sleep(); - - mutex_acquire(&lock->dep_map, subclass, 0, _RET_IP_); - rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, rt_mutex_slowlock); + rt_mutex_lock_state(lock, subclass, TASK_UNINTERRUPTIBLE); } #ifdef CONFIG_DEBUG_LOCK_ALLOC @@ -1508,16 +2017,7 @@ EXPORT_SYMBOL_GPL(rt_mutex_lock); */ int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock) { - int ret; - - might_sleep(); - - mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); - ret = rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE, rt_mutex_slowlock); - if (ret) - mutex_release(&lock->dep_map, _RET_IP_); - - return ret; + return rt_mutex_lock_state(lock, 0, TASK_INTERRUPTIBLE); } EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible); @@ -1534,36 +2034,17 @@ int __sched __rt_mutex_futex_trylock(struct rt_mutex *lock) return __rt_mutex_slowtrylock(lock); } -/** - * rt_mutex_timed_lock - lock a rt_mutex interruptible - * the timeout structure is provided - * by the caller - * - * @lock: the rt_mutex to be locked - * @timeout: timeout structure or NULL (no timeout) - * - * Returns: - * 0 on success - * -EINTR when interrupted by a signal - * -ETIMEDOUT when the timeout expired - */ -int -rt_mutex_timed_lock(struct rt_mutex *lock, struct hrtimer_sleeper *timeout) +int __sched __rt_mutex_trylock(struct rt_mutex *lock) { - int ret; +#ifdef CONFIG_PREEMPT_RT + if (WARN_ON_ONCE(in_irq() || in_nmi())) +#else + if (WARN_ON_ONCE(in_irq() || in_nmi() || in_serving_softirq())) +#endif + return 0; - might_sleep(); - - mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_); - ret = rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout, - RT_MUTEX_MIN_CHAINWALK, - rt_mutex_slowlock); - if (ret) - mutex_release(&lock->dep_map, _RET_IP_); - - return ret; + return rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock); } -EXPORT_SYMBOL_GPL(rt_mutex_timed_lock); /** * rt_mutex_trylock - try to lock a rt_mutex @@ -1580,10 +2061,7 @@ int __sched rt_mutex_trylock(struct rt_mutex *lock) { int ret; - if (WARN_ON_ONCE(in_irq() || in_nmi() || in_serving_softirq())) - return 0; - - ret = rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock); + ret = __rt_mutex_trylock(lock); if (ret) mutex_acquire(&lock->dep_map, 0, 1, _RET_IP_); @@ -1591,6 +2069,11 @@ int __sched rt_mutex_trylock(struct rt_mutex *lock) } EXPORT_SYMBOL_GPL(rt_mutex_trylock); +void __sched __rt_mutex_unlock(struct rt_mutex *lock) +{ + rt_mutex_fastunlock(lock, rt_mutex_slowunlock); +} + /** * rt_mutex_unlock - unlock a rt_mutex * @@ -1599,16 +2082,13 @@ EXPORT_SYMBOL_GPL(rt_mutex_trylock); void __sched rt_mutex_unlock(struct rt_mutex *lock) { mutex_release(&lock->dep_map, _RET_IP_); - rt_mutex_fastunlock(lock, rt_mutex_slowunlock); + __rt_mutex_unlock(lock); } EXPORT_SYMBOL_GPL(rt_mutex_unlock); -/** - * Futex variant, that since futex variants do not use the fast-path, can be - * simple and will not need to retry. - */ -bool __sched __rt_mutex_futex_unlock(struct rt_mutex *lock, - struct wake_q_head *wake_q) +static bool __sched __rt_mutex_unlock_common(struct rt_mutex *lock, + struct wake_q_head *wake_q, + struct wake_q_head *wq_sleeper) { lockdep_assert_held(&lock->wait_lock); @@ -1625,23 +2105,35 @@ bool __sched __rt_mutex_futex_unlock(struct rt_mutex *lock, * avoid inversion prior to the wakeup. preempt_disable() * therein pairs with rt_mutex_postunlock(). */ - mark_wakeup_next_waiter(wake_q, lock); + mark_wakeup_next_waiter(wake_q, wq_sleeper, lock); return true; /* call postunlock() */ } +/** + * Futex variant, that since futex variants do not use the fast-path, can be + * simple and will not need to retry. + */ +bool __sched __rt_mutex_futex_unlock(struct rt_mutex *lock, + struct wake_q_head *wake_q, + struct wake_q_head *wq_sleeper) +{ + return __rt_mutex_unlock_common(lock, wake_q, wq_sleeper); +} + void __sched rt_mutex_futex_unlock(struct rt_mutex *lock) { DEFINE_WAKE_Q(wake_q); + DEFINE_WAKE_Q(wake_sleeper_q); unsigned long flags; bool postunlock; raw_spin_lock_irqsave(&lock->wait_lock, flags); - postunlock = __rt_mutex_futex_unlock(lock, &wake_q); + postunlock = __rt_mutex_futex_unlock(lock, &wake_q, &wake_sleeper_q); raw_spin_unlock_irqrestore(&lock->wait_lock, flags); if (postunlock) - rt_mutex_postunlock(&wake_q); + rt_mutex_postunlock(&wake_q, &wake_sleeper_q); } /** @@ -1655,9 +2147,6 @@ void __sched rt_mutex_futex_unlock(struct rt_mutex *lock) void rt_mutex_destroy(struct rt_mutex *lock) { WARN_ON(rt_mutex_is_locked(lock)); -#ifdef CONFIG_DEBUG_RT_MUTEXES - lock->magic = NULL; -#endif } EXPORT_SYMBOL_GPL(rt_mutex_destroy); @@ -1680,7 +2169,7 @@ void __rt_mutex_init(struct rt_mutex *lock, const char *name, if (name && key) debug_rt_mutex_init(lock, name, key); } -EXPORT_SYMBOL_GPL(__rt_mutex_init); +EXPORT_SYMBOL(__rt_mutex_init); /** * rt_mutex_init_proxy_locked - initialize and lock a rt_mutex on behalf of a @@ -1700,6 +2189,14 @@ void rt_mutex_init_proxy_locked(struct rt_mutex *lock, struct task_struct *proxy_owner) { __rt_mutex_init(lock, NULL, NULL); +#ifdef CONFIG_DEBUG_SPINLOCK + /* + * get another key class for the wait_lock. LOCK_PI and UNLOCK_PI is + * holding the ->wait_lock of the proxy_lock while unlocking a sleeping + * lock. + */ + raw_spin_lock_init(&lock->wait_lock); +#endif debug_rt_mutex_proxy_lock(lock, proxy_owner); rt_mutex_set_owner(lock, proxy_owner); } @@ -1723,6 +2220,26 @@ void rt_mutex_proxy_unlock(struct rt_mutex *lock, rt_mutex_set_owner(lock, NULL); } +static void fixup_rt_mutex_blocked(struct rt_mutex *lock) +{ + struct task_struct *tsk = current; + /* + * RT has a problem here when the wait got interrupted by a timeout + * or a signal. task->pi_blocked_on is still set. The task must + * acquire the hash bucket lock when returning from this function. + * + * If the hash bucket lock is contended then the + * BUG_ON(rt_mutex_real_waiter(task->pi_blocked_on)) in + * task_blocks_on_rt_mutex() will trigger. This can be avoided by + * clearing task->pi_blocked_on which removes the task from the + * boosting chain of the rtmutex. That's correct because the task + * is not longer blocked on it. + */ + raw_spin_lock(&tsk->pi_lock); + tsk->pi_blocked_on = NULL; + raw_spin_unlock(&tsk->pi_lock); +} + /** * __rt_mutex_start_proxy_lock() - Start lock acquisition for another task * @lock: the rt_mutex to take @@ -1753,6 +2270,34 @@ int __rt_mutex_start_proxy_lock(struct rt_mutex *lock, if (try_to_take_rt_mutex(lock, task, NULL)) return 1; +#ifdef CONFIG_PREEMPT_RT + /* + * In PREEMPT_RT there's an added race. + * If the task, that we are about to requeue, times out, + * it can set the PI_WAKEUP_INPROGRESS. This tells the requeue + * to skip this task. But right after the task sets + * its pi_blocked_on to PI_WAKEUP_INPROGRESS it can then + * block on the spin_lock(&hb->lock), which in RT is an rtmutex. + * This will replace the PI_WAKEUP_INPROGRESS with the actual + * lock that it blocks on. We *must not* place this task + * on this proxy lock in that case. + * + * To prevent this race, we first take the task's pi_lock + * and check if it has updated its pi_blocked_on. If it has, + * we assume that it woke up and we return -EAGAIN. + * Otherwise, we set the task's pi_blocked_on to + * PI_REQUEUE_INPROGRESS, so that if the task is waking up + * it will know that we are in the process of requeuing it. + */ + raw_spin_lock(&task->pi_lock); + if (task->pi_blocked_on) { + raw_spin_unlock(&task->pi_lock); + return -EAGAIN; + } + task->pi_blocked_on = PI_REQUEUE_INPROGRESS; + raw_spin_unlock(&task->pi_lock); +#endif + /* We enforce deadlock detection for futexes */ ret = task_blocks_on_rt_mutex(lock, waiter, task, RT_MUTEX_FULL_CHAINWALK); @@ -1767,7 +2312,8 @@ int __rt_mutex_start_proxy_lock(struct rt_mutex *lock, ret = 0; } - debug_rt_mutex_print_deadlock(waiter); + if (ret) + fixup_rt_mutex_blocked(lock); return ret; } @@ -1852,12 +2398,15 @@ int rt_mutex_wait_proxy_lock(struct rt_mutex *lock, raw_spin_lock_irq(&lock->wait_lock); /* sleep on the mutex */ set_current_state(TASK_INTERRUPTIBLE); - ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter); + ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter, NULL); /* * try_to_take_rt_mutex() sets the waiter bit unconditionally. We might * have to fix that up. */ fixup_rt_mutex_waiters(lock); + if (ret) + fixup_rt_mutex_blocked(lock); + raw_spin_unlock_irq(&lock->wait_lock); return ret; @@ -1919,3 +2468,97 @@ bool rt_mutex_cleanup_proxy_lock(struct rt_mutex *lock, return cleanup; } + +static inline int +ww_mutex_deadlock_injection(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +{ +#ifdef CONFIG_DEBUG_WW_MUTEX_SLOWPATH + unsigned int tmp; + + if (ctx->deadlock_inject_countdown-- == 0) { + tmp = ctx->deadlock_inject_interval; + if (tmp > UINT_MAX/4) + tmp = UINT_MAX; + else + tmp = tmp*2 + tmp + tmp/2; + + ctx->deadlock_inject_interval = tmp; + ctx->deadlock_inject_countdown = tmp; + ctx->contending_lock = lock; + + ww_mutex_unlock(lock); + + return -EDEADLK; + } +#endif + + return 0; +} + +#ifdef CONFIG_PREEMPT_RT +int __sched +ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +{ + int ret; + + might_sleep(); + + mutex_acquire_nest(&lock->base.dep_map, 0, 0, + ctx ? &ctx->dep_map : NULL, _RET_IP_); + ret = rt_mutex_slowlock(&lock->base.lock, TASK_INTERRUPTIBLE, NULL, 0, + ctx); + if (ret) + mutex_release(&lock->base.dep_map, _RET_IP_); + else if (!ret && ctx && ctx->acquired > 1) + return ww_mutex_deadlock_injection(lock, ctx); + + return ret; +} +EXPORT_SYMBOL_GPL(ww_mutex_lock_interruptible); + +int __sched +ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ctx) +{ + int ret; + + might_sleep(); + + mutex_acquire_nest(&lock->base.dep_map, 0, 0, + ctx ? &ctx->dep_map : NULL, _RET_IP_); + ret = rt_mutex_slowlock(&lock->base.lock, TASK_UNINTERRUPTIBLE, NULL, 0, + ctx); + if (ret) + mutex_release(&lock->base.dep_map, _RET_IP_); + else if (!ret && ctx && ctx->acquired > 1) + return ww_mutex_deadlock_injection(lock, ctx); + + return ret; +} +EXPORT_SYMBOL_GPL(ww_mutex_lock); + +void __sched ww_mutex_unlock(struct ww_mutex *lock) +{ + /* + * The unlocking fastpath is the 0->1 transition from 'locked' + * into 'unlocked' state: + */ + if (lock->ctx) { +#ifdef CONFIG_DEBUG_MUTEXES + DEBUG_LOCKS_WARN_ON(!lock->ctx->acquired); +#endif + if (lock->ctx->acquired > 0) + lock->ctx->acquired--; + lock->ctx = NULL; + } + + mutex_release(&lock->base.dep_map, _RET_IP_); + __rt_mutex_unlock(&lock->base.lock); +} +EXPORT_SYMBOL(ww_mutex_unlock); + +int __rt_mutex_owner_current(struct rt_mutex *lock) +{ + return rt_mutex_owner(lock) == current; +} +EXPORT_SYMBOL(__rt_mutex_owner_current); +#endif diff --git a/kernel/locking/rtmutex.h b/kernel/locking/rtmutex.h index 732f96abf4623..338ccd29119a1 100644 --- a/kernel/locking/rtmutex.h +++ b/kernel/locking/rtmutex.h @@ -19,15 +19,8 @@ #define debug_rt_mutex_proxy_unlock(l) do { } while (0) #define debug_rt_mutex_unlock(l) do { } while (0) #define debug_rt_mutex_init(m, n, k) do { } while (0) -#define debug_rt_mutex_deadlock(d, a ,l) do { } while (0) -#define debug_rt_mutex_print_deadlock(w) do { } while (0) #define debug_rt_mutex_reset_waiter(w) do { } while (0) -static inline void rt_mutex_print_deadlock(struct rt_mutex_waiter *w) -{ - WARN(1, "rtmutex deadlock detected\n"); -} - static inline bool debug_rt_mutex_detect_deadlock(struct rt_mutex_waiter *w, enum rtmutex_chainwalk walk) { diff --git a/kernel/locking/rtmutex_common.h b/kernel/locking/rtmutex_common.h index d1d62f942be22..407f3da146cb2 100644 --- a/kernel/locking/rtmutex_common.h +++ b/kernel/locking/rtmutex_common.h @@ -15,6 +15,7 @@ #include #include +#include /* * This is the control structure for tasks blocked on a rt_mutex, @@ -29,12 +30,8 @@ struct rt_mutex_waiter { struct rb_node pi_tree_entry; struct task_struct *task; struct rt_mutex *lock; -#ifdef CONFIG_DEBUG_RT_MUTEXES - unsigned long ip; - struct pid *deadlock_task_pid; - struct rt_mutex *deadlock_lock; -#endif int prio; + bool savestate; u64 deadline; }; @@ -130,12 +127,15 @@ enum rtmutex_chainwalk { /* * PI-futex support (proxy locking functions, etc.): */ +#define PI_WAKEUP_INPROGRESS ((struct rt_mutex_waiter *) 1) +#define PI_REQUEUE_INPROGRESS ((struct rt_mutex_waiter *) 2) + extern struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock); extern void rt_mutex_init_proxy_locked(struct rt_mutex *lock, struct task_struct *proxy_owner); extern void rt_mutex_proxy_unlock(struct rt_mutex *lock, struct task_struct *proxy_owner); -extern void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter); +extern void rt_mutex_init_waiter(struct rt_mutex_waiter *waiter, bool savetate); extern int __rt_mutex_start_proxy_lock(struct rt_mutex *lock, struct rt_mutex_waiter *waiter, struct task_struct *task); @@ -153,9 +153,27 @@ extern int __rt_mutex_futex_trylock(struct rt_mutex *l); extern void rt_mutex_futex_unlock(struct rt_mutex *lock); extern bool __rt_mutex_futex_unlock(struct rt_mutex *lock, - struct wake_q_head *wqh); + struct wake_q_head *wqh, + struct wake_q_head *wq_sleeper); -extern void rt_mutex_postunlock(struct wake_q_head *wake_q); +extern void rt_mutex_postunlock(struct wake_q_head *wake_q, + struct wake_q_head *wake_sleeper_q); + +/* RW semaphore special interface */ +struct ww_acquire_ctx; + +extern int __rt_mutex_lock_state(struct rt_mutex *lock, int state); +extern int __rt_mutex_trylock(struct rt_mutex *lock); +extern void __rt_mutex_unlock(struct rt_mutex *lock); +int __sched rt_mutex_slowlock_locked(struct rt_mutex *lock, int state, + struct hrtimer_sleeper *timeout, + enum rtmutex_chainwalk chwalk, + struct ww_acquire_ctx *ww_ctx, + struct rt_mutex_waiter *waiter); +void __sched rt_spin_lock_slowlock_locked(struct rt_mutex *lock, + struct rt_mutex_waiter *waiter, + unsigned long flags); +void __sched rt_spin_lock_slowunlock(struct rt_mutex *lock); #ifdef CONFIG_DEBUG_RT_MUTEXES # include "rtmutex-debug.h" diff --git a/kernel/locking/rwlock-rt.c b/kernel/locking/rwlock-rt.c new file mode 100644 index 0000000000000..3d2d1f14b5135 --- /dev/null +++ b/kernel/locking/rwlock-rt.c @@ -0,0 +1,334 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include + +#include "rtmutex_common.h" +#include + +/* + * RT-specific reader/writer locks + * + * write_lock() + * 1) Lock lock->rtmutex + * 2) Remove the reader BIAS to force readers into the slow path + * 3) Wait until all readers have left the critical region + * 4) Mark it write locked + * + * write_unlock() + * 1) Remove the write locked marker + * 2) Set the reader BIAS so readers can use the fast path again + * 3) Unlock lock->rtmutex to release blocked readers + * + * read_lock() + * 1) Try fast path acquisition (reader BIAS is set) + * 2) Take lock->rtmutex.wait_lock which protects the writelocked flag + * 3) If !writelocked, acquire it for read + * 4) If writelocked, block on lock->rtmutex + * 5) unlock lock->rtmutex, goto 1) + * + * read_unlock() + * 1) Try fast path release (reader count != 1) + * 2) Wake the writer waiting in write_lock()#3 + * + * read_lock()#3 has the consequence, that rw locks on RT are not writer + * fair, but writers, which should be avoided in RT tasks (think tasklist + * lock), are subject to the rtmutex priority/DL inheritance mechanism. + * + * It's possible to make the rw locks writer fair by keeping a list of + * active readers. A blocked writer would force all newly incoming readers + * to block on the rtmutex, but the rtmutex would have to be proxy locked + * for one reader after the other. We can't use multi-reader inheritance + * because there is no way to support that with + * SCHED_DEADLINE. Implementing the one by one reader boosting/handover + * mechanism is a major surgery for a very dubious value. + * + * The risk of writer starvation is there, but the pathological use cases + * which trigger it are not necessarily the typical RT workloads. + */ + +void __rwlock_biased_rt_init(struct rt_rw_lock *lock, const char *name, + struct lock_class_key *key) +{ +#ifdef CONFIG_DEBUG_LOCK_ALLOC + /* + * Make sure we are not reinitializing a held semaphore: + */ + debug_check_no_locks_freed((void *)lock, sizeof(*lock)); + lockdep_init_map(&lock->dep_map, name, key, 0); +#endif + atomic_set(&lock->readers, READER_BIAS); + rt_mutex_init(&lock->rtmutex); + lock->rtmutex.save_state = 1; +} + +static int __read_rt_trylock(struct rt_rw_lock *lock) +{ + int r, old; + + /* + * Increment reader count, if lock->readers < 0, i.e. READER_BIAS is + * set. + */ + for (r = atomic_read(&lock->readers); r < 0;) { + old = atomic_cmpxchg(&lock->readers, r, r + 1); + if (likely(old == r)) + return 1; + r = old; + } + return 0; +} + +static void __read_rt_lock(struct rt_rw_lock *lock) +{ + struct rt_mutex *m = &lock->rtmutex; + struct rt_mutex_waiter waiter; + unsigned long flags; + + if (__read_rt_trylock(lock)) + return; + + raw_spin_lock_irqsave(&m->wait_lock, flags); + /* + * Allow readers as long as the writer has not completely + * acquired the semaphore for write. + */ + if (atomic_read(&lock->readers) != WRITER_BIAS) { + atomic_inc(&lock->readers); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + return; + } + + /* + * Call into the slow lock path with the rtmutex->wait_lock + * held, so this can't result in the following race: + * + * Reader1 Reader2 Writer + * read_lock() + * write_lock() + * rtmutex_lock(m) + * swait() + * read_lock() + * unlock(m->wait_lock) + * read_unlock() + * swake() + * lock(m->wait_lock) + * lock->writelocked=true + * unlock(m->wait_lock) + * + * write_unlock() + * lock->writelocked=false + * rtmutex_unlock(m) + * read_lock() + * write_lock() + * rtmutex_lock(m) + * swait() + * rtmutex_lock(m) + * + * That would put Reader1 behind the writer waiting on + * Reader2 to call read_unlock() which might be unbound. + */ + rt_mutex_init_waiter(&waiter, true); + rt_spin_lock_slowlock_locked(m, &waiter, flags); + /* + * The slowlock() above is guaranteed to return with the rtmutex is + * now held, so there can't be a writer active. Increment the reader + * count and immediately drop the rtmutex again. + */ + atomic_inc(&lock->readers); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + rt_spin_lock_slowunlock(m); + + debug_rt_mutex_free_waiter(&waiter); +} + +static void __read_rt_unlock(struct rt_rw_lock *lock) +{ + struct rt_mutex *m = &lock->rtmutex; + struct task_struct *tsk; + + /* + * sem->readers can only hit 0 when a writer is waiting for the + * active readers to leave the critical region. + */ + if (!atomic_dec_and_test(&lock->readers)) + return; + + raw_spin_lock_irq(&m->wait_lock); + /* + * Wake the writer, i.e. the rtmutex owner. It might release the + * rtmutex concurrently in the fast path, but to clean up the rw + * lock it needs to acquire m->wait_lock. The worst case which can + * happen is a spurious wakeup. + */ + tsk = rt_mutex_owner(m); + if (tsk) + wake_up_process(tsk); + + raw_spin_unlock_irq(&m->wait_lock); +} + +static void __write_unlock_common(struct rt_rw_lock *lock, int bias, + unsigned long flags) +{ + struct rt_mutex *m = &lock->rtmutex; + + atomic_add(READER_BIAS - bias, &lock->readers); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + rt_spin_lock_slowunlock(m); +} + +static void __write_rt_lock(struct rt_rw_lock *lock) +{ + struct rt_mutex *m = &lock->rtmutex; + struct task_struct *self = current; + unsigned long flags; + + /* Take the rtmutex as a first step */ + __rt_spin_lock(m); + + /* Force readers into slow path */ + atomic_sub(READER_BIAS, &lock->readers); + + raw_spin_lock_irqsave(&m->wait_lock, flags); + + raw_spin_lock(&self->pi_lock); + self->saved_state = self->state; + __set_current_state_no_track(TASK_UNINTERRUPTIBLE); + raw_spin_unlock(&self->pi_lock); + + for (;;) { + /* Have all readers left the critical region? */ + if (!atomic_read(&lock->readers)) { + atomic_set(&lock->readers, WRITER_BIAS); + raw_spin_lock(&self->pi_lock); + __set_current_state_no_track(self->saved_state); + self->saved_state = TASK_RUNNING; + raw_spin_unlock(&self->pi_lock); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + return; + } + + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + + if (atomic_read(&lock->readers) != 0) + preempt_schedule_lock(); + + raw_spin_lock_irqsave(&m->wait_lock, flags); + + raw_spin_lock(&self->pi_lock); + __set_current_state_no_track(TASK_UNINTERRUPTIBLE); + raw_spin_unlock(&self->pi_lock); + } +} + +static int __write_rt_trylock(struct rt_rw_lock *lock) +{ + struct rt_mutex *m = &lock->rtmutex; + unsigned long flags; + + if (!__rt_mutex_trylock(m)) + return 0; + + atomic_sub(READER_BIAS, &lock->readers); + + raw_spin_lock_irqsave(&m->wait_lock, flags); + if (!atomic_read(&lock->readers)) { + atomic_set(&lock->readers, WRITER_BIAS); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + return 1; + } + __write_unlock_common(lock, 0, flags); + return 0; +} + +static void __write_rt_unlock(struct rt_rw_lock *lock) +{ + struct rt_mutex *m = &lock->rtmutex; + unsigned long flags; + + raw_spin_lock_irqsave(&m->wait_lock, flags); + __write_unlock_common(lock, WRITER_BIAS, flags); +} + +int __lockfunc rt_read_can_lock(rwlock_t *rwlock) +{ + return atomic_read(&rwlock->readers) < 0; +} + +int __lockfunc rt_write_can_lock(rwlock_t *rwlock) +{ + return atomic_read(&rwlock->readers) == READER_BIAS; +} + +/* + * The common functions which get wrapped into the rwlock API. + */ +int __lockfunc rt_read_trylock(rwlock_t *rwlock) +{ + int ret; + + ret = __read_rt_trylock(rwlock); + if (ret) { + rwlock_acquire_read(&rwlock->dep_map, 0, 1, _RET_IP_); + rcu_read_lock(); + migrate_disable(); + } + return ret; +} +EXPORT_SYMBOL(rt_read_trylock); + +int __lockfunc rt_write_trylock(rwlock_t *rwlock) +{ + int ret; + + ret = __write_rt_trylock(rwlock); + if (ret) { + rwlock_acquire(&rwlock->dep_map, 0, 1, _RET_IP_); + rcu_read_lock(); + migrate_disable(); + } + return ret; +} +EXPORT_SYMBOL(rt_write_trylock); + +void __lockfunc rt_read_lock(rwlock_t *rwlock) +{ + rwlock_acquire_read(&rwlock->dep_map, 0, 0, _RET_IP_); + __read_rt_lock(rwlock); + rcu_read_lock(); + migrate_disable(); +} +EXPORT_SYMBOL(rt_read_lock); + +void __lockfunc rt_write_lock(rwlock_t *rwlock) +{ + rwlock_acquire(&rwlock->dep_map, 0, 0, _RET_IP_); + __write_rt_lock(rwlock); + rcu_read_lock(); + migrate_disable(); +} +EXPORT_SYMBOL(rt_write_lock); + +void __lockfunc rt_read_unlock(rwlock_t *rwlock) +{ + rwlock_release(&rwlock->dep_map, _RET_IP_); + migrate_enable(); + rcu_read_unlock(); + __read_rt_unlock(rwlock); +} +EXPORT_SYMBOL(rt_read_unlock); + +void __lockfunc rt_write_unlock(rwlock_t *rwlock) +{ + rwlock_release(&rwlock->dep_map, _RET_IP_); + migrate_enable(); + rcu_read_unlock(); + __write_rt_unlock(rwlock); +} +EXPORT_SYMBOL(rt_write_unlock); + +void __rt_rwlock_init(rwlock_t *rwlock, char *name, struct lock_class_key *key) +{ + __rwlock_biased_rt_init(rwlock, name, key); +} +EXPORT_SYMBOL(__rt_rwlock_init); diff --git a/kernel/locking/rwsem-rt.c b/kernel/locking/rwsem-rt.c new file mode 100644 index 0000000000000..ab05ce0903537 --- /dev/null +++ b/kernel/locking/rwsem-rt.c @@ -0,0 +1,307 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include +#include +#include +#include +#include + +#include "rtmutex_common.h" + +/* + * RT-specific reader/writer semaphores + * + * down_write() + * 1) Lock sem->rtmutex + * 2) Remove the reader BIAS to force readers into the slow path + * 3) Wait until all readers have left the critical region + * 4) Mark it write locked + * + * up_write() + * 1) Remove the write locked marker + * 2) Set the reader BIAS so readers can use the fast path again + * 3) Unlock sem->rtmutex to release blocked readers + * + * down_read() + * 1) Try fast path acquisition (reader BIAS is set) + * 2) Take sem->rtmutex.wait_lock which protects the writelocked flag + * 3) If !writelocked, acquire it for read + * 4) If writelocked, block on sem->rtmutex + * 5) unlock sem->rtmutex, goto 1) + * + * up_read() + * 1) Try fast path release (reader count != 1) + * 2) Wake the writer waiting in down_write()#3 + * + * down_read()#3 has the consequence, that rw semaphores on RT are not writer + * fair, but writers, which should be avoided in RT tasks (think mmap_sem), + * are subject to the rtmutex priority/DL inheritance mechanism. + * + * It's possible to make the rw semaphores writer fair by keeping a list of + * active readers. A blocked writer would force all newly incoming readers to + * block on the rtmutex, but the rtmutex would have to be proxy locked for one + * reader after the other. We can't use multi-reader inheritance because there + * is no way to support that with SCHED_DEADLINE. Implementing the one by one + * reader boosting/handover mechanism is a major surgery for a very dubious + * value. + * + * The risk of writer starvation is there, but the pathological use cases + * which trigger it are not necessarily the typical RT workloads. + */ + +void __rwsem_init(struct rw_semaphore *sem, const char *name, + struct lock_class_key *key) +{ +#ifdef CONFIG_DEBUG_LOCK_ALLOC + /* + * Make sure we are not reinitializing a held semaphore: + */ + debug_check_no_locks_freed((void *)sem, sizeof(*sem)); + lockdep_init_map(&sem->dep_map, name, key, 0); +#endif + atomic_set(&sem->readers, READER_BIAS); +} +EXPORT_SYMBOL(__rwsem_init); + +int __down_read_trylock(struct rw_semaphore *sem) +{ + int r, old; + + /* + * Increment reader count, if sem->readers < 0, i.e. READER_BIAS is + * set. + */ + for (r = atomic_read(&sem->readers); r < 0;) { + old = atomic_cmpxchg(&sem->readers, r, r + 1); + if (likely(old == r)) + return 1; + r = old; + } + return 0; +} + +static int __sched __down_read_common(struct rw_semaphore *sem, int state) +{ + struct rt_mutex *m = &sem->rtmutex; + struct rt_mutex_waiter waiter; + int ret; + + if (__down_read_trylock(sem)) + return 0; + + /* + * Flush blk before ->pi_blocked_on is set. At schedule() time it is too + * late if one of the callbacks needs to acquire a sleeping lock. + */ + if (blk_needs_flush_plug(current)) + blk_schedule_flush_plug(current); + + might_sleep(); + raw_spin_lock_irq(&m->wait_lock); + /* + * Allow readers as long as the writer has not completely + * acquired the semaphore for write. + */ + if (atomic_read(&sem->readers) != WRITER_BIAS) { + atomic_inc(&sem->readers); + raw_spin_unlock_irq(&m->wait_lock); + return 0; + } + + /* + * Call into the slow lock path with the rtmutex->wait_lock + * held, so this can't result in the following race: + * + * Reader1 Reader2 Writer + * down_read() + * down_write() + * rtmutex_lock(m) + * swait() + * down_read() + * unlock(m->wait_lock) + * up_read() + * swake() + * lock(m->wait_lock) + * sem->writelocked=true + * unlock(m->wait_lock) + * + * up_write() + * sem->writelocked=false + * rtmutex_unlock(m) + * down_read() + * down_write() + * rtmutex_lock(m) + * swait() + * rtmutex_lock(m) + * + * That would put Reader1 behind the writer waiting on + * Reader2 to call up_read() which might be unbound. + */ + rt_mutex_init_waiter(&waiter, false); + ret = rt_mutex_slowlock_locked(m, state, NULL, RT_MUTEX_MIN_CHAINWALK, + NULL, &waiter); + /* + * The slowlock() above is guaranteed to return with the rtmutex (for + * ret = 0) is now held, so there can't be a writer active. Increment + * the reader count and immediately drop the rtmutex again. + * For ret != 0 we don't hold the rtmutex and need unlock the wait_lock. + * We don't own the lock then. + */ + if (!ret) + atomic_inc(&sem->readers); + raw_spin_unlock_irq(&m->wait_lock); + if (!ret) + __rt_mutex_unlock(m); + + debug_rt_mutex_free_waiter(&waiter); + return ret; +} + +void __down_read(struct rw_semaphore *sem) +{ + int ret; + + ret = __down_read_common(sem, TASK_UNINTERRUPTIBLE); + WARN_ON_ONCE(ret); +} + +int __down_read_killable(struct rw_semaphore *sem) +{ + int ret; + + ret = __down_read_common(sem, TASK_KILLABLE); + if (likely(!ret)) + return ret; + WARN_ONCE(ret != -EINTR, "Unexpected state: %d\n", ret); + return -EINTR; +} + +void __up_read(struct rw_semaphore *sem) +{ + struct rt_mutex *m = &sem->rtmutex; + struct task_struct *tsk; + + /* + * sem->readers can only hit 0 when a writer is waiting for the + * active readers to leave the critical region. + */ + if (!atomic_dec_and_test(&sem->readers)) + return; + + might_sleep(); + raw_spin_lock_irq(&m->wait_lock); + /* + * Wake the writer, i.e. the rtmutex owner. It might release the + * rtmutex concurrently in the fast path (due to a signal), but to + * clean up the rwsem it needs to acquire m->wait_lock. The worst + * case which can happen is a spurious wakeup. + */ + tsk = rt_mutex_owner(m); + if (tsk) + wake_up_process(tsk); + + raw_spin_unlock_irq(&m->wait_lock); +} + +static void __up_write_unlock(struct rw_semaphore *sem, int bias, + unsigned long flags) +{ + struct rt_mutex *m = &sem->rtmutex; + + atomic_add(READER_BIAS - bias, &sem->readers); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + __rt_mutex_unlock(m); +} + +static int __sched __down_write_common(struct rw_semaphore *sem, int state) +{ + struct rt_mutex *m = &sem->rtmutex; + unsigned long flags; + + /* + * Flush blk before ->pi_blocked_on is set. At schedule() time it is too + * late if one of the callbacks needs to acquire a sleeping lock. + */ + if (blk_needs_flush_plug(current)) + blk_schedule_flush_plug(current); + + /* Take the rtmutex as a first step */ + if (__rt_mutex_lock_state(m, state)) + return -EINTR; + + /* Force readers into slow path */ + atomic_sub(READER_BIAS, &sem->readers); + might_sleep(); + + set_current_state(state); + for (;;) { + raw_spin_lock_irqsave(&m->wait_lock, flags); + /* Have all readers left the critical region? */ + if (!atomic_read(&sem->readers)) { + atomic_set(&sem->readers, WRITER_BIAS); + __set_current_state(TASK_RUNNING); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + return 0; + } + + if (signal_pending_state(state, current)) { + __set_current_state(TASK_RUNNING); + __up_write_unlock(sem, 0, flags); + return -EINTR; + } + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + + if (atomic_read(&sem->readers) != 0) { + schedule(); + set_current_state(state); + } + } +} + +void __sched __down_write(struct rw_semaphore *sem) +{ + __down_write_common(sem, TASK_UNINTERRUPTIBLE); +} + +int __sched __down_write_killable(struct rw_semaphore *sem) +{ + return __down_write_common(sem, TASK_KILLABLE); +} + +int __down_write_trylock(struct rw_semaphore *sem) +{ + struct rt_mutex *m = &sem->rtmutex; + unsigned long flags; + + if (!__rt_mutex_trylock(m)) + return 0; + + atomic_sub(READER_BIAS, &sem->readers); + + raw_spin_lock_irqsave(&m->wait_lock, flags); + if (!atomic_read(&sem->readers)) { + atomic_set(&sem->readers, WRITER_BIAS); + raw_spin_unlock_irqrestore(&m->wait_lock, flags); + return 1; + } + __up_write_unlock(sem, 0, flags); + return 0; +} + +void __up_write(struct rw_semaphore *sem) +{ + struct rt_mutex *m = &sem->rtmutex; + unsigned long flags; + + raw_spin_lock_irqsave(&m->wait_lock, flags); + __up_write_unlock(sem, WRITER_BIAS, flags); +} + +void __downgrade_write(struct rw_semaphore *sem) +{ + struct rt_mutex *m = &sem->rtmutex; + unsigned long flags; + + raw_spin_lock_irqsave(&m->wait_lock, flags); + /* Release it and account current as reader */ + __up_write_unlock(sem, WRITER_BIAS - 1, flags); +} diff --git a/kernel/locking/rwsem.c b/kernel/locking/rwsem.c index f11b9bd3431d2..fce8a6e3fa7c3 100644 --- a/kernel/locking/rwsem.c +++ b/kernel/locking/rwsem.c @@ -28,6 +28,7 @@ #include #include +#ifndef CONFIG_PREEMPT_RT #include "lock_events.h" /* @@ -1482,6 +1483,7 @@ static inline void __downgrade_write(struct rw_semaphore *sem) if (tmp & RWSEM_FLAG_WAITERS) rwsem_downgrade_wake(sem); } +#endif /* * lock for reading @@ -1617,7 +1619,9 @@ void down_read_non_owner(struct rw_semaphore *sem) { might_sleep(); __down_read(sem); +#ifndef CONFIG_PREEMPT_RT __rwsem_set_reader_owned(sem, NULL); +#endif } EXPORT_SYMBOL(down_read_non_owner); @@ -1646,7 +1650,9 @@ EXPORT_SYMBOL(down_write_killable_nested); void up_read_non_owner(struct rw_semaphore *sem) { +#ifndef CONFIG_PREEMPT_RT DEBUG_RWSEMS_WARN_ON(!is_rwsem_reader_owned(sem), sem); +#endif __up_read(sem); } EXPORT_SYMBOL(up_read_non_owner); diff --git a/kernel/locking/spinlock.c b/kernel/locking/spinlock.c index 0ff08380f5318..45445a2f1799a 100644 --- a/kernel/locking/spinlock.c +++ b/kernel/locking/spinlock.c @@ -124,8 +124,11 @@ void __lockfunc __raw_##op##_lock_bh(locktype##_t *lock) \ * __[spin|read|write]_lock_bh() */ BUILD_LOCK_OPS(spin, raw_spinlock); + +#ifndef CONFIG_PREEMPT_RT BUILD_LOCK_OPS(read, rwlock); BUILD_LOCK_OPS(write, rwlock); +#endif #endif @@ -209,6 +212,8 @@ void __lockfunc _raw_spin_unlock_bh(raw_spinlock_t *lock) EXPORT_SYMBOL(_raw_spin_unlock_bh); #endif +#ifndef CONFIG_PREEMPT_RT + #ifndef CONFIG_INLINE_READ_TRYLOCK int __lockfunc _raw_read_trylock(rwlock_t *lock) { @@ -353,6 +358,8 @@ void __lockfunc _raw_write_unlock_bh(rwlock_t *lock) EXPORT_SYMBOL(_raw_write_unlock_bh); #endif +#endif /* !PREEMPT_RT */ + #ifdef CONFIG_DEBUG_LOCK_ALLOC void __lockfunc _raw_spin_lock_nested(raw_spinlock_t *lock, int subclass) diff --git a/kernel/locking/spinlock_debug.c b/kernel/locking/spinlock_debug.c index b9d93087ee669..72e306e0e8a3b 100644 --- a/kernel/locking/spinlock_debug.c +++ b/kernel/locking/spinlock_debug.c @@ -31,6 +31,7 @@ void __raw_spin_lock_init(raw_spinlock_t *lock, const char *name, EXPORT_SYMBOL(__raw_spin_lock_init); +#ifndef CONFIG_PREEMPT_RT void __rwlock_init(rwlock_t *lock, const char *name, struct lock_class_key *key) { @@ -48,6 +49,7 @@ void __rwlock_init(rwlock_t *lock, const char *name, } EXPORT_SYMBOL(__rwlock_init); +#endif static void spin_dump(raw_spinlock_t *lock, const char *msg) { @@ -139,6 +141,7 @@ void do_raw_spin_unlock(raw_spinlock_t *lock) arch_spin_unlock(&lock->raw_lock); } +#ifndef CONFIG_PREEMPT_RT static void rwlock_bug(rwlock_t *lock, const char *msg) { if (!debug_locks_off()) @@ -228,3 +231,5 @@ void do_raw_write_unlock(rwlock_t *lock) debug_write_unlock(lock); arch_write_unlock(&lock->raw_lock); } + +#endif diff --git a/kernel/panic.c b/kernel/panic.c index aef8872ba8435..d563542bc7eb1 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -247,7 +247,6 @@ void panic(const char *fmt, ...) * Bypass the panic_cpu check and call __crash_kexec directly. */ if (!_crash_kexec_post_notifiers) { - printk_safe_flush_on_panic(); __crash_kexec(NULL); /* @@ -271,8 +270,6 @@ void panic(const char *fmt, ...) */ atomic_notifier_call_chain(&panic_notifier_list, 0, buf); - /* Call flush even twice. It tries harder with a single online CPU */ - printk_safe_flush_on_panic(); kmsg_dump(KMSG_DUMP_PANIC); /* @@ -542,9 +539,11 @@ static u64 oops_id; static int init_oops_id(void) { +#ifndef CONFIG_PREEMPT_RT if (!oops_id) get_random_bytes(&oops_id, sizeof(oops_id)); else +#endif oops_id++; return 0; diff --git a/kernel/printk/Makefile b/kernel/printk/Makefile index 4d052fc6bcde7..59cb24e25f004 100644 --- a/kernel/printk/Makefile +++ b/kernel/printk/Makefile @@ -1,4 +1,4 @@ # SPDX-License-Identifier: GPL-2.0-only obj-y = printk.o -obj-$(CONFIG_PRINTK) += printk_safe.o obj-$(CONFIG_A11Y_BRAILLE_CONSOLE) += braille.o +obj-$(CONFIG_PRINTK) += printk_ringbuffer.o diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h deleted file mode 100644 index 660f9a6bf73a1..0000000000000 --- a/kernel/printk/internal.h +++ /dev/null @@ -1,74 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-or-later */ -/* - * internal.h - printk internal definitions - */ -#include - -#ifdef CONFIG_PRINTK - -#define PRINTK_SAFE_CONTEXT_MASK 0x007ffffff -#define PRINTK_NMI_DIRECT_CONTEXT_MASK 0x008000000 -#define PRINTK_NMI_CONTEXT_MASK 0xff0000000 - -#define PRINTK_NMI_CONTEXT_OFFSET 0x010000000 - -extern raw_spinlock_t logbuf_lock; - -__printf(5, 0) -int vprintk_store(int facility, int level, - const char *dict, size_t dictlen, - const char *fmt, va_list args); - -__printf(1, 0) int vprintk_default(const char *fmt, va_list args); -__printf(1, 0) int vprintk_deferred(const char *fmt, va_list args); -__printf(1, 0) int vprintk_func(const char *fmt, va_list args); -void __printk_safe_enter(void); -void __printk_safe_exit(void); - -void printk_safe_init(void); -bool printk_percpu_data_ready(void); - -#define printk_safe_enter_irqsave(flags) \ - do { \ - local_irq_save(flags); \ - __printk_safe_enter(); \ - } while (0) - -#define printk_safe_exit_irqrestore(flags) \ - do { \ - __printk_safe_exit(); \ - local_irq_restore(flags); \ - } while (0) - -#define printk_safe_enter_irq() \ - do { \ - local_irq_disable(); \ - __printk_safe_enter(); \ - } while (0) - -#define printk_safe_exit_irq() \ - do { \ - __printk_safe_exit(); \ - local_irq_enable(); \ - } while (0) - -void defer_console_output(void); - -#else - -__printf(1, 0) int vprintk_func(const char *fmt, va_list args) { return 0; } - -/* - * In !PRINTK builds we still export logbuf_lock spin_lock, console_sem - * semaphore and some of console functions (console_unlock()/etc.), so - * printk-safe must preserve the existing local IRQ guarantees. - */ -#define printk_safe_enter_irqsave(flags) local_irq_save(flags) -#define printk_safe_exit_irqrestore(flags) local_irq_restore(flags) - -#define printk_safe_enter_irq() local_irq_disable() -#define printk_safe_exit_irq() local_irq_enable() - -static inline void printk_safe_init(void) { } -static inline bool printk_percpu_data_ready(void) { return false; } -#endif /* CONFIG_PRINTK */ diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 9b75f6bfc3333..0c56873396a99 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -44,9 +44,12 @@ #include #include #include +#include +#include #include #include #include +#include #include #include @@ -55,9 +58,9 @@ #define CREATE_TRACE_POINTS #include +#include "printk_ringbuffer.h" #include "console_cmdline.h" #include "braille.h" -#include "internal.h" int console_printk[4] = { CONSOLE_LOGLEVEL_DEFAULT, /* console_loglevel */ @@ -77,6 +80,9 @@ EXPORT_SYMBOL(ignore_console_lock_warning); int oops_in_progress; EXPORT_SYMBOL(oops_in_progress); +/* Set to enable sync mode. Once set, it is never cleared. */ +static bool sync_mode; + /* * console_sem protects the console_drivers list, and also * provides serialisation for access to the entire console @@ -224,19 +230,7 @@ static int nr_ext_console_drivers; static int __down_trylock_console_sem(unsigned long ip) { - int lock_failed; - unsigned long flags; - - /* - * Here and in __up_console_sem() we need to be in safe mode, - * because spindump/WARN/etc from under console ->lock will - * deadlock in printk()->down_trylock_console_sem() otherwise. - */ - printk_safe_enter_irqsave(flags); - lock_failed = down_trylock(&console_sem); - printk_safe_exit_irqrestore(flags); - - if (lock_failed) + if (down_trylock(&console_sem)) return 1; mutex_acquire(&console_lock_dep_map, 0, 1, ip); return 0; @@ -245,13 +239,9 @@ static int __down_trylock_console_sem(unsigned long ip) static void __up_console_sem(unsigned long ip) { - unsigned long flags; - mutex_release(&console_lock_dep_map, ip); - printk_safe_enter_irqsave(flags); up(&console_sem); - printk_safe_exit_irqrestore(flags); } #define up_console_sem() __up_console_sem(_RET_IP_) @@ -265,11 +255,6 @@ static void __up_console_sem(unsigned long ip) */ static int console_locked, console_suspended; -/* - * If exclusive_console is non-NULL then only this console is to be printed to. - */ -static struct console *exclusive_console; - /* * Array of consoles built from command line options (console=) */ @@ -294,30 +279,22 @@ enum con_msg_format_flags { static int console_msg_format = MSG_FORMAT_DEFAULT; /* - * The printk log buffer consists of a chain of concatenated variable - * length records. Every record starts with a record header, containing - * the overall length of the record. + * The printk log buffer consists of a sequenced collection of records, each + * containing variable length message text. Every record also contains its + * own meta-data (@info). * - * The heads to the first and last entry in the buffer, as well as the - * sequence numbers of these entries are maintained when messages are - * stored. + * Every record meta-data carries the timestamp in microseconds, as well as + * the standard userspace syslog level and syslog facility. The usual kernel + * messages use LOG_KERN; userspace-injected messages always carry a matching + * syslog facility, by default LOG_USER. The origin of every message can be + * reliably determined that way. * - * If the heads indicate available messages, the length in the header - * tells the start next message. A length == 0 for the next message - * indicates a wrap-around to the beginning of the buffer. + * The human readable log message of a record is available in @text, the + * length of the message text in @text_len. The stored message is not + * terminated. * - * Every record carries the monotonic timestamp in microseconds, as well as - * the standard userspace syslog level and syslog facility. The usual - * kernel messages use LOG_KERN; userspace-injected messages always carry - * a matching syslog facility, by default LOG_USER. The origin of every - * message can be reliably determined that way. - * - * The human readable log message directly follows the message header. The - * length of the message text is stored in the header, the stored message - * is not terminated. - * - * Optionally, a message can carry a dictionary of properties (key/value pairs), - * to provide userspace with a machine-readable message context. + * Optionally, a record can carry a dictionary of properties (key/value + * pairs), to provide userspace with a machine-readable message context. * * Examples for well-defined, commonly used property names are: * DEVICE=b12:8 device identifier @@ -327,25 +304,22 @@ static int console_msg_format = MSG_FORMAT_DEFAULT; * +sound:card0 subsystem:devname * SUBSYSTEM=pci driver-core subsystem name * - * Valid characters in property names are [a-zA-Z0-9.-_]. The plain text value - * follows directly after a '=' character. Every property is terminated by - * a '\0' character. The last property is not terminated. + * Valid characters in property names are [a-zA-Z0-9.-_]. Property names + * and values are terminated by a '\0' character. * - * Example of a message structure: - * 0000 ff 8f 00 00 00 00 00 00 monotonic time in nsec - * 0008 34 00 record is 52 bytes long - * 000a 0b 00 text is 11 bytes long - * 000c 1f 00 dictionary is 23 bytes long - * 000e 03 00 LOG_KERN (facility) LOG_ERR (level) - * 0010 69 74 27 73 20 61 20 6c "it's a l" - * 69 6e 65 "ine" - * 001b 44 45 56 49 43 "DEVIC" - * 45 3d 62 38 3a 32 00 44 "E=b8:2\0D" - * 52 49 56 45 52 3d 62 75 "RIVER=bu" - * 67 "g" - * 0032 00 00 00 padding to next message header + * Example of record values: + * record.text_buf = "it's a line" (unterminated) + * record.info.seq = 56 + * record.info.ts_nsec = 36863 + * record.info.text_len = 11 + * record.info.facility = 0 (LOG_KERN) + * record.info.flags = 0 + * record.info.level = 3 (LOG_ERR) + * record.info.caller_id = 299 (task 299) + * record.info.dev_info.subsystem = "pci" (terminated) + * record.info.dev_info.device = "+pci:0000:00:01.0" (terminated) * - * The 'struct printk_log' buffer header must never be directly exported to + * The 'struct printk_info' buffer must never be directly exported to * userspace, it is a kernel-private implementation detail that might * need to be changed in the future, when the requirements change. * @@ -365,82 +339,23 @@ enum log_flags { LOG_CONT = 8, /* text is a fragment of a continuation line */ }; -struct printk_log { - u64 ts_nsec; /* timestamp in nanoseconds */ - u16 len; /* length of entire record */ - u16 text_len; /* length of text buffer */ - u16 dict_len; /* length of dictionary buffer */ - u8 facility; /* syslog facility */ - u8 flags:5; /* internal record flags */ - u8 level:3; /* syslog level */ -#ifdef CONFIG_PRINTK_CALLER - u32 caller_id; /* thread id or processor id */ -#endif -} -#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS -__packed __aligned(4) -#endif -; - -/* - * The logbuf_lock protects kmsg buffer, indices, counters. This can be taken - * within the scheduler's rq lock. It must be released before calling - * console_unlock() or anything else that might wake up a process. - */ -DEFINE_RAW_SPINLOCK(logbuf_lock); - -/* - * Helper macros to lock/unlock logbuf_lock and switch between - * printk-safe/unsafe modes. - */ -#define logbuf_lock_irq() \ - do { \ - printk_safe_enter_irq(); \ - raw_spin_lock(&logbuf_lock); \ - } while (0) - -#define logbuf_unlock_irq() \ - do { \ - raw_spin_unlock(&logbuf_lock); \ - printk_safe_exit_irq(); \ - } while (0) - -#define logbuf_lock_irqsave(flags) \ - do { \ - printk_safe_enter_irqsave(flags); \ - raw_spin_lock(&logbuf_lock); \ - } while (0) - -#define logbuf_unlock_irqrestore(flags) \ - do { \ - raw_spin_unlock(&logbuf_lock); \ - printk_safe_exit_irqrestore(flags); \ - } while (0) +/* The syslog_lock protects syslog_* variables. */ +static DEFINE_SPINLOCK(syslog_lock); +#define syslog_lock_irq() spin_lock_irq(&syslog_lock) +#define syslog_unlock_irq() spin_unlock_irq(&syslog_lock) +#define syslog_lock_irqsave(flags) spin_lock_irqsave(&syslog_lock, flags) +#define syslog_unlock_irqrestore(flags) spin_unlock_irqrestore(&syslog_lock, flags) #ifdef CONFIG_PRINTK DECLARE_WAIT_QUEUE_HEAD(log_wait); +/* All 3 protected by @syslog_lock. */ /* the next printk record to read by syslog(READ) or /proc/kmsg */ static u64 syslog_seq; -static u32 syslog_idx; static size_t syslog_partial; static bool syslog_time; -/* index and sequence number of the first record stored in the buffer */ -static u64 log_first_seq; -static u32 log_first_idx; - -/* index and sequence number of the next record to store in the buffer */ -static u64 log_next_seq; -static u32 log_next_idx; - -/* the next printk record to write to the console */ -static u64 console_seq; -static u32 console_idx; -static u64 exclusive_console_stop_seq; - /* the next printk record to read after the last 'clear' command */ -static u64 clear_seq; -static u32 clear_idx; +static atomic64_t clear_seq = ATOMIC64_INIT(0); #ifdef CONFIG_PRINTK_CALLER #define PREFIX_MAX 48 @@ -453,13 +368,30 @@ static u32 clear_idx; #define LOG_FACILITY(v) ((v) >> 3 & 0xff) /* record buffer */ -#define LOG_ALIGN __alignof__(struct printk_log) +#define LOG_ALIGN __alignof__(unsigned long) #define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT) #define LOG_BUF_LEN_MAX (u32)(1 << 31) static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN); static char *log_buf = __log_buf; static u32 log_buf_len = __LOG_BUF_LEN; +/* + * Define the average message size. This only affects the number of + * descriptors that will be available. Underestimating is better than + * overestimating (too many available descriptors is better than not enough). + */ +#define PRB_AVGBITS 5 /* 32 character average length */ + +#if CONFIG_LOG_BUF_SHIFT <= PRB_AVGBITS +#error CONFIG_LOG_BUF_SHIFT value too small. +#endif +_DEFINE_PRINTKRB(printk_rb_static, CONFIG_LOG_BUF_SHIFT - PRB_AVGBITS, + PRB_AVGBITS, &__log_buf[0]); + +static struct printk_ringbuffer printk_rb_dynamic; + +static struct printk_ringbuffer *prb = &printk_rb_static; + /* * We cannot access per-CPU data (e.g. per-CPU flush irq_work) before * per_cpu_areas are initialised. This variable is set to true when @@ -467,7 +399,7 @@ static u32 log_buf_len = __LOG_BUF_LEN; */ static bool __printk_percpu_data_ready __read_mostly; -bool printk_percpu_data_ready(void) +static bool printk_percpu_data_ready(void) { return __printk_percpu_data_ready; } @@ -484,108 +416,6 @@ u32 log_buf_len_get(void) return log_buf_len; } -/* human readable text of the record */ -static char *log_text(const struct printk_log *msg) -{ - return (char *)msg + sizeof(struct printk_log); -} - -/* optional key/value pair dictionary attached to the record */ -static char *log_dict(const struct printk_log *msg) -{ - return (char *)msg + sizeof(struct printk_log) + msg->text_len; -} - -/* get record by index; idx must point to valid msg */ -static struct printk_log *log_from_idx(u32 idx) -{ - struct printk_log *msg = (struct printk_log *)(log_buf + idx); - - /* - * A length == 0 record is the end of buffer marker. Wrap around and - * read the message at the start of the buffer. - */ - if (!msg->len) - return (struct printk_log *)log_buf; - return msg; -} - -/* get next record; idx must point to valid msg */ -static u32 log_next(u32 idx) -{ - struct printk_log *msg = (struct printk_log *)(log_buf + idx); - - /* length == 0 indicates the end of the buffer; wrap */ - /* - * A length == 0 record is the end of buffer marker. Wrap around and - * read the message at the start of the buffer as *this* one, and - * return the one after that. - */ - if (!msg->len) { - msg = (struct printk_log *)log_buf; - return msg->len; - } - return idx + msg->len; -} - -/* - * Check whether there is enough free space for the given message. - * - * The same values of first_idx and next_idx mean that the buffer - * is either empty or full. - * - * If the buffer is empty, we must respect the position of the indexes. - * They cannot be reset to the beginning of the buffer. - */ -static int logbuf_has_space(u32 msg_size, bool empty) -{ - u32 free; - - if (log_next_idx > log_first_idx || empty) - free = max(log_buf_len - log_next_idx, log_first_idx); - else - free = log_first_idx - log_next_idx; - - /* - * We need space also for an empty header that signalizes wrapping - * of the buffer. - */ - return free >= msg_size + sizeof(struct printk_log); -} - -static int log_make_free_space(u32 msg_size) -{ - while (log_first_seq < log_next_seq && - !logbuf_has_space(msg_size, false)) { - /* drop old messages until we have enough contiguous space */ - log_first_idx = log_next(log_first_idx); - log_first_seq++; - } - - if (clear_seq < log_first_seq) { - clear_seq = log_first_seq; - clear_idx = log_first_idx; - } - - /* sequence numbers are equal, so the log buffer is empty */ - if (logbuf_has_space(msg_size, log_first_seq == log_next_seq)) - return 0; - - return -ENOMEM; -} - -/* compute the message size including the padding bytes */ -static u32 msg_used_size(u16 text_len, u16 dict_len, u32 *pad_len) -{ - u32 size; - - size = sizeof(struct printk_log) + text_len + dict_len; - *pad_len = (-size) & (LOG_ALIGN - 1); - size += *pad_len; - - return size; -} - /* * Define how much of the log buffer we could take at maximum. The value * must be greater than two. Note that only half of the buffer is available @@ -594,84 +424,23 @@ static u32 msg_used_size(u16 text_len, u16 dict_len, u32 *pad_len) #define MAX_LOG_TAKE_PART 4 static const char trunc_msg[] = ""; -static u32 truncate_msg(u16 *text_len, u16 *trunc_msg_len, - u16 *dict_len, u32 *pad_len) +static void truncate_msg(u16 *text_len, u16 *trunc_msg_len) { /* * The message should not take the whole buffer. Otherwise, it might * get removed too soon. */ u32 max_text_len = log_buf_len / MAX_LOG_TAKE_PART; + if (*text_len > max_text_len) *text_len = max_text_len; - /* enable the warning message */ + + /* enable the warning message (if there is room) */ *trunc_msg_len = strlen(trunc_msg); - /* disable the "dict" completely */ - *dict_len = 0; - /* compute the size again, count also the warning message */ - return msg_used_size(*text_len + *trunc_msg_len, 0, pad_len); -} - -/* insert record into the buffer, discard old ones, update heads */ -static int log_store(u32 caller_id, int facility, int level, - enum log_flags flags, u64 ts_nsec, - const char *dict, u16 dict_len, - const char *text, u16 text_len) -{ - struct printk_log *msg; - u32 size, pad_len; - u16 trunc_msg_len = 0; - - /* number of '\0' padding bytes to next message */ - size = msg_used_size(text_len, dict_len, &pad_len); - - if (log_make_free_space(size)) { - /* truncate the message if it is too long for empty buffer */ - size = truncate_msg(&text_len, &trunc_msg_len, - &dict_len, &pad_len); - /* survive when the log buffer is too small for trunc_msg */ - if (log_make_free_space(size)) - return 0; - } - - if (log_next_idx + size + sizeof(struct printk_log) > log_buf_len) { - /* - * This message + an additional empty header does not fit - * at the end of the buffer. Add an empty header with len == 0 - * to signify a wrap around. - */ - memset(log_buf + log_next_idx, 0, sizeof(struct printk_log)); - log_next_idx = 0; - } - - /* fill message */ - msg = (struct printk_log *)(log_buf + log_next_idx); - memcpy(log_text(msg), text, text_len); - msg->text_len = text_len; - if (trunc_msg_len) { - memcpy(log_text(msg) + text_len, trunc_msg, trunc_msg_len); - msg->text_len += trunc_msg_len; - } - memcpy(log_dict(msg), dict, dict_len); - msg->dict_len = dict_len; - msg->facility = facility; - msg->level = level & 7; - msg->flags = flags & 0x1f; - if (ts_nsec > 0) - msg->ts_nsec = ts_nsec; + if (*text_len >= *trunc_msg_len) + *text_len -= *trunc_msg_len; else - msg->ts_nsec = local_clock(); -#ifdef CONFIG_PRINTK_CALLER - msg->caller_id = caller_id; -#endif - memset(log_dict(msg) + dict_len, 0, pad_len); - msg->len = size; - - /* insert message */ - log_next_idx += msg->len; - log_next_seq++; - - return msg->text_len; + *trunc_msg_len = 0; } int dmesg_restrict = IS_ENABLED(CONFIG_SECURITY_DMESG_RESTRICT); @@ -723,13 +492,13 @@ static void append_char(char **pp, char *e, char c) *(*pp)++ = c; } -static ssize_t msg_print_ext_header(char *buf, size_t size, - struct printk_log *msg, u64 seq) +static ssize_t info_print_ext_header(char *buf, size_t size, + struct printk_info *info) { - u64 ts_usec = msg->ts_nsec; + u64 ts_usec = info->ts_nsec; char caller[20]; #ifdef CONFIG_PRINTK_CALLER - u32 id = msg->caller_id; + u32 id = info->caller_id; snprintf(caller, sizeof(caller), ",caller=%c%u", id & 0x80000000 ? 'C' : 'T', id & ~0x80000000); @@ -740,13 +509,13 @@ static ssize_t msg_print_ext_header(char *buf, size_t size, do_div(ts_usec, 1000); return scnprintf(buf, size, "%u,%llu,%llu,%c%s;", - (msg->facility << 3) | msg->level, seq, ts_usec, - msg->flags & LOG_CONT ? 'c' : '-', caller); + (info->facility << 3) | info->level, info->seq, + ts_usec, info->flags & LOG_CONT ? 'c' : '-', caller); } -static ssize_t msg_print_ext_body(char *buf, size_t size, - char *dict, size_t dict_len, - char *text, size_t text_len) +static ssize_t msg_add_ext_text(char *buf, size_t size, + const char *text, size_t text_len, + unsigned char endc) { char *p = buf, *e = buf + size; size_t i; @@ -760,45 +529,56 @@ static ssize_t msg_print_ext_body(char *buf, size_t size, else append_char(&p, e, c); } - append_char(&p, e, '\n'); - - if (dict_len) { - bool line = true; - - for (i = 0; i < dict_len; i++) { - unsigned char c = dict[i]; - - if (line) { - append_char(&p, e, ' '); - line = false; - } - - if (c == '\0') { - append_char(&p, e, '\n'); - line = true; - continue; - } - - if (c < ' ' || c >= 127 || c == '\\') { - p += scnprintf(p, e - p, "\\x%02x", c); - continue; - } - - append_char(&p, e, c); - } - append_char(&p, e, '\n'); - } + append_char(&p, e, endc); return p - buf; } +static ssize_t msg_add_dict_text(char *buf, size_t size, + const char *key, const char *val) +{ + size_t val_len = strlen(val); + ssize_t len; + + if (!val_len) + return 0; + + len = msg_add_ext_text(buf, size, "", 0, ' '); /* dict prefix */ + len += msg_add_ext_text(buf + len, size - len, key, strlen(key), '='); + len += msg_add_ext_text(buf + len, size - len, val, val_len, '\n'); + + return len; +} + +static ssize_t msg_print_ext_body(char *buf, size_t size, + char *text, size_t text_len, + struct dev_printk_info *dev_info) +{ + ssize_t len; + + len = msg_add_ext_text(buf, size, text, text_len, '\n'); + + if (!dev_info) + goto out; + + len += msg_add_dict_text(buf + len, size - len, "SUBSYSTEM", + dev_info->subsystem); + len += msg_add_dict_text(buf + len, size - len, "DEVICE", + dev_info->device); +out: + return len; +} + /* /dev/kmsg - userspace message inject/listen interface */ struct devkmsg_user { u64 seq; - u32 idx; struct ratelimit_state rs; struct mutex lock; char buf[CONSOLE_EXT_LOG_MAX]; + + struct printk_info info; + char text_buf[CONSOLE_EXT_LOG_MAX]; + struct printk_record record; }; static __printf(3, 4) __cold @@ -808,7 +588,7 @@ int devkmsg_emit(int facility, int level, const char *fmt, ...) int r; va_start(args, fmt); - r = vprintk_emit(facility, level, NULL, 0, fmt, args); + r = vprintk_emit(facility, level, NULL, fmt, args); va_end(args); return r; @@ -881,7 +661,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { struct devkmsg_user *user = file->private_data; - struct printk_log *msg; + struct printk_record *r = &user->record; size_t len; ssize_t ret; @@ -892,41 +672,31 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, if (ret) return ret; - logbuf_lock_irq(); - while (user->seq == log_next_seq) { + if (!prb_read_valid(prb, user->seq, r)) { if (file->f_flags & O_NONBLOCK) { ret = -EAGAIN; - logbuf_unlock_irq(); goto out; } - logbuf_unlock_irq(); ret = wait_event_interruptible(log_wait, - user->seq != log_next_seq); + prb_read_valid(prb, user->seq, r)); if (ret) goto out; - logbuf_lock_irq(); } - if (user->seq < log_first_seq) { + if (user->seq < prb_first_valid_seq(prb)) { /* our last seen message is gone, return error and reset */ - user->idx = log_first_idx; - user->seq = log_first_seq; + user->seq = prb_first_valid_seq(prb); ret = -EPIPE; - logbuf_unlock_irq(); goto out; } - msg = log_from_idx(user->idx); - len = msg_print_ext_header(user->buf, sizeof(user->buf), - msg, user->seq); + len = info_print_ext_header(user->buf, sizeof(user->buf), r->info); len += msg_print_ext_body(user->buf + len, sizeof(user->buf) - len, - log_dict(msg), msg->dict_len, - log_text(msg), msg->text_len); + &r->text_buf[0], r->info->text_len, + &r->info->dev_info); - user->idx = log_next(user->idx); - user->seq++; - logbuf_unlock_irq(); + user->seq = r->info->seq + 1; if (len > count) { ret = -EINVAL; @@ -961,12 +731,10 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) if (offset) return -ESPIPE; - logbuf_lock_irq(); switch (whence) { case SEEK_SET: /* the first record */ - user->idx = log_first_idx; - user->seq = log_first_seq; + user->seq = prb_first_valid_seq(prb); break; case SEEK_DATA: /* @@ -974,18 +742,15 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) * like issued by 'dmesg -c'. Reading /dev/kmsg itself * changes no global state, and does not clear anything. */ - user->idx = clear_idx; - user->seq = clear_seq; + user->seq = atomic64_read(&clear_seq); break; case SEEK_END: /* after the last record */ - user->idx = log_next_idx; - user->seq = log_next_seq; + user->seq = prb_next_seq(prb); break; default: ret = -EINVAL; } - logbuf_unlock_irq(); return ret; } @@ -999,15 +764,13 @@ static __poll_t devkmsg_poll(struct file *file, poll_table *wait) poll_wait(file, &log_wait, wait); - logbuf_lock_irq(); - if (user->seq < log_next_seq) { + if (prb_read_valid(prb, user->seq, NULL)) { /* return error when data has vanished underneath us */ - if (user->seq < log_first_seq) + if (user->seq < prb_first_valid_seq(prb)) ret = EPOLLIN|EPOLLRDNORM|EPOLLERR|EPOLLPRI; else ret = EPOLLIN|EPOLLRDNORM; } - logbuf_unlock_irq(); return ret; } @@ -1037,10 +800,10 @@ static int devkmsg_open(struct inode *inode, struct file *file) mutex_init(&user->lock); - logbuf_lock_irq(); - user->idx = log_first_idx; - user->seq = log_first_seq; - logbuf_unlock_irq(); + prb_rec_init_rd(&user->record, &user->info, + &user->text_buf[0], sizeof(user->text_buf)); + + user->seq = prb_first_valid_seq(prb); file->private_data = user; return 0; @@ -1080,23 +843,61 @@ const struct file_operations kmsg_fops = { */ void log_buf_vmcoreinfo_setup(void) { - VMCOREINFO_SYMBOL(log_buf); - VMCOREINFO_SYMBOL(log_buf_len); - VMCOREINFO_SYMBOL(log_first_idx); - VMCOREINFO_SYMBOL(clear_idx); - VMCOREINFO_SYMBOL(log_next_idx); + struct dev_printk_info *dev_info = NULL; + + VMCOREINFO_SYMBOL(prb); + VMCOREINFO_SYMBOL(printk_rb_static); + VMCOREINFO_SYMBOL(clear_seq); + /* - * Export struct printk_log size and field offsets. User space tools can + * Export struct size and field offsets. User space tools can * parse it and detect any changes to structure down the line. */ - VMCOREINFO_STRUCT_SIZE(printk_log); - VMCOREINFO_OFFSET(printk_log, ts_nsec); - VMCOREINFO_OFFSET(printk_log, len); - VMCOREINFO_OFFSET(printk_log, text_len); - VMCOREINFO_OFFSET(printk_log, dict_len); -#ifdef CONFIG_PRINTK_CALLER - VMCOREINFO_OFFSET(printk_log, caller_id); -#endif + + VMCOREINFO_SIZE(atomic64_t); + VMCOREINFO_TYPE_OFFSET(atomic64_t, counter); + + VMCOREINFO_STRUCT_SIZE(printk_ringbuffer); + VMCOREINFO_OFFSET(printk_ringbuffer, desc_ring); + VMCOREINFO_OFFSET(printk_ringbuffer, text_data_ring); + VMCOREINFO_OFFSET(printk_ringbuffer, fail); + + VMCOREINFO_STRUCT_SIZE(prb_desc_ring); + VMCOREINFO_OFFSET(prb_desc_ring, count_bits); + VMCOREINFO_OFFSET(prb_desc_ring, descs); + VMCOREINFO_OFFSET(prb_desc_ring, infos); + VMCOREINFO_OFFSET(prb_desc_ring, head_id); + VMCOREINFO_OFFSET(prb_desc_ring, tail_id); + + VMCOREINFO_STRUCT_SIZE(prb_desc); + VMCOREINFO_OFFSET(prb_desc, state_var); + VMCOREINFO_OFFSET(prb_desc, text_blk_lpos); + + VMCOREINFO_STRUCT_SIZE(prb_data_blk_lpos); + VMCOREINFO_OFFSET(prb_data_blk_lpos, begin); + VMCOREINFO_OFFSET(prb_data_blk_lpos, next); + + VMCOREINFO_STRUCT_SIZE(printk_info); + VMCOREINFO_OFFSET(printk_info, seq); + VMCOREINFO_OFFSET(printk_info, ts_nsec); + VMCOREINFO_OFFSET(printk_info, text_len); + VMCOREINFO_OFFSET(printk_info, caller_id); + VMCOREINFO_OFFSET(printk_info, dev_info); + + VMCOREINFO_STRUCT_SIZE(dev_printk_info); + VMCOREINFO_OFFSET(dev_printk_info, subsystem); + VMCOREINFO_LENGTH(printk_info_subsystem, sizeof(dev_info->subsystem)); + VMCOREINFO_OFFSET(dev_printk_info, device); + VMCOREINFO_LENGTH(printk_info_device, sizeof(dev_info->device)); + + VMCOREINFO_STRUCT_SIZE(prb_data_ring); + VMCOREINFO_OFFSET(prb_data_ring, size_bits); + VMCOREINFO_OFFSET(prb_data_ring, data); + VMCOREINFO_OFFSET(prb_data_ring, head_lpos); + VMCOREINFO_OFFSET(prb_data_ring, tail_lpos); + + VMCOREINFO_SIZE(atomic_long_t); + VMCOREINFO_TYPE_OFFSET(atomic_long_t, counter); } #endif @@ -1168,17 +969,48 @@ static inline void log_buf_add_cpu(void) {} static void __init set_percpu_data_ready(void) { - printk_safe_init(); - /* Make sure we set this flag only after printk_safe() init is done */ - barrier(); __printk_percpu_data_ready = true; } +static unsigned int __init add_to_rb(struct printk_ringbuffer *rb, + struct printk_record *r) +{ + struct prb_reserved_entry e; + struct printk_record dest_r; + + prb_rec_init_wr(&dest_r, r->info->text_len); + + if (!prb_reserve(&e, rb, &dest_r)) + return 0; + + memcpy(&dest_r.text_buf[0], &r->text_buf[0], r->info->text_len); + dest_r.info->text_len = r->info->text_len; + dest_r.info->facility = r->info->facility; + dest_r.info->level = r->info->level; + dest_r.info->flags = r->info->flags; + dest_r.info->ts_nsec = r->info->ts_nsec; + dest_r.info->caller_id = r->info->caller_id; + memcpy(&dest_r.info->dev_info, &r->info->dev_info, sizeof(dest_r.info->dev_info)); + + prb_final_commit(&e); + + return prb_record_text_space(&e); +} + +static char setup_text_buf[LOG_LINE_MAX] __initdata; + void __init setup_log_buf(int early) { - unsigned long flags; + struct printk_info *new_infos; + unsigned int new_descs_count; + struct prb_desc *new_descs; + struct printk_info info; + struct printk_record r; + size_t new_descs_size; + size_t new_infos_size; char *new_log_buf; unsigned int free; + u64 seq; /* * Some archs call setup_log_buf() multiple times - first is very @@ -1197,24 +1029,71 @@ void __init setup_log_buf(int early) if (!new_log_buf_len) return; - new_log_buf = memblock_alloc(new_log_buf_len, LOG_ALIGN); - if (unlikely(!new_log_buf)) { - pr_err("log_buf_len: %lu bytes not available\n", - new_log_buf_len); + new_descs_count = new_log_buf_len >> PRB_AVGBITS; + if (new_descs_count == 0) { + pr_err("new_log_buf_len: %lu too small\n", new_log_buf_len); return; } - logbuf_lock_irqsave(flags); + new_log_buf = memblock_alloc(new_log_buf_len, LOG_ALIGN); + if (unlikely(!new_log_buf)) { + pr_err("log_buf_len: %lu text bytes not available\n", + new_log_buf_len); + return; + } + + new_descs_size = new_descs_count * sizeof(struct prb_desc); + new_descs = memblock_alloc(new_descs_size, LOG_ALIGN); + if (unlikely(!new_descs)) { + pr_err("log_buf_len: %zu desc bytes not available\n", + new_descs_size); + goto err_free_log_buf; + } + + new_infos_size = new_descs_count * sizeof(struct printk_info); + new_infos = memblock_alloc(new_infos_size, LOG_ALIGN); + if (unlikely(!new_infos)) { + pr_err("log_buf_len: %zu info bytes not available\n", + new_infos_size); + goto err_free_descs; + } + + prb_rec_init_rd(&r, &info, &setup_text_buf[0], sizeof(setup_text_buf)); + + prb_init(&printk_rb_dynamic, + new_log_buf, ilog2(new_log_buf_len), + new_descs, ilog2(new_descs_count), + new_infos); + log_buf_len = new_log_buf_len; log_buf = new_log_buf; new_log_buf_len = 0; - free = __LOG_BUF_LEN - log_next_idx; - memcpy(log_buf, __log_buf, __LOG_BUF_LEN); - logbuf_unlock_irqrestore(flags); + + free = __LOG_BUF_LEN; + prb_for_each_record(0, &printk_rb_static, seq, &r) + free -= add_to_rb(&printk_rb_dynamic, &r); + + /* + * This is early enough that everything is still running on the + * boot CPU and interrupts are disabled. So no new messages will + * appear during the transition to the dynamic buffer. + */ + prb = &printk_rb_dynamic; + + if (seq != prb_next_seq(&printk_rb_static)) { + pr_err("dropped %llu messages\n", + prb_next_seq(&printk_rb_static) - seq); + } pr_info("log_buf_len: %u bytes\n", log_buf_len); pr_info("early log buf free: %u(%u%%)\n", free, (free * 100) / __LOG_BUF_LEN); + return; + +err_free_descs: + memblock_free(__pa(new_descs), new_descs_size); +err_free_log_buf: + memblock_free(__pa(new_log_buf), new_log_buf_len); } static bool __read_mostly ignore_loglevel; @@ -1321,18 +1200,18 @@ static size_t print_caller(u32 id, char *buf) #define print_caller(id, buf) 0 #endif -static size_t print_prefix(const struct printk_log *msg, bool syslog, - bool time, char *buf) +static size_t info_print_prefix(const struct printk_info *info, bool syslog, + bool time, char *buf) { size_t len = 0; if (syslog) - len = print_syslog((msg->facility << 3) | msg->level, buf); + len = print_syslog((info->facility << 3) | info->level, buf); if (time) - len += print_time(msg->ts_nsec, buf + len); + len += print_time(info->ts_nsec, buf + len); - len += print_caller(msg->caller_id, buf + len); + len += print_caller(info->caller_id, buf + len); if (IS_ENABLED(CONFIG_PRINTK_CALLER) || time) { buf[len++] = ' '; @@ -1342,72 +1221,150 @@ static size_t print_prefix(const struct printk_log *msg, bool syslog, return len; } -static size_t msg_print_text(const struct printk_log *msg, bool syslog, - bool time, char *buf, size_t size) +/* + * Prepare the record for printing. The text is shifted within the given + * buffer to avoid a need for another one. The following operations are + * done: + * + * - Add prefix for each line. + * - Add the trailing newline that has been removed in vprintk_store(). + * - Drop truncated lines that do not longer fit into the buffer. + * + * Return: The length of the updated/prepared text, including the added + * prefixes and the newline. The dropped line(s) are not counted. + */ +static size_t record_print_text(struct printk_record *r, bool syslog, + bool time) { - const char *text = log_text(msg); - size_t text_size = msg->text_len; - size_t len = 0; + size_t text_len = r->info->text_len; + size_t buf_size = r->text_buf_size; + char *text = r->text_buf; char prefix[PREFIX_MAX]; - const size_t prefix_len = print_prefix(msg, syslog, time, prefix); + bool truncated = false; + size_t prefix_len; + size_t line_len; + size_t len = 0; + char *next; - do { - const char *next = memchr(text, '\n', text_size); - size_t text_len; + /* + * If the message was truncated because the buffer was not large + * enough, treat the available text as if it were the full text. + */ + if (text_len > buf_size) + text_len = buf_size; + prefix_len = info_print_prefix(r->info, syslog, time, prefix); + + /* + * @text_len: bytes of unprocessed text + * @line_len: bytes of current line _without_ newline + * @text: pointer to beginning of current line + * @len: number of bytes prepared in r->text_buf + */ + for (;;) { + next = memchr(text, '\n', text_len); if (next) { - text_len = next - text; - next++; - text_size -= next - text; + line_len = next - text; } else { - text_len = text_size; + /* Drop truncated line(s). */ + if (truncated) + break; + line_len = text_len; } - if (buf) { - if (prefix_len + text_len + 1 >= size - len) + /* + * Truncate the text if there is not enough space to add the + * prefix and a trailing newline. + */ + if (len + prefix_len + text_len + 1 > buf_size) { + /* Drop even the current line if no space. */ + if (len + prefix_len + line_len + 1 > buf_size) break; - memcpy(buf + len, prefix, prefix_len); - len += prefix_len; - memcpy(buf + len, text, text_len); - len += text_len; - buf[len++] = '\n'; - } else { - /* SYSLOG_ACTION_* buffer size only calculation */ - len += prefix_len + text_len + 1; + text_len = buf_size - len - prefix_len - 1; + truncated = true; } - text = next; - } while (text); + memmove(text + prefix_len, text, text_len); + memcpy(text, prefix, prefix_len); + + len += prefix_len + line_len + 1; + + if (text_len == line_len) { + /* + * Add the trailing newline removed in + * vprintk_store(). + */ + text[prefix_len + line_len] = '\n'; + break; + } + + /* + * Advance beyond the added prefix and the related line with + * its newline. + */ + text += prefix_len + line_len + 1; + + /* + * The remaining text has only decreased by the line with its + * newline. + * + * Note that @text_len can become zero. It happens when @text + * ended with a newline (either due to truncation or the + * original string ending with "\n\n"). The loop is correctly + * repeated and (if not truncated) an empty line with a prefix + * will be prepared. + */ + text_len -= line_len + 1; + } return len; } +static size_t get_record_print_text_size(struct printk_info *info, + unsigned int line_count, + bool syslog, bool time) +{ + char prefix[PREFIX_MAX]; + size_t prefix_len; + + prefix_len = info_print_prefix(info, syslog, time, prefix); + + /* + * Each line will be preceded with a prefix. The intermediate + * newlines are already within the text, but a final trailing + * newline will be added. + */ + return ((prefix_len * line_count) + info->text_len + 1); +} + static int syslog_print(char __user *buf, int size) { + struct printk_info info; + struct printk_record r; char *text; - struct printk_log *msg; int len = 0; text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL); if (!text) return -ENOMEM; + prb_rec_init_rd(&r, &info, text, LOG_LINE_MAX + PREFIX_MAX); + while (size > 0) { size_t n; size_t skip; - logbuf_lock_irq(); - if (syslog_seq < log_first_seq) { - /* messages are gone, move to first one */ - syslog_seq = log_first_seq; - syslog_idx = log_first_idx; - syslog_partial = 0; - } - if (syslog_seq == log_next_seq) { - logbuf_unlock_irq(); + syslog_lock_irq(); + if (!prb_read_valid(prb, syslog_seq, &r)) { + syslog_unlock_irq(); break; } + if (r.info->seq != syslog_seq) { + /* message is gone, move to next valid one */ + syslog_seq = r.info->seq; + syslog_partial = 0; + } /* * To keep reading/counting partial line consistent, @@ -1417,13 +1374,10 @@ static int syslog_print(char __user *buf, int size) syslog_time = printk_time; skip = syslog_partial; - msg = log_from_idx(syslog_idx); - n = msg_print_text(msg, true, syslog_time, text, - LOG_LINE_MAX + PREFIX_MAX); + n = record_print_text(&r, true, syslog_time); if (n - syslog_partial <= size) { /* message fits into buffer, move forward */ - syslog_idx = log_next(syslog_idx); - syslog_seq++; + syslog_seq = r.info->seq + 1; n -= syslog_partial; syslog_partial = 0; } else if (!len){ @@ -1432,7 +1386,7 @@ static int syslog_print(char __user *buf, int size) syslog_partial += n; } else n = 0; - logbuf_unlock_irq(); + syslog_unlock_irq(); if (!n) break; @@ -1454,11 +1408,14 @@ static int syslog_print(char __user *buf, int size) static int syslog_print_all(char __user *buf, int size, bool clear) { + struct printk_info info; + unsigned int line_count; + struct printk_record r; + u64 newest_seq; + u64 clr_seq; char *text; int len = 0; - u64 next_seq; u64 seq; - u32 idx; bool time; text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL); @@ -1466,63 +1423,58 @@ static int syslog_print_all(char __user *buf, int size, bool clear) return -ENOMEM; time = printk_time; - logbuf_lock_irq(); + clr_seq = atomic64_read(&clear_seq); + /* * Find first record that fits, including all following records, * into the user-provided buffer for this dump. */ - seq = clear_seq; - idx = clear_idx; - while (seq < log_next_seq) { - struct printk_log *msg = log_from_idx(idx); - len += msg_print_text(msg, true, time, NULL, 0); - idx = log_next(idx); - seq++; + prb_for_each_info(clr_seq, prb, seq, &info, &line_count) + len += get_record_print_text_size(&info, line_count, true, time); + + /* + * Keep track of the latest in case new records are coming in fast + * and overwriting the older records. + */ + newest_seq = seq; + + /* + * Move first record forward until length fits into the buffer. This + * is a best effort attempt. If @newest_seq is reached because the + * ringbuffer is wrapping too fast, just start filling the buffer + * from there. + */ + prb_for_each_info(clr_seq, prb, seq, &info, &line_count) { + if (len <= size || info.seq > newest_seq) + break; + len -= get_record_print_text_size(&info, line_count, true, time); } - /* move first record forward until length fits into the buffer */ - seq = clear_seq; - idx = clear_idx; - while (len > size && seq < log_next_seq) { - struct printk_log *msg = log_from_idx(idx); - - len -= msg_print_text(msg, true, time, NULL, 0); - idx = log_next(idx); - seq++; - } - - /* last message fitting into this dump */ - next_seq = log_next_seq; + prb_rec_init_rd(&r, &info, text, LOG_LINE_MAX + PREFIX_MAX); len = 0; - while (len >= 0 && seq < next_seq) { - struct printk_log *msg = log_from_idx(idx); - int textlen = msg_print_text(msg, true, time, text, - LOG_LINE_MAX + PREFIX_MAX); + prb_for_each_record(seq, prb, seq, &r) { + int textlen; - idx = log_next(idx); - seq++; + textlen = record_print_text(&r, true, time); + + if (len + textlen > size) { + seq--; + break; + } - logbuf_unlock_irq(); if (copy_to_user(buf + len, text, textlen)) len = -EFAULT; else len += textlen; - logbuf_lock_irq(); - if (seq < log_first_seq) { - /* messages are gone, move to next one */ - seq = log_first_seq; - idx = log_first_idx; - } + if (len < 0) + break; } - if (clear) { - clear_seq = log_next_seq; - clear_idx = log_next_idx; - } - logbuf_unlock_irq(); + if (clear) + atomic64_set(&clear_seq, seq); kfree(text); return len; @@ -1530,10 +1482,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) static void syslog_clear(void) { - logbuf_lock_irq(); - clear_seq = log_next_seq; - clear_idx = log_next_idx; - logbuf_unlock_irq(); + atomic64_set(&clear_seq, prb_next_seq(prb)); } int do_syslog(int type, char __user *buf, int len, int source) @@ -1541,6 +1490,7 @@ int do_syslog(int type, char __user *buf, int len, int source) bool clear = false; static int saved_console_loglevel = LOGLEVEL_DEFAULT; int error; + u64 seq; error = check_syslog_permissions(type, source); if (error) @@ -1558,8 +1508,11 @@ int do_syslog(int type, char __user *buf, int len, int source) return 0; if (!access_ok(buf, len)) return -EFAULT; + syslog_lock_irq(); + seq = syslog_seq; + syslog_unlock_irq(); error = wait_event_interruptible(log_wait, - syslog_seq != log_next_seq); + prb_read_valid(prb, seq, NULL)); if (error) return error; error = syslog_print(buf, len); @@ -1567,7 +1520,7 @@ int do_syslog(int type, char __user *buf, int len, int source) /* Read/clear last kernel messages */ case SYSLOG_ACTION_READ_CLEAR: clear = true; - /* FALL THRU */ + fallthrough; /* Read last kernel messages */ case SYSLOG_ACTION_READ_ALL: if (!buf || len < 0) @@ -1607,11 +1560,10 @@ int do_syslog(int type, char __user *buf, int len, int source) break; /* Number of chars in the log buffer */ case SYSLOG_ACTION_SIZE_UNREAD: - logbuf_lock_irq(); - if (syslog_seq < log_first_seq) { + syslog_lock_irq(); + if (syslog_seq < prb_first_valid_seq(prb)) { /* messages are gone, move to first one */ - syslog_seq = log_first_seq; - syslog_idx = log_first_idx; + syslog_seq = prb_first_valid_seq(prb); syslog_partial = 0; } if (source == SYSLOG_FROM_PROC) { @@ -1620,24 +1572,22 @@ int do_syslog(int type, char __user *buf, int len, int source) * for pending data, not the size; return the count of * records, not the length. */ - error = log_next_seq - syslog_seq; + error = prb_next_seq(prb) - syslog_seq; } else { - u64 seq = syslog_seq; - u32 idx = syslog_idx; bool time = syslog_partial ? syslog_time : printk_time; + struct printk_info info; + unsigned int line_count; + u64 seq; - while (seq < log_next_seq) { - struct printk_log *msg = log_from_idx(idx); - - error += msg_print_text(msg, true, time, NULL, - 0); + prb_for_each_info(syslog_seq, prb, seq, &info, + &line_count) { + error += get_record_print_text_size(&info, line_count, + true, time); time = printk_time; - idx = log_next(idx); - seq++; } error -= syslog_partial; } - logbuf_unlock_irq(); + syslog_unlock_irq(); break; /* Size of the log buffer */ case SYSLOG_ACTION_SIZE_BUFFER: @@ -1657,178 +1607,134 @@ SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len) } /* - * Special console_lock variants that help to reduce the risk of soft-lockups. - * They allow to pass console_lock to another printk() call using a busy wait. + * The per-cpu sprint buffers are used with interrupts disabled, so each CPU + * only requires 2 buffers: for non-NMI and NMI contexts. Recursive printk() + * calls are handled by the global sprint buffers. */ +#define SPRINT_CTX_DEPTH 2 -#ifdef CONFIG_LOCKDEP -static struct lockdep_map console_owner_dep_map = { - .name = "console_owner" +/* Static sprint buffers for early boot (only 1 CPU) and recursion. */ +static DECLARE_BITMAP(sprint_global_buffer_map, SPRINT_CTX_DEPTH); +static char sprint_global_buffer[SPRINT_CTX_DEPTH][PREFIX_MAX + LOG_LINE_MAX]; + +struct sprint_buffers { + char buf[SPRINT_CTX_DEPTH][PREFIX_MAX + LOG_LINE_MAX]; + atomic_t index; }; -#endif -static DEFINE_RAW_SPINLOCK(console_owner_lock); -static struct task_struct *console_owner; -static bool console_waiter; +static DEFINE_PER_CPU(struct sprint_buffers, percpu_sprint_buffers); -/** - * console_lock_spinning_enable - mark beginning of code where another - * thread might safely busy wait - * - * This basically converts console_lock into a spinlock. This marks - * the section where the console_lock owner can not sleep, because - * there may be a waiter spinning (like a spinlock). Also it must be - * ready to hand over the lock at the end of the section. +/* + * Acquire an unused buffer, returning its index. If no buffer is + * available, @count is returned. */ -static void console_lock_spinning_enable(void) +static int _get_sprint_buf(unsigned long *map, int count) { - raw_spin_lock(&console_owner_lock); - console_owner = current; - raw_spin_unlock(&console_owner_lock); + int index; - /* The waiter may spin on us after setting console_owner */ - spin_acquire(&console_owner_dep_map, 0, 0, _THIS_IP_); + do { + index = find_first_zero_bit(map, count); + if (index == count) + break; + /* + * Guarantee map changes are ordered for the other CPUs. + * Pairs with clear_bit() in _put_sprint_buf(). + */ + } while (test_and_set_bit(index, map)); + + return index; } -/** - * console_lock_spinning_disable_and_check - mark end of code where another - * thread was able to busy wait and check if there is a waiter - * - * This is called at the end of the section where spinning is allowed. - * It has two functions. First, it is a signal that it is no longer - * safe to start busy waiting for the lock. Second, it checks if - * there is a busy waiter and passes the lock rights to her. - * - * Important: Callers lose the lock if there was a busy waiter. - * They must not touch items synchronized by console_lock - * in this case. - * - * Return: 1 if the lock rights were passed, 0 otherwise. - */ -static int console_lock_spinning_disable_and_check(void) +/* Mark the buffer @index as unused. */ +static void _put_sprint_buf(unsigned long *map, unsigned int count, unsigned int index) { - int waiter; - - raw_spin_lock(&console_owner_lock); - waiter = READ_ONCE(console_waiter); - console_owner = NULL; - raw_spin_unlock(&console_owner_lock); - - if (!waiter) { - spin_release(&console_owner_dep_map, _THIS_IP_); - return 0; - } - - /* The waiter is now free to continue */ - WRITE_ONCE(console_waiter, false); - - spin_release(&console_owner_dep_map, _THIS_IP_); - /* - * Hand off console_lock to waiter. The waiter will perform - * the up(). After this, the waiter is the console_lock owner. + * Guarantee map changes are ordered for the other CPUs. + * Pairs with test_and_set_bit() in _get_sprint_buf(). */ - mutex_release(&console_lock_dep_map, _THIS_IP_); - return 1; -} - -/** - * console_trylock_spinning - try to get console_lock by busy waiting - * - * This allows to busy wait for the console_lock when the current - * owner is running in specially marked sections. It means that - * the current owner is running and cannot reschedule until it - * is ready to lose the lock. - * - * Return: 1 if we got the lock, 0 othrewise - */ -static int console_trylock_spinning(void) -{ - struct task_struct *owner = NULL; - bool waiter; - bool spin = false; - unsigned long flags; - - if (console_trylock()) - return 1; - - printk_safe_enter_irqsave(flags); - - raw_spin_lock(&console_owner_lock); - owner = READ_ONCE(console_owner); - waiter = READ_ONCE(console_waiter); - if (!waiter && owner && owner != current) { - WRITE_ONCE(console_waiter, true); - spin = true; - } - raw_spin_unlock(&console_owner_lock); - - /* - * If there is an active printk() writing to the - * consoles, instead of having it write our data too, - * see if we can offload that load from the active - * printer, and do some printing ourselves. - * Go into a spin only if there isn't already a waiter - * spinning, and there is an active printer, and - * that active printer isn't us (recursive printk?). - */ - if (!spin) { - printk_safe_exit_irqrestore(flags); - return 0; - } - - /* We spin waiting for the owner to release us */ - spin_acquire(&console_owner_dep_map, 0, 0, _THIS_IP_); - /* Owner will clear console_waiter on hand off */ - while (READ_ONCE(console_waiter)) - cpu_relax(); - spin_release(&console_owner_dep_map, _THIS_IP_); - - printk_safe_exit_irqrestore(flags); - /* - * The owner passed the console lock to us. - * Since we did not spin on console lock, annotate - * this as a trylock. Otherwise lockdep will - * complain. - */ - mutex_acquire(&console_lock_dep_map, 0, 1, _THIS_IP_); - - return 1; + clear_bit(index, map); } /* - * Call the console drivers, asking them to write out - * log_buf[start] to log_buf[end - 1]. - * The console_lock must be held. + * Get a buffer sized PREFIX_MAX+LOG_LINE_MAX for sprinting. On success, @id + * is set and interrupts are disabled. @id is used to put back the buffer. + * + * @id is non-negative for per-cpu buffers, negative for global buffers. */ -static void call_console_drivers(const char *ext_text, size_t ext_len, - const char *text, size_t len) +static char *get_sprint_buf(int *id, unsigned long *flags) { - struct console *con; + struct sprint_buffers *bufs; + unsigned int index; + unsigned int cpu; - trace_console_rcuidle(text, len); + local_irq_save(*flags); + cpu = get_cpu(); - for_each_console(con) { - if (exclusive_console && con != exclusive_console) - continue; - if (!(con->flags & CON_ENABLED)) - continue; - if (!con->write) - continue; - if (!cpu_online(smp_processor_id()) && - !(con->flags & CON_ANYTIME)) - continue; - if (con->flags & CON_EXTENDED) - con->write(con, ext_text, ext_len); - else - con->write(con, text, len); + if (printk_percpu_data_ready()) { + + /* + * First try with per-cpu pool. Note that the last + * buffer is reserved for NMI context. + */ + bufs = per_cpu_ptr(&percpu_sprint_buffers, cpu); + index = atomic_read(&bufs->index); + if (index < (SPRINT_CTX_DEPTH - 1) || + (in_nmi() && index < SPRINT_CTX_DEPTH)) { + atomic_set(&bufs->index, index + 1); + *id = cpu; + return &bufs->buf[index][0]; + } } + + /* + * Fallback to global pool. + * + * The global pool will only ever be used if per-cpu data is not ready + * yet or printk recurses. Recursion will not occur unless printk is + * having internal issues. + */ + index = _get_sprint_buf(sprint_global_buffer_map, SPRINT_CTX_DEPTH); + if (index != SPRINT_CTX_DEPTH) { + /* Convert to global buffer representation. */ + *id = -index - 1; + return &sprint_global_buffer[index][0]; + } + + /* Failed to get a buffer. */ + put_cpu(); + local_irq_restore(*flags); + return NULL; +} + +/* Put back an sprint buffer and restore interrupts. */ +static void put_sprint_buf(int id, unsigned long flags) +{ + struct sprint_buffers *bufs; + unsigned int index; + unsigned int cpu; + + if (id >= 0) { + cpu = id; + bufs = per_cpu_ptr(&percpu_sprint_buffers, cpu); + index = atomic_read(&bufs->index); + atomic_set(&bufs->index, index - 1); + } else { + /* Convert from global buffer representation. */ + index = -id - 1; + _put_sprint_buf(sprint_global_buffer_map, + SPRINT_CTX_DEPTH, index); + } + + put_cpu(); + local_irq_restore(flags); } int printk_delay_msec __read_mostly; -static inline void printk_delay(void) +static inline void printk_delay(int level) { + boot_delay_msec(level); + if (unlikely(printk_delay_msec)) { int m = printk_delay_msec; @@ -1839,115 +1745,155 @@ static inline void printk_delay(void) } } +static bool kernel_sync_mode(void) +{ + return (oops_in_progress || sync_mode); +} + +static bool console_can_sync(struct console *con) +{ + if (!(con->flags & CON_ENABLED)) + return false; + if (con->write_atomic && kernel_sync_mode()) + return true; + if (con->write_atomic && (con->flags & CON_HANDOVER) && !con->thread) + return true; + if (con->write && (con->flags & CON_BOOT) && !con->thread) + return true; + return false; +} + +static bool call_sync_console_driver(struct console *con, const char *text, size_t text_len) +{ + if (!(con->flags & CON_ENABLED)) + return false; + if (con->write_atomic && kernel_sync_mode()) + con->write_atomic(con, text, text_len); + else if (con->write_atomic && (con->flags & CON_HANDOVER) && !con->thread) + con->write_atomic(con, text, text_len); + else if (con->write && (con->flags & CON_BOOT) && !con->thread) + con->write(con, text, text_len); + else + return false; + + return true; +} + +static bool any_console_can_sync(void) +{ + struct console *con; + + for_each_console(con) { + if (console_can_sync(con)) + return true; + } + return false; +} + +static bool have_atomic_console(void) +{ + struct console *con; + + for_each_console(con) { + if (!(con->flags & CON_ENABLED)) + continue; + if (con->write_atomic) + return true; + } + return false; +} + +static bool print_sync(struct console *con, char *buf, size_t buf_size, u64 *seq) +{ + struct printk_info info; + struct printk_record r; + size_t text_len; + + prb_rec_init_rd(&r, &info, buf, buf_size); + + if (!prb_read_valid(prb, *seq, &r)) + return false; + + text_len = record_print_text(&r, console_msg_format & MSG_FORMAT_SYSLOG, printk_time); + + if (!call_sync_console_driver(con, buf, text_len)) + return false; + + *seq = r.info->seq; + + touch_softlockup_watchdog_sync(); + clocksource_touch_watchdog(); + rcu_cpu_stall_reset(); + touch_nmi_watchdog(); + + if (text_len) + printk_delay(r.info->level); + + return true; +} + +static void print_sync_until(u64 seq, struct console *con, char *buf, size_t buf_size) +{ + unsigned int flags; + u64 printk_seq; + + if (!con) { + for_each_console(con) { + if (console_can_sync(con)) + print_sync_until(seq, con, buf, buf_size); + } + return; + } + + console_atomic_lock(&flags); + for (;;) { + printk_seq = atomic64_read(&con->printk_seq); + if (printk_seq >= seq) + break; + if (!print_sync(con, buf, buf_size, &printk_seq)) + break; + atomic64_set(&con->printk_seq, printk_seq + 1); + } + console_atomic_unlock(flags); +} + static inline u32 printk_caller_id(void) { return in_task() ? task_pid_nr(current) : 0x80000000 + raw_smp_processor_id(); } -/* - * Continuation lines are buffered, and not committed to the record buffer - * until the line is complete, or a race forces it. The line fragments - * though, are printed immediately to the consoles to ensure everything has - * reached the console in case of a kernel crash. - */ -static struct cont { - char buf[LOG_LINE_MAX]; - size_t len; /* length == 0 means unused buffer */ - u32 caller_id; /* printk_caller_id() of first print */ - u64 ts_nsec; /* time of first print */ - u8 level; /* log level of first message */ - u8 facility; /* log facility of first message */ - enum log_flags flags; /* prefix, newline flags */ -} cont; - -static void cont_flush(void) -{ - if (cont.len == 0) - return; - - log_store(cont.caller_id, cont.facility, cont.level, cont.flags, - cont.ts_nsec, NULL, 0, cont.buf, cont.len); - cont.len = 0; -} - -static bool cont_add(u32 caller_id, int facility, int level, - enum log_flags flags, const char *text, size_t len) -{ - /* If the line gets too long, split it up in separate records. */ - if (cont.len + len > sizeof(cont.buf)) { - cont_flush(); - return false; - } - - if (!cont.len) { - cont.facility = facility; - cont.level = level; - cont.caller_id = caller_id; - cont.ts_nsec = local_clock(); - cont.flags = flags; - } - - memcpy(cont.buf + cont.len, text, len); - cont.len += len; - - // The original flags come from the first line, - // but later continuations can add a newline. - if (flags & LOG_NEWLINE) { - cont.flags |= LOG_NEWLINE; - cont_flush(); - } - - return true; -} - -static size_t log_output(int facility, int level, enum log_flags lflags, const char *dict, size_t dictlen, char *text, size_t text_len) +__printf(4, 0) +static int vprintk_store(int facility, int level, + const struct dev_printk_info *dev_info, + const char *fmt, va_list args) { const u32 caller_id = printk_caller_id(); - - /* - * If an earlier line was buffered, and we're a continuation - * write from the same context, try to add it to the buffer. - */ - if (cont.len) { - if (cont.caller_id == caller_id && (lflags & LOG_CONT)) { - if (cont_add(caller_id, facility, level, lflags, text, text_len)) - return text_len; - } - /* Otherwise, make sure it's flushed */ - cont_flush(); - } - - /* Skip empty continuation lines that couldn't be added - they just flush */ - if (!text_len && (lflags & LOG_CONT)) - return 0; - - /* If it doesn't end in a newline, try to buffer the current line */ - if (!(lflags & LOG_NEWLINE)) { - if (cont_add(caller_id, facility, level, lflags, text, text_len)) - return text_len; - } - - /* Store it in the record log */ - return log_store(caller_id, facility, level, lflags, 0, - dict, dictlen, text, text_len); -} - -/* Must be called under logbuf_lock. */ -int vprintk_store(int facility, int level, - const char *dict, size_t dictlen, - const char *fmt, va_list args) -{ - static char textbuf[LOG_LINE_MAX]; - char *text = textbuf; - size_t text_len; + struct prb_reserved_entry e; enum log_flags lflags = 0; + bool final_commit = false; + unsigned long irqflags; + struct printk_record r; + u16 trunc_msg_len = 0; + int sprint_id; + u16 text_len; + u64 ts_nsec; + int ret = 0; + char *text; + u64 seq; + + ts_nsec = local_clock(); + + /* No buffer is available if printk has recursed too much. */ + text = get_sprint_buf(&sprint_id, &irqflags); + if (!text) + return 0; /* * The printf needs to come first; we need the syslog * prefix which might be passed-in as a parameter. */ - text_len = vscnprintf(text, sizeof(textbuf), fmt, args); + text_len = vscnprintf(text, LOG_LINE_MAX, fmt, args); /* mark and strip a trailing newline */ if (text_len && text[text_len-1] == '\n') { @@ -1977,77 +1923,116 @@ int vprintk_store(int facility, int level, if (level == LOGLEVEL_DEFAULT) level = default_message_loglevel; - if (dict) + if (dev_info) lflags |= LOG_NEWLINE; - return log_output(facility, level, lflags, - dict, dictlen, text, text_len); + if (lflags & LOG_CONT) { + prb_rec_init_wr(&r, text_len); + if (prb_reserve_in_last(&e, prb, &r, caller_id, LOG_LINE_MAX)) { + seq = r.info->seq; + memcpy(&r.text_buf[r.info->text_len], text, text_len); + r.info->text_len += text_len; + if (lflags & LOG_NEWLINE) { + r.info->flags |= LOG_NEWLINE; + prb_final_commit(&e); + final_commit = true; + } else { + prb_commit(&e); + } + ret = text_len; + goto out; + } + } + + /* Store it in the record log */ + + prb_rec_init_wr(&r, text_len); + + if (!prb_reserve(&e, prb, &r)) { + /* truncate the message if it is too long for empty buffer */ + truncate_msg(&text_len, &trunc_msg_len); + prb_rec_init_wr(&r, text_len + trunc_msg_len); + /* survive when the log buffer is too small for trunc_msg */ + if (!prb_reserve(&e, prb, &r)) + goto out; + } + + seq = r.info->seq; + + /* fill message */ + memcpy(&r.text_buf[0], text, text_len); + if (trunc_msg_len) + memcpy(&r.text_buf[text_len], trunc_msg, trunc_msg_len); + r.info->text_len = text_len + trunc_msg_len; + r.info->facility = facility; + r.info->level = level & 7; + r.info->flags = lflags & 0x1f; + r.info->ts_nsec = ts_nsec; + r.info->caller_id = caller_id; + if (dev_info) + memcpy(&r.info->dev_info, dev_info, sizeof(r.info->dev_info)); + + /* insert message */ + if ((lflags & LOG_CONT) || !(lflags & LOG_NEWLINE)) { + prb_commit(&e); + } else { + prb_final_commit(&e); + final_commit = true; + } + + ret = text_len + trunc_msg_len; +out: + /* only the kernel may perform synchronous printing */ + if (facility == 0 && final_commit && any_console_can_sync()) + print_sync_until(seq + 1, NULL, text, PREFIX_MAX + LOG_LINE_MAX); + + put_sprint_buf(sprint_id, irqflags); + return ret; } asmlinkage int vprintk_emit(int facility, int level, - const char *dict, size_t dictlen, + const struct dev_printk_info *dev_info, const char *fmt, va_list args) { int printed_len; - bool in_sched = false, pending_output; - unsigned long flags; - u64 curr_log_seq; /* Suppress unimportant messages after panic happens */ if (unlikely(suppress_printk)) return 0; - if (level == LOGLEVEL_SCHED) { + if (level == LOGLEVEL_SCHED) level = LOGLEVEL_DEFAULT; - in_sched = true; - } - boot_delay_msec(level); - printk_delay(); + printed_len = vprintk_store(facility, level, dev_info, fmt, args); - /* This stops the holder of console_sem just where we want him */ - logbuf_lock_irqsave(flags); - curr_log_seq = log_next_seq; - printed_len = vprintk_store(facility, level, dict, dictlen, fmt, args); - pending_output = (curr_log_seq != log_next_seq); - logbuf_unlock_irqrestore(flags); - - /* If called from the scheduler, we can not call up(). */ - if (!in_sched && pending_output) { - /* - * Disable preemption to avoid being preempted while holding - * console_sem which would prevent anyone from printing to - * console - */ - preempt_disable(); - /* - * Try to acquire and then immediately release the console - * semaphore. The release will print out buffers and wake up - * /dev/kmsg and syslog() users. - */ - if (console_trylock_spinning()) - console_unlock(); - preempt_enable(); - } - - if (pending_output) - wake_up_klogd(); + wake_up_klogd(); return printed_len; } EXPORT_SYMBOL(vprintk_emit); + __printf(1, 0) +static int vprintk_default(const char *fmt, va_list args) +{ + return vprintk_emit(0, LOGLEVEL_DEFAULT, NULL, fmt, args); +} + +__printf(1, 0) +static int vprintk_func(const char *fmt, va_list args) +{ +#ifdef CONFIG_KGDB_KDB + /* Allow to pass printk() to kdb but avoid a recursion. */ + if (unlikely(kdb_trap_printk && kdb_printf_cpu < 0)) + return vkdb_printf(KDB_MSGSRC_PRINTK, fmt, args); +#endif + return vprintk_default(fmt, args); +} + asmlinkage int vprintk(const char *fmt, va_list args) { return vprintk_func(fmt, args); } EXPORT_SYMBOL(vprintk); -int vprintk_default(const char *fmt, va_list args) -{ - return vprintk_emit(0, LOGLEVEL_DEFAULT, NULL, 0, fmt, args); -} -EXPORT_SYMBOL_GPL(vprintk_default); - /** * printk - print a kernel message * @fmt: format string @@ -2088,30 +2073,26 @@ EXPORT_SYMBOL(printk); #define PREFIX_MAX 0 #define printk_time false +#define prb_read_valid(rb, seq, r) false +#define prb_first_valid_seq(rb) 0 + static u64 syslog_seq; -static u32 syslog_idx; -static u64 console_seq; -static u32 console_idx; -static u64 exclusive_console_stop_seq; -static u64 log_first_seq; -static u32 log_first_idx; -static u64 log_next_seq; -static char *log_text(const struct printk_log *msg) { return NULL; } -static char *log_dict(const struct printk_log *msg) { return NULL; } -static struct printk_log *log_from_idx(u32 idx) { return NULL; } -static u32 log_next(u32 idx) { return 0; } -static ssize_t msg_print_ext_header(char *buf, size_t size, - struct printk_log *msg, - u64 seq) { return 0; } + +static size_t record_print_text(const struct printk_record *r, + bool syslog, bool time) +{ + return 0; +} +static ssize_t info_print_ext_header(char *buf, size_t size, + struct printk_info *info) +{ + return 0; +} static ssize_t msg_print_ext_body(char *buf, size_t size, - char *dict, size_t dict_len, - char *text, size_t text_len) { return 0; } -static void console_lock_spinning_enable(void) { } -static int console_lock_spinning_disable_and_check(void) { return 0; } + char *text, size_t text_len, + struct dev_printk_info *dev_info) { return 0; } static void call_console_drivers(const char *ext_text, size_t ext_len, const char *text, size_t len) {} -static size_t msg_print_text(const struct printk_log *msg, bool syslog, - bool time, char *buf, size_t size) { return 0; } static bool suppress_message_printing(int level) { return false; } #endif /* CONFIG_PRINTK */ @@ -2350,34 +2331,6 @@ int is_console_locked(void) } EXPORT_SYMBOL(is_console_locked); -/* - * Check if we have any console that is capable of printing while cpu is - * booting or shutting down. Requires console_sem. - */ -static int have_callable_console(void) -{ - struct console *con; - - for_each_console(con) - if ((con->flags & CON_ENABLED) && - (con->flags & CON_ANYTIME)) - return 1; - - return 0; -} - -/* - * Can we actually use the console at this time on this cpu? - * - * Console drivers may assume that per-cpu resources have been allocated. So - * unless they're explicitly marked as being able to cope (CON_ANYTIME) don't - * call them until this CPU is officially up. - */ -static inline int can_use_console(void) -{ - return cpu_online(raw_smp_processor_id()) || have_callable_console(); -} - /** * console_unlock - unlock the console system * @@ -2394,143 +2347,14 @@ static inline int can_use_console(void) */ void console_unlock(void) { - static char ext_text[CONSOLE_EXT_LOG_MAX]; - static char text[LOG_LINE_MAX + PREFIX_MAX]; - unsigned long flags; - bool do_cond_resched, retry; - if (console_suspended) { up_console_sem(); return; } - /* - * Console drivers are called with interrupts disabled, so - * @console_may_schedule should be cleared before; however, we may - * end up dumping a lot of lines, for example, if called from - * console registration path, and should invoke cond_resched() - * between lines if allowable. Not doing so can cause a very long - * scheduling stall on a slow console leading to RCU stall and - * softlockup warnings which exacerbate the issue with more - * messages practically incapacitating the system. - * - * console_trylock() is not able to detect the preemptive - * context reliably. Therefore the value must be stored before - * and cleared after the the "again" goto label. - */ - do_cond_resched = console_may_schedule; -again: - console_may_schedule = 0; - - /* - * We released the console_sem lock, so we need to recheck if - * cpu is online and (if not) is there at least one CON_ANYTIME - * console. - */ - if (!can_use_console()) { - console_locked = 0; - up_console_sem(); - return; - } - - for (;;) { - struct printk_log *msg; - size_t ext_len = 0; - size_t len; - - printk_safe_enter_irqsave(flags); - raw_spin_lock(&logbuf_lock); - if (console_seq < log_first_seq) { - len = snprintf(text, sizeof(text), - "** %llu printk messages dropped **\n", - log_first_seq - console_seq); - - /* messages are gone, move to first one */ - console_seq = log_first_seq; - console_idx = log_first_idx; - } else { - len = 0; - } -skip: - if (console_seq == log_next_seq) - break; - - msg = log_from_idx(console_idx); - if (suppress_message_printing(msg->level)) { - /* - * Skip record we have buffered and already printed - * directly to the console when we received it, and - * record that has level above the console loglevel. - */ - console_idx = log_next(console_idx); - console_seq++; - goto skip; - } - - /* Output to all consoles once old messages replayed. */ - if (unlikely(exclusive_console && - console_seq >= exclusive_console_stop_seq)) { - exclusive_console = NULL; - } - - len += msg_print_text(msg, - console_msg_format & MSG_FORMAT_SYSLOG, - printk_time, text + len, sizeof(text) - len); - if (nr_ext_console_drivers) { - ext_len = msg_print_ext_header(ext_text, - sizeof(ext_text), - msg, console_seq); - ext_len += msg_print_ext_body(ext_text + ext_len, - sizeof(ext_text) - ext_len, - log_dict(msg), msg->dict_len, - log_text(msg), msg->text_len); - } - console_idx = log_next(console_idx); - console_seq++; - raw_spin_unlock(&logbuf_lock); - - /* - * While actively printing out messages, if another printk() - * were to occur on another CPU, it may wait for this one to - * finish. This task can not be preempted if there is a - * waiter waiting to take over. - */ - console_lock_spinning_enable(); - - stop_critical_timings(); /* don't trace print latency */ - call_console_drivers(ext_text, ext_len, text, len); - start_critical_timings(); - - if (console_lock_spinning_disable_and_check()) { - printk_safe_exit_irqrestore(flags); - return; - } - - printk_safe_exit_irqrestore(flags); - - if (do_cond_resched) - cond_resched(); - } - console_locked = 0; - raw_spin_unlock(&logbuf_lock); - up_console_sem(); - - /* - * Someone could have filled up the buffer again, so re-check if there's - * something to flush. In case we cannot trylock the console_sem again, - * there's a new owner and the console_unlock() from them will do the - * flush, no worries. - */ - raw_spin_lock(&logbuf_lock); - retry = console_seq != log_next_seq; - raw_spin_unlock(&logbuf_lock); - printk_safe_exit_irqrestore(flags); - - if (retry && console_trylock()) - goto again; } EXPORT_SYMBOL(console_unlock); @@ -2580,24 +2404,20 @@ void console_unblank(void) */ void console_flush_on_panic(enum con_flush_mode mode) { - /* - * If someone else is holding the console lock, trylock will fail - * and may_schedule may be set. Ignore and proceed to unlock so - * that messages are flushed out. As this can be called from any - * context and we don't want to get preempted while flushing, - * ensure may_schedule is cleared. - */ - console_trylock(); + struct console *c; + u64 seq; + + if (!console_trylock()) + return; + console_may_schedule = 0; if (mode == CONSOLE_REPLAY_ALL) { - unsigned long flags; - - logbuf_lock_irqsave(flags); - console_seq = log_first_seq; - console_idx = log_first_idx; - logbuf_unlock_irqrestore(flags); + seq = prb_first_valid_seq(prb); + for_each_console(c) + atomic64_set(&c->printk_seq, seq); } + console_unlock(); } @@ -2711,6 +2531,8 @@ static int try_enable_new_console(struct console *newcon, bool user_specified) return -ENOENT; } +static void console_try_thread(struct console *con); + /* * The console driver calls this routine during kernel initialization * to register the console printing procedure with printk() and to @@ -2732,7 +2554,6 @@ static int try_enable_new_console(struct console *newcon, bool user_specified) */ void register_console(struct console *newcon) { - unsigned long flags; struct console *bcon = NULL; int err; @@ -2756,6 +2577,8 @@ void register_console(struct console *newcon) } } + newcon->thread = NULL; + if (console_drivers && console_drivers->flags & CON_BOOT) bcon = console_drivers; @@ -2797,8 +2620,10 @@ void register_console(struct console *newcon) * the real console are the same physical device, it's annoying to * see the beginning boot messages twice */ - if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV)) + if (bcon && ((newcon->flags & (CON_CONSDEV | CON_BOOT)) == CON_CONSDEV)) { newcon->flags &= ~CON_PRINTBUFFER; + newcon->flags |= CON_HANDOVER; + } /* * Put this console in the list - keep the @@ -2820,27 +2645,12 @@ void register_console(struct console *newcon) if (newcon->flags & CON_EXTENDED) nr_ext_console_drivers++; - if (newcon->flags & CON_PRINTBUFFER) { - /* - * console_unlock(); will print out the buffered messages - * for us. - */ - logbuf_lock_irqsave(flags); - /* - * We're about to replay the log buffer. Only do this to the - * just-registered console to avoid excessive message spam to - * the already-registered consoles. - * - * Set exclusive_console with disabled interrupts to reduce - * race window with eventual console_flush_on_panic() that - * ignores console_lock. - */ - exclusive_console = newcon; - exclusive_console_stop_seq = console_seq; - console_seq = syslog_seq; - console_idx = syslog_idx; - logbuf_unlock_irqrestore(flags); - } + if (newcon->flags & CON_PRINTBUFFER) + atomic64_set(&newcon->printk_seq, 0); + else + atomic64_set(&newcon->printk_seq, prb_next_seq(prb)); + + console_try_thread(newcon); console_unlock(); console_sysfs_notify(); @@ -2914,6 +2724,9 @@ int unregister_console(struct console *console) console_unlock(); console_sysfs_notify(); + if (console->thread && !IS_ERR(console->thread)) + kthread_stop(console->thread); + if (console->exit) res = console->exit(console); @@ -2957,6 +2770,154 @@ void __init console_init(void) } } +static int printk_kthread_func(void *data) +{ + struct console *con = data; + unsigned long dropped = 0; + struct printk_info info; + struct printk_record r; + char *ext_text = NULL; + size_t dropped_len; + char *dropped_text; + int ret = -ENOMEM; + char *write_text; + u64 printk_seq; + size_t len; + char *text; + int error; + u64 seq; + + if (con->flags & CON_EXTENDED) { + ext_text = kmalloc(CONSOLE_EXT_LOG_MAX, GFP_KERNEL); + if (!ext_text) + return ret; + } + text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL); + dropped_text = kmalloc(64, GFP_KERNEL); + if (!text || !dropped_text) + goto out; + + if (con->flags & CON_EXTENDED) + write_text = ext_text; + else + write_text = text; + + seq = atomic64_read(&con->printk_seq); + + prb_rec_init_rd(&r, &info, text, LOG_LINE_MAX + PREFIX_MAX); + + for (;;) { + error = wait_event_interruptible(log_wait, + prb_read_valid(prb, seq, &r) || kthread_should_stop()); + + if (kthread_should_stop()) + break; + + if (error) + continue; + + if (seq != r.info->seq) { + dropped += r.info->seq - seq; + seq = r.info->seq; + } + + seq++; + + if (!(con->flags & CON_ENABLED)) + continue; + + if (suppress_message_printing(r.info->level)) + continue; + + if (con->flags & CON_EXTENDED) { + len = info_print_ext_header(ext_text, + CONSOLE_EXT_LOG_MAX, + r.info); + len += msg_print_ext_body(ext_text + len, + CONSOLE_EXT_LOG_MAX - len, + &r.text_buf[0], r.info->text_len, + &r.info->dev_info); + } else { + len = record_print_text(&r, + console_msg_format & MSG_FORMAT_SYSLOG, + printk_time); + } + + printk_seq = atomic64_read(&con->printk_seq); + + console_lock(); + console_may_schedule = 0; + + if (kernel_sync_mode() && con->write_atomic) { + console_unlock(); + break; + } + + if (!(con->flags & CON_EXTENDED) && dropped) { + dropped_len = snprintf(dropped_text, 64, + "** %lu printk messages dropped **\n", + dropped); + dropped = 0; + + con->write(con, dropped_text, dropped_len); + printk_delay(r.info->level); + } + + con->write(con, write_text, len); + if (len) + printk_delay(r.info->level); + + atomic64_cmpxchg_relaxed(&con->printk_seq, printk_seq, seq); + + console_unlock(); + } +out: + kfree(dropped_text); + kfree(text); + kfree(ext_text); + pr_info("%sconsole [%s%d]: printing thread stopped\n", + (con->flags & CON_BOOT) ? "boot" : "" , + con->name, con->index); + return ret; +} + +static void start_printk_kthread(struct console *con) +{ + con->thread = kthread_run(printk_kthread_func, con, + "pr/%s%d", con->name, con->index); + if (IS_ERR(con->thread)) { + pr_err("%sconsole [%s%d]: unable to start printing thread\n", + (con->flags & CON_BOOT) ? "boot" : "" , + con->name, con->index); + return; + } + pr_info("%sconsole [%s%d]: printing thread started\n", + (con->flags & CON_BOOT) ? "boot" : "" , + con->name, con->index); +} + +static bool kthreads_started; + +static void console_try_thread(struct console *con) +{ + unsigned long irqflags; + int sprint_id; + char *buf; + + if (kthreads_started) { + start_printk_kthread(con); + return; + } + + buf = get_sprint_buf(&sprint_id, &irqflags); + if (!buf) + return; + + print_sync_until(prb_next_seq(prb), con, buf, PREFIX_MAX + LOG_LINE_MAX); + + put_sprint_buf(sprint_id, irqflags); +} + /* * Some boot consoles access data that is in the init section and which will * be discarded after the initcalls have been run. To make sure that no code @@ -2996,6 +2957,13 @@ static int __init printk_late_init(void) unregister_console(con); } } + + console_lock(); + for_each_console(con) + start_printk_kthread(con); + kthreads_started = true; + console_unlock(); + ret = cpuhp_setup_state_nocalls(CPUHP_PRINTK_DEAD, "printk:dead", NULL, console_cpu_notify); WARN_ON(ret < 0); @@ -3011,7 +2979,6 @@ late_initcall(printk_late_init); * Delayed printk version, for scheduler-internal messages: */ #define PRINTK_PENDING_WAKEUP 0x01 -#define PRINTK_PENDING_OUTPUT 0x02 static DEFINE_PER_CPU(int, printk_pending); @@ -3019,12 +2986,6 @@ static void wake_up_klogd_work_func(struct irq_work *irq_work) { int pending = __this_cpu_xchg(printk_pending, 0); - if (pending & PRINTK_PENDING_OUTPUT) { - /* If trylock fails, someone else is doing the printing */ - if (console_trylock()) - console_unlock(); - } - if (pending & PRINTK_PENDING_WAKEUP) wake_up_interruptible(&log_wait); } @@ -3047,25 +3008,10 @@ void wake_up_klogd(void) preempt_enable(); } -void defer_console_output(void) +__printf(1, 0) +static int vprintk_deferred(const char *fmt, va_list args) { - if (!printk_percpu_data_ready()) - return; - - preempt_disable(); - __this_cpu_or(printk_pending, PRINTK_PENDING_OUTPUT); - irq_work_queue(this_cpu_ptr(&wake_up_klogd_work)); - preempt_enable(); -} - -int vprintk_deferred(const char *fmt, va_list args) -{ - int r; - - r = vprintk_emit(0, LOGLEVEL_SCHED, NULL, 0, fmt, args); - defer_console_output(); - - return r; + return vprintk_emit(0, LOGLEVEL_DEFAULT, NULL, fmt, args); } int printk_deferred(const char *fmt, ...) @@ -3205,7 +3151,18 @@ EXPORT_SYMBOL_GPL(kmsg_dump_reason_str); void kmsg_dump(enum kmsg_dump_reason reason) { struct kmsg_dumper *dumper; - unsigned long flags; + + if (!oops_in_progress) { + /* + * If atomic consoles are available, activate kernel sync mode + * to make sure any final messages are visible. The trailing + * printk message is important to flush any pending messages. + */ + if (have_atomic_console()) { + sync_mode = true; + pr_info("enabled sync mode\n"); + } + } rcu_read_lock(); list_for_each_entry_rcu(dumper, &dump_list, list) { @@ -3225,12 +3182,7 @@ void kmsg_dump(enum kmsg_dump_reason reason) /* initialize iterator with data about the stored records */ dumper->active = true; - logbuf_lock_irqsave(flags); - dumper->cur_seq = clear_seq; - dumper->cur_idx = clear_idx; - dumper->next_seq = log_next_seq; - dumper->next_idx = log_next_idx; - logbuf_unlock_irqrestore(flags); + kmsg_dump_rewind_nolock(dumper); /* invoke dumper which will iterate over records */ dumper->dump(dumper, reason); @@ -3263,28 +3215,33 @@ void kmsg_dump(enum kmsg_dump_reason reason) bool kmsg_dump_get_line_nolock(struct kmsg_dumper *dumper, bool syslog, char *line, size_t size, size_t *len) { - struct printk_log *msg; + struct printk_info info; + unsigned int line_count; + struct printk_record r; size_t l = 0; bool ret = false; + prb_rec_init_rd(&r, &info, line, size); + if (!dumper->active) goto out; - if (dumper->cur_seq < log_first_seq) { - /* messages are gone, move to first available one */ - dumper->cur_seq = log_first_seq; - dumper->cur_idx = log_first_idx; + /* Read text or count text lines? */ + if (line) { + if (!prb_read_valid(prb, dumper->cur_seq, &r)) + goto out; + l = record_print_text(&r, syslog, printk_time); + } else { + if (!prb_read_valid_info(prb, dumper->cur_seq, + &info, &line_count)) { + goto out; + } + l = get_record_print_text_size(&info, line_count, syslog, + printk_time); + } - /* last entry */ - if (dumper->cur_seq >= log_next_seq) - goto out; - - msg = log_from_idx(dumper->cur_idx); - l = msg_print_text(msg, syslog, printk_time, line, size); - - dumper->cur_idx = log_next(dumper->cur_idx); - dumper->cur_seq++; + dumper->cur_seq = r.info->seq + 1; ret = true; out: if (len) @@ -3312,14 +3269,7 @@ bool kmsg_dump_get_line_nolock(struct kmsg_dumper *dumper, bool syslog, bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, char *line, size_t size, size_t *len) { - unsigned long flags; - bool ret; - - logbuf_lock_irqsave(flags); - ret = kmsg_dump_get_line_nolock(dumper, syslog, line, size, len); - logbuf_unlock_irqrestore(flags); - - return ret; + return kmsg_dump_get_line_nolock(dumper, syslog, line, size, len); } EXPORT_SYMBOL_GPL(kmsg_dump_get_line); @@ -3329,7 +3279,7 @@ EXPORT_SYMBOL_GPL(kmsg_dump_get_line); * @syslog: include the "<4>" prefixes * @buf: buffer to copy the line to * @size: maximum size of the buffer - * @len: length of line placed into buffer + * @len_out: length of line placed into buffer * * Start at the end of the kmsg buffer and fill the provided buffer * with as many of the the *youngest* kmsg records that fit into it. @@ -3343,75 +3293,73 @@ EXPORT_SYMBOL_GPL(kmsg_dump_get_line); * read. */ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, - char *buf, size_t size, size_t *len) + char *buf, size_t size, size_t *len_out) { - unsigned long flags; + struct printk_info info; + unsigned int line_count; + struct printk_record r; u64 seq; - u32 idx; u64 next_seq; - u32 next_idx; - size_t l = 0; + size_t len = 0; bool ret = false; bool time = printk_time; - if (!dumper->active) + if (!dumper->active || !buf || !size) goto out; - logbuf_lock_irqsave(flags); - if (dumper->cur_seq < log_first_seq) { + if (dumper->cur_seq < prb_first_valid_seq(prb)) { /* messages are gone, move to first available one */ - dumper->cur_seq = log_first_seq; - dumper->cur_idx = log_first_idx; + dumper->cur_seq = prb_first_valid_seq(prb); } /* last entry */ - if (dumper->cur_seq >= dumper->next_seq) { - logbuf_unlock_irqrestore(flags); + if (dumper->cur_seq >= dumper->next_seq) goto out; + + /* + * Find first record that fits, including all following records, + * into the user-provided buffer for this dump. + */ + + prb_for_each_info(dumper->cur_seq, prb, seq, &info, &line_count) { + if (info.seq >= dumper->next_seq) + break; + len += get_record_print_text_size(&info, line_count, true, time); } - /* calculate length of entire buffer */ - seq = dumper->cur_seq; - idx = dumper->cur_idx; - while (seq < dumper->next_seq) { - struct printk_log *msg = log_from_idx(idx); - - l += msg_print_text(msg, true, time, NULL, 0); - idx = log_next(idx); - seq++; + /* + * Move first record forward until length fits into the buffer. This + * is a best effort attempt. If @dumper->next_seq is reached because + * the ringbuffer is wrapping too fast, just start filling the buffer + * from there. + */ + prb_for_each_info(dumper->cur_seq, prb, seq, &info, &line_count) { + if (len <= size || info.seq >= dumper->next_seq) + break; + len -= get_record_print_text_size(&info, line_count, true, time); } - /* move first record forward until length fits into the buffer */ - seq = dumper->cur_seq; - idx = dumper->cur_idx; - while (l >= size && seq < dumper->next_seq) { - struct printk_log *msg = log_from_idx(idx); - - l -= msg_print_text(msg, true, time, NULL, 0); - idx = log_next(idx); - seq++; - } - - /* last message in next interation */ + /* Keep track of the last message for the next interation. */ next_seq = seq; - next_idx = idx; - l = 0; - while (seq < dumper->next_seq) { - struct printk_log *msg = log_from_idx(idx); + prb_rec_init_rd(&r, &info, buf, size); - l += msg_print_text(msg, syslog, time, buf + l, size - l); - idx = log_next(idx); - seq++; + len = 0; + prb_for_each_record(seq, prb, seq, &r) { + if (r.info->seq >= dumper->next_seq) + break; + + len += record_print_text(&r, syslog, time); + + /* Adjust record to store to remaining buffer space. */ + prb_rec_init_rd(&r, &info, buf + len, size - len); } dumper->next_seq = next_seq; - dumper->next_idx = next_idx; ret = true; - logbuf_unlock_irqrestore(flags); out: - if (len) - *len = l; + if (len_out) + *len_out = len; return ret; } EXPORT_SYMBOL_GPL(kmsg_dump_get_buffer); @@ -3423,15 +3371,11 @@ EXPORT_SYMBOL_GPL(kmsg_dump_get_buffer); * Reset the dumper's iterator so that kmsg_dump_get_line() and * kmsg_dump_get_buffer() can be called again and used multiple * times within the same dumper.dump() callback. - * - * The function is similar to kmsg_dump_rewind(), but grabs no locks. */ void kmsg_dump_rewind_nolock(struct kmsg_dumper *dumper) { - dumper->cur_seq = clear_seq; - dumper->cur_idx = clear_idx; - dumper->next_seq = log_next_seq; - dumper->next_idx = log_next_idx; + dumper->cur_seq = atomic64_read(&clear_seq); + dumper->next_seq = prb_next_seq(prb); } /** @@ -3444,12 +3388,108 @@ void kmsg_dump_rewind_nolock(struct kmsg_dumper *dumper) */ void kmsg_dump_rewind(struct kmsg_dumper *dumper) { - unsigned long flags; - - logbuf_lock_irqsave(flags); kmsg_dump_rewind_nolock(dumper); - logbuf_unlock_irqrestore(flags); } EXPORT_SYMBOL_GPL(kmsg_dump_rewind); #endif + +struct prb_cpulock { + atomic_t owner; + unsigned long __percpu *irqflags; +}; + +#define DECLARE_STATIC_PRINTKRB_CPULOCK(name) \ +static DEFINE_PER_CPU(unsigned long, _##name##_percpu_irqflags); \ +static struct prb_cpulock name = { \ + .owner = ATOMIC_INIT(-1), \ + .irqflags = &_##name##_percpu_irqflags, \ +} + +static bool __prb_trylock(struct prb_cpulock *cpu_lock, + unsigned int *cpu_store) +{ + unsigned long *flags; + unsigned int cpu; + + cpu = get_cpu(); + + *cpu_store = atomic_read(&cpu_lock->owner); + /* memory barrier to ensure the current lock owner is visible */ + smp_rmb(); + if (*cpu_store == -1) { + flags = per_cpu_ptr(cpu_lock->irqflags, cpu); + local_irq_save(*flags); + if (atomic_try_cmpxchg_acquire(&cpu_lock->owner, + cpu_store, cpu)) { + return true; + } + local_irq_restore(*flags); + } else if (*cpu_store == cpu) { + return true; + } + + put_cpu(); + return false; +} + +/* + * prb_lock: Perform a processor-reentrant spin lock. + * @cpu_lock: A pointer to the lock object. + * @cpu_store: A "flags" pointer to store lock status information. + * + * If no processor has the lock, the calling processor takes the lock and + * becomes the owner. If the calling processor is already the owner of the + * lock, this function succeeds immediately. If lock is locked by another + * processor, this function spins until the calling processor becomes the + * owner. + * + * It is safe to call this function from any context and state. + */ +static void prb_lock(struct prb_cpulock *cpu_lock, unsigned int *cpu_store) +{ + for (;;) { + if (__prb_trylock(cpu_lock, cpu_store)) + break; + cpu_relax(); + } +} + +/* + * prb_unlock: Perform a processor-reentrant spin unlock. + * @cpu_lock: A pointer to the lock object. + * @cpu_store: A "flags" object storing lock status information. + * + * Release the lock. The calling processor must be the owner of the lock. + * + * It is safe to call this function from any context and state. + */ +static void prb_unlock(struct prb_cpulock *cpu_lock, unsigned int cpu_store) +{ + unsigned long *flags; + unsigned int cpu; + + cpu = atomic_read(&cpu_lock->owner); + atomic_set_release(&cpu_lock->owner, cpu_store); + + if (cpu_store == -1) { + flags = per_cpu_ptr(cpu_lock->irqflags, cpu); + local_irq_restore(*flags); + } + + put_cpu(); +} + +DECLARE_STATIC_PRINTKRB_CPULOCK(printk_cpulock); + +void console_atomic_lock(unsigned int *flags) +{ + prb_lock(&printk_cpulock, flags); +} +EXPORT_SYMBOL(console_atomic_lock); + +void console_atomic_unlock(unsigned int flags) +{ + prb_unlock(&printk_cpulock, flags); +} +EXPORT_SYMBOL(console_atomic_unlock); diff --git a/kernel/printk/printk_ringbuffer.c b/kernel/printk/printk_ringbuffer.c new file mode 100644 index 0000000000000..24a960a89aa89 --- /dev/null +++ b/kernel/printk/printk_ringbuffer.c @@ -0,0 +1,2086 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include +#include +#include +#include +#include +#include "printk_ringbuffer.h" + +/** + * DOC: printk_ringbuffer overview + * + * Data Structure + * -------------- + * The printk_ringbuffer is made up of 3 internal ringbuffers: + * + * desc_ring + * A ring of descriptors and their meta data (such as sequence number, + * timestamp, loglevel, etc.) as well as internal state information about + * the record and logical positions specifying where in the other + * ringbuffer the text strings are located. + * + * text_data_ring + * A ring of data blocks. A data block consists of an unsigned long + * integer (ID) that maps to a desc_ring index followed by the text + * string of the record. + * + * The internal state information of a descriptor is the key element to allow + * readers and writers to locklessly synchronize access to the data. + * + * Implementation + * -------------- + * + * Descriptor Ring + * ~~~~~~~~~~~~~~~ + * The descriptor ring is an array of descriptors. A descriptor contains + * essential meta data to track the data of a printk record using + * blk_lpos structs pointing to associated text data blocks (see + * "Data Rings" below). Each descriptor is assigned an ID that maps + * directly to index values of the descriptor array and has a state. The ID + * and the state are bitwise combined into a single descriptor field named + * @state_var, allowing ID and state to be synchronously and atomically + * updated. + * + * Descriptors have four states: + * + * reserved + * A writer is modifying the record. + * + * committed + * The record and all its data are written. A writer can reopen the + * descriptor (transitioning it back to reserved), but in the committed + * state the data is consistent. + * + * finalized + * The record and all its data are complete and available for reading. A + * writer cannot reopen the descriptor. + * + * reusable + * The record exists, but its text and/or meta data may no longer be + * available. + * + * Querying the @state_var of a record requires providing the ID of the + * descriptor to query. This can yield a possible fifth (pseudo) state: + * + * miss + * The descriptor being queried has an unexpected ID. + * + * The descriptor ring has a @tail_id that contains the ID of the oldest + * descriptor and @head_id that contains the ID of the newest descriptor. + * + * When a new descriptor should be created (and the ring is full), the tail + * descriptor is invalidated by first transitioning to the reusable state and + * then invalidating all tail data blocks up to and including the data blocks + * associated with the tail descriptor (for the text ring). Then + * @tail_id is advanced, followed by advancing @head_id. And finally the + * @state_var of the new descriptor is initialized to the new ID and reserved + * state. + * + * The @tail_id can only be advanced if the new @tail_id would be in the + * committed or reusable queried state. This makes it possible that a valid + * sequence number of the tail is always available. + * + * Descriptor Finalization + * ~~~~~~~~~~~~~~~~~~~~~~~ + * When a writer calls the commit function prb_commit(), record data is + * fully stored and is consistent within the ringbuffer. However, a writer can + * reopen that record, claiming exclusive access (as with prb_reserve()), and + * modify that record. When finished, the writer must again commit the record. + * + * In order for a record to be made available to readers (and also become + * recyclable for writers), it must be finalized. A finalized record cannot be + * reopened and can never become "unfinalized". Record finalization can occur + * in three different scenarios: + * + * 1) A writer can simultaneously commit and finalize its record by calling + * prb_final_commit() instead of prb_commit(). + * + * 2) When a new record is reserved and the previous record has been + * committed via prb_commit(), that previous record is automatically + * finalized. + * + * 3) When a record is committed via prb_commit() and a newer record + * already exists, the record being committed is automatically finalized. + * + * Data Ring + * ~~~~~~~~~ + * The text data ring is a byte array composed of data blocks. Data blocks are + * referenced by blk_lpos structs that point to the logical position of the + * beginning of a data block and the beginning of the next adjacent data + * block. Logical positions are mapped directly to index values of the byte + * array ringbuffer. + * + * Each data block consists of an ID followed by the writer data. The ID is + * the identifier of a descriptor that is associated with the data block. A + * given data block is considered valid if all of the following conditions + * are met: + * + * 1) The descriptor associated with the data block is in the committed + * or finalized queried state. + * + * 2) The blk_lpos struct within the descriptor associated with the data + * block references back to the same data block. + * + * 3) The data block is within the head/tail logical position range. + * + * If the writer data of a data block would extend beyond the end of the + * byte array, only the ID of the data block is stored at the logical + * position and the full data block (ID and writer data) is stored at the + * beginning of the byte array. The referencing blk_lpos will point to the + * ID before the wrap and the next data block will be at the logical + * position adjacent the full data block after the wrap. + * + * Data rings have a @tail_lpos that points to the beginning of the oldest + * data block and a @head_lpos that points to the logical position of the + * next (not yet existing) data block. + * + * When a new data block should be created (and the ring is full), tail data + * blocks will first be invalidated by putting their associated descriptors + * into the reusable state and then pushing the @tail_lpos forward beyond + * them. Then the @head_lpos is pushed forward and is associated with a new + * descriptor. If a data block is not valid, the @tail_lpos cannot be + * advanced beyond it. + * + * Info Array + * ~~~~~~~~~~ + * The general meta data of printk records are stored in printk_info structs, + * stored in an array with the same number of elements as the descriptor ring. + * Each info corresponds to the descriptor of the same index in the + * descriptor ring. Info validity is confirmed by evaluating the corresponding + * descriptor before and after loading the info. + * + * Usage + * ----- + * Here are some simple examples demonstrating writers and readers. For the + * examples a global ringbuffer (test_rb) is available (which is not the + * actual ringbuffer used by printk):: + * + * DEFINE_PRINTKRB(test_rb, 15, 5); + * + * This ringbuffer allows up to 32768 records (2 ^ 15) and has a size of + * 1 MiB (2 ^ (15 + 5)) for text data. + * + * Sample writer code:: + * + * const char *textstr = "message text"; + * struct prb_reserved_entry e; + * struct printk_record r; + * + * // specify how much to allocate + * prb_rec_init_wr(&r, strlen(textstr) + 1); + * + * if (prb_reserve(&e, &test_rb, &r)) { + * snprintf(r.text_buf, r.text_buf_size, "%s", textstr); + * + * r.info->text_len = strlen(textstr); + * r.info->ts_nsec = local_clock(); + * r.info->caller_id = printk_caller_id(); + * + * // commit and finalize the record + * prb_final_commit(&e); + * } + * + * Note that additional writer functions are available to extend a record + * after it has been committed but not yet finalized. This can be done as + * long as no new records have been reserved and the caller is the same. + * + * Sample writer code (record extending):: + * + * // alternate rest of previous example + * + * r.info->text_len = strlen(textstr); + * r.info->ts_nsec = local_clock(); + * r.info->caller_id = printk_caller_id(); + * + * // commit the record (but do not finalize yet) + * prb_commit(&e); + * } + * + * ... + * + * // specify additional 5 bytes text space to extend + * prb_rec_init_wr(&r, 5); + * + * // try to extend, but only if it does not exceed 32 bytes + * if (prb_reserve_in_last(&e, &test_rb, &r, printk_caller_id()), 32) { + * snprintf(&r.text_buf[r.info->text_len], + * r.text_buf_size - r.info->text_len, "hello"); + * + * r.info->text_len += 5; + * + * // commit and finalize the record + * prb_final_commit(&e); + * } + * + * Sample reader code:: + * + * struct printk_info info; + * struct printk_record r; + * char text_buf[32]; + * u64 seq; + * + * prb_rec_init_rd(&r, &info, &text_buf[0], sizeof(text_buf)); + * + * prb_for_each_record(0, &test_rb, &seq, &r) { + * if (info.seq != seq) + * pr_warn("lost %llu records\n", info.seq - seq); + * + * if (info.text_len > r.text_buf_size) { + * pr_warn("record %llu text truncated\n", info.seq); + * text_buf[r.text_buf_size - 1] = 0; + * } + * + * pr_info("%llu: %llu: %s\n", info.seq, info.ts_nsec, + * &text_buf[0]); + * } + * + * Note that additional less convenient reader functions are available to + * allow complex record access. + * + * ABA Issues + * ~~~~~~~~~~ + * To help avoid ABA issues, descriptors are referenced by IDs (array index + * values combined with tagged bits counting array wraps) and data blocks are + * referenced by logical positions (array index values combined with tagged + * bits counting array wraps). However, on 32-bit systems the number of + * tagged bits is relatively small such that an ABA incident is (at least + * theoretically) possible. For example, if 4 million maximally sized (1KiB) + * printk messages were to occur in NMI context on a 32-bit system, the + * interrupted context would not be able to recognize that the 32-bit integer + * completely wrapped and thus represents a different data block than the one + * the interrupted context expects. + * + * To help combat this possibility, additional state checking is performed + * (such as using cmpxchg() even though set() would suffice). These extra + * checks are commented as such and will hopefully catch any ABA issue that + * a 32-bit system might experience. + * + * Memory Barriers + * ~~~~~~~~~~~~~~~ + * Multiple memory barriers are used. To simplify proving correctness and + * generating litmus tests, lines of code related to memory barriers + * (loads, stores, and the associated memory barriers) are labeled:: + * + * LMM(function:letter) + * + * Comments reference the labels using only the "function:letter" part. + * + * The memory barrier pairs and their ordering are: + * + * desc_reserve:D / desc_reserve:B + * push descriptor tail (id), then push descriptor head (id) + * + * desc_reserve:D / data_push_tail:B + * push data tail (lpos), then set new descriptor reserved (state) + * + * desc_reserve:D / desc_push_tail:C + * push descriptor tail (id), then set new descriptor reserved (state) + * + * desc_reserve:D / prb_first_seq:C + * push descriptor tail (id), then set new descriptor reserved (state) + * + * desc_reserve:F / desc_read:D + * set new descriptor id and reserved (state), then allow writer changes + * + * data_alloc:A (or data_realloc:A) / desc_read:D + * set old descriptor reusable (state), then modify new data block area + * + * data_alloc:A (or data_realloc:A) / data_push_tail:B + * push data tail (lpos), then modify new data block area + * + * _prb_commit:B / desc_read:B + * store writer changes, then set new descriptor committed (state) + * + * desc_reopen_last:A / _prb_commit:B + * set descriptor reserved (state), then read descriptor data + * + * _prb_commit:B / desc_reserve:D + * set new descriptor committed (state), then check descriptor head (id) + * + * data_push_tail:D / data_push_tail:A + * set descriptor reusable (state), then push data tail (lpos) + * + * desc_push_tail:B / desc_reserve:D + * set descriptor reusable (state), then push descriptor tail (id) + */ + +#define DATA_SIZE(data_ring) _DATA_SIZE((data_ring)->size_bits) +#define DATA_SIZE_MASK(data_ring) (DATA_SIZE(data_ring) - 1) + +#define DESCS_COUNT(desc_ring) _DESCS_COUNT((desc_ring)->count_bits) +#define DESCS_COUNT_MASK(desc_ring) (DESCS_COUNT(desc_ring) - 1) + +/* Determine the data array index from a logical position. */ +#define DATA_INDEX(data_ring, lpos) ((lpos) & DATA_SIZE_MASK(data_ring)) + +/* Determine the desc array index from an ID or sequence number. */ +#define DESC_INDEX(desc_ring, n) ((n) & DESCS_COUNT_MASK(desc_ring)) + +/* Determine how many times the data array has wrapped. */ +#define DATA_WRAPS(data_ring, lpos) ((lpos) >> (data_ring)->size_bits) + +/* Determine if a logical position refers to a data-less block. */ +#define LPOS_DATALESS(lpos) ((lpos) & 1UL) +#define BLK_DATALESS(blk) (LPOS_DATALESS((blk)->begin) && \ + LPOS_DATALESS((blk)->next)) + +/* Get the logical position at index 0 of the current wrap. */ +#define DATA_THIS_WRAP_START_LPOS(data_ring, lpos) \ +((lpos) & ~DATA_SIZE_MASK(data_ring)) + +/* Get the ID for the same index of the previous wrap as the given ID. */ +#define DESC_ID_PREV_WRAP(desc_ring, id) \ +DESC_ID((id) - DESCS_COUNT(desc_ring)) + +/* + * A data block: mapped directly to the beginning of the data block area + * specified as a logical position within the data ring. + * + * @id: the ID of the associated descriptor + * @data: the writer data + * + * Note that the size of a data block is only known by its associated + * descriptor. + */ +struct prb_data_block { + unsigned long id; + char data[0]; +}; + +/* + * Return the descriptor associated with @n. @n can be either a + * descriptor ID or a sequence number. + */ +static struct prb_desc *to_desc(struct prb_desc_ring *desc_ring, u64 n) +{ + return &desc_ring->descs[DESC_INDEX(desc_ring, n)]; +} + +/* + * Return the printk_info associated with @n. @n can be either a + * descriptor ID or a sequence number. + */ +static struct printk_info *to_info(struct prb_desc_ring *desc_ring, u64 n) +{ + return &desc_ring->infos[DESC_INDEX(desc_ring, n)]; +} + +static struct prb_data_block *to_block(struct prb_data_ring *data_ring, + unsigned long begin_lpos) +{ + return (void *)&data_ring->data[DATA_INDEX(data_ring, begin_lpos)]; +} + +/* + * Increase the data size to account for data block meta data plus any + * padding so that the adjacent data block is aligned on the ID size. + */ +static unsigned int to_blk_size(unsigned int size) +{ + struct prb_data_block *db = NULL; + + size += sizeof(*db); + size = ALIGN(size, sizeof(db->id)); + return size; +} + +/* + * Sanity checker for reserve size. The ringbuffer code assumes that a data + * block does not exceed the maximum possible size that could fit within the + * ringbuffer. This function provides that basic size check so that the + * assumption is safe. + */ +static bool data_check_size(struct prb_data_ring *data_ring, unsigned int size) +{ + struct prb_data_block *db = NULL; + + if (size == 0) + return true; + + /* + * Ensure the alignment padded size could possibly fit in the data + * array. The largest possible data block must still leave room for + * at least the ID of the next block. + */ + size = to_blk_size(size); + if (size > DATA_SIZE(data_ring) - sizeof(db->id)) + return false; + + return true; +} + +/* Query the state of a descriptor. */ +static enum desc_state get_desc_state(unsigned long id, + unsigned long state_val) +{ + if (id != DESC_ID(state_val)) + return desc_miss; + + return DESC_STATE(state_val); +} + +/* + * Get a copy of a specified descriptor and return its queried state. If the + * descriptor is in an inconsistent state (miss or reserved), the caller can + * only expect the descriptor's @state_var field to be valid. + * + * The sequence number and caller_id can be optionally retrieved. Like all + * non-state_var data, they are only valid if the descriptor is in a + * consistent state. + */ +static enum desc_state desc_read(struct prb_desc_ring *desc_ring, + unsigned long id, struct prb_desc *desc_out, + u64 *seq_out, u32 *caller_id_out) +{ + struct printk_info *info = to_info(desc_ring, id); + struct prb_desc *desc = to_desc(desc_ring, id); + atomic_long_t *state_var = &desc->state_var; + enum desc_state d_state; + unsigned long state_val; + + /* Check the descriptor state. */ + state_val = atomic_long_read(state_var); /* LMM(desc_read:A) */ + d_state = get_desc_state(id, state_val); + if (d_state == desc_miss || d_state == desc_reserved) { + /* + * The descriptor is in an inconsistent state. Set at least + * @state_var so that the caller can see the details of + * the inconsistent state. + */ + goto out; + } + + /* + * Guarantee the state is loaded before copying the descriptor + * content. This avoids copying obsolete descriptor content that might + * not apply to the descriptor state. This pairs with _prb_commit:B. + * + * Memory barrier involvement: + * + * If desc_read:A reads from _prb_commit:B, then desc_read:C reads + * from _prb_commit:A. + * + * Relies on: + * + * WMB from _prb_commit:A to _prb_commit:B + * matching + * RMB from desc_read:A to desc_read:C + */ + smp_rmb(); /* LMM(desc_read:B) */ + + /* + * Copy the descriptor data. The data is not valid until the + * state has been re-checked. A memcpy() for all of @desc + * cannot be used because of the atomic_t @state_var field. + */ + memcpy(&desc_out->text_blk_lpos, &desc->text_blk_lpos, + sizeof(desc_out->text_blk_lpos)); /* LMM(desc_read:C) */ + if (seq_out) + *seq_out = info->seq; /* also part of desc_read:C */ + if (caller_id_out) + *caller_id_out = info->caller_id; /* also part of desc_read:C */ + + /* + * 1. Guarantee the descriptor content is loaded before re-checking + * the state. This avoids reading an obsolete descriptor state + * that may not apply to the copied content. This pairs with + * desc_reserve:F. + * + * Memory barrier involvement: + * + * If desc_read:C reads from desc_reserve:G, then desc_read:E + * reads from desc_reserve:F. + * + * Relies on: + * + * WMB from desc_reserve:F to desc_reserve:G + * matching + * RMB from desc_read:C to desc_read:E + * + * 2. Guarantee the record data is loaded before re-checking the + * state. This avoids reading an obsolete descriptor state that may + * not apply to the copied data. This pairs with data_alloc:A and + * data_realloc:A. + * + * Memory barrier involvement: + * + * If copy_data:A reads from data_alloc:B, then desc_read:E + * reads from desc_make_reusable:A. + * + * Relies on: + * + * MB from desc_make_reusable:A to data_alloc:B + * matching + * RMB from desc_read:C to desc_read:E + * + * Note: desc_make_reusable:A and data_alloc:B can be different + * CPUs. However, the data_alloc:B CPU (which performs the + * full memory barrier) must have previously seen + * desc_make_reusable:A. + */ + smp_rmb(); /* LMM(desc_read:D) */ + + /* + * The data has been copied. Return the current descriptor state, + * which may have changed since the load above. + */ + state_val = atomic_long_read(state_var); /* LMM(desc_read:E) */ + d_state = get_desc_state(id, state_val); +out: + atomic_long_set(&desc_out->state_var, state_val); + return d_state; +} + +/* + * Take a specified descriptor out of the finalized state by attempting + * the transition from finalized to reusable. Either this context or some + * other context will have been successful. + */ +static void desc_make_reusable(struct prb_desc_ring *desc_ring, + unsigned long id) +{ + unsigned long val_finalized = DESC_SV(id, desc_finalized); + unsigned long val_reusable = DESC_SV(id, desc_reusable); + struct prb_desc *desc = to_desc(desc_ring, id); + atomic_long_t *state_var = &desc->state_var; + + atomic_long_cmpxchg_relaxed(state_var, val_finalized, + val_reusable); /* LMM(desc_make_reusable:A) */ +} + +/* + * Given the text data ring, put the associated descriptor of each + * data block from @lpos_begin until @lpos_end into the reusable state. + * + * If there is any problem making the associated descriptor reusable, either + * the descriptor has not yet been finalized or another writer context has + * already pushed the tail lpos past the problematic data block. Regardless, + * on error the caller can re-load the tail lpos to determine the situation. + */ +static bool data_make_reusable(struct printk_ringbuffer *rb, + struct prb_data_ring *data_ring, + unsigned long lpos_begin, + unsigned long lpos_end, + unsigned long *lpos_out) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + struct prb_data_block *blk; + enum desc_state d_state; + struct prb_desc desc; + struct prb_data_blk_lpos *blk_lpos = &desc.text_blk_lpos; + unsigned long id; + + /* Loop until @lpos_begin has advanced to or beyond @lpos_end. */ + while ((lpos_end - lpos_begin) - 1 < DATA_SIZE(data_ring)) { + blk = to_block(data_ring, lpos_begin); + + /* + * Load the block ID from the data block. This is a data race + * against a writer that may have newly reserved this data + * area. If the loaded value matches a valid descriptor ID, + * the blk_lpos of that descriptor will be checked to make + * sure it points back to this data block. If the check fails, + * the data area has been recycled by another writer. + */ + id = blk->id; /* LMM(data_make_reusable:A) */ + + d_state = desc_read(desc_ring, id, &desc, + NULL, NULL); /* LMM(data_make_reusable:B) */ + + switch (d_state) { + case desc_miss: + case desc_reserved: + case desc_committed: + return false; + case desc_finalized: + /* + * This data block is invalid if the descriptor + * does not point back to it. + */ + if (blk_lpos->begin != lpos_begin) + return false; + desc_make_reusable(desc_ring, id); + break; + case desc_reusable: + /* + * This data block is invalid if the descriptor + * does not point back to it. + */ + if (blk_lpos->begin != lpos_begin) + return false; + break; + } + + /* Advance @lpos_begin to the next data block. */ + lpos_begin = blk_lpos->next; + } + + *lpos_out = lpos_begin; + return true; +} + +/* + * Advance the data ring tail to at least @lpos. This function puts + * descriptors into the reusable state if the tail is pushed beyond + * their associated data block. + */ +static bool data_push_tail(struct printk_ringbuffer *rb, + struct prb_data_ring *data_ring, + unsigned long lpos) +{ + unsigned long tail_lpos_new; + unsigned long tail_lpos; + unsigned long next_lpos; + + /* If @lpos is from a data-less block, there is nothing to do. */ + if (LPOS_DATALESS(lpos)) + return true; + + /* + * Any descriptor states that have transitioned to reusable due to the + * data tail being pushed to this loaded value will be visible to this + * CPU. This pairs with data_push_tail:D. + * + * Memory barrier involvement: + * + * If data_push_tail:A reads from data_push_tail:D, then this CPU can + * see desc_make_reusable:A. + * + * Relies on: + * + * MB from desc_make_reusable:A to data_push_tail:D + * matches + * READFROM from data_push_tail:D to data_push_tail:A + * thus + * READFROM from desc_make_reusable:A to this CPU + */ + tail_lpos = atomic_long_read(&data_ring->tail_lpos); /* LMM(data_push_tail:A) */ + + /* + * Loop until the tail lpos is at or beyond @lpos. This condition + * may already be satisfied, resulting in no full memory barrier + * from data_push_tail:D being performed. However, since this CPU + * sees the new tail lpos, any descriptor states that transitioned to + * the reusable state must already be visible. + */ + while ((lpos - tail_lpos) - 1 < DATA_SIZE(data_ring)) { + /* + * Make all descriptors reusable that are associated with + * data blocks before @lpos. + */ + if (!data_make_reusable(rb, data_ring, tail_lpos, lpos, + &next_lpos)) { + /* + * 1. Guarantee the block ID loaded in + * data_make_reusable() is performed before + * reloading the tail lpos. The failed + * data_make_reusable() may be due to a newly + * recycled data area causing the tail lpos to + * have been previously pushed. This pairs with + * data_alloc:A and data_realloc:A. + * + * Memory barrier involvement: + * + * If data_make_reusable:A reads from data_alloc:B, + * then data_push_tail:C reads from + * data_push_tail:D. + * + * Relies on: + * + * MB from data_push_tail:D to data_alloc:B + * matching + * RMB from data_make_reusable:A to + * data_push_tail:C + * + * Note: data_push_tail:D and data_alloc:B can be + * different CPUs. However, the data_alloc:B + * CPU (which performs the full memory + * barrier) must have previously seen + * data_push_tail:D. + * + * 2. Guarantee the descriptor state loaded in + * data_make_reusable() is performed before + * reloading the tail lpos. The failed + * data_make_reusable() may be due to a newly + * recycled descriptor causing the tail lpos to + * have been previously pushed. This pairs with + * desc_reserve:D. + * + * Memory barrier involvement: + * + * If data_make_reusable:B reads from + * desc_reserve:F, then data_push_tail:C reads + * from data_push_tail:D. + * + * Relies on: + * + * MB from data_push_tail:D to desc_reserve:F + * matching + * RMB from data_make_reusable:B to + * data_push_tail:C + * + * Note: data_push_tail:D and desc_reserve:F can + * be different CPUs. However, the + * desc_reserve:F CPU (which performs the + * full memory barrier) must have previously + * seen data_push_tail:D. + */ + smp_rmb(); /* LMM(data_push_tail:B) */ + + tail_lpos_new = atomic_long_read(&data_ring->tail_lpos + ); /* LMM(data_push_tail:C) */ + if (tail_lpos_new == tail_lpos) + return false; + + /* Another CPU pushed the tail. Try again. */ + tail_lpos = tail_lpos_new; + continue; + } + + /* + * Guarantee any descriptor states that have transitioned to + * reusable are stored before pushing the tail lpos. A full + * memory barrier is needed since other CPUs may have made + * the descriptor states reusable. This pairs with + * data_push_tail:A. + */ + if (atomic_long_try_cmpxchg(&data_ring->tail_lpos, &tail_lpos, + next_lpos)) { /* LMM(data_push_tail:D) */ + break; + } + } + + return true; +} + +/* + * Advance the desc ring tail. This function advances the tail by one + * descriptor, thus invalidating the oldest descriptor. Before advancing + * the tail, the tail descriptor is made reusable and all data blocks up to + * and including the descriptor's data block are invalidated (i.e. the data + * ring tail is pushed past the data block of the descriptor being made + * reusable). + */ +static bool desc_push_tail(struct printk_ringbuffer *rb, + unsigned long tail_id) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + enum desc_state d_state; + struct prb_desc desc; + + d_state = desc_read(desc_ring, tail_id, &desc, NULL, NULL); + + switch (d_state) { + case desc_miss: + /* + * If the ID is exactly 1 wrap behind the expected, it is + * in the process of being reserved by another writer and + * must be considered reserved. + */ + if (DESC_ID(atomic_long_read(&desc.state_var)) == + DESC_ID_PREV_WRAP(desc_ring, tail_id)) { + return false; + } + + /* + * The ID has changed. Another writer must have pushed the + * tail and recycled the descriptor already. Success is + * returned because the caller is only interested in the + * specified tail being pushed, which it was. + */ + return true; + case desc_reserved: + case desc_committed: + return false; + case desc_finalized: + desc_make_reusable(desc_ring, tail_id); + break; + case desc_reusable: + break; + } + + /* + * Data blocks must be invalidated before their associated + * descriptor can be made available for recycling. Invalidating + * them later is not possible because there is no way to trust + * data blocks once their associated descriptor is gone. + */ + + if (!data_push_tail(rb, &rb->text_data_ring, desc.text_blk_lpos.next)) + return false; + + /* + * Check the next descriptor after @tail_id before pushing the tail + * to it because the tail must always be in a finalized or reusable + * state. The implementation of prb_first_seq() relies on this. + * + * A successful read implies that the next descriptor is less than or + * equal to @head_id so there is no risk of pushing the tail past the + * head. + */ + d_state = desc_read(desc_ring, DESC_ID(tail_id + 1), &desc, + NULL, NULL); /* LMM(desc_push_tail:A) */ + + if (d_state == desc_finalized || d_state == desc_reusable) { + /* + * Guarantee any descriptor states that have transitioned to + * reusable are stored before pushing the tail ID. This allows + * verifying the recycled descriptor state. A full memory + * barrier is needed since other CPUs may have made the + * descriptor states reusable. This pairs with desc_reserve:D. + */ + atomic_long_cmpxchg(&desc_ring->tail_id, tail_id, + DESC_ID(tail_id + 1)); /* LMM(desc_push_tail:B) */ + } else { + /* + * Guarantee the last state load from desc_read() is before + * reloading @tail_id in order to see a new tail ID in the + * case that the descriptor has been recycled. This pairs + * with desc_reserve:D. + * + * Memory barrier involvement: + * + * If desc_push_tail:A reads from desc_reserve:F, then + * desc_push_tail:D reads from desc_push_tail:B. + * + * Relies on: + * + * MB from desc_push_tail:B to desc_reserve:F + * matching + * RMB from desc_push_tail:A to desc_push_tail:D + * + * Note: desc_push_tail:B and desc_reserve:F can be different + * CPUs. However, the desc_reserve:F CPU (which performs + * the full memory barrier) must have previously seen + * desc_push_tail:B. + */ + smp_rmb(); /* LMM(desc_push_tail:C) */ + + /* + * Re-check the tail ID. The descriptor following @tail_id is + * not in an allowed tail state. But if the tail has since + * been moved by another CPU, then it does not matter. + */ + if (atomic_long_read(&desc_ring->tail_id) == tail_id) /* LMM(desc_push_tail:D) */ + return false; + } + + return true; +} + +/* Reserve a new descriptor, invalidating the oldest if necessary. */ +static bool desc_reserve(struct printk_ringbuffer *rb, unsigned long *id_out) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + unsigned long prev_state_val; + unsigned long id_prev_wrap; + struct prb_desc *desc; + unsigned long head_id; + unsigned long id; + + head_id = atomic_long_read(&desc_ring->head_id); /* LMM(desc_reserve:A) */ + + do { + desc = to_desc(desc_ring, head_id); + + id = DESC_ID(head_id + 1); + id_prev_wrap = DESC_ID_PREV_WRAP(desc_ring, id); + + /* + * Guarantee the head ID is read before reading the tail ID. + * Since the tail ID is updated before the head ID, this + * guarantees that @id_prev_wrap is never ahead of the tail + * ID. This pairs with desc_reserve:D. + * + * Memory barrier involvement: + * + * If desc_reserve:A reads from desc_reserve:D, then + * desc_reserve:C reads from desc_push_tail:B. + * + * Relies on: + * + * MB from desc_push_tail:B to desc_reserve:D + * matching + * RMB from desc_reserve:A to desc_reserve:C + * + * Note: desc_push_tail:B and desc_reserve:D can be different + * CPUs. However, the desc_reserve:D CPU (which performs + * the full memory barrier) must have previously seen + * desc_push_tail:B. + */ + smp_rmb(); /* LMM(desc_reserve:B) */ + + if (id_prev_wrap == atomic_long_read(&desc_ring->tail_id + )) { /* LMM(desc_reserve:C) */ + /* + * Make space for the new descriptor by + * advancing the tail. + */ + if (!desc_push_tail(rb, id_prev_wrap)) + return false; + } + + /* + * 1. Guarantee the tail ID is read before validating the + * recycled descriptor state. A read memory barrier is + * sufficient for this. This pairs with desc_push_tail:B. + * + * Memory barrier involvement: + * + * If desc_reserve:C reads from desc_push_tail:B, then + * desc_reserve:E reads from desc_make_reusable:A. + * + * Relies on: + * + * MB from desc_make_reusable:A to desc_push_tail:B + * matching + * RMB from desc_reserve:C to desc_reserve:E + * + * Note: desc_make_reusable:A and desc_push_tail:B can be + * different CPUs. However, the desc_push_tail:B CPU + * (which performs the full memory barrier) must have + * previously seen desc_make_reusable:A. + * + * 2. Guarantee the tail ID is stored before storing the head + * ID. This pairs with desc_reserve:B. + * + * 3. Guarantee any data ring tail changes are stored before + * recycling the descriptor. Data ring tail changes can + * happen via desc_push_tail()->data_push_tail(). A full + * memory barrier is needed since another CPU may have + * pushed the data ring tails. This pairs with + * data_push_tail:B. + * + * 4. Guarantee a new tail ID is stored before recycling the + * descriptor. A full memory barrier is needed since + * another CPU may have pushed the tail ID. This pairs + * with desc_push_tail:C and this also pairs with + * prb_first_seq:C. + * + * 5. Guarantee the head ID is stored before trying to + * finalize the previous descriptor. This pairs with + * _prb_commit:B. + */ + } while (!atomic_long_try_cmpxchg(&desc_ring->head_id, &head_id, + id)); /* LMM(desc_reserve:D) */ + + desc = to_desc(desc_ring, id); + + /* + * If the descriptor has been recycled, verify the old state val. + * See "ABA Issues" about why this verification is performed. + */ + prev_state_val = atomic_long_read(&desc->state_var); /* LMM(desc_reserve:E) */ + if (prev_state_val && + get_desc_state(id_prev_wrap, prev_state_val) != desc_reusable) { + WARN_ON_ONCE(1); + return false; + } + + /* + * Assign the descriptor a new ID and set its state to reserved. + * See "ABA Issues" about why cmpxchg() instead of set() is used. + * + * Guarantee the new descriptor ID and state is stored before making + * any other changes. A write memory barrier is sufficient for this. + * This pairs with desc_read:D. + */ + if (!atomic_long_try_cmpxchg(&desc->state_var, &prev_state_val, + DESC_SV(id, desc_reserved))) { /* LMM(desc_reserve:F) */ + WARN_ON_ONCE(1); + return false; + } + + /* Now data in @desc can be modified: LMM(desc_reserve:G) */ + + *id_out = id; + return true; +} + +/* Determine the end of a data block. */ +static unsigned long get_next_lpos(struct prb_data_ring *data_ring, + unsigned long lpos, unsigned int size) +{ + unsigned long begin_lpos; + unsigned long next_lpos; + + begin_lpos = lpos; + next_lpos = lpos + size; + + /* First check if the data block does not wrap. */ + if (DATA_WRAPS(data_ring, begin_lpos) == DATA_WRAPS(data_ring, next_lpos)) + return next_lpos; + + /* Wrapping data blocks store their data at the beginning. */ + return (DATA_THIS_WRAP_START_LPOS(data_ring, next_lpos) + size); +} + +/* + * Allocate a new data block, invalidating the oldest data block(s) + * if necessary. This function also associates the data block with + * a specified descriptor. + */ +static char *data_alloc(struct printk_ringbuffer *rb, + struct prb_data_ring *data_ring, unsigned int size, + struct prb_data_blk_lpos *blk_lpos, unsigned long id) +{ + struct prb_data_block *blk; + unsigned long begin_lpos; + unsigned long next_lpos; + + if (size == 0) { + /* Specify a data-less block. */ + blk_lpos->begin = NO_LPOS; + blk_lpos->next = NO_LPOS; + return NULL; + } + + size = to_blk_size(size); + + begin_lpos = atomic_long_read(&data_ring->head_lpos); + + do { + next_lpos = get_next_lpos(data_ring, begin_lpos, size); + + if (!data_push_tail(rb, data_ring, next_lpos - DATA_SIZE(data_ring))) { + /* Failed to allocate, specify a data-less block. */ + blk_lpos->begin = FAILED_LPOS; + blk_lpos->next = FAILED_LPOS; + return NULL; + } + + /* + * 1. Guarantee any descriptor states that have transitioned + * to reusable are stored before modifying the newly + * allocated data area. A full memory barrier is needed + * since other CPUs may have made the descriptor states + * reusable. See data_push_tail:A about why the reusable + * states are visible. This pairs with desc_read:D. + * + * 2. Guarantee any updated tail lpos is stored before + * modifying the newly allocated data area. Another CPU may + * be in data_make_reusable() and is reading a block ID + * from this area. data_make_reusable() can handle reading + * a garbage block ID value, but then it must be able to + * load a new tail lpos. A full memory barrier is needed + * since other CPUs may have updated the tail lpos. This + * pairs with data_push_tail:B. + */ + } while (!atomic_long_try_cmpxchg(&data_ring->head_lpos, &begin_lpos, + next_lpos)); /* LMM(data_alloc:A) */ + + blk = to_block(data_ring, begin_lpos); + blk->id = id; /* LMM(data_alloc:B) */ + + if (DATA_WRAPS(data_ring, begin_lpos) != DATA_WRAPS(data_ring, next_lpos)) { + /* Wrapping data blocks store their data at the beginning. */ + blk = to_block(data_ring, 0); + + /* + * Store the ID on the wrapped block for consistency. + * The printk_ringbuffer does not actually use it. + */ + blk->id = id; + } + + blk_lpos->begin = begin_lpos; + blk_lpos->next = next_lpos; + + return &blk->data[0]; +} + +/* + * Try to resize an existing data block associated with the descriptor + * specified by @id. If the resized data block should become wrapped, it + * copies the old data to the new data block. If @size yields a data block + * with the same or less size, the data block is left as is. + * + * Fail if this is not the last allocated data block or if there is not + * enough space or it is not possible make enough space. + * + * Return a pointer to the beginning of the entire data buffer or NULL on + * failure. + */ +static char *data_realloc(struct printk_ringbuffer *rb, + struct prb_data_ring *data_ring, unsigned int size, + struct prb_data_blk_lpos *blk_lpos, unsigned long id) +{ + struct prb_data_block *blk; + unsigned long head_lpos; + unsigned long next_lpos; + bool wrapped; + + /* Reallocation only works if @blk_lpos is the newest data block. */ + head_lpos = atomic_long_read(&data_ring->head_lpos); + if (head_lpos != blk_lpos->next) + return NULL; + + /* Keep track if @blk_lpos was a wrapping data block. */ + wrapped = (DATA_WRAPS(data_ring, blk_lpos->begin) != DATA_WRAPS(data_ring, blk_lpos->next)); + + size = to_blk_size(size); + + next_lpos = get_next_lpos(data_ring, blk_lpos->begin, size); + + /* If the data block does not increase, there is nothing to do. */ + if (head_lpos - next_lpos < DATA_SIZE(data_ring)) { + if (wrapped) + blk = to_block(data_ring, 0); + else + blk = to_block(data_ring, blk_lpos->begin); + return &blk->data[0]; + } + + if (!data_push_tail(rb, data_ring, next_lpos - DATA_SIZE(data_ring))) + return NULL; + + /* The memory barrier involvement is the same as data_alloc:A. */ + if (!atomic_long_try_cmpxchg(&data_ring->head_lpos, &head_lpos, + next_lpos)) { /* LMM(data_realloc:A) */ + return NULL; + } + + blk = to_block(data_ring, blk_lpos->begin); + + if (DATA_WRAPS(data_ring, blk_lpos->begin) != DATA_WRAPS(data_ring, next_lpos)) { + struct prb_data_block *old_blk = blk; + + /* Wrapping data blocks store their data at the beginning. */ + blk = to_block(data_ring, 0); + + /* + * Store the ID on the wrapped block for consistency. + * The printk_ringbuffer does not actually use it. + */ + blk->id = id; + + if (!wrapped) { + /* + * Since the allocated space is now in the newly + * created wrapping data block, copy the content + * from the old data block. + */ + memcpy(&blk->data[0], &old_blk->data[0], + (blk_lpos->next - blk_lpos->begin) - sizeof(blk->id)); + } + } + + blk_lpos->next = next_lpos; + + return &blk->data[0]; +} + +/* Return the number of bytes used by a data block. */ +static unsigned int space_used(struct prb_data_ring *data_ring, + struct prb_data_blk_lpos *blk_lpos) +{ + /* Data-less blocks take no space. */ + if (BLK_DATALESS(blk_lpos)) + return 0; + + if (DATA_WRAPS(data_ring, blk_lpos->begin) == DATA_WRAPS(data_ring, blk_lpos->next)) { + /* Data block does not wrap. */ + return (DATA_INDEX(data_ring, blk_lpos->next) - + DATA_INDEX(data_ring, blk_lpos->begin)); + } + + /* + * For wrapping data blocks, the trailing (wasted) space is + * also counted. + */ + return (DATA_INDEX(data_ring, blk_lpos->next) + + DATA_SIZE(data_ring) - DATA_INDEX(data_ring, blk_lpos->begin)); +} + +/* + * Given @blk_lpos, return a pointer to the writer data from the data block + * and calculate the size of the data part. A NULL pointer is returned if + * @blk_lpos specifies values that could never be legal. + * + * This function (used by readers) performs strict validation on the lpos + * values to possibly detect bugs in the writer code. A WARN_ON_ONCE() is + * triggered if an internal error is detected. + */ +static const char *get_data(struct prb_data_ring *data_ring, + struct prb_data_blk_lpos *blk_lpos, + unsigned int *data_size) +{ + struct prb_data_block *db; + + /* Data-less data block description. */ + if (BLK_DATALESS(blk_lpos)) { + if (blk_lpos->begin == NO_LPOS && blk_lpos->next == NO_LPOS) { + *data_size = 0; + return ""; + } + return NULL; + } + + /* Regular data block: @begin less than @next and in same wrap. */ + if (DATA_WRAPS(data_ring, blk_lpos->begin) == DATA_WRAPS(data_ring, blk_lpos->next) && + blk_lpos->begin < blk_lpos->next) { + db = to_block(data_ring, blk_lpos->begin); + *data_size = blk_lpos->next - blk_lpos->begin; + + /* Wrapping data block: @begin is one wrap behind @next. */ + } else if (DATA_WRAPS(data_ring, blk_lpos->begin + DATA_SIZE(data_ring)) == + DATA_WRAPS(data_ring, blk_lpos->next)) { + db = to_block(data_ring, 0); + *data_size = DATA_INDEX(data_ring, blk_lpos->next); + + /* Illegal block description. */ + } else { + WARN_ON_ONCE(1); + return NULL; + } + + /* A valid data block will always be aligned to the ID size. */ + if (WARN_ON_ONCE(blk_lpos->begin != ALIGN(blk_lpos->begin, sizeof(db->id))) || + WARN_ON_ONCE(blk_lpos->next != ALIGN(blk_lpos->next, sizeof(db->id)))) { + return NULL; + } + + /* A valid data block will always have at least an ID. */ + if (WARN_ON_ONCE(*data_size < sizeof(db->id))) + return NULL; + + /* Subtract block ID space from size to reflect data size. */ + *data_size -= sizeof(db->id); + + return &db->data[0]; +} + +/* + * Attempt to transition the newest descriptor from committed back to reserved + * so that the record can be modified by a writer again. This is only possible + * if the descriptor is not yet finalized and the provided @caller_id matches. + */ +static struct prb_desc *desc_reopen_last(struct prb_desc_ring *desc_ring, + u32 caller_id, unsigned long *id_out) +{ + unsigned long prev_state_val; + enum desc_state d_state; + struct prb_desc desc; + struct prb_desc *d; + unsigned long id; + u32 cid; + + id = atomic_long_read(&desc_ring->head_id); + + /* + * To reduce unnecessarily reopening, first check if the descriptor + * state and caller ID are correct. + */ + d_state = desc_read(desc_ring, id, &desc, NULL, &cid); + if (d_state != desc_committed || cid != caller_id) + return NULL; + + d = to_desc(desc_ring, id); + + prev_state_val = DESC_SV(id, desc_committed); + + /* + * Guarantee the reserved state is stored before reading any + * record data. A full memory barrier is needed because @state_var + * modification is followed by reading. This pairs with _prb_commit:B. + * + * Memory barrier involvement: + * + * If desc_reopen_last:A reads from _prb_commit:B, then + * prb_reserve_in_last:A reads from _prb_commit:A. + * + * Relies on: + * + * WMB from _prb_commit:A to _prb_commit:B + * matching + * MB If desc_reopen_last:A to prb_reserve_in_last:A + */ + if (!atomic_long_try_cmpxchg(&d->state_var, &prev_state_val, + DESC_SV(id, desc_reserved))) { /* LMM(desc_reopen_last:A) */ + return NULL; + } + + *id_out = id; + return d; +} + +/** + * prb_reserve_in_last() - Re-reserve and extend the space in the ringbuffer + * used by the newest record. + * + * @e: The entry structure to setup. + * @rb: The ringbuffer to re-reserve and extend data in. + * @r: The record structure to allocate buffers for. + * @caller_id: The caller ID of the caller (reserving writer). + * @max_size: Fail if the extended size would be greater than this. + * + * This is the public function available to writers to re-reserve and extend + * data. + * + * The writer specifies the text size to extend (not the new total size) by + * setting the @text_buf_size field of @r. To ensure proper initialization + * of @r, prb_rec_init_wr() should be used. + * + * This function will fail if @caller_id does not match the caller ID of the + * newest record. In that case the caller must reserve new data using + * prb_reserve(). + * + * Context: Any context. Disables local interrupts on success. + * Return: true if text data could be extended, otherwise false. + * + * On success: + * + * - @r->text_buf points to the beginning of the entire text buffer. + * + * - @r->text_buf_size is set to the new total size of the buffer. + * + * - @r->info is not touched so that @r->info->text_len could be used + * to append the text. + * + * - prb_record_text_space() can be used on @e to query the new + * actually used space. + * + * Important: All @r->info fields will already be set with the current values + * for the record. I.e. @r->info->text_len will be less than + * @text_buf_size. Writers can use @r->info->text_len to know + * where concatenation begins and writers should update + * @r->info->text_len after concatenating. + */ +bool prb_reserve_in_last(struct prb_reserved_entry *e, struct printk_ringbuffer *rb, + struct printk_record *r, u32 caller_id, unsigned int max_size) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + struct printk_info *info; + unsigned int data_size; + struct prb_desc *d; + unsigned long id; + + local_irq_save(e->irqflags); + + /* Transition the newest descriptor back to the reserved state. */ + d = desc_reopen_last(desc_ring, caller_id, &id); + if (!d) { + local_irq_restore(e->irqflags); + goto fail_reopen; + } + + /* Now the writer has exclusive access: LMM(prb_reserve_in_last:A) */ + + info = to_info(desc_ring, id); + + /* + * Set the @e fields here so that prb_commit() can be used if + * anything fails from now on. + */ + e->rb = rb; + e->id = id; + + /* + * desc_reopen_last() checked the caller_id, but there was no + * exclusive access at that point. The descriptor may have + * changed since then. + */ + if (caller_id != info->caller_id) + goto fail; + + if (BLK_DATALESS(&d->text_blk_lpos)) { + if (WARN_ON_ONCE(info->text_len != 0)) { + pr_warn_once("wrong text_len value (%hu, expecting 0)\n", + info->text_len); + info->text_len = 0; + } + + if (!data_check_size(&rb->text_data_ring, r->text_buf_size)) + goto fail; + + if (r->text_buf_size > max_size) + goto fail; + + r->text_buf = data_alloc(rb, &rb->text_data_ring, r->text_buf_size, + &d->text_blk_lpos, id); + } else { + if (!get_data(&rb->text_data_ring, &d->text_blk_lpos, &data_size)) + goto fail; + + /* + * Increase the buffer size to include the original size. If + * the meta data (@text_len) is not sane, use the full data + * block size. + */ + if (WARN_ON_ONCE(info->text_len > data_size)) { + pr_warn_once("wrong text_len value (%hu, expecting <=%u)\n", + info->text_len, data_size); + info->text_len = data_size; + } + r->text_buf_size += info->text_len; + + if (!data_check_size(&rb->text_data_ring, r->text_buf_size)) + goto fail; + + if (r->text_buf_size > max_size) + goto fail; + + r->text_buf = data_realloc(rb, &rb->text_data_ring, r->text_buf_size, + &d->text_blk_lpos, id); + } + if (r->text_buf_size && !r->text_buf) + goto fail; + + r->info = info; + + e->text_space = space_used(&rb->text_data_ring, &d->text_blk_lpos); + + return true; +fail: + prb_commit(e); + /* prb_commit() re-enabled interrupts. */ +fail_reopen: + /* Make it clear to the caller that the re-reserve failed. */ + memset(r, 0, sizeof(*r)); + return false; +} + +/* + * Attempt to finalize a specified descriptor. If this fails, the descriptor + * is either already final or it will finalize itself when the writer commits. + */ +static void desc_make_final(struct prb_desc_ring *desc_ring, unsigned long id) +{ + unsigned long prev_state_val = DESC_SV(id, desc_committed); + struct prb_desc *d = to_desc(desc_ring, id); + + atomic_long_cmpxchg_relaxed(&d->state_var, prev_state_val, + DESC_SV(id, desc_finalized)); /* LMM(desc_make_final:A) */ +} + +/** + * prb_reserve() - Reserve space in the ringbuffer. + * + * @e: The entry structure to setup. + * @rb: The ringbuffer to reserve data in. + * @r: The record structure to allocate buffers for. + * + * This is the public function available to writers to reserve data. + * + * The writer specifies the text size to reserve by setting the + * @text_buf_size field of @r. To ensure proper initialization of @r, + * prb_rec_init_wr() should be used. + * + * Context: Any context. Disables local interrupts on success. + * Return: true if at least text data could be allocated, otherwise false. + * + * On success, the fields @info and @text_buf of @r will be set by this + * function and should be filled in by the writer before committing. Also + * on success, prb_record_text_space() can be used on @e to query the actual + * space used for the text data block. + * + * Important: @info->text_len needs to be set correctly by the writer in + * order for data to be readable and/or extended. Its value + * is initialized to 0. + */ +bool prb_reserve(struct prb_reserved_entry *e, struct printk_ringbuffer *rb, + struct printk_record *r) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + struct printk_info *info; + struct prb_desc *d; + unsigned long id; + u64 seq; + + if (!data_check_size(&rb->text_data_ring, r->text_buf_size)) + goto fail; + + /* + * Descriptors in the reserved state act as blockers to all further + * reservations once the desc_ring has fully wrapped. Disable + * interrupts during the reserve/commit window in order to minimize + * the likelihood of this happening. + */ + local_irq_save(e->irqflags); + + if (!desc_reserve(rb, &id)) { + /* Descriptor reservation failures are tracked. */ + atomic_long_inc(&rb->fail); + local_irq_restore(e->irqflags); + goto fail; + } + + d = to_desc(desc_ring, id); + info = to_info(desc_ring, id); + + /* + * All @info fields (except @seq) are cleared and must be filled in + * by the writer. Save @seq before clearing because it is used to + * determine the new sequence number. + */ + seq = info->seq; + memset(info, 0, sizeof(*info)); + + /* + * Set the @e fields here so that prb_commit() can be used if + * text data allocation fails. + */ + e->rb = rb; + e->id = id; + + /* + * Initialize the sequence number if it has "never been set". + * Otherwise just increment it by a full wrap. + * + * @seq is considered "never been set" if it has a value of 0, + * _except_ for @infos[0], which was specially setup by the ringbuffer + * initializer and therefore is always considered as set. + * + * See the "Bootstrap" comment block in printk_ringbuffer.h for + * details about how the initializer bootstraps the descriptors. + */ + if (seq == 0 && DESC_INDEX(desc_ring, id) != 0) + info->seq = DESC_INDEX(desc_ring, id); + else + info->seq = seq + DESCS_COUNT(desc_ring); + + /* + * New data is about to be reserved. Once that happens, previous + * descriptors are no longer able to be extended. Finalize the + * previous descriptor now so that it can be made available to + * readers. (For seq==0 there is no previous descriptor.) + */ + if (info->seq > 0) + desc_make_final(desc_ring, DESC_ID(id - 1)); + + r->text_buf = data_alloc(rb, &rb->text_data_ring, r->text_buf_size, + &d->text_blk_lpos, id); + /* If text data allocation fails, a data-less record is committed. */ + if (r->text_buf_size && !r->text_buf) { + prb_commit(e); + /* prb_commit() re-enabled interrupts. */ + goto fail; + } + + r->info = info; + + /* Record full text space used by record. */ + e->text_space = space_used(&rb->text_data_ring, &d->text_blk_lpos); + + return true; +fail: + /* Make it clear to the caller that the reserve failed. */ + memset(r, 0, sizeof(*r)); + return false; +} + +/* Commit the data (possibly finalizing it) and restore interrupts. */ +static void _prb_commit(struct prb_reserved_entry *e, unsigned long state_val) +{ + struct prb_desc_ring *desc_ring = &e->rb->desc_ring; + struct prb_desc *d = to_desc(desc_ring, e->id); + unsigned long prev_state_val = DESC_SV(e->id, desc_reserved); + + /* Now the writer has finished all writing: LMM(_prb_commit:A) */ + + /* + * Set the descriptor as committed. See "ABA Issues" about why + * cmpxchg() instead of set() is used. + * + * 1 Guarantee all record data is stored before the descriptor state + * is stored as committed. A write memory barrier is sufficient + * for this. This pairs with desc_read:B and desc_reopen_last:A. + * + * 2. Guarantee the descriptor state is stored as committed before + * re-checking the head ID in order to possibly finalize this + * descriptor. This pairs with desc_reserve:D. + * + * Memory barrier involvement: + * + * If prb_commit:A reads from desc_reserve:D, then + * desc_make_final:A reads from _prb_commit:B. + * + * Relies on: + * + * MB _prb_commit:B to prb_commit:A + * matching + * MB desc_reserve:D to desc_make_final:A + */ + if (!atomic_long_try_cmpxchg(&d->state_var, &prev_state_val, + DESC_SV(e->id, state_val))) { /* LMM(_prb_commit:B) */ + WARN_ON_ONCE(1); + } + + /* Restore interrupts, the reserve/commit window is finished. */ + local_irq_restore(e->irqflags); +} + +/** + * prb_commit() - Commit (previously reserved) data to the ringbuffer. + * + * @e: The entry containing the reserved data information. + * + * This is the public function available to writers to commit data. + * + * Note that the data is not yet available to readers until it is finalized. + * Finalizing happens automatically when space for the next record is + * reserved. + * + * See prb_final_commit() for a version of this function that finalizes + * immediately. + * + * Context: Any context. Enables local interrupts. + */ +void prb_commit(struct prb_reserved_entry *e) +{ + struct prb_desc_ring *desc_ring = &e->rb->desc_ring; + unsigned long head_id; + + _prb_commit(e, desc_committed); + + /* + * If this descriptor is no longer the head (i.e. a new record has + * been allocated), extending the data for this record is no longer + * allowed and therefore it must be finalized. + */ + head_id = atomic_long_read(&desc_ring->head_id); /* LMM(prb_commit:A) */ + if (head_id != e->id) + desc_make_final(desc_ring, e->id); +} + +/** + * prb_final_commit() - Commit and finalize (previously reserved) data to + * the ringbuffer. + * + * @e: The entry containing the reserved data information. + * + * This is the public function available to writers to commit+finalize data. + * + * By finalizing, the data is made immediately available to readers. + * + * This function should only be used if there are no intentions of extending + * this data using prb_reserve_in_last(). + * + * Context: Any context. Enables local interrupts. + */ +void prb_final_commit(struct prb_reserved_entry *e) +{ + _prb_commit(e, desc_finalized); +} + +/* + * Count the number of lines in provided text. All text has at least 1 line + * (even if @text_size is 0). Each '\n' processed is counted as an additional + * line. + */ +static unsigned int count_lines(const char *text, unsigned int text_size) +{ + unsigned int next_size = text_size; + unsigned int line_count = 1; + const char *next = text; + + while (next_size) { + next = memchr(next, '\n', next_size); + if (!next) + break; + line_count++; + next++; + next_size = text_size - (next - text); + } + + return line_count; +} + +/* + * Given @blk_lpos, copy an expected @len of data into the provided buffer. + * If @line_count is provided, count the number of lines in the data. + * + * This function (used by readers) performs strict validation on the data + * size to possibly detect bugs in the writer code. A WARN_ON_ONCE() is + * triggered if an internal error is detected. + */ +static bool copy_data(struct prb_data_ring *data_ring, + struct prb_data_blk_lpos *blk_lpos, u16 len, char *buf, + unsigned int buf_size, unsigned int *line_count) +{ + unsigned int data_size; + const char *data; + + /* Caller might not want any data. */ + if ((!buf || !buf_size) && !line_count) + return true; + + data = get_data(data_ring, blk_lpos, &data_size); + if (!data) + return false; + + /* + * Actual cannot be less than expected. It can be more than expected + * because of the trailing alignment padding. + * + * Note that invalid @len values can occur because the caller loads + * the value during an allowed data race. + */ + if (data_size < (unsigned int)len) + return false; + + /* Caller interested in the line count? */ + if (line_count) + *line_count = count_lines(data, data_size); + + /* Caller interested in the data content? */ + if (!buf || !buf_size) + return true; + + data_size = min_t(u16, buf_size, len); + + memcpy(&buf[0], data, data_size); /* LMM(copy_data:A) */ + return true; +} + +/* + * This is an extended version of desc_read(). It gets a copy of a specified + * descriptor. However, it also verifies that the record is finalized and has + * the sequence number @seq. On success, 0 is returned. + * + * Error return values: + * -EINVAL: A finalized record with sequence number @seq does not exist. + * -ENOENT: A finalized record with sequence number @seq exists, but its data + * is not available. This is a valid record, so readers should + * continue with the next record. + */ +static int desc_read_finalized_seq(struct prb_desc_ring *desc_ring, + unsigned long id, u64 seq, + struct prb_desc *desc_out) +{ + struct prb_data_blk_lpos *blk_lpos = &desc_out->text_blk_lpos; + enum desc_state d_state; + u64 s; + + d_state = desc_read(desc_ring, id, desc_out, &s, NULL); + + /* + * An unexpected @id (desc_miss) or @seq mismatch means the record + * does not exist. A descriptor in the reserved or committed state + * means the record does not yet exist for the reader. + */ + if (d_state == desc_miss || + d_state == desc_reserved || + d_state == desc_committed || + s != seq) { + return -EINVAL; + } + + /* + * A descriptor in the reusable state may no longer have its data + * available; report it as existing but with lost data. Or the record + * may actually be a record with lost data. + */ + if (d_state == desc_reusable || + (blk_lpos->begin == FAILED_LPOS && blk_lpos->next == FAILED_LPOS)) { + return -ENOENT; + } + + return 0; +} + +/* + * Copy the ringbuffer data from the record with @seq to the provided + * @r buffer. On success, 0 is returned. + * + * See desc_read_finalized_seq() for error return values. + */ +static int prb_read(struct printk_ringbuffer *rb, u64 seq, + struct printk_record *r, unsigned int *line_count) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + struct printk_info *info = to_info(desc_ring, seq); + struct prb_desc *rdesc = to_desc(desc_ring, seq); + atomic_long_t *state_var = &rdesc->state_var; + struct prb_desc desc; + unsigned long id; + int err; + + /* Extract the ID, used to specify the descriptor to read. */ + id = DESC_ID(atomic_long_read(state_var)); + + /* Get a local copy of the correct descriptor (if available). */ + err = desc_read_finalized_seq(desc_ring, id, seq, &desc); + + /* + * If @r is NULL, the caller is only interested in the availability + * of the record. + */ + if (err || !r) + return err; + + /* If requested, copy meta data. */ + if (r->info) + memcpy(r->info, info, sizeof(*(r->info))); + + /* Copy text data. If it fails, this is a data-less record. */ + if (!copy_data(&rb->text_data_ring, &desc.text_blk_lpos, info->text_len, + r->text_buf, r->text_buf_size, line_count)) { + return -ENOENT; + } + + /* Ensure the record is still finalized and has the same @seq. */ + return desc_read_finalized_seq(desc_ring, id, seq, &desc); +} + +/* Get the sequence number of the tail descriptor. */ +static u64 prb_first_seq(struct printk_ringbuffer *rb) +{ + struct prb_desc_ring *desc_ring = &rb->desc_ring; + enum desc_state d_state; + struct prb_desc desc; + unsigned long id; + u64 seq; + + for (;;) { + id = atomic_long_read(&rb->desc_ring.tail_id); /* LMM(prb_first_seq:A) */ + + d_state = desc_read(desc_ring, id, &desc, &seq, NULL); /* LMM(prb_first_seq:B) */ + + /* + * This loop will not be infinite because the tail is + * _always_ in the finalized or reusable state. + */ + if (d_state == desc_finalized || d_state == desc_reusable) + break; + + /* + * Guarantee the last state load from desc_read() is before + * reloading @tail_id in order to see a new tail in the case + * that the descriptor has been recycled. This pairs with + * desc_reserve:D. + * + * Memory barrier involvement: + * + * If prb_first_seq:B reads from desc_reserve:F, then + * prb_first_seq:A reads from desc_push_tail:B. + * + * Relies on: + * + * MB from desc_push_tail:B to desc_reserve:F + * matching + * RMB prb_first_seq:B to prb_first_seq:A + */ + smp_rmb(); /* LMM(prb_first_seq:C) */ + } + + return seq; +} + +/* + * Non-blocking read of a record. Updates @seq to the last finalized record + * (which may have no data available). + * + * See the description of prb_read_valid() and prb_read_valid_info() + * for details. + */ +static bool _prb_read_valid(struct printk_ringbuffer *rb, u64 *seq, + struct printk_record *r, unsigned int *line_count) +{ + u64 tail_seq; + int err; + + while ((err = prb_read(rb, *seq, r, line_count))) { + tail_seq = prb_first_seq(rb); + + if (*seq < tail_seq) { + /* + * Behind the tail. Catch up and try again. This + * can happen for -ENOENT and -EINVAL cases. + */ + *seq = tail_seq; + + } else if (err == -ENOENT) { + /* Record exists, but no data available. Skip. */ + (*seq)++; + + } else { + /* Non-existent/non-finalized record. Must stop. */ + return false; + } + } + + return true; +} + +/** + * prb_read_valid() - Non-blocking read of a requested record or (if gone) + * the next available record. + * + * @rb: The ringbuffer to read from. + * @seq: The sequence number of the record to read. + * @r: A record data buffer to store the read record to. + * + * This is the public function available to readers to read a record. + * + * The reader provides the @info and @text_buf buffers of @r to be + * filled in. Any of the buffer pointers can be set to NULL if the reader + * is not interested in that data. To ensure proper initialization of @r, + * prb_rec_init_rd() should be used. + * + * Context: Any context. + * Return: true if a record was read, otherwise false. + * + * On success, the reader must check r->info.seq to see which record was + * actually read. This allows the reader to detect dropped records. + * + * Failure means @seq refers to a not yet written record. + */ +bool prb_read_valid(struct printk_ringbuffer *rb, u64 seq, + struct printk_record *r) +{ + return _prb_read_valid(rb, &seq, r, NULL); +} + +/** + * prb_read_valid_info() - Non-blocking read of meta data for a requested + * record or (if gone) the next available record. + * + * @rb: The ringbuffer to read from. + * @seq: The sequence number of the record to read. + * @info: A buffer to store the read record meta data to. + * @line_count: A buffer to store the number of lines in the record text. + * + * This is the public function available to readers to read only the + * meta data of a record. + * + * The reader provides the @info, @line_count buffers to be filled in. + * Either of the buffer pointers can be set to NULL if the reader is not + * interested in that data. + * + * Context: Any context. + * Return: true if a record's meta data was read, otherwise false. + * + * On success, the reader must check info->seq to see which record meta data + * was actually read. This allows the reader to detect dropped records. + * + * Failure means @seq refers to a not yet written record. + */ +bool prb_read_valid_info(struct printk_ringbuffer *rb, u64 seq, + struct printk_info *info, unsigned int *line_count) +{ + struct printk_record r; + + prb_rec_init_rd(&r, info, NULL, 0); + + return _prb_read_valid(rb, &seq, &r, line_count); +} + +/** + * prb_first_valid_seq() - Get the sequence number of the oldest available + * record. + * + * @rb: The ringbuffer to get the sequence number from. + * + * This is the public function available to readers to see what the + * first/oldest valid sequence number is. + * + * This provides readers a starting point to begin iterating the ringbuffer. + * + * Context: Any context. + * Return: The sequence number of the first/oldest record or, if the + * ringbuffer is empty, 0 is returned. + */ +u64 prb_first_valid_seq(struct printk_ringbuffer *rb) +{ + u64 seq = 0; + + if (!_prb_read_valid(rb, &seq, NULL, NULL)) + return 0; + + return seq; +} + +/** + * prb_next_seq() - Get the sequence number after the last available record. + * + * @rb: The ringbuffer to get the sequence number from. + * + * This is the public function available to readers to see what the next + * newest sequence number available to readers will be. + * + * This provides readers a sequence number to jump to if all currently + * available records should be skipped. + * + * Context: Any context. + * Return: The sequence number of the next newest (not yet available) record + * for readers. + */ +u64 prb_next_seq(struct printk_ringbuffer *rb) +{ + u64 seq = 0; + + /* Search forward from the oldest descriptor. */ + while (_prb_read_valid(rb, &seq, NULL, NULL)) + seq++; + + return seq; +} + +/** + * prb_init() - Initialize a ringbuffer to use provided external buffers. + * + * @rb: The ringbuffer to initialize. + * @text_buf: The data buffer for text data. + * @textbits: The size of @text_buf as a power-of-2 value. + * @descs: The descriptor buffer for ringbuffer records. + * @descbits: The count of @descs items as a power-of-2 value. + * @infos: The printk_info buffer for ringbuffer records. + * + * This is the public function available to writers to setup a ringbuffer + * during runtime using provided buffers. + * + * This must match the initialization of DEFINE_PRINTKRB(). + * + * Context: Any context. + */ +void prb_init(struct printk_ringbuffer *rb, + char *text_buf, unsigned int textbits, + struct prb_desc *descs, unsigned int descbits, + struct printk_info *infos) +{ + memset(descs, 0, _DESCS_COUNT(descbits) * sizeof(descs[0])); + memset(infos, 0, _DESCS_COUNT(descbits) * sizeof(infos[0])); + + rb->desc_ring.count_bits = descbits; + rb->desc_ring.descs = descs; + rb->desc_ring.infos = infos; + atomic_long_set(&rb->desc_ring.head_id, DESC0_ID(descbits)); + atomic_long_set(&rb->desc_ring.tail_id, DESC0_ID(descbits)); + + rb->text_data_ring.size_bits = textbits; + rb->text_data_ring.data = text_buf; + atomic_long_set(&rb->text_data_ring.head_lpos, BLK0_LPOS(textbits)); + atomic_long_set(&rb->text_data_ring.tail_lpos, BLK0_LPOS(textbits)); + + atomic_long_set(&rb->fail, 0); + + atomic_long_set(&(descs[_DESCS_COUNT(descbits) - 1].state_var), DESC0_SV(descbits)); + descs[_DESCS_COUNT(descbits) - 1].text_blk_lpos.begin = FAILED_LPOS; + descs[_DESCS_COUNT(descbits) - 1].text_blk_lpos.next = FAILED_LPOS; + + infos[0].seq = -(u64)_DESCS_COUNT(descbits); + infos[_DESCS_COUNT(descbits) - 1].seq = 0; +} + +/** + * prb_record_text_space() - Query the full actual used ringbuffer space for + * the text data of a reserved entry. + * + * @e: The successfully reserved entry to query. + * + * This is the public function available to writers to see how much actual + * space is used in the ringbuffer to store the text data of the specified + * entry. + * + * This function is only valid if @e has been successfully reserved using + * prb_reserve(). + * + * Context: Any context. + * Return: The size in bytes used by the text data of the associated record. + */ +unsigned int prb_record_text_space(struct prb_reserved_entry *e) +{ + return e->text_space; +} diff --git a/kernel/printk/printk_ringbuffer.h b/kernel/printk/printk_ringbuffer.h new file mode 100644 index 0000000000000..5dc9d022db070 --- /dev/null +++ b/kernel/printk/printk_ringbuffer.h @@ -0,0 +1,382 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _KERNEL_PRINTK_RINGBUFFER_H +#define _KERNEL_PRINTK_RINGBUFFER_H + +#include +#include + +/* + * Meta information about each stored message. + * + * All fields are set by the printk code except for @seq, which is + * set by the ringbuffer code. + */ +struct printk_info { + u64 seq; /* sequence number */ + u64 ts_nsec; /* timestamp in nanoseconds */ + u16 text_len; /* length of text message */ + u8 facility; /* syslog facility */ + u8 flags:5; /* internal record flags */ + u8 level:3; /* syslog level */ + u32 caller_id; /* thread id or processor id */ + + struct dev_printk_info dev_info; +}; + +/* + * A structure providing the buffers, used by writers and readers. + * + * Writers: + * Using prb_rec_init_wr(), a writer sets @text_buf_size before calling + * prb_reserve(). On success, prb_reserve() sets @info and @text_buf to + * buffers reserved for that writer. + * + * Readers: + * Using prb_rec_init_rd(), a reader sets all fields before calling + * prb_read_valid(). Note that the reader provides the @info and @text_buf, + * buffers. On success, the struct pointed to by @info will be filled and + * the char array pointed to by @text_buf will be filled with text data. + */ +struct printk_record { + struct printk_info *info; + char *text_buf; + unsigned int text_buf_size; +}; + +/* Specifies the logical position and span of a data block. */ +struct prb_data_blk_lpos { + unsigned long begin; + unsigned long next; +}; + +/* + * A descriptor: the complete meta-data for a record. + * + * @state_var: A bitwise combination of descriptor ID and descriptor state. + */ +struct prb_desc { + atomic_long_t state_var; + struct prb_data_blk_lpos text_blk_lpos; +}; + +/* A ringbuffer of "ID + data" elements. */ +struct prb_data_ring { + unsigned int size_bits; + char *data; + atomic_long_t head_lpos; + atomic_long_t tail_lpos; +}; + +/* A ringbuffer of "struct prb_desc" elements. */ +struct prb_desc_ring { + unsigned int count_bits; + struct prb_desc *descs; + struct printk_info *infos; + atomic_long_t head_id; + atomic_long_t tail_id; +}; + +/* + * The high level structure representing the printk ringbuffer. + * + * @fail: Count of failed prb_reserve() calls where not even a data-less + * record was created. + */ +struct printk_ringbuffer { + struct prb_desc_ring desc_ring; + struct prb_data_ring text_data_ring; + atomic_long_t fail; +}; + +/* + * Used by writers as a reserve/commit handle. + * + * @rb: Ringbuffer where the entry is reserved. + * @irqflags: Saved irq flags to restore on entry commit. + * @id: ID of the reserved descriptor. + * @text_space: Total occupied buffer space in the text data ring, including + * ID, alignment padding, and wrapping data blocks. + * + * This structure is an opaque handle for writers. Its contents are only + * to be used by the ringbuffer implementation. + */ +struct prb_reserved_entry { + struct printk_ringbuffer *rb; + unsigned long irqflags; + unsigned long id; + unsigned int text_space; +}; + +/* The possible responses of a descriptor state-query. */ +enum desc_state { + desc_miss = -1, /* ID mismatch (pseudo state) */ + desc_reserved = 0x0, /* reserved, in use by writer */ + desc_committed = 0x1, /* committed by writer, could get reopened */ + desc_finalized = 0x2, /* committed, no further modification allowed */ + desc_reusable = 0x3, /* free, not yet used by any writer */ +}; + +#define _DATA_SIZE(sz_bits) (1UL << (sz_bits)) +#define _DESCS_COUNT(ct_bits) (1U << (ct_bits)) +#define DESC_SV_BITS (sizeof(unsigned long) * 8) +#define DESC_FLAGS_SHIFT (DESC_SV_BITS - 2) +#define DESC_FLAGS_MASK (3UL << DESC_FLAGS_SHIFT) +#define DESC_STATE(sv) (3UL & (sv >> DESC_FLAGS_SHIFT)) +#define DESC_SV(id, state) (((unsigned long)state << DESC_FLAGS_SHIFT) | id) +#define DESC_ID_MASK (~DESC_FLAGS_MASK) +#define DESC_ID(sv) ((sv) & DESC_ID_MASK) +#define FAILED_LPOS 0x1 +#define NO_LPOS 0x3 + +#define FAILED_BLK_LPOS \ +{ \ + .begin = FAILED_LPOS, \ + .next = FAILED_LPOS, \ +} + +/* + * Descriptor Bootstrap + * + * The descriptor array is minimally initialized to allow immediate usage + * by readers and writers. The requirements that the descriptor array + * initialization must satisfy: + * + * Req1 + * The tail must point to an existing (committed or reusable) descriptor. + * This is required by the implementation of prb_first_seq(). + * + * Req2 + * Readers must see that the ringbuffer is initially empty. + * + * Req3 + * The first record reserved by a writer is assigned sequence number 0. + * + * To satisfy Req1, the tail initially points to a descriptor that is + * minimally initialized (having no data block, i.e. data-less with the + * data block's lpos @begin and @next values set to FAILED_LPOS). + * + * To satisfy Req2, the initial tail descriptor is initialized to the + * reusable state. Readers recognize reusable descriptors as existing + * records, but skip over them. + * + * To satisfy Req3, the last descriptor in the array is used as the initial + * head (and tail) descriptor. This allows the first record reserved by a + * writer (head + 1) to be the first descriptor in the array. (Only the first + * descriptor in the array could have a valid sequence number of 0.) + * + * The first time a descriptor is reserved, it is assigned a sequence number + * with the value of the array index. A "first time reserved" descriptor can + * be recognized because it has a sequence number of 0 but does not have an + * index of 0. (Only the first descriptor in the array could have a valid + * sequence number of 0.) After the first reservation, all future reservations + * (recycling) simply involve incrementing the sequence number by the array + * count. + * + * Hack #1 + * Only the first descriptor in the array is allowed to have the sequence + * number 0. In this case it is not possible to recognize if it is being + * reserved the first time (set to index value) or has been reserved + * previously (increment by the array count). This is handled by _always_ + * incrementing the sequence number by the array count when reserving the + * first descriptor in the array. In order to satisfy Req3, the sequence + * number of the first descriptor in the array is initialized to minus + * the array count. Then, upon the first reservation, it is incremented + * to 0, thus satisfying Req3. + * + * Hack #2 + * prb_first_seq() can be called at any time by readers to retrieve the + * sequence number of the tail descriptor. However, due to Req2 and Req3, + * initially there are no records to report the sequence number of + * (sequence numbers are u64 and there is nothing less than 0). To handle + * this, the sequence number of the initial tail descriptor is initialized + * to 0. Technically this is incorrect, because there is no record with + * sequence number 0 (yet) and the tail descriptor is not the first + * descriptor in the array. But it allows prb_read_valid() to correctly + * report the existence of a record for _any_ given sequence number at all + * times. Bootstrapping is complete when the tail is pushed the first + * time, thus finally pointing to the first descriptor reserved by a + * writer, which has the assigned sequence number 0. + */ + +/* + * Initiating Logical Value Overflows + * + * Both logical position (lpos) and ID values can be mapped to array indexes + * but may experience overflows during the lifetime of the system. To ensure + * that printk_ringbuffer can handle the overflows for these types, initial + * values are chosen that map to the correct initial array indexes, but will + * result in overflows soon. + * + * BLK0_LPOS + * The initial @head_lpos and @tail_lpos for data rings. It is at index + * 0 and the lpos value is such that it will overflow on the first wrap. + * + * DESC0_ID + * The initial @head_id and @tail_id for the desc ring. It is at the last + * index of the descriptor array (see Req3 above) and the ID value is such + * that it will overflow on the second wrap. + */ +#define BLK0_LPOS(sz_bits) (-(_DATA_SIZE(sz_bits))) +#define DESC0_ID(ct_bits) DESC_ID(-(_DESCS_COUNT(ct_bits) + 1)) +#define DESC0_SV(ct_bits) DESC_SV(DESC0_ID(ct_bits), desc_reusable) + +/* + * Define a ringbuffer with an external text data buffer. The same as + * DEFINE_PRINTKRB() but requires specifying an external buffer for the + * text data. + * + * Note: The specified external buffer must be of the size: + * 2 ^ (descbits + avgtextbits) + */ +#define _DEFINE_PRINTKRB(name, descbits, avgtextbits, text_buf) \ +static struct prb_desc _##name##_descs[_DESCS_COUNT(descbits)] = { \ + /* the initial head and tail */ \ + [_DESCS_COUNT(descbits) - 1] = { \ + /* reusable */ \ + .state_var = ATOMIC_INIT(DESC0_SV(descbits)), \ + /* no associated data block */ \ + .text_blk_lpos = FAILED_BLK_LPOS, \ + }, \ +}; \ +static struct printk_info _##name##_infos[_DESCS_COUNT(descbits)] = { \ + /* this will be the first record reserved by a writer */ \ + [0] = { \ + /* will be incremented to 0 on the first reservation */ \ + .seq = -(u64)_DESCS_COUNT(descbits), \ + }, \ + /* the initial head and tail */ \ + [_DESCS_COUNT(descbits) - 1] = { \ + /* reports the first seq value during the bootstrap phase */ \ + .seq = 0, \ + }, \ +}; \ +static struct printk_ringbuffer name = { \ + .desc_ring = { \ + .count_bits = descbits, \ + .descs = &_##name##_descs[0], \ + .infos = &_##name##_infos[0], \ + .head_id = ATOMIC_INIT(DESC0_ID(descbits)), \ + .tail_id = ATOMIC_INIT(DESC0_ID(descbits)), \ + }, \ + .text_data_ring = { \ + .size_bits = (avgtextbits) + (descbits), \ + .data = text_buf, \ + .head_lpos = ATOMIC_LONG_INIT(BLK0_LPOS((avgtextbits) + (descbits))), \ + .tail_lpos = ATOMIC_LONG_INIT(BLK0_LPOS((avgtextbits) + (descbits))), \ + }, \ + .fail = ATOMIC_LONG_INIT(0), \ +} + +/** + * DEFINE_PRINTKRB() - Define a ringbuffer. + * + * @name: The name of the ringbuffer variable. + * @descbits: The number of descriptors as a power-of-2 value. + * @avgtextbits: The average text data size per record as a power-of-2 value. + * + * This is a macro for defining a ringbuffer and all internal structures + * such that it is ready for immediate use. See _DEFINE_PRINTKRB() for a + * variant where the text data buffer can be specified externally. + */ +#define DEFINE_PRINTKRB(name, descbits, avgtextbits) \ +static char _##name##_text[1U << ((avgtextbits) + (descbits))] \ + __aligned(__alignof__(unsigned long)); \ +_DEFINE_PRINTKRB(name, descbits, avgtextbits, &_##name##_text[0]) + +/* Writer Interface */ + +/** + * prb_rec_init_wd() - Initialize a buffer for writing records. + * + * @r: The record to initialize. + * @text_buf_size: The needed text buffer size. + */ +static inline void prb_rec_init_wr(struct printk_record *r, + unsigned int text_buf_size) +{ + r->info = NULL; + r->text_buf = NULL; + r->text_buf_size = text_buf_size; +} + +bool prb_reserve(struct prb_reserved_entry *e, struct printk_ringbuffer *rb, + struct printk_record *r); +bool prb_reserve_in_last(struct prb_reserved_entry *e, struct printk_ringbuffer *rb, + struct printk_record *r, u32 caller_id, unsigned int max_size); +void prb_commit(struct prb_reserved_entry *e); +void prb_final_commit(struct prb_reserved_entry *e); + +void prb_init(struct printk_ringbuffer *rb, + char *text_buf, unsigned int text_buf_size, + struct prb_desc *descs, unsigned int descs_count_bits, + struct printk_info *infos); +unsigned int prb_record_text_space(struct prb_reserved_entry *e); + +/* Reader Interface */ + +/** + * prb_rec_init_rd() - Initialize a buffer for reading records. + * + * @r: The record to initialize. + * @info: A buffer to store record meta-data. + * @text_buf: A buffer to store text data. + * @text_buf_size: The size of @text_buf. + * + * Initialize all the fields that a reader is interested in. All arguments + * (except @r) are optional. Only record data for arguments that are + * non-NULL or non-zero will be read. + */ +static inline void prb_rec_init_rd(struct printk_record *r, + struct printk_info *info, + char *text_buf, unsigned int text_buf_size) +{ + r->info = info; + r->text_buf = text_buf; + r->text_buf_size = text_buf_size; +} + +/** + * prb_for_each_record() - Iterate over the records of a ringbuffer. + * + * @from: The sequence number to begin with. + * @rb: The ringbuffer to iterate over. + * @s: A u64 to store the sequence number on each iteration. + * @r: A printk_record to store the record on each iteration. + * + * This is a macro for conveniently iterating over a ringbuffer. + * Note that @s may not be the sequence number of the record on each + * iteration. For the sequence number, @r->info->seq should be checked. + * + * Context: Any context. + */ +#define prb_for_each_record(from, rb, s, r) \ +for ((s) = from; prb_read_valid(rb, s, r); (s) = (r)->info->seq + 1) + +/** + * prb_for_each_info() - Iterate over the meta data of a ringbuffer. + * + * @from: The sequence number to begin with. + * @rb: The ringbuffer to iterate over. + * @s: A u64 to store the sequence number on each iteration. + * @i: A printk_info to store the record meta data on each iteration. + * @lc: An unsigned int to store the text line count of each record. + * + * This is a macro for conveniently iterating over a ringbuffer. + * Note that @s may not be the sequence number of the record on each + * iteration. For the sequence number, @r->info->seq should be checked. + * + * Context: Any context. + */ +#define prb_for_each_info(from, rb, s, i, lc) \ +for ((s) = from; prb_read_valid_info(rb, s, i, lc); (s) = (i)->seq + 1) + +bool prb_read_valid(struct printk_ringbuffer *rb, u64 seq, + struct printk_record *r); +bool prb_read_valid_info(struct printk_ringbuffer *rb, u64 seq, + struct printk_info *info, unsigned int *line_count); + +u64 prb_first_valid_seq(struct printk_ringbuffer *rb); +u64 prb_next_seq(struct printk_ringbuffer *rb); + +#endif /* _KERNEL_PRINTK_RINGBUFFER_H */ diff --git a/kernel/printk/printk_safe.c b/kernel/printk/printk_safe.c deleted file mode 100644 index 50aeae7704346..0000000000000 --- a/kernel/printk/printk_safe.c +++ /dev/null @@ -1,414 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* - * printk_safe.c - Safe printk for printk-deadlock-prone contexts - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include "internal.h" - -/* - * printk() could not take logbuf_lock in NMI context. Instead, - * it uses an alternative implementation that temporary stores - * the strings into a per-CPU buffer. The content of the buffer - * is later flushed into the main ring buffer via IRQ work. - * - * The alternative implementation is chosen transparently - * by examinig current printk() context mask stored in @printk_context - * per-CPU variable. - * - * The implementation allows to flush the strings also from another CPU. - * There are situations when we want to make sure that all buffers - * were handled or when IRQs are blocked. - */ - -#define SAFE_LOG_BUF_LEN ((1 << CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT) - \ - sizeof(atomic_t) - \ - sizeof(atomic_t) - \ - sizeof(struct irq_work)) - -struct printk_safe_seq_buf { - atomic_t len; /* length of written data */ - atomic_t message_lost; - struct irq_work work; /* IRQ work that flushes the buffer */ - unsigned char buffer[SAFE_LOG_BUF_LEN]; -}; - -static DEFINE_PER_CPU(struct printk_safe_seq_buf, safe_print_seq); -static DEFINE_PER_CPU(int, printk_context); - -#ifdef CONFIG_PRINTK_NMI -static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq); -#endif - -/* Get flushed in a more safe context. */ -static void queue_flush_work(struct printk_safe_seq_buf *s) -{ - if (printk_percpu_data_ready()) - irq_work_queue(&s->work); -} - -/* - * Add a message to per-CPU context-dependent buffer. NMI and printk-safe - * have dedicated buffers, because otherwise printk-safe preempted by - * NMI-printk would have overwritten the NMI messages. - * - * The messages are flushed from irq work (or from panic()), possibly, - * from other CPU, concurrently with printk_safe_log_store(). Should this - * happen, printk_safe_log_store() will notice the buffer->len mismatch - * and repeat the write. - */ -static __printf(2, 0) int printk_safe_log_store(struct printk_safe_seq_buf *s, - const char *fmt, va_list args) -{ - int add; - size_t len; - va_list ap; - -again: - len = atomic_read(&s->len); - - /* The trailing '\0' is not counted into len. */ - if (len >= sizeof(s->buffer) - 1) { - atomic_inc(&s->message_lost); - queue_flush_work(s); - return 0; - } - - /* - * Make sure that all old data have been read before the buffer - * was reset. This is not needed when we just append data. - */ - if (!len) - smp_rmb(); - - va_copy(ap, args); - add = vscnprintf(s->buffer + len, sizeof(s->buffer) - len, fmt, ap); - va_end(ap); - if (!add) - return 0; - - /* - * Do it once again if the buffer has been flushed in the meantime. - * Note that atomic_cmpxchg() is an implicit memory barrier that - * makes sure that the data were written before updating s->len. - */ - if (atomic_cmpxchg(&s->len, len, len + add) != len) - goto again; - - queue_flush_work(s); - return add; -} - -static inline void printk_safe_flush_line(const char *text, int len) -{ - /* - * Avoid any console drivers calls from here, because we may be - * in NMI or printk_safe context (when in panic). The messages - * must go only into the ring buffer at this stage. Consoles will - * get explicitly called later when a crashdump is not generated. - */ - printk_deferred("%.*s", len, text); -} - -/* printk part of the temporary buffer line by line */ -static int printk_safe_flush_buffer(const char *start, size_t len) -{ - const char *c, *end; - bool header; - - c = start; - end = start + len; - header = true; - - /* Print line by line. */ - while (c < end) { - if (*c == '\n') { - printk_safe_flush_line(start, c - start + 1); - start = ++c; - header = true; - continue; - } - - /* Handle continuous lines or missing new line. */ - if ((c + 1 < end) && printk_get_level(c)) { - if (header) { - c = printk_skip_level(c); - continue; - } - - printk_safe_flush_line(start, c - start); - start = c++; - header = true; - continue; - } - - header = false; - c++; - } - - /* Check if there was a partial line. Ignore pure header. */ - if (start < end && !header) { - static const char newline[] = KERN_CONT "\n"; - - printk_safe_flush_line(start, end - start); - printk_safe_flush_line(newline, strlen(newline)); - } - - return len; -} - -static void report_message_lost(struct printk_safe_seq_buf *s) -{ - int lost = atomic_xchg(&s->message_lost, 0); - - if (lost) - printk_deferred("Lost %d message(s)!\n", lost); -} - -/* - * Flush data from the associated per-CPU buffer. The function - * can be called either via IRQ work or independently. - */ -static void __printk_safe_flush(struct irq_work *work) -{ - static raw_spinlock_t read_lock = - __RAW_SPIN_LOCK_INITIALIZER(read_lock); - struct printk_safe_seq_buf *s = - container_of(work, struct printk_safe_seq_buf, work); - unsigned long flags; - size_t len; - int i; - - /* - * The lock has two functions. First, one reader has to flush all - * available message to make the lockless synchronization with - * writers easier. Second, we do not want to mix messages from - * different CPUs. This is especially important when printing - * a backtrace. - */ - raw_spin_lock_irqsave(&read_lock, flags); - - i = 0; -more: - len = atomic_read(&s->len); - - /* - * This is just a paranoid check that nobody has manipulated - * the buffer an unexpected way. If we printed something then - * @len must only increase. Also it should never overflow the - * buffer size. - */ - if ((i && i >= len) || len > sizeof(s->buffer)) { - const char *msg = "printk_safe_flush: internal error\n"; - - printk_safe_flush_line(msg, strlen(msg)); - len = 0; - } - - if (!len) - goto out; /* Someone else has already flushed the buffer. */ - - /* Make sure that data has been written up to the @len */ - smp_rmb(); - i += printk_safe_flush_buffer(s->buffer + i, len - i); - - /* - * Check that nothing has got added in the meantime and truncate - * the buffer. Note that atomic_cmpxchg() is an implicit memory - * barrier that makes sure that the data were copied before - * updating s->len. - */ - if (atomic_cmpxchg(&s->len, len, 0) != len) - goto more; - -out: - report_message_lost(s); - raw_spin_unlock_irqrestore(&read_lock, flags); -} - -/** - * printk_safe_flush - flush all per-cpu nmi buffers. - * - * The buffers are flushed automatically via IRQ work. This function - * is useful only when someone wants to be sure that all buffers have - * been flushed at some point. - */ -void printk_safe_flush(void) -{ - int cpu; - - for_each_possible_cpu(cpu) { -#ifdef CONFIG_PRINTK_NMI - __printk_safe_flush(&per_cpu(nmi_print_seq, cpu).work); -#endif - __printk_safe_flush(&per_cpu(safe_print_seq, cpu).work); - } -} - -/** - * printk_safe_flush_on_panic - flush all per-cpu nmi buffers when the system - * goes down. - * - * Similar to printk_safe_flush() but it can be called even in NMI context when - * the system goes down. It does the best effort to get NMI messages into - * the main ring buffer. - * - * Note that it could try harder when there is only one CPU online. - */ -void printk_safe_flush_on_panic(void) -{ - /* - * Make sure that we could access the main ring buffer. - * Do not risk a double release when more CPUs are up. - */ - if (raw_spin_is_locked(&logbuf_lock)) { - if (num_online_cpus() > 1) - return; - - debug_locks_off(); - raw_spin_lock_init(&logbuf_lock); - } - - printk_safe_flush(); -} - -#ifdef CONFIG_PRINTK_NMI -/* - * Safe printk() for NMI context. It uses a per-CPU buffer to - * store the message. NMIs are not nested, so there is always only - * one writer running. But the buffer might get flushed from another - * CPU, so we need to be careful. - */ -static __printf(1, 0) int vprintk_nmi(const char *fmt, va_list args) -{ - struct printk_safe_seq_buf *s = this_cpu_ptr(&nmi_print_seq); - - return printk_safe_log_store(s, fmt, args); -} - -void noinstr printk_nmi_enter(void) -{ - this_cpu_add(printk_context, PRINTK_NMI_CONTEXT_OFFSET); -} - -void noinstr printk_nmi_exit(void) -{ - this_cpu_sub(printk_context, PRINTK_NMI_CONTEXT_OFFSET); -} - -/* - * Marks a code that might produce many messages in NMI context - * and the risk of losing them is more critical than eventual - * reordering. - * - * It has effect only when called in NMI context. Then printk() - * will try to store the messages into the main logbuf directly - * and use the per-CPU buffers only as a fallback when the lock - * is not available. - */ -void printk_nmi_direct_enter(void) -{ - if (this_cpu_read(printk_context) & PRINTK_NMI_CONTEXT_MASK) - this_cpu_or(printk_context, PRINTK_NMI_DIRECT_CONTEXT_MASK); -} - -void printk_nmi_direct_exit(void) -{ - this_cpu_and(printk_context, ~PRINTK_NMI_DIRECT_CONTEXT_MASK); -} - -#else - -static __printf(1, 0) int vprintk_nmi(const char *fmt, va_list args) -{ - return 0; -} - -#endif /* CONFIG_PRINTK_NMI */ - -/* - * Lock-less printk(), to avoid deadlocks should the printk() recurse - * into itself. It uses a per-CPU buffer to store the message, just like - * NMI. - */ -static __printf(1, 0) int vprintk_safe(const char *fmt, va_list args) -{ - struct printk_safe_seq_buf *s = this_cpu_ptr(&safe_print_seq); - - return printk_safe_log_store(s, fmt, args); -} - -/* Can be preempted by NMI. */ -void __printk_safe_enter(void) -{ - this_cpu_inc(printk_context); -} - -/* Can be preempted by NMI. */ -void __printk_safe_exit(void) -{ - this_cpu_dec(printk_context); -} - -__printf(1, 0) int vprintk_func(const char *fmt, va_list args) -{ -#ifdef CONFIG_KGDB_KDB - /* Allow to pass printk() to kdb but avoid a recursion. */ - if (unlikely(kdb_trap_printk && kdb_printf_cpu < 0)) - return vkdb_printf(KDB_MSGSRC_PRINTK, fmt, args); -#endif - - /* - * Try to use the main logbuf even in NMI. But avoid calling console - * drivers that might have their own locks. - */ - if ((this_cpu_read(printk_context) & PRINTK_NMI_DIRECT_CONTEXT_MASK) && - raw_spin_trylock(&logbuf_lock)) { - int len; - - len = vprintk_store(0, LOGLEVEL_DEFAULT, NULL, 0, fmt, args); - raw_spin_unlock(&logbuf_lock); - defer_console_output(); - return len; - } - - /* Use extra buffer in NMI when logbuf_lock is taken or in safe mode. */ - if (this_cpu_read(printk_context) & PRINTK_NMI_CONTEXT_MASK) - return vprintk_nmi(fmt, args); - - /* Use extra buffer to prevent a recursion deadlock in safe mode. */ - if (this_cpu_read(printk_context) & PRINTK_SAFE_CONTEXT_MASK) - return vprintk_safe(fmt, args); - - /* No obstacles. */ - return vprintk_default(fmt, args); -} - -void __init printk_safe_init(void) -{ - int cpu; - - for_each_possible_cpu(cpu) { - struct printk_safe_seq_buf *s; - - s = &per_cpu(safe_print_seq, cpu); - init_irq_work(&s->work, __printk_safe_flush); - -#ifdef CONFIG_PRINTK_NMI - s = &per_cpu(nmi_print_seq, cpu); - init_irq_work(&s->work, __printk_safe_flush); -#endif - } - - /* Flush pending messages that did not have scheduled IRQ works. */ - printk_safe_flush(); -} diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 43d6179508d64..3075006d720e6 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -180,7 +180,14 @@ static bool ptrace_freeze_traced(struct task_struct *task) spin_lock_irq(&task->sighand->siglock); if (task_is_traced(task) && !__fatal_signal_pending(task)) { - task->state = __TASK_TRACED; + unsigned long flags; + + raw_spin_lock_irqsave(&task->pi_lock, flags); + if (task->state & __TASK_TRACED) + task->state = __TASK_TRACED; + else + task->saved_state = __TASK_TRACED; + raw_spin_unlock_irqrestore(&task->pi_lock, flags); ret = true; } spin_unlock_irq(&task->sighand->siglock); diff --git a/kernel/rcu/Kconfig b/kernel/rcu/Kconfig index 0ebe15a849856..046974b22b8f6 100644 --- a/kernel/rcu/Kconfig +++ b/kernel/rcu/Kconfig @@ -186,8 +186,8 @@ config RCU_FAST_NO_HZ config RCU_BOOST bool "Enable RCU priority boosting" - depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT - default n + depends on (RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT) || PREEMPT_RT + default y if PREEMPT_RT help This option boosts the priority of preempted RCU readers that block the current preemptible RCU grace period for too long. diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index f453bf8d2f1ef..a046e0c84db96 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -74,10 +74,13 @@ MODULE_AUTHOR("Paul E. McKenney and Josh Triplett > RCUTORTURE_RDR_SHIFT) > 1); rtrsp->rt_readstate = newstate; - /* First, put new protection in place to avoid critical-section gap. */ + /* + * First, put new protection in place to avoid critical-section gap. + * Disable preemption around the ATOM disables to ensure that + * in_atomic() is true. + */ if (statesnew & RCUTORTURE_RDR_BH) local_bh_disable(); + if (statesnew & RCUTORTURE_RDR_RBH) + rcu_read_lock_bh(); if (statesnew & RCUTORTURE_RDR_IRQ) local_irq_disable(); if (statesnew & RCUTORTURE_RDR_PREEMPT) preempt_disable(); - if (statesnew & RCUTORTURE_RDR_RBH) - rcu_read_lock_bh(); if (statesnew & RCUTORTURE_RDR_SCHED) rcu_read_lock_sched(); + preempt_disable(); + if (statesnew & RCUTORTURE_RDR_ATOM_BH) + local_bh_disable(); + if (statesnew & RCUTORTURE_RDR_ATOM_RBH) + rcu_read_lock_bh(); + preempt_enable(); if (statesnew & RCUTORTURE_RDR_RCU) idxnew = cur_ops->readlock() << RCUTORTURE_RDR_SHIFT; - /* Next, remove old protection, irq first due to bh conflict. */ + /* + * Next, remove old protection, in decreasing order of strength + * to avoid unlock paths that aren't safe in the stronger + * context. Disable preemption around the ATOM enables in + * case the context was only atomic due to IRQ disabling. + */ + preempt_disable(); if (statesold & RCUTORTURE_RDR_IRQ) local_irq_enable(); - if (statesold & RCUTORTURE_RDR_BH) + if (statesold & RCUTORTURE_RDR_ATOM_BH) local_bh_enable(); + if (statesold & RCUTORTURE_RDR_ATOM_RBH) + rcu_read_unlock_bh(); + preempt_enable(); if (statesold & RCUTORTURE_RDR_PREEMPT) preempt_enable(); - if (statesold & RCUTORTURE_RDR_RBH) - rcu_read_unlock_bh(); if (statesold & RCUTORTURE_RDR_SCHED) rcu_read_unlock_sched(); + if (statesold & RCUTORTURE_RDR_BH) + local_bh_enable(); + if (statesold & RCUTORTURE_RDR_RBH) + rcu_read_unlock_bh(); + if (statesold & RCUTORTURE_RDR_RCU) { bool lockit = !statesnew && !(torture_random(trsp) & 0xffff); @@ -1313,6 +1338,12 @@ rcutorture_extend_mask(int oldmask, struct torture_random_state *trsp) int mask = rcutorture_extend_mask_max(); unsigned long randmask1 = torture_random(trsp) >> 8; unsigned long randmask2 = randmask1 >> 3; + unsigned long preempts = RCUTORTURE_RDR_PREEMPT | RCUTORTURE_RDR_SCHED; + unsigned long preempts_irq = preempts | RCUTORTURE_RDR_IRQ; + unsigned long nonatomic_bhs = RCUTORTURE_RDR_BH | RCUTORTURE_RDR_RBH; + unsigned long atomic_bhs = RCUTORTURE_RDR_ATOM_BH | + RCUTORTURE_RDR_ATOM_RBH; + unsigned long tmp; WARN_ON_ONCE(mask >> RCUTORTURE_RDR_SHIFT); /* Mostly only one bit (need preemption!), sometimes lots of bits. */ @@ -1320,11 +1351,49 @@ rcutorture_extend_mask(int oldmask, struct torture_random_state *trsp) mask = mask & randmask2; else mask = mask & (1 << (randmask2 % RCUTORTURE_RDR_NBITS)); - /* Can't enable bh w/irq disabled. */ - if ((mask & RCUTORTURE_RDR_IRQ) && - ((!(mask & RCUTORTURE_RDR_BH) && (oldmask & RCUTORTURE_RDR_BH)) || - (!(mask & RCUTORTURE_RDR_RBH) && (oldmask & RCUTORTURE_RDR_RBH)))) - mask |= RCUTORTURE_RDR_BH | RCUTORTURE_RDR_RBH; + + /* + * Can't enable bh w/irq disabled. + */ + tmp = atomic_bhs | nonatomic_bhs; + if (mask & RCUTORTURE_RDR_IRQ) + mask |= oldmask & tmp; + + /* + * Ideally these sequences would be detected in debug builds + * (regardless of RT), but until then don't stop testing + * them on non-RT. + */ + if (IS_ENABLED(CONFIG_PREEMPT_RT)) { + /* + * Can't release the outermost rcu lock in an irq disabled + * section without preemption also being disabled, if irqs + * had ever been enabled during this RCU critical section + * (could leak a special flag and delay reporting the qs). + */ + if ((oldmask & RCUTORTURE_RDR_RCU) && + (mask & RCUTORTURE_RDR_IRQ) && + !(mask & preempts)) + mask |= RCUTORTURE_RDR_RCU; + + /* Can't modify atomic bh in non-atomic context */ + if ((oldmask & atomic_bhs) && (mask & atomic_bhs) && + !(mask & preempts_irq)) { + mask |= oldmask & preempts_irq; + if (mask & RCUTORTURE_RDR_IRQ) + mask |= oldmask & tmp; + } + if ((mask & atomic_bhs) && !(mask & preempts_irq)) + mask |= RCUTORTURE_RDR_PREEMPT; + + /* Can't modify non-atomic bh in atomic context */ + tmp = nonatomic_bhs; + if (oldmask & preempts_irq) + mask &= ~tmp; + if ((oldmask | mask) & preempts_irq) + mask |= oldmask & tmp; + } + return mask ?: RCUTORTURE_RDR_RCU; } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index f78ee759af9cb..367165074e5f3 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -113,8 +113,10 @@ static struct rcu_state rcu_state = { static bool dump_tree; module_param(dump_tree, bool, 0444); /* By default, use RCU_SOFTIRQ instead of rcuc kthreads. */ -static bool use_softirq = true; +static bool use_softirq = !IS_ENABLED(CONFIG_PREEMPT_RT); +#ifndef CONFIG_PREEMPT_RT module_param(use_softirq, bool, 0444); +#endif /* Control rcu_node-tree auto-balancing at boot time. */ static bool rcu_fanout_exact; module_param(rcu_fanout_exact, bool, 0444); diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 2de49b5d8dd29..294f7021a4591 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -69,8 +69,10 @@ #ifndef CONFIG_TINY_RCU module_param(rcu_expedited, int, 0); module_param(rcu_normal, int, 0); -static int rcu_normal_after_boot; +static int rcu_normal_after_boot = IS_ENABLED(CONFIG_PREEMPT_RT); +#ifndef CONFIG_PREEMPT_RT module_param(rcu_normal_after_boot, int, 0); +#endif #endif /* #ifndef CONFIG_TINY_RCU */ #ifdef CONFIG_DEBUG_LOCK_ALLOC diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2d95dc3f46444..34158d8ecf194 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -63,7 +63,11 @@ const_debug unsigned int sysctl_sched_features = * Number of tasks to iterate in a single balance run. * Limited because this is done with IRQs disabled. */ +#ifdef CONFIG_PREEMPT_RT +const_debug unsigned int sysctl_sched_nr_migrate = 8; +#else const_debug unsigned int sysctl_sched_nr_migrate = 32; +#endif /* * period over which we measure -rt task CPU usage in us. @@ -511,9 +515,15 @@ static bool set_nr_if_polling(struct task_struct *p) #endif #endif -static bool __wake_q_add(struct wake_q_head *head, struct task_struct *task) +static bool __wake_q_add(struct wake_q_head *head, struct task_struct *task, + bool sleeper) { - struct wake_q_node *node = &task->wake_q; + struct wake_q_node *node; + + if (sleeper) + node = &task->wake_q_sleeper; + else + node = &task->wake_q; /* * Atomically grab the task, if ->wake_q is !nil already it means @@ -549,7 +559,13 @@ static bool __wake_q_add(struct wake_q_head *head, struct task_struct *task) */ void wake_q_add(struct wake_q_head *head, struct task_struct *task) { - if (__wake_q_add(head, task)) + if (__wake_q_add(head, task, false)) + get_task_struct(task); +} + +void wake_q_add_sleeper(struct wake_q_head *head, struct task_struct *task) +{ + if (__wake_q_add(head, task, true)) get_task_struct(task); } @@ -572,28 +588,39 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task) */ void wake_q_add_safe(struct wake_q_head *head, struct task_struct *task) { - if (!__wake_q_add(head, task)) + if (!__wake_q_add(head, task, false)) put_task_struct(task); } -void wake_up_q(struct wake_q_head *head) +void __wake_up_q(struct wake_q_head *head, bool sleeper) { struct wake_q_node *node = head->first; while (node != WAKE_Q_TAIL) { struct task_struct *task; - task = container_of(node, struct task_struct, wake_q); + if (sleeper) + task = container_of(node, struct task_struct, wake_q_sleeper); + else + task = container_of(node, struct task_struct, wake_q); + BUG_ON(!task); /* Task can safely be re-inserted now: */ node = node->next; - task->wake_q.next = NULL; + if (sleeper) + task->wake_q_sleeper.next = NULL; + else + task->wake_q.next = NULL; /* * wake_up_process() executes a full barrier, which pairs with * the queueing in wake_q_add() so as not to miss wakeups. */ - wake_up_process(task); + if (sleeper) + wake_up_lock_sleeper(task); + else + wake_up_process(task); + put_task_struct(task); } } @@ -629,6 +656,48 @@ void resched_curr(struct rq *rq) trace_sched_wake_idle_without_ipi(cpu); } +#ifdef CONFIG_PREEMPT_LAZY + +static int tsk_is_polling(struct task_struct *p) +{ +#ifdef TIF_POLLING_NRFLAG + return test_tsk_thread_flag(p, TIF_POLLING_NRFLAG); +#else + return 0; +#endif +} + +void resched_curr_lazy(struct rq *rq) +{ + struct task_struct *curr = rq->curr; + int cpu; + + if (!sched_feat(PREEMPT_LAZY)) { + resched_curr(rq); + return; + } + + lockdep_assert_held(&rq->lock); + + if (test_tsk_need_resched(curr)) + return; + + if (test_tsk_need_resched_lazy(curr)) + return; + + set_tsk_need_resched_lazy(curr); + + cpu = cpu_of(rq); + if (cpu == smp_processor_id()) + return; + + /* NEED_RESCHED_LAZY must be visible before we test polling */ + smp_mb(); + if (!tsk_is_polling(curr)) + smp_send_reschedule(cpu); +} +#endif + void resched_cpu(int cpu) { struct rq *rq = cpu_rq(cpu); @@ -1700,6 +1769,86 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) #ifdef CONFIG_SMP +#ifdef CONFIG_PREEMPT_RT + +static void +__do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask, u32 flags); + +static int __set_cpus_allowed_ptr(struct task_struct *p, + const struct cpumask *new_mask, + u32 flags); + +static void migrate_disable_switch(struct rq *rq, struct task_struct *p) +{ + if (likely(!p->migration_disabled)) + return; + + if (p->cpus_ptr != &p->cpus_mask) + return; + + /* + * Violates locking rules! see comment in __do_set_cpus_allowed(). + */ + __do_set_cpus_allowed(p, cpumask_of(rq->cpu), SCA_MIGRATE_DISABLE); +} + +void migrate_disable(void) +{ + struct task_struct *p = current; + + if (p->migration_disabled) { + p->migration_disabled++; + return; + } + + trace_sched_migrate_disable_tp(p); + + preempt_disable(); + this_rq()->nr_pinned++; + p->migration_disabled = 1; + preempt_lazy_disable(); + preempt_enable(); +} +EXPORT_SYMBOL_GPL(migrate_disable); + +void migrate_enable(void) +{ + struct task_struct *p = current; + + if (p->migration_disabled > 1) { + p->migration_disabled--; + return; + } + + /* + * Ensure stop_task runs either before or after this, and that + * __set_cpus_allowed_ptr(SCA_MIGRATE_ENABLE) doesn't schedule(). + */ + preempt_disable(); + if (p->cpus_ptr != &p->cpus_mask) + __set_cpus_allowed_ptr(p, &p->cpus_mask, SCA_MIGRATE_ENABLE); + /* + * Mustn't clear migration_disabled() until cpus_ptr points back at the + * regular cpus_mask, otherwise things that race (eg. + * select_fallback_rq) get confused. + */ + barrier(); + p->migration_disabled = 0; + this_rq()->nr_pinned--; + preempt_lazy_enable(); + preempt_enable(); + + trace_sched_migrate_enable_tp(p); +} +EXPORT_SYMBOL_GPL(migrate_enable); + +static inline bool rq_has_pinned_tasks(struct rq *rq) +{ + return rq->nr_pinned; +} + +#endif + /* * Per-CPU kthreads are allowed to run on !active && online CPUs, see * __set_cpus_allowed_ptr() and select_fallback_rq(). @@ -1709,7 +1858,7 @@ static inline bool is_cpu_allowed(struct task_struct *p, int cpu) if (!cpumask_test_cpu(cpu, p->cpus_ptr)) return false; - if (is_per_cpu_kthread(p)) + if (is_per_cpu_kthread(p) || is_migration_disabled(p)) return cpu_online(cpu); return cpu_active(cpu); @@ -1754,8 +1903,16 @@ static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, } struct migration_arg { - struct task_struct *task; - int dest_cpu; + struct task_struct *task; + int dest_cpu; + struct set_affinity_pending *pending; +}; + +struct set_affinity_pending { + refcount_t refs; + struct completion done; + struct cpu_stop_work stop_work; + struct migration_arg arg; }; /* @@ -1787,16 +1944,19 @@ static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf, */ static int migration_cpu_stop(void *data) { + struct set_affinity_pending *pending; struct migration_arg *arg = data; struct task_struct *p = arg->task; + int dest_cpu = arg->dest_cpu; struct rq *rq = this_rq(); + bool complete = false; struct rq_flags rf; /* * The original target CPU might have gone down and we might * be on another CPU but it doesn't matter. */ - local_irq_disable(); + local_irq_save(rf.flags); /* * We need to explicitly wake pending tasks before running * __migrate_task() such that we will not miss enforcing cpus_ptr @@ -1806,21 +1966,126 @@ static int migration_cpu_stop(void *data) raw_spin_lock(&p->pi_lock); rq_lock(rq, &rf); + + pending = p->migration_pending; /* * If task_rq(p) != rq, it cannot be migrated here, because we're * holding rq->lock, if p->on_rq == 0 it cannot get enqueued because * we're holding p->pi_lock. */ if (task_rq(p) == rq) { - if (task_on_rq_queued(p)) - rq = __migrate_task(rq, &rf, p, arg->dest_cpu); - else - p->wake_cpu = arg->dest_cpu; - } - rq_unlock(rq, &rf); - raw_spin_unlock(&p->pi_lock); + if (is_migration_disabled(p)) + goto out; - local_irq_enable(); + if (pending) { + p->migration_pending = NULL; + complete = true; + } + + /* migrate_enable() -- we must not race against SCA */ + if (dest_cpu < 0) { + /* + * When this was migrate_enable() but we no longer + * have a @pending, a concurrent SCA 'fixed' things + * and we should be valid again. Nothing to do. + */ + if (!pending) { + WARN_ON_ONCE(!is_cpu_allowed(p, cpu_of(rq))); + goto out; + } + + dest_cpu = cpumask_any_distribute(&p->cpus_mask); + } + + if (task_on_rq_queued(p)) + rq = __migrate_task(rq, &rf, p, dest_cpu); + else + p->wake_cpu = dest_cpu; + + } else if (dest_cpu < 0) { + /* + * This happens when we get migrated between migrate_enable()'s + * preempt_enable() and scheduling the stopper task. At that + * point we're a regular task again and not current anymore. + * + * A !PREEMPT kernel has a giant hole here, which makes it far + * more likely. + */ + + /* + * When this was migrate_enable() but we no longer have an + * @pending, a concurrent SCA 'fixed' things and we should be + * valid again. Nothing to do. + */ + if (!pending) { + WARN_ON_ONCE(!is_cpu_allowed(p, cpu_of(rq))); + goto out; + } + + /* + * When migrate_enable() hits a rq mis-match we can't reliably + * determine is_migration_disabled() and so have to chase after + * it. + */ + task_rq_unlock(rq, p, &rf); + stop_one_cpu_nowait(task_cpu(p), migration_cpu_stop, + &pending->arg, &pending->stop_work); + return 0; + } +out: + task_rq_unlock(rq, p, &rf); + + if (complete) + complete_all(&pending->done); + + /* For pending->{arg,stop_work} */ + pending = arg->pending; + if (pending && refcount_dec_and_test(&pending->refs)) + wake_up_var(&pending->refs); + + return 0; +} + +int push_cpu_stop(void *arg) +{ + struct rq *lowest_rq = NULL, *rq = this_rq(); + struct task_struct *p = arg; + + raw_spin_lock_irq(&p->pi_lock); + raw_spin_lock(&rq->lock); + + if (task_rq(p) != rq) + goto out_unlock; + + if (is_migration_disabled(p)) { + p->migration_flags |= MDF_PUSH; + goto out_unlock; + } + + p->migration_flags &= ~MDF_PUSH; + + if (p->sched_class->find_lock_rq) + lowest_rq = p->sched_class->find_lock_rq(p, rq); + + if (!lowest_rq) + goto out_unlock; + + // XXX validate p is still the highest prio task + if (task_rq(p) == rq) { + deactivate_task(rq, p, 0); + set_task_cpu(p, lowest_rq->cpu); + activate_task(lowest_rq, p, 0); + resched_curr(lowest_rq); + } + + double_unlock_balance(rq, lowest_rq); + +out_unlock: + rq->push_busy = false; + raw_spin_unlock(&rq->lock); + raw_spin_unlock_irq(&p->pi_lock); + + put_task_struct(p); return 0; } @@ -1828,18 +2093,39 @@ static int migration_cpu_stop(void *data) * sched_class::set_cpus_allowed must do the below, but is not required to * actually call this function. */ -void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask) +void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask, u32 flags) { + if (flags & (SCA_MIGRATE_ENABLE | SCA_MIGRATE_DISABLE)) { + p->cpus_ptr = new_mask; + return; + } + cpumask_copy(&p->cpus_mask, new_mask); p->nr_cpus_allowed = cpumask_weight(new_mask); } -void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) +static void +__do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask, u32 flags) { struct rq *rq = task_rq(p); bool queued, running; - lockdep_assert_held(&p->pi_lock); + /* + * This here violates the locking rules for affinity, since we're only + * supposed to change these variables while holding both rq->lock and + * p->pi_lock. + * + * HOWEVER, it magically works, because ttwu() is the only code that + * accesses these variables under p->pi_lock and only does so after + * smp_cond_load_acquire(&p->on_cpu, !VAL), and we're in __schedule() + * before finish_task(). + * + * XXX do further audits, this smells like something putrid. + */ + if (flags & SCA_MIGRATE_DISABLE) + SCHED_WARN_ON(!p->on_cpu); + else + lockdep_assert_held(&p->pi_lock); queued = task_on_rq_queued(p); running = task_current(rq, p); @@ -1855,7 +2141,7 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) if (running) put_prev_task(rq, p); - p->sched_class->set_cpus_allowed(p, new_mask); + p->sched_class->set_cpus_allowed(p, new_mask, flags); if (queued) enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); @@ -1863,6 +2149,206 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) set_next_task(rq, p); } +void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) +{ + __do_set_cpus_allowed(p, new_mask, 0); +} + +/* + * This function is wildly self concurrent; here be dragons. + * + * + * When given a valid mask, __set_cpus_allowed_ptr() must block until the + * designated task is enqueued on an allowed CPU. If that task is currently + * running, we have to kick it out using the CPU stopper. + * + * Migrate-Disable comes along and tramples all over our nice sandcastle. + * Consider: + * + * Initial conditions: P0->cpus_mask = [0, 1] + * + * P0@CPU0 P1 + * + * migrate_disable(); + * + * set_cpus_allowed_ptr(P0, [1]); + * + * P1 *cannot* return from this set_cpus_allowed_ptr() call until P0 executes + * its outermost migrate_enable() (i.e. it exits its Migrate-Disable region). + * This means we need the following scheme: + * + * P0@CPU0 P1 + * + * migrate_disable(); + * + * set_cpus_allowed_ptr(P0, [1]); + * + * + * migrate_enable(); + * __set_cpus_allowed_ptr(); + * + * `--> + * + * Now the fun stuff: there may be several P1-like tasks, i.e. multiple + * concurrent set_cpus_allowed_ptr(P0, [*]) calls. CPU affinity changes of any + * task p are serialized by p->pi_lock, which we can leverage: the one that + * should come into effect at the end of the Migrate-Disable region is the last + * one. This means we only need to track a single cpumask (i.e. p->cpus_mask), + * but we still need to properly signal those waiting tasks at the appropriate + * moment. + * + * This is implemented using struct set_affinity_pending. The first + * __set_cpus_allowed_ptr() caller within a given Migrate-Disable region will + * setup an instance of that struct and install it on the targeted task_struct. + * Any and all further callers will reuse that instance. Those then wait for + * a completion signaled at the tail of the CPU stopper callback (1), triggered + * on the end of the Migrate-Disable region (i.e. outermost migrate_enable()). + * + * + * (1) In the cases covered above. There is one more where the completion is + * signaled within affine_move_task() itself: when a subsequent affinity request + * cancels the need for an active migration. Consider: + * + * Initial conditions: P0->cpus_mask = [0, 1] + * + * P0@CPU0 P1 P2 + * + * migrate_disable(); + * + * set_cpus_allowed_ptr(P0, [1]); + * + * set_cpus_allowed_ptr(P0, [0, 1]); + * + * + * + * Note that the above is safe vs a concurrent migrate_enable(), as any + * pending affinity completion is preceded an uninstallion of + * p->migration_pending done with p->pi_lock held. + */ +static int affine_move_task(struct rq *rq, struct task_struct *p, struct rq_flags *rf, + int dest_cpu, unsigned int flags) +{ + struct set_affinity_pending my_pending = { }, *pending = NULL; + struct migration_arg arg = { + .task = p, + .dest_cpu = dest_cpu, + }; + bool complete = false; + + /* Can the task run on the task's current CPU? If so, we're done */ + if (cpumask_test_cpu(task_cpu(p), &p->cpus_mask)) { + struct task_struct *push_task = NULL; + + if ((flags & SCA_MIGRATE_ENABLE) && + (p->migration_flags & MDF_PUSH) && !rq->push_busy) { + rq->push_busy = true; + push_task = get_task_struct(p); + } + + pending = p->migration_pending; + if (pending) { + refcount_inc(&pending->refs); + p->migration_pending = NULL; + complete = true; + } + task_rq_unlock(rq, p, rf); + + if (push_task) { + stop_one_cpu_nowait(rq->cpu, push_cpu_stop, + p, &rq->push_work); + } + + if (complete) + goto do_complete; + + return 0; + } + + if (!(flags & SCA_MIGRATE_ENABLE)) { + /* serialized by p->pi_lock */ + if (!p->migration_pending) { + /* Install the request */ + refcount_set(&my_pending.refs, 1); + init_completion(&my_pending.done); + p->migration_pending = &my_pending; + } else { + pending = p->migration_pending; + refcount_inc(&pending->refs); + } + } + pending = p->migration_pending; + /* + * - !MIGRATE_ENABLE: + * we'll have installed a pending if there wasn't one already. + * + * - MIGRATE_ENABLE: + * we're here because the current CPU isn't matching anymore, + * the only way that can happen is because of a concurrent + * set_cpus_allowed_ptr() call, which should then still be + * pending completion. + * + * Either way, we really should have a @pending here. + */ + if (WARN_ON_ONCE(!pending)) + return -EINVAL; + + if (flags & SCA_MIGRATE_ENABLE) { + + refcount_inc(&pending->refs); /* pending->{arg,stop_work} */ + p->migration_flags &= ~MDF_PUSH; + task_rq_unlock(rq, p, rf); + + pending->arg = (struct migration_arg) { + .task = p, + .dest_cpu = -1, + .pending = pending, + }; + + stop_one_cpu_nowait(cpu_of(rq), migration_cpu_stop, + &pending->arg, &pending->stop_work); + + return 0; + } + + if (task_running(rq, p) || p->state == TASK_WAKING) { + /* + * Lessen races (and headaches) by delegating + * is_migration_disabled(p) checks to the stopper, which will + * run on the same CPU as said p. + */ + task_rq_unlock(rq, p, rf); + stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg); + + } else { + + if (!is_migration_disabled(p)) { + if (task_on_rq_queued(p)) + rq = move_queued_task(rq, rf, p, dest_cpu); + + p->migration_pending = NULL; + complete = true; + } + task_rq_unlock(rq, p, rf); + +do_complete: + if (complete) + complete_all(&pending->done); + } + + wait_for_completion(&pending->done); + + if (refcount_dec_and_test(&pending->refs)) + wake_up_var(&pending->refs); + + /* + * Block the original owner of &pending until all subsequent callers + * have seen the completion and decremented the refcount + */ + wait_var_event(&my_pending.refs, !refcount_read(&my_pending.refs)); + + return 0; +} + /* * Change a given task's CPU affinity. Migrate the thread to a * proper CPU and schedule it away if the CPU it's executing on @@ -1873,7 +2359,8 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) * call is not atomic; no spinlocks may be held. */ static int __set_cpus_allowed_ptr(struct task_struct *p, - const struct cpumask *new_mask, bool check) + const struct cpumask *new_mask, + u32 flags) { const struct cpumask *cpu_valid_mask = cpu_active_mask; unsigned int dest_cpu; @@ -1884,9 +2371,14 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, rq = task_rq_lock(p, &rf); update_rq_clock(rq); - if (p->flags & PF_KTHREAD) { + if (p->flags & PF_KTHREAD || is_migration_disabled(p)) { /* - * Kernel threads are allowed on online && !active CPUs + * Kernel threads are allowed on online && !active CPUs. + * + * Specifically, migration_disabled() tasks must not fail the + * cpumask_any_and_distribute() pick below, esp. so on + * SCA_MIGRATE_ENABLE, otherwise we'll not call + * set_cpus_allowed_common() and actually reset p->cpus_ptr. */ cpu_valid_mask = cpu_online_mask; } @@ -1895,13 +2387,22 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, * Must re-check here, to close a race against __kthread_bind(), * sched_setaffinity() is not guaranteed to observe the flag. */ - if (check && (p->flags & PF_NO_SETAFFINITY)) { + if ((flags & SCA_CHECK) && (p->flags & PF_NO_SETAFFINITY)) { ret = -EINVAL; goto out; } - if (cpumask_equal(&p->cpus_mask, new_mask)) - goto out; + if (!(flags & SCA_MIGRATE_ENABLE)) { + if (cpumask_equal(&p->cpus_mask, new_mask)) + goto out; + + if (WARN_ON_ONCE(p == current && + is_migration_disabled(p) && + !cpumask_test_cpu(task_cpu(p), new_mask))) { + ret = -EBUSY; + goto out; + } + } /* * Picking a ~random cpu helps in cases where we are changing affinity @@ -1914,7 +2415,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, goto out; } - do_set_cpus_allowed(p, new_mask); + __do_set_cpus_allowed(p, new_mask, flags); if (p->flags & PF_KTHREAD) { /* @@ -1926,23 +2427,8 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, p->nr_cpus_allowed != 1); } - /* Can the task run on the task's current CPU? If so, we're done */ - if (cpumask_test_cpu(task_cpu(p), new_mask)) - goto out; + return affine_move_task(rq, p, &rf, dest_cpu, flags); - if (task_running(rq, p) || p->state == TASK_WAKING) { - struct migration_arg arg = { p, dest_cpu }; - /* Need help from migration thread: drop lock and wait. */ - task_rq_unlock(rq, p, &rf); - stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg); - return 0; - } else if (task_on_rq_queued(p)) { - /* - * OK, since we're going to drop the lock immediately - * afterwards anyway. - */ - rq = move_queued_task(rq, &rf, p, dest_cpu); - } out: task_rq_unlock(rq, p, &rf); @@ -1951,7 +2437,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p, int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) { - return __set_cpus_allowed_ptr(p, new_mask, false); + return __set_cpus_allowed_ptr(p, new_mask, 0); } EXPORT_SYMBOL_GPL(set_cpus_allowed_ptr); @@ -1992,6 +2478,8 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) * Clearly, migrating tasks to offline CPUs is a fairly daft thing. */ WARN_ON_ONCE(!cpu_online(new_cpu)); + + WARN_ON_ONCE(is_migration_disabled(p)); #endif trace_sched_migrate_task(p, new_cpu); @@ -2124,6 +2612,18 @@ int migrate_swap(struct task_struct *cur, struct task_struct *p, } #endif /* CONFIG_NUMA_BALANCING */ +static bool check_task_state(struct task_struct *p, long match_state) +{ + bool match = false; + + raw_spin_lock_irq(&p->pi_lock); + if (p->state == match_state || p->saved_state == match_state) + match = true; + raw_spin_unlock_irq(&p->pi_lock); + + return match; +} + /* * wait_task_inactive - wait for a thread to unschedule. * @@ -2168,7 +2668,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) * is actually now running somewhere else! */ while (task_running(rq, p)) { - if (match_state && unlikely(p->state != match_state)) + if (match_state && !check_task_state(p, match_state)) return 0; cpu_relax(); } @@ -2183,7 +2683,8 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) running = task_running(rq, p); queued = task_on_rq_queued(p); ncsw = 0; - if (!match_state || p->state == match_state) + if (!match_state || p->state == match_state || + p->saved_state == match_state) ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ task_rq_unlock(rq, p, &rf); @@ -2322,6 +2823,12 @@ static int select_fallback_rq(int cpu, struct task_struct *p) } fallthrough; case possible: + /* + * XXX When called from select_task_rq() we only + * hold p->pi_lock and again violate locking order. + * + * More yuck to audit. + */ do_set_cpus_allowed(p, cpu_possible_mask); state = fail; break; @@ -2356,7 +2863,7 @@ int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags) { lockdep_assert_held(&p->pi_lock); - if (p->nr_cpus_allowed > 1) + if (p->nr_cpus_allowed > 1 && !is_migration_disabled(p)) cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags); else cpu = cpumask_any(p->cpus_ptr); @@ -2379,6 +2886,7 @@ int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags) void sched_set_stop_task(int cpu, struct task_struct *stop) { + static struct lock_class_key stop_pi_lock; struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; struct task_struct *old_stop = cpu_rq(cpu)->stop; @@ -2394,6 +2902,20 @@ void sched_set_stop_task(int cpu, struct task_struct *stop) sched_setscheduler_nocheck(stop, SCHED_FIFO, ¶m); stop->sched_class = &stop_sched_class; + + /* + * The PI code calls rt_mutex_setprio() with ->pi_lock held to + * adjust the effective priority of a task. As a result, + * rt_mutex_setprio() can trigger (RT) balancing operations, + * which can then trigger wakeups of the stop thread to push + * around the current task. + * + * The stop task itself will never be part of the PI-chain, it + * never blocks, therefore that ->pi_lock recursion is safe. + * Tell lockdep about this by placing the stop->pi_lock in its + * own class. + */ + lockdep_set_class(&stop->pi_lock, &stop_pi_lock); } cpu_rq(cpu)->stop = stop; @@ -2410,13 +2932,25 @@ void sched_set_stop_task(int cpu, struct task_struct *stop) #else static inline int __set_cpus_allowed_ptr(struct task_struct *p, - const struct cpumask *new_mask, bool check) + const struct cpumask *new_mask, + u32 flags) { return set_cpus_allowed_ptr(p, new_mask); } #endif /* CONFIG_SMP */ +#if !defined(CONFIG_SMP) || !defined(CONFIG_PREEMPT_RT) + +static inline void migrate_disable_switch(struct rq *rq, struct task_struct *p) { } + +static inline bool rq_has_pinned_tasks(struct rq *rq) +{ + return false; +} + +#endif + static void ttwu_stat(struct task_struct *p, int cpu, int wake_flags) { @@ -2828,7 +3362,7 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) int cpu, success = 0; preempt_disable(); - if (p == current) { + if (!IS_ENABLED(CONFIG_PREEMPT_RT) && p == current) { /* * We're waking current, this means 'p->on_rq' and 'task_cpu(p) * == smp_processor_id()'. Together this means we can special @@ -2858,8 +3392,26 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) */ raw_spin_lock_irqsave(&p->pi_lock, flags); smp_mb__after_spinlock(); - if (!(p->state & state)) + if (!(p->state & state)) { + /* + * The task might be running due to a spinlock sleeper + * wakeup. Check the saved state and set it to running + * if the wakeup condition is true. + */ + if (!(wake_flags & WF_LOCK_SLEEPER)) { + if (p->saved_state & state) { + p->saved_state = TASK_RUNNING; + success = 1; + } + } goto unlock; + } + /* + * If this is a regular wakeup, then we can unconditionally + * clear the saved state of a "lock sleeper". + */ + if (!(wake_flags & WF_LOCK_SLEEPER)) + p->saved_state = TASK_RUNNING; trace_sched_waking(p); @@ -3049,6 +3601,18 @@ int wake_up_process(struct task_struct *p) } EXPORT_SYMBOL(wake_up_process); +/** + * wake_up_lock_sleeper - Wake up a specific process blocked on a "sleeping lock" + * @p: The process to be woken up. + * + * Same as wake_up_process() above, but wake_flags=WF_LOCK_SLEEPER to indicate + * the nature of the wakeup. + */ +int wake_up_lock_sleeper(struct task_struct *p) +{ + return try_to_wake_up(p, TASK_UNINTERRUPTIBLE, WF_LOCK_SLEEPER); +} + int wake_up_state(struct task_struct *p, unsigned int state) { return try_to_wake_up(p, state, 0); @@ -3102,6 +3666,7 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p) init_numa_balancing(clone_flags, p); #ifdef CONFIG_SMP p->wake_entry.u_flags = CSD_TYPE_TTWU; + p->migration_pending = NULL; #endif } @@ -3295,6 +3860,9 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p) p->on_cpu = 0; #endif init_task_preempt_count(p); +#ifdef CONFIG_HAVE_PREEMPT_LAZY + task_thread_info(p)->preempt_lazy_count = 0; +#endif #ifdef CONFIG_SMP plist_node_init(&p->pushable_tasks, MAX_PRIO); RB_CLEAR_NODE(&p->pushable_dl_tasks); @@ -3489,6 +4057,90 @@ static inline void finish_task(struct task_struct *prev) #endif } +#ifdef CONFIG_SMP + +static void do_balance_callbacks(struct rq *rq, struct callback_head *head) +{ + void (*func)(struct rq *rq); + struct callback_head *next; + + lockdep_assert_held(&rq->lock); + + while (head) { + func = (void (*)(struct rq *))head->func; + next = head->next; + head->next = NULL; + head = next; + + func(rq); + } +} + +static inline struct callback_head *splice_balance_callbacks(struct rq *rq) +{ + struct callback_head *head = rq->balance_callback; + + lockdep_assert_held(&rq->lock); + if (head) { + rq->balance_callback = NULL; + rq->balance_flags &= ~BALANCE_WORK; + } + + return head; +} + +static void __balance_callbacks(struct rq *rq) +{ + do_balance_callbacks(rq, splice_balance_callbacks(rq)); +} + +static inline void balance_callbacks(struct rq *rq, struct callback_head *head) +{ + unsigned long flags; + + if (unlikely(head)) { + raw_spin_lock_irqsave(&rq->lock, flags); + do_balance_callbacks(rq, head); + raw_spin_unlock_irqrestore(&rq->lock, flags); + } +} + +static void balance_push(struct rq *rq); + +static inline void balance_switch(struct rq *rq) +{ + if (likely(!rq->balance_flags)) + return; + + if (rq->balance_flags & BALANCE_PUSH) { + balance_push(rq); + return; + } + + __balance_callbacks(rq); +} + +#else + +static inline void __balance_callbacks(struct rq *rq) +{ +} + +static inline struct callback_head *splice_balance_callbacks(struct rq *rq) +{ + return NULL; +} + +static inline void balance_callbacks(struct rq *rq, struct callback_head *head) +{ +} + +static inline void balance_switch(struct rq *rq) +{ +} + +#endif + static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf) { @@ -3514,6 +4166,7 @@ static inline void finish_lock_switch(struct rq *rq) * prev into current: */ spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_); + balance_switch(rq); raw_spin_unlock_irq(&rq->lock); } @@ -3631,23 +4284,18 @@ static struct rq *finish_task_switch(struct task_struct *prev) * provided by mmdrop(), * - a sync_core for SYNC_CORE. */ + /* + * We use mmdrop_delayed() here so we don't have to do the + * full __mmdrop() when we are the last user. + */ if (mm) { membarrier_mm_sync_core_before_usermode(mm); - mmdrop(mm); + mmdrop_delayed(mm); } if (unlikely(prev_state == TASK_DEAD)) { if (prev->sched_class->task_dead) prev->sched_class->task_dead(prev); - /* - * Remove function-return probe instances associated with this - * task and put them back on the free list. - */ - kprobe_flush_task(prev); - - /* Task is done with its stack. */ - put_task_stack(prev); - put_task_struct_rcu_user(prev); } @@ -3655,43 +4303,6 @@ static struct rq *finish_task_switch(struct task_struct *prev) return rq; } -#ifdef CONFIG_SMP - -/* rq->lock is NOT held, but preemption is disabled */ -static void __balance_callback(struct rq *rq) -{ - struct callback_head *head, *next; - void (*func)(struct rq *rq); - unsigned long flags; - - raw_spin_lock_irqsave(&rq->lock, flags); - head = rq->balance_callback; - rq->balance_callback = NULL; - while (head) { - func = (void (*)(struct rq *))head->func; - next = head->next; - head->next = NULL; - head = next; - - func(rq); - } - raw_spin_unlock_irqrestore(&rq->lock, flags); -} - -static inline void balance_callback(struct rq *rq) -{ - if (unlikely(rq->balance_callback)) - __balance_callback(rq); -} - -#else - -static inline void balance_callback(struct rq *rq) -{ -} - -#endif - /** * schedule_tail - first thing a freshly forked thread must call. * @prev: the thread we just switched away from. @@ -3711,7 +4322,6 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev) */ rq = finish_task_switch(prev); - balance_callback(rq); preempt_enable(); if (current->set_child_tid) @@ -4406,7 +5016,7 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) * * WARNING: must be called with preemption disabled! */ -static void __sched notrace __schedule(bool preempt) +static void __sched notrace __schedule(bool preempt, bool spinning_lock) { struct task_struct *prev, *next; unsigned long *switch_count; @@ -4459,7 +5069,7 @@ static void __sched notrace __schedule(bool preempt) * - ptrace_{,un}freeze_traced() can change ->state underneath us. */ prev_state = prev->state; - if (!preempt && prev_state) { + if ((!preempt || spinning_lock) && prev_state) { if (signal_pending_state(prev_state, prev)) { prev->state = TASK_RUNNING; } else { @@ -4494,6 +5104,7 @@ static void __sched notrace __schedule(bool preempt) next = pick_next_task(rq, prev, &rf); clear_tsk_need_resched(prev); + clear_tsk_need_resched_lazy(prev); clear_preempt_need_resched(); if (likely(prev != next)) { @@ -4519,6 +5130,7 @@ static void __sched notrace __schedule(bool preempt) */ ++*switch_count; + migrate_disable_switch(rq, prev); psi_sched_switch(prev, next, !task_on_rq_queued(prev)); trace_sched_switch(preempt, prev, next); @@ -4527,10 +5139,11 @@ static void __sched notrace __schedule(bool preempt) rq = context_switch(rq, prev, next, &rf); } else { rq->clock_update_flags &= ~(RQCF_ACT_SKIP|RQCF_REQ_SKIP); - rq_unlock_irq(rq, &rf); - } - balance_callback(rq); + rq_unpin_lock(rq, &rf); + __balance_callbacks(rq); + raw_spin_unlock_irq(&rq->lock); + } } void __noreturn do_task_dead(void) @@ -4541,7 +5154,7 @@ void __noreturn do_task_dead(void) /* Tell freezer to ignore us: */ current->flags |= PF_NOFREEZE; - __schedule(false); + __schedule(false, false); BUG(); /* Avoid "noreturn function does return" - but don't continue if BUG() is a NOP: */ @@ -4571,9 +5184,6 @@ static inline void sched_submit_work(struct task_struct *tsk) preempt_enable_no_resched(); } - if (tsk_is_pi_blocked(tsk)) - return; - /* * If we are going to sleep and we have plugged IO queued, * make sure to submit it to avoid deadlocks. @@ -4599,7 +5209,7 @@ asmlinkage __visible void __sched schedule(void) sched_submit_work(tsk); do { preempt_disable(); - __schedule(false); + __schedule(false, false); sched_preempt_enable_no_resched(); } while (need_resched()); sched_update_worker(tsk); @@ -4627,7 +5237,7 @@ void __sched schedule_idle(void) */ WARN_ON_ONCE(current->state); do { - __schedule(false); + __schedule(false, false); } while (need_resched()); } @@ -4680,7 +5290,7 @@ static void __sched notrace preempt_schedule_common(void) */ preempt_disable_notrace(); preempt_latency_start(1); - __schedule(true); + __schedule(true, false); preempt_latency_stop(1); preempt_enable_no_resched_notrace(); @@ -4691,6 +5301,30 @@ static void __sched notrace preempt_schedule_common(void) } while (need_resched()); } +#ifdef CONFIG_PREEMPT_LAZY +/* + * If TIF_NEED_RESCHED is then we allow to be scheduled away since this is + * set by a RT task. Oterwise we try to avoid beeing scheduled out as long as + * preempt_lazy_count counter >0. + */ +static __always_inline int preemptible_lazy(void) +{ + if (test_thread_flag(TIF_NEED_RESCHED)) + return 1; + if (current_thread_info()->preempt_lazy_count) + return 0; + return 1; +} + +#else + +static inline int preemptible_lazy(void) +{ + return 1; +} + +#endif + #ifdef CONFIG_PREEMPTION /* * This is the entry point to schedule() from in-kernel preemption @@ -4704,12 +5338,26 @@ asmlinkage __visible void __sched notrace preempt_schedule(void) */ if (likely(!preemptible())) return; - + if (!preemptible_lazy()) + return; preempt_schedule_common(); } NOKPROBE_SYMBOL(preempt_schedule); EXPORT_SYMBOL(preempt_schedule); +#ifdef CONFIG_PREEMPT_RT +void __sched notrace preempt_schedule_lock(void) +{ + do { + preempt_disable(); + __schedule(true, true); + sched_preempt_enable_no_resched(); + } while (need_resched()); +} +NOKPROBE_SYMBOL(preempt_schedule_lock); +EXPORT_SYMBOL(preempt_schedule_lock); +#endif + /** * preempt_schedule_notrace - preempt_schedule called by tracing * @@ -4731,6 +5379,9 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void) if (likely(!preemptible())) return; + if (!preemptible_lazy()) + return; + do { /* * Because the function tracer can trace preempt_count_sub() @@ -4753,7 +5404,7 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void) * an infinite recursion. */ prev_ctx = exception_enter(); - __schedule(true); + __schedule(true, false); exception_exit(prev_ctx); preempt_latency_stop(1); @@ -4782,7 +5433,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void) do { preempt_disable(); local_irq_enable(); - __schedule(true); + __schedule(true, false); local_irq_disable(); sched_preempt_enable_no_resched(); } while (need_resched()); @@ -4938,9 +5589,11 @@ void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task) out_unlock: /* Avoid rq from going away on us: */ preempt_disable(); - __task_rq_unlock(rq, &rf); - balance_callback(rq); + rq_unpin_lock(rq, &rf); + __balance_callbacks(rq); + raw_spin_unlock(&rq->lock); + preempt_enable(); } #else @@ -5214,6 +5867,7 @@ static int __sched_setscheduler(struct task_struct *p, int retval, oldprio, oldpolicy = -1, queued, running; int new_effective_prio, policy = attr->sched_policy; const struct sched_class *prev_class; + struct callback_head *head; struct rq_flags rf; int reset_on_fork; int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; @@ -5452,6 +6106,7 @@ static int __sched_setscheduler(struct task_struct *p, /* Avoid rq from going away on us: */ preempt_disable(); + head = splice_balance_callbacks(rq); task_rq_unlock(rq, p, &rf); if (pi) { @@ -5460,7 +6115,7 @@ static int __sched_setscheduler(struct task_struct *p, } /* Run balance callbacks after we've adjusted the PI chain: */ - balance_callback(rq); + balance_callbacks(rq, head); preempt_enable(); return 0; @@ -5955,7 +6610,7 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) } #endif again: - retval = __set_cpus_allowed_ptr(p, new_mask, true); + retval = __set_cpus_allowed_ptr(p, new_mask, SCA_CHECK); if (!retval) { cpuset_cpus_allowed(p, cpus_allowed); @@ -6538,7 +7193,7 @@ void init_idle(struct task_struct *idle, int cpu) * * And since this is boot we can forgo the serialization. */ - set_cpus_allowed_common(idle, cpumask_of(cpu)); + set_cpus_allowed_common(idle, cpumask_of(cpu), 0); #endif /* * We're having a chicken and egg problem, even though we are @@ -6565,7 +7220,9 @@ void init_idle(struct task_struct *idle, int cpu) /* Set the preempt count _outside_ the spinlocks! */ init_idle_preempt_count(idle, cpu); - +#ifdef CONFIG_HAVE_PREEMPT_LAZY + task_thread_info(idle)->preempt_lazy_count = 0; +#endif /* * The idle tasks have their own, simple scheduling class: */ @@ -6670,6 +7327,7 @@ void sched_setnuma(struct task_struct *p, int nid) #endif /* CONFIG_NUMA_BALANCING */ #ifdef CONFIG_HOTPLUG_CPU + /* * Ensure that the idle task is using init_mm right before its CPU goes * offline. @@ -6689,119 +7347,126 @@ void idle_task_exit(void) /* finish_cpu(), as ran on the BP, will clean up the active_mm state */ } -/* - * Since this CPU is going 'away' for a while, fold any nr_active delta - * we might have. Assumes we're called after migrate_tasks() so that the - * nr_active count is stable. We need to take the teardown thread which - * is calling this into account, so we hand in adjust = 1 to the load - * calculation. - * - * Also see the comment "Global load-average calculations". - */ -static void calc_load_migrate(struct rq *rq) +static int __balance_push_cpu_stop(void *arg) { - long delta = calc_load_fold_active(rq, 1); - if (delta) - atomic_long_add(delta, &calc_load_tasks); -} + struct task_struct *p = arg; + struct rq *rq = this_rq(); + struct rq_flags rf; + int cpu; -static struct task_struct *__pick_migrate_task(struct rq *rq) -{ - const struct sched_class *class; - struct task_struct *next; + raw_spin_lock_irq(&p->pi_lock); + rq_lock(rq, &rf); - for_each_class(class) { - next = class->pick_next_task(rq); - if (next) { - next->sched_class->put_prev_task(rq, next); - return next; - } - } - - /* The idle class should always have a runnable task */ - BUG(); -} - -/* - * Migrate all tasks from the rq, sleeping tasks will be migrated by - * try_to_wake_up()->select_task_rq(). - * - * Called with rq->lock held even though we'er in stop_machine() and - * there's no concurrency possible, we hold the required locks anyway - * because of lock validation efforts. - */ -static void migrate_tasks(struct rq *dead_rq, struct rq_flags *rf) -{ - struct rq *rq = dead_rq; - struct task_struct *next, *stop = rq->stop; - struct rq_flags orf = *rf; - int dest_cpu; - - /* - * Fudge the rq selection such that the below task selection loop - * doesn't get stuck on the currently eligible stop task. - * - * We're currently inside stop_machine() and the rq is either stuck - * in the stop_machine_cpu_stop() loop, or we're executing this code, - * either way we should never end up calling schedule() until we're - * done here. - */ - rq->stop = NULL; - - /* - * put_prev_task() and pick_next_task() sched - * class method both need to have an up-to-date - * value of rq->clock[_task] - */ update_rq_clock(rq); - for (;;) { - /* - * There's this thread running, bail when that's the only - * remaining thread: - */ - if (rq->nr_running == 1) - break; - - next = __pick_migrate_task(rq); - - /* - * Rules for changing task_struct::cpus_mask are holding - * both pi_lock and rq->lock, such that holding either - * stabilizes the mask. - * - * Drop rq->lock is not quite as disastrous as it usually is - * because !cpu_active at this point, which means load-balance - * will not interfere. Also, stop-machine. - */ - rq_unlock(rq, rf); - raw_spin_lock(&next->pi_lock); - rq_relock(rq, rf); - - /* - * Since we're inside stop-machine, _nothing_ should have - * changed the task, WARN if weird stuff happened, because in - * that case the above rq->lock drop is a fail too. - */ - if (WARN_ON(task_rq(next) != rq || !task_on_rq_queued(next))) { - raw_spin_unlock(&next->pi_lock); - continue; - } - - /* Find suitable destination for @next, with force if needed. */ - dest_cpu = select_fallback_rq(dead_rq->cpu, next); - rq = __migrate_task(rq, rf, next, dest_cpu); - if (rq != dead_rq) { - rq_unlock(rq, rf); - rq = dead_rq; - *rf = orf; - rq_relock(rq, rf); - } - raw_spin_unlock(&next->pi_lock); + if (task_rq(p) == rq && task_on_rq_queued(p)) { + cpu = select_fallback_rq(rq->cpu, p); + rq = __migrate_task(rq, &rf, p, cpu); } - rq->stop = stop; + rq_unlock(rq, &rf); + raw_spin_unlock_irq(&p->pi_lock); + + put_task_struct(p); + + return 0; } + +static DEFINE_PER_CPU(struct cpu_stop_work, push_work); + +/* + * Ensure we only run per-cpu kthreads once the CPU goes !active. + */ +static void balance_push(struct rq *rq) +{ + struct task_struct *push_task = rq->curr; + + lockdep_assert_held(&rq->lock); + SCHED_WARN_ON(rq->cpu != smp_processor_id()); + + /* + * Both the cpu-hotplug and stop task are in this case and are + * required to complete the hotplug process. + */ + if (is_per_cpu_kthread(push_task) || is_migration_disabled(push_task)) { + /* + * If this is the idle task on the outgoing CPU try to wake + * up the hotplug control thread which might wait for the + * last task to vanish. The rcuwait_active() check is + * accurate here because the waiter is pinned on this CPU + * and can't obviously be running in parallel. + * + * On RT kernels this also has to check whether there are + * pinned and scheduled out tasks on the runqueue. They + * need to leave the migrate disabled section first. + */ + if (!rq->nr_running && !rq_has_pinned_tasks(rq) && + rcuwait_active(&rq->hotplug_wait)) { + raw_spin_unlock(&rq->lock); + rcuwait_wake_up(&rq->hotplug_wait); + raw_spin_lock(&rq->lock); + } + return; + } + + get_task_struct(push_task); + /* + * Temporarily drop rq->lock such that we can wake-up the stop task. + * Both preemption and IRQs are still disabled. + */ + raw_spin_unlock(&rq->lock); + stop_one_cpu_nowait(rq->cpu, __balance_push_cpu_stop, push_task, + this_cpu_ptr(&push_work)); + /* + * At this point need_resched() is true and we'll take the loop in + * schedule(). The next pick is obviously going to be the stop task + * which is_per_cpu_kthread() and will push this task away. + */ + raw_spin_lock(&rq->lock); +} + +static void balance_push_set(int cpu, bool on) +{ + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; + + rq_lock_irqsave(rq, &rf); + if (on) + rq->balance_flags |= BALANCE_PUSH; + else + rq->balance_flags &= ~BALANCE_PUSH; + rq_unlock_irqrestore(rq, &rf); +} + +/* + * Invoked from a CPUs hotplug control thread after the CPU has been marked + * inactive. All tasks which are not per CPU kernel threads are either + * pushed off this CPU now via balance_push() or placed on a different CPU + * during wakeup. Wait until the CPU is quiescent. + */ +static void balance_hotplug_wait(void) +{ + struct rq *rq = this_rq(); + + rcuwait_wait_event(&rq->hotplug_wait, + rq->nr_running == 1 && !rq_has_pinned_tasks(rq), + TASK_UNINTERRUPTIBLE); +} + +#else + +static inline void balance_push(struct rq *rq) +{ +} + +static inline void balance_push_set(int cpu, bool on) +{ +} + +static inline void balance_hotplug_wait(void) +{ +} + #endif /* CONFIG_HOTPLUG_CPU */ void set_rq_online(struct rq *rq) @@ -6887,6 +7552,8 @@ int sched_cpu_activate(unsigned int cpu) struct rq *rq = cpu_rq(cpu); struct rq_flags rf; + balance_push_set(cpu, false); + #ifdef CONFIG_SCHED_SMT /* * When going up, increment the number of cores with SMT present. @@ -6922,6 +7589,8 @@ int sched_cpu_activate(unsigned int cpu) int sched_cpu_deactivate(unsigned int cpu) { + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; int ret; set_cpu_active(cpu, false); @@ -6934,6 +7603,16 @@ int sched_cpu_deactivate(unsigned int cpu) */ synchronize_rcu(); + balance_push_set(cpu, true); + + rq_lock_irqsave(rq, &rf); + if (rq->rd) { + update_rq_clock(rq); + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); + set_rq_offline(rq); + } + rq_unlock_irqrestore(rq, &rf); + #ifdef CONFIG_SCHED_SMT /* * When going down, decrement the number of cores with SMT present. @@ -6947,6 +7626,7 @@ int sched_cpu_deactivate(unsigned int cpu) ret = cpuset_cpu_inactive(cpu); if (ret) { + balance_push_set(cpu, false); set_cpu_active(cpu, true); return ret; } @@ -6970,6 +7650,41 @@ int sched_cpu_starting(unsigned int cpu) } #ifdef CONFIG_HOTPLUG_CPU + +/* + * Invoked immediately before the stopper thread is invoked to bring the + * CPU down completely. At this point all per CPU kthreads except the + * hotplug thread (current) and the stopper thread (inactive) have been + * either parked or have been unbound from the outgoing CPU. Ensure that + * any of those which might be on the way out are gone. + * + * If after this point a bound task is being woken on this CPU then the + * responsible hotplug callback has failed to do it's job. + * sched_cpu_dying() will catch it with the appropriate fireworks. + */ +int sched_cpu_wait_empty(unsigned int cpu) +{ + balance_hotplug_wait(); + return 0; +} + +/* + * Since this CPU is going 'away' for a while, fold any nr_active delta we + * might have. Called from the CPU stopper task after ensuring that the + * stopper is the last running task on the CPU, so nr_active count is + * stable. We need to take the teardown thread which is calling this into + * account, so we hand in adjust = 1 to the load calculation. + * + * Also see the comment "Global load-average calculations". + */ +static void calc_load_migrate(struct rq *rq) +{ + long delta = calc_load_fold_active(rq, 1); + + if (delta) + atomic_long_add(delta, &calc_load_tasks); +} + int sched_cpu_dying(unsigned int cpu) { struct rq *rq = cpu_rq(cpu); @@ -6979,12 +7694,7 @@ int sched_cpu_dying(unsigned int cpu) sched_tick_stop(cpu); rq_lock_irqsave(rq, &rf); - if (rq->rd) { - BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); - set_rq_offline(rq); - } - migrate_tasks(rq, &rf); - BUG_ON(rq->nr_running != 1); + BUG_ON(rq->nr_running != 1 || rq_has_pinned_tasks(rq)); rq_unlock_irqrestore(rq, &rf); calc_load_migrate(rq); @@ -7191,6 +7901,9 @@ void __init sched_init(void) rq_csd_init(rq, &rq->nohz_csd, nohz_csd_func); #endif +#ifdef CONFIG_HOTPLUG_CPU + rcuwait_init(&rq->hotplug_wait); +#endif #endif /* CONFIG_SMP */ hrtick_rq_init(rq); atomic_set(&rq->nr_iowait, 0); @@ -7231,7 +7944,7 @@ void __init sched_init(void) #ifdef CONFIG_DEBUG_ATOMIC_SLEEP static inline int preempt_count_equals(int preempt_offset) { - int nested = preempt_count() + rcu_preempt_depth(); + int nested = preempt_count() + sched_rcu_preempt_depth(); return (nested == preempt_offset); } diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c index 8cb06c8c7eb11..ceb03d76c0ccd 100644 --- a/kernel/sched/cpudeadline.c +++ b/kernel/sched/cpudeadline.c @@ -120,7 +120,7 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p, const struct sched_dl_entity *dl_se = &p->dl; if (later_mask && - cpumask_and(later_mask, cp->free_cpus, p->cpus_ptr)) { + cpumask_and(later_mask, cp->free_cpus, &p->cpus_mask)) { unsigned long cap, max_cap = 0; int cpu, max_cpu = -1; @@ -151,7 +151,7 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p, WARN_ON(best_cpu != -1 && !cpu_present(best_cpu)); - if (cpumask_test_cpu(best_cpu, p->cpus_ptr) && + if (cpumask_test_cpu(best_cpu, &p->cpus_mask) && dl_time_before(dl_se->deadline, cp->elements[0].dl)) { if (later_mask) cpumask_set_cpu(best_cpu, later_mask); diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index 0033731a07976..11c4df2010de5 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -73,11 +73,11 @@ static inline int __cpupri_find(struct cpupri *cp, struct task_struct *p, if (skip) return 0; - if (cpumask_any_and(p->cpus_ptr, vec->mask) >= nr_cpu_ids) + if (cpumask_any_and(&p->cpus_mask, vec->mask) >= nr_cpu_ids) return 0; if (lowest_mask) { - cpumask_and(lowest_mask, p->cpus_ptr, vec->mask); + cpumask_and(lowest_mask, &p->cpus_mask, vec->mask); /* * We have to ensure that we have at least one bit diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 3862a28cd05d0..6df71d487ed06 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -543,7 +543,7 @@ static int push_dl_task(struct rq *rq); static inline bool need_pull_dl_task(struct rq *rq, struct task_struct *prev) { - return dl_task(prev); + return rq->online && dl_task(prev); } static DEFINE_PER_CPU(struct callback_head, dl_push_head); @@ -1888,7 +1888,7 @@ static void task_fork_dl(struct task_struct *p) static int pick_dl_task(struct rq *rq, struct task_struct *p, int cpu) { if (!task_running(rq, p) && - cpumask_test_cpu(cpu, p->cpus_ptr)) + cpumask_test_cpu(cpu, &p->cpus_mask)) return 1; return 0; } @@ -1978,8 +1978,8 @@ static int find_later_rq(struct task_struct *task) return this_cpu; } - best_cpu = cpumask_first_and(later_mask, - sched_domain_span(sd)); + best_cpu = cpumask_any_and_distribute(later_mask, + sched_domain_span(sd)); /* * Last chance: if a CPU being in both later_mask * and current sd span is valid, that becomes our @@ -2001,7 +2001,7 @@ static int find_later_rq(struct task_struct *task) if (this_cpu != -1) return this_cpu; - cpu = cpumask_any(later_mask); + cpu = cpumask_any_distribute(later_mask); if (cpu < nr_cpu_ids) return cpu; @@ -2038,7 +2038,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) /* Retry if something changed. */ if (double_lock_balance(rq, later_rq)) { if (unlikely(task_rq(task) != rq || - !cpumask_test_cpu(later_rq->cpu, task->cpus_ptr) || + !cpumask_test_cpu(later_rq->cpu, &task->cpus_mask) || task_running(rq, task) || !dl_task(task) || !task_on_rq_queued(task))) { @@ -2105,6 +2105,9 @@ static int push_dl_task(struct rq *rq) return 0; retry: + if (is_migration_disabled(next_task)) + return 0; + if (WARN_ON(next_task == rq->curr)) return 0; @@ -2182,7 +2185,7 @@ static void push_dl_tasks(struct rq *rq) static void pull_dl_task(struct rq *this_rq) { int this_cpu = this_rq->cpu, cpu; - struct task_struct *p; + struct task_struct *p, *push_task; bool resched = false; struct rq *src_rq; u64 dmin = LONG_MAX; @@ -2212,6 +2215,7 @@ static void pull_dl_task(struct rq *this_rq) continue; /* Might drop this_rq->lock */ + push_task = NULL; double_lock_balance(this_rq, src_rq); /* @@ -2243,17 +2247,28 @@ static void pull_dl_task(struct rq *this_rq) src_rq->curr->dl.deadline)) goto skip; - resched = true; - - deactivate_task(src_rq, p, 0); - set_task_cpu(p, this_cpu); - activate_task(this_rq, p, 0); - dmin = p->dl.deadline; + if (is_migration_disabled(p)) { + trace_sched_migrate_pull_tp(p); + push_task = get_push_task(src_rq); + } else { + deactivate_task(src_rq, p, 0); + set_task_cpu(p, this_cpu); + activate_task(this_rq, p, 0); + dmin = p->dl.deadline; + resched = true; + } /* Is there any other task even earlier? */ } skip: double_unlock_balance(this_rq, src_rq); + + if (push_task) { + raw_spin_unlock(&this_rq->lock); + stop_one_cpu_nowait(src_rq->cpu, push_cpu_stop, + push_task, &src_rq->push_work); + raw_spin_lock(&this_rq->lock); + } } if (resched) @@ -2277,7 +2292,8 @@ static void task_woken_dl(struct rq *rq, struct task_struct *p) } static void set_cpus_allowed_dl(struct task_struct *p, - const struct cpumask *new_mask) + const struct cpumask *new_mask, + u32 flags) { struct root_domain *src_rd; struct rq *rq; @@ -2306,7 +2322,7 @@ static void set_cpus_allowed_dl(struct task_struct *p, raw_spin_unlock(&src_dl_b->lock); } - set_cpus_allowed_common(p, new_mask); + set_cpus_allowed_common(p, new_mask, flags); } /* Assumes rq->lock is held */ @@ -2499,6 +2515,7 @@ const struct sched_class dl_sched_class .rq_online = rq_online_dl, .rq_offline = rq_offline_dl, .task_woken = task_woken_dl, + .find_lock_rq = find_lock_later_rq, #endif .task_tick = task_tick_dl, diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 1a68a0536adda..d31aab136644c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4357,7 +4357,7 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) ideal_runtime = sched_slice(cfs_rq, curr); delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime; if (delta_exec > ideal_runtime) { - resched_curr(rq_of(cfs_rq)); + resched_curr_lazy(rq_of(cfs_rq)); /* * The current task ran long enough, ensure it doesn't get * re-elected due to buddy favours. @@ -4381,7 +4381,7 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) return; if (delta > ideal_runtime) - resched_curr(rq_of(cfs_rq)); + resched_curr_lazy(rq_of(cfs_rq)); } static void @@ -4524,7 +4524,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) * validating it and just reschedule. */ if (queued) { - resched_curr(rq_of(cfs_rq)); + resched_curr_lazy(rq_of(cfs_rq)); return; } /* @@ -4661,7 +4661,7 @@ static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec) * hierarchy can be throttled */ if (!assign_cfs_rq_runtime(cfs_rq) && likely(cfs_rq->curr)) - resched_curr(rq_of(cfs_rq)); + resched_curr_lazy(rq_of(cfs_rq)); } static __always_inline @@ -5396,7 +5396,7 @@ static void hrtick_start_fair(struct rq *rq, struct task_struct *p) if (delta < 0) { if (rq->curr == p) - resched_curr(rq); + resched_curr_lazy(rq); return; } hrtick_start(rq, delta); @@ -6953,7 +6953,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ return; preempt: - resched_curr(rq); + resched_curr_lazy(rq); /* * Only set the backward buddy when the current task is still * on the rq. This can happen when a wakeup gets interleaved @@ -10694,7 +10694,7 @@ static void task_fork_fair(struct task_struct *p) * 'current' within the tree based on its new key value. */ swap(curr->vruntime, se->vruntime); - resched_curr(rq); + resched_curr_lazy(rq); } se->vruntime -= cfs_rq->min_vruntime; @@ -10721,7 +10721,7 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) */ if (rq->curr == p) { if (p->prio > oldprio) - resched_curr(rq); + resched_curr_lazy(rq); } else check_preempt_curr(rq, p, 0); } diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 7481cd96f3915..862e8b3c7d280 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -45,11 +45,19 @@ SCHED_FEAT(DOUBLE_TICK, false) */ SCHED_FEAT(NONTASK_CAPACITY, true) +#ifdef CONFIG_PREEMPT_RT +SCHED_FEAT(TTWU_QUEUE, false) +# ifdef CONFIG_PREEMPT_LAZY +SCHED_FEAT(PREEMPT_LAZY, true) +# endif +#else + /* * Queue remote wakeups on the target CPU and process them * using the scheduler IPI. Reduces rq->lock contention/bounces. */ SCHED_FEAT(TTWU_QUEUE, true) +#endif /* * When doing wakeups, attempt to limit superfluous scans of the LLC domain. diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index f215eea6a9661..03f7b397716dd 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -265,7 +265,7 @@ static void pull_rt_task(struct rq *this_rq); static inline bool need_pull_rt_task(struct rq *rq, struct task_struct *prev) { /* Try to pull RT tasks here if we lower this rq's prio */ - return rq->rt.highest_prio.curr > prev->prio; + return rq->online && rq->rt.highest_prio.curr > prev->prio; } static inline int rt_overloaded(struct rq *rq) @@ -1658,7 +1658,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu) { if (!task_running(rq, p) && - cpumask_test_cpu(cpu, p->cpus_ptr)) + cpumask_test_cpu(cpu, &p->cpus_mask)) return 1; return 0; @@ -1752,8 +1752,8 @@ static int find_lowest_rq(struct task_struct *task) return this_cpu; } - best_cpu = cpumask_first_and(lowest_mask, - sched_domain_span(sd)); + best_cpu = cpumask_any_and_distribute(lowest_mask, + sched_domain_span(sd)); if (best_cpu < nr_cpu_ids) { rcu_read_unlock(); return best_cpu; @@ -1770,7 +1770,7 @@ static int find_lowest_rq(struct task_struct *task) if (this_cpu != -1) return this_cpu; - cpu = cpumask_any(lowest_mask); + cpu = cpumask_any_distribute(lowest_mask); if (cpu < nr_cpu_ids) return cpu; @@ -1811,7 +1811,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) * Also make sure that it wasn't scheduled on its rq. */ if (unlikely(task_rq(task) != rq || - !cpumask_test_cpu(lowest_rq->cpu, task->cpus_ptr) || + !cpumask_test_cpu(lowest_rq->cpu, &task->cpus_mask) || task_running(rq, task) || !rt_task(task) || !task_on_rq_queued(task))) { @@ -1859,7 +1859,7 @@ static struct task_struct *pick_next_pushable_task(struct rq *rq) * running task can migrate over to a CPU that is running a task * of lesser priority. */ -static int push_rt_task(struct rq *rq) +static int push_rt_task(struct rq *rq, bool pull) { struct task_struct *next_task; struct rq *lowest_rq; @@ -1873,6 +1873,39 @@ static int push_rt_task(struct rq *rq) return 0; retry: + if (is_migration_disabled(next_task)) { + struct task_struct *push_task = NULL; + int cpu; + + if (!pull) + return 0; + + trace_sched_migrate_pull_tp(next_task); + + if (rq->push_busy) + return 0; + + cpu = find_lowest_rq(rq->curr); + if (cpu == -1 || cpu == rq->cpu) + return 0; + + /* + * Given we found a CPU with lower priority than @next_task, + * therefore it should be running. However we cannot migrate it + * to this other CPU, instead attempt to push the current + * running task on this CPU away. + */ + push_task = get_push_task(rq); + if (push_task) { + raw_spin_unlock(&rq->lock); + stop_one_cpu_nowait(rq->cpu, push_cpu_stop, + push_task, &rq->push_work); + raw_spin_lock(&rq->lock); + } + + return 0; + } + if (WARN_ON(next_task == rq->curr)) return 0; @@ -1927,12 +1960,10 @@ static int push_rt_task(struct rq *rq) deactivate_task(rq, next_task, 0); set_task_cpu(next_task, lowest_rq->cpu); activate_task(lowest_rq, next_task, 0); + resched_curr(lowest_rq); ret = 1; - resched_curr(lowest_rq); - double_unlock_balance(rq, lowest_rq); - out: put_task_struct(next_task); @@ -1942,7 +1973,7 @@ static int push_rt_task(struct rq *rq) static void push_rt_tasks(struct rq *rq) { /* push_rt_task will return true if it moved an RT */ - while (push_rt_task(rq)) + while (push_rt_task(rq, false)) ; } @@ -2095,7 +2126,8 @@ void rto_push_irq_work_func(struct irq_work *work) */ if (has_pushable_tasks(rq)) { raw_spin_lock(&rq->lock); - push_rt_tasks(rq); + while (push_rt_task(rq, true)) + ; raw_spin_unlock(&rq->lock); } @@ -2120,7 +2152,7 @@ static void pull_rt_task(struct rq *this_rq) { int this_cpu = this_rq->cpu, cpu; bool resched = false; - struct task_struct *p; + struct task_struct *p, *push_task; struct rq *src_rq; int rt_overload_count = rt_overloaded(this_rq); @@ -2167,6 +2199,7 @@ static void pull_rt_task(struct rq *this_rq) * double_lock_balance, and another CPU could * alter this_rq */ + push_task = NULL; double_lock_balance(this_rq, src_rq); /* @@ -2194,11 +2227,15 @@ static void pull_rt_task(struct rq *this_rq) if (p->prio < src_rq->curr->prio) goto skip; - resched = true; - - deactivate_task(src_rq, p, 0); - set_task_cpu(p, this_cpu); - activate_task(this_rq, p, 0); + if (is_migration_disabled(p)) { + trace_sched_migrate_pull_tp(p); + push_task = get_push_task(src_rq); + } else { + deactivate_task(src_rq, p, 0); + set_task_cpu(p, this_cpu); + activate_task(this_rq, p, 0); + resched = true; + } /* * We continue with the search, just in * case there's an even higher prio task @@ -2208,6 +2245,13 @@ static void pull_rt_task(struct rq *this_rq) } skip: double_unlock_balance(this_rq, src_rq); + + if (push_task) { + raw_spin_unlock(&this_rq->lock); + stop_one_cpu_nowait(src_rq->cpu, push_cpu_stop, + push_task, &src_rq->push_work); + raw_spin_lock(&this_rq->lock); + } } if (resched) @@ -2449,6 +2493,7 @@ const struct sched_class rt_sched_class .rq_offline = rq_offline_rt, .task_woken = task_woken_rt, .switched_from = switched_from_rt, + .find_lock_rq = find_lock_lowest_rq, #endif .task_tick = task_tick_rt, diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 28709f6b0975c..19847e4ae1323 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -973,6 +973,7 @@ struct rq { unsigned long cpu_capacity_orig; struct callback_head *balance_callback; + unsigned char balance_flags; unsigned char nohz_idle_balance; unsigned char idle_balance; @@ -1003,6 +1004,10 @@ struct rq { /* This is used to determine avg_idle's max value */ u64 max_idle_balance_cost; + +#ifdef CONFIG_HOTPLUG_CPU + struct rcuwait hotplug_wait; +#endif #endif /* CONFIG_SMP */ #ifdef CONFIG_IRQ_TIME_ACCOUNTING @@ -1048,6 +1053,12 @@ struct rq { /* Must be inspected within a rcu lock section */ struct cpuidle_state *idle_state; #endif + +#if defined(CONFIG_PREEMPT_RT) && defined(CONFIG_SMP) + unsigned int nr_pinned; +#endif + unsigned int push_busy; + struct cpu_stop_work push_work; }; #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1075,6 +1086,16 @@ static inline int cpu_of(struct rq *rq) #endif } +#define MDF_PUSH 0x01 + +static inline bool is_migration_disabled(struct task_struct *p) +{ +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + return p->migration_disabled; +#else + return false; +#endif +} #ifdef CONFIG_SCHED_SMT extern void __update_idle_core(struct rq *rq); @@ -1221,6 +1242,9 @@ static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf) rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP); rf->clock_update_flags = 0; #endif +#ifdef CONFIG_SMP + SCHED_WARN_ON(rq->balance_callback); +#endif } static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf) @@ -1382,6 +1406,9 @@ init_numa_balancing(unsigned long clone_flags, struct task_struct *p) #ifdef CONFIG_SMP +#define BALANCE_WORK 0x01 +#define BALANCE_PUSH 0x02 + static inline void queue_balance_callback(struct rq *rq, struct callback_head *head, @@ -1389,12 +1416,13 @@ queue_balance_callback(struct rq *rq, { lockdep_assert_held(&rq->lock); - if (unlikely(head->next)) + if (unlikely(head->next || (rq->balance_flags & BALANCE_PUSH))) return; head->func = (void (*)(struct callback_head *))func; head->next = rq->balance_callback; rq->balance_callback = head; + rq->balance_flags |= BALANCE_WORK; } #define rcu_dereference_check_sched_domain(p) \ @@ -1714,6 +1742,7 @@ static inline int task_on_rq_migrating(struct task_struct *p) #define WF_FORK 0x02 /* Child wakeup after fork */ #define WF_MIGRATED 0x04 /* Internal use, task got migrated */ #define WF_ON_CPU 0x08 /* Wakee is on_cpu */ +#define WF_LOCK_SLEEPER 0x10 /* Wakeup spinlock "sleeper" */ /* * To aid in avoiding the subversion of "niceness" due to uneven distribution @@ -1795,10 +1824,13 @@ struct sched_class { void (*task_woken)(struct rq *this_rq, struct task_struct *task); void (*set_cpus_allowed)(struct task_struct *p, - const struct cpumask *newmask); + const struct cpumask *newmask, + u32 flags); void (*rq_online)(struct rq *rq); void (*rq_offline)(struct rq *rq); + + struct rq *(*find_lock_rq)(struct task_struct *p, struct rq *rq); #endif void (*task_tick)(struct rq *rq, struct task_struct *p, int queued); @@ -1882,13 +1914,35 @@ static inline bool sched_fair_runnable(struct rq *rq) extern struct task_struct *pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf); extern struct task_struct *pick_next_task_idle(struct rq *rq); +#define SCA_CHECK 0x01 +#define SCA_MIGRATE_DISABLE 0x02 +#define SCA_MIGRATE_ENABLE 0x04 + #ifdef CONFIG_SMP extern void update_group_capacity(struct sched_domain *sd, int cpu); extern void trigger_load_balance(struct rq *rq); -extern void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask); +extern void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask, u32 flags); + +static inline struct task_struct *get_push_task(struct rq *rq) +{ + struct task_struct *p = rq->curr; + + lockdep_assert_held(&rq->lock); + + if (rq->push_busy) + return NULL; + + if (p->nr_cpus_allowed == 1) + return NULL; + + rq->push_busy = true; + return get_task_struct(p); +} + +extern int push_cpu_stop(void *arg); #endif @@ -1932,6 +1986,15 @@ extern void reweight_task(struct task_struct *p, int prio); extern void resched_curr(struct rq *rq); extern void resched_cpu(int cpu); +#ifdef CONFIG_PREEMPT_LAZY +extern void resched_curr_lazy(struct rq *rq); +#else +static inline void resched_curr_lazy(struct rq *rq) +{ + resched_curr(rq); +} +#endif + extern struct rt_bandwidth def_rt_bandwidth; extern void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime); diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c index e1c655f928c74..f230b1ac7f910 100644 --- a/kernel/sched/swait.c +++ b/kernel/sched/swait.c @@ -64,6 +64,7 @@ void swake_up_all(struct swait_queue_head *q) struct swait_queue *curr; LIST_HEAD(tmp); + WARN_ON(irqs_disabled()); raw_spin_lock_irq(&q->lock); list_splice_init(&q->task_list, &tmp); while (!list_empty(&tmp)) { diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index 1bd7e3af904f6..04ead8eeb07c6 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -500,6 +500,7 @@ static int init_rootdomain(struct root_domain *rd) rd->rto_cpu = -1; raw_spin_lock_init(&rd->rto_lock); init_irq_work(&rd->rto_push_work, rto_push_irq_work_func); + atomic_or(IRQ_WORK_HARD_IRQ, &rd->rto_push_work.flags); #endif init_dl_bw(&rd->dl_bw); diff --git a/kernel/signal.c b/kernel/signal.c index a38b3edc68511..c3d7abd1f6aea 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include @@ -403,13 +404,30 @@ void task_join_group_stop(struct task_struct *task) } } +static inline struct sigqueue *get_task_cache(struct task_struct *t) +{ + struct sigqueue *q = t->sigqueue_cache; + + if (cmpxchg(&t->sigqueue_cache, q, NULL) != q) + return NULL; + return q; +} + +static inline int put_task_cache(struct task_struct *t, struct sigqueue *q) +{ + if (cmpxchg(&t->sigqueue_cache, NULL, q) == NULL) + return 0; + return 1; +} + /* * allocate a new signal queue record * - this may be called without locks if and only if t == current, otherwise an * appropriate lock must be held to stop the target task from exiting */ static struct sigqueue * -__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimit) +__sigqueue_do_alloc(int sig, struct task_struct *t, gfp_t flags, + int override_rlimit, int fromslab) { struct sigqueue *q = NULL; struct user_struct *user; @@ -431,7 +449,10 @@ __sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimi rcu_read_unlock(); if (override_rlimit || likely(sigpending <= task_rlimit(t, RLIMIT_SIGPENDING))) { - q = kmem_cache_alloc(sigqueue_cachep, flags); + if (!fromslab) + q = get_task_cache(t); + if (!q) + q = kmem_cache_alloc(sigqueue_cachep, flags); } else { print_dropped_signal(sig); } @@ -448,6 +469,13 @@ __sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimi return q; } +static struct sigqueue * +__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, + int override_rlimit) +{ + return __sigqueue_do_alloc(sig, t, flags, override_rlimit, 0); +} + static void __sigqueue_free(struct sigqueue *q) { if (q->flags & SIGQUEUE_PREALLOC) @@ -457,6 +485,21 @@ static void __sigqueue_free(struct sigqueue *q) kmem_cache_free(sigqueue_cachep, q); } +static void sigqueue_free_current(struct sigqueue *q) +{ + struct user_struct *up; + + if (q->flags & SIGQUEUE_PREALLOC) + return; + + up = q->user; + if (rt_prio(current->normal_prio) && !put_task_cache(current, q)) { + if (atomic_dec_and_test(&up->sigpending)) + free_uid(up); + } else + __sigqueue_free(q); +} + void flush_sigqueue(struct sigpending *queue) { struct sigqueue *q; @@ -469,6 +512,21 @@ void flush_sigqueue(struct sigpending *queue) } } +/* + * Called from __exit_signal. Flush tsk->pending and + * tsk->sigqueue_cache + */ +void flush_task_sigqueue(struct task_struct *tsk) +{ + struct sigqueue *q; + + flush_sigqueue(&tsk->pending); + + q = get_task_cache(tsk); + if (q) + kmem_cache_free(sigqueue_cachep, q); +} + /* * Flush all pending signals for this kthread. */ @@ -593,7 +651,7 @@ static void collect_signal(int sig, struct sigpending *list, kernel_siginfo_t *i (info->si_code == SI_TIMER) && (info->si_sys_private); - __sigqueue_free(first); + sigqueue_free_current(first); } else { /* * Ok, it wasn't in the queue. This must be @@ -630,6 +688,8 @@ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, kernel_siginfo_t *in bool resched_timer = false; int signr; + WARN_ON_ONCE(tsk != current); + /* We only dequeue private signals from ourselves, we don't let * signalfd steal them */ @@ -1313,6 +1373,34 @@ force_sig_info_to_task(struct kernel_siginfo *info, struct task_struct *t) struct k_sigaction *action; int sig = info->si_signo; + /* + * On some archs, PREEMPT_RT has to delay sending a signal from a trap + * since it can not enable preemption, and the signal code's spin_locks + * turn into mutexes. Instead, it must set TIF_NOTIFY_RESUME which will + * send the signal on exit of the trap. + */ +#ifdef ARCH_RT_DELAYS_SIGNAL_SEND + if (in_atomic()) { + struct task_struct *t = current; + + if (WARN_ON_ONCE(t->forced_info.si_signo)) + return 0; + + if (is_si_special(info)) { + WARN_ON_ONCE(info != SEND_SIG_PRIV); + t->forced_info.si_signo = info->si_signo; + t->forced_info.si_errno = 0; + t->forced_info.si_code = SI_KERNEL; + t->forced_info.si_pid = 0; + t->forced_info.si_uid = 0; + } else { + t->forced_info = *info; + } + + set_tsk_thread_flag(t, TIF_NOTIFY_RESUME); + return 0; + } +#endif spin_lock_irqsave(&t->sighand->siglock, flags); action = &t->sighand->action[sig-1]; ignored = action->sa.sa_handler == SIG_IGN; @@ -1806,7 +1894,8 @@ EXPORT_SYMBOL(kill_pid); */ struct sigqueue *sigqueue_alloc(void) { - struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0); + /* Preallocated sigqueue objects always from the slabcache ! */ + struct sigqueue *q = __sigqueue_do_alloc(-1, current, GFP_KERNEL, 0, 1); if (q) q->flags |= SIGQUEUE_PREALLOC; @@ -2202,16 +2291,8 @@ static void ptrace_stop(int exit_code, int why, int clear_code, kernel_siginfo_t if (gstop_done && ptrace_reparented(current)) do_notify_parent_cldstop(current, false, why); - /* - * Don't want to allow preemption here, because - * sys_ptrace() needs this task to be inactive. - * - * XXX: implement read_unlock_no_resched(). - */ - preempt_disable(); read_unlock(&tasklist_lock); cgroup_enter_frozen(); - preempt_enable_no_resched(); freezable_schedule(); cgroup_leave_frozen(true); } else { diff --git a/kernel/softirq.c b/kernel/softirq.c index bf88d7f624332..102f5553884c5 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -92,15 +93,225 @@ static bool ksoftirqd_running(unsigned long pending) !__kthread_should_park(tsk); } +#ifdef CONFIG_TRACE_IRQFLAGS +DEFINE_PER_CPU(int, hardirqs_enabled); +DEFINE_PER_CPU(int, hardirq_context); +EXPORT_PER_CPU_SYMBOL_GPL(hardirqs_enabled); +EXPORT_PER_CPU_SYMBOL_GPL(hardirq_context); +#endif + /* - * preempt_count and SOFTIRQ_OFFSET usage: - * - preempt_count is changed by SOFTIRQ_OFFSET on entering or leaving - * softirq processing. - * - preempt_count is changed by SOFTIRQ_DISABLE_OFFSET (= 2 * SOFTIRQ_OFFSET) + * SOFTIRQ_OFFSET usage: + * + * On !RT kernels 'count' is the preempt counter, on RT kernels this applies + * to a per CPU counter and to task::softirqs_disabled_cnt. + * + * - count is changed by SOFTIRQ_OFFSET on entering or leaving softirq + * processing. + * + * - count is changed by SOFTIRQ_DISABLE_OFFSET (= 2 * SOFTIRQ_OFFSET) * on local_bh_disable or local_bh_enable. + * * This lets us distinguish between whether we are currently processing * softirq and whether we just have bh disabled. */ +#ifdef CONFIG_PREEMPT_RT + +/* + * RT accounts for BH disabled sections in task::softirqs_disabled_cnt and + * also in per CPU softirq_ctrl::cnt. This is necessary to allow tasks in a + * softirq disabled section to be preempted. + * + * The per task counter is used for softirq_count(), in_softirq() and + * in_serving_softirqs() because these counts are only valid when the task + * holding softirq_ctrl::lock is running. + * + * The per CPU counter prevents pointless wakeups of ksoftirqd in case that + * the task which is in a softirq disabled section is preempted or blocks. + */ +struct softirq_ctrl { + local_lock_t lock; + int cnt; +}; + +static DEFINE_PER_CPU(struct softirq_ctrl, softirq_ctrl) = { + .lock = INIT_LOCAL_LOCK(softirq_ctrl.lock), +}; + +/** + * local_bh_blocked() - Check for idle whether BH processing is blocked + * + * Returns false if the per CPU softirq::cnt is 0 otherwise true. + * + * This is invoked from the idle task to guard against false positive + * softirq pending warnings, which would happen when the task which holds + * softirq_ctrl::lock was the only running task on the CPU and blocks on + * some other lock. + */ +bool local_bh_blocked(void) +{ + return this_cpu_read(softirq_ctrl.cnt) != 0; +} + +void __local_bh_disable_ip(unsigned long ip, unsigned int cnt) +{ + unsigned long flags; + int newcnt; + + WARN_ON_ONCE(in_irq()); + + /* First entry of a task into a BH disabled section? */ + if (!current->softirq_disable_cnt) { + if (preemptible()) { + local_lock(&softirq_ctrl.lock); + rcu_read_lock(); + } else { + DEBUG_LOCKS_WARN_ON(this_cpu_read(softirq_ctrl.cnt)); + } + } + + preempt_disable(); + /* + * Track the per CPU softirq disabled state. On RT this is per CPU + * state to allow preemption of bottom half disabled sections. + */ + newcnt = this_cpu_add_return(softirq_ctrl.cnt, cnt); + /* + * Reflect the result in the task state to prevent recursion on the + * local lock and to make softirq_count() & al work. + */ + current->softirq_disable_cnt = newcnt; + + if (IS_ENABLED(CONFIG_TRACE_IRQFLAGS) && newcnt == cnt) { + raw_local_irq_save(flags); + lockdep_softirqs_off(ip); + raw_local_irq_restore(flags); + } + preempt_enable(); +} +EXPORT_SYMBOL(__local_bh_disable_ip); + +static void __local_bh_enable(unsigned int cnt, bool unlock) +{ + unsigned long flags; + int newcnt; + + DEBUG_LOCKS_WARN_ON(current->softirq_disable_cnt != + this_cpu_read(softirq_ctrl.cnt)); + + preempt_disable(); + if (IS_ENABLED(CONFIG_TRACE_IRQFLAGS) && softirq_count() == cnt) { + raw_local_irq_save(flags); + lockdep_softirqs_on(_RET_IP_); + raw_local_irq_restore(flags); + } + + newcnt = this_cpu_sub_return(softirq_ctrl.cnt, cnt); + current->softirq_disable_cnt = newcnt; + preempt_enable(); + + if (!newcnt && unlock) { + rcu_read_unlock(); + local_unlock(&softirq_ctrl.lock); + } +} + +void __local_bh_enable_ip(unsigned long ip, unsigned int cnt) +{ + bool preempt_on = preemptible(); + unsigned long flags; + u32 pending; + int curcnt; + + WARN_ON_ONCE(in_irq()); + lockdep_assert_irqs_enabled(); + + local_irq_save(flags); + curcnt = this_cpu_read(softirq_ctrl.cnt); + + /* + * If this is not reenabling soft interrupts, no point in trying to + * run pending ones. + */ + if (curcnt != cnt) + goto out; + + pending = local_softirq_pending(); + if (!pending || ksoftirqd_running(pending)) + goto out; + + /* + * If this was called from non preemptible context, wake up the + * softirq daemon. + */ + if (!preempt_on) { + wakeup_softirqd(); + goto out; + } + + /* + * Adjust softirq count to SOFTIRQ_OFFSET which makes + * in_serving_softirq() become true. + */ + cnt = SOFTIRQ_OFFSET; + __local_bh_enable(cnt, false); + __do_softirq(); + +out: + __local_bh_enable(cnt, preempt_on); + local_irq_restore(flags); +} +EXPORT_SYMBOL(__local_bh_enable_ip); + +/* + * Invoked from irq_enter_rcu() to prevent that tick_irq_enter() + * pointlessly wakes the softirq daemon. That's handled in __irq_exit_rcu(). + * None of the above logic in the regular bh_disable/enable functions is + * required here. + */ +static inline void local_bh_disable_irq_enter(void) +{ + this_cpu_add(softirq_ctrl.cnt, SOFTIRQ_DISABLE_OFFSET); +} + +static inline void local_bh_enable_irq_enter(void) +{ + this_cpu_sub(softirq_ctrl.cnt, SOFTIRQ_DISABLE_OFFSET); +} + +/* + * Invoked from ksoftirqd_run() outside of the interrupt disabled section + * to acquire the per CPU local lock for reentrancy protection. + */ +static inline void ksoftirqd_run_begin(void) +{ + __local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET); + local_irq_disable(); +} + +/* Counterpart to ksoftirqd_run_begin() */ +static inline void ksoftirqd_run_end(void) +{ + __local_bh_enable(SOFTIRQ_OFFSET, true); + WARN_ON_ONCE(in_interrupt()); + local_irq_enable(); +} + +static inline void softirq_handle_begin(void) { } +static inline void softirq_handle_end(void) { } + +static inline void invoke_softirq(void) +{ + if (!this_cpu_read(softirq_ctrl.cnt)) + wakeup_softirqd(); +} + +static inline bool should_wake_ksoftirqd(void) +{ + return !this_cpu_read(softirq_ctrl.cnt); +} + +#else /* CONFIG_PREEMPT_RT */ /* * This one is for softirq.c-internal use, @@ -108,11 +319,6 @@ static bool ksoftirqd_running(unsigned long pending) */ #ifdef CONFIG_TRACE_IRQFLAGS -DEFINE_PER_CPU(int, hardirqs_enabled); -DEFINE_PER_CPU(int, hardirq_context); -EXPORT_PER_CPU_SYMBOL_GPL(hardirqs_enabled); -EXPORT_PER_CPU_SYMBOL_GPL(hardirq_context); - void __local_bh_disable_ip(unsigned long ip, unsigned int cnt) { unsigned long flags; @@ -203,6 +409,67 @@ void __local_bh_enable_ip(unsigned long ip, unsigned int cnt) } EXPORT_SYMBOL(__local_bh_enable_ip); +static inline void local_bh_disable_irq_enter(void) +{ + local_bh_disable(); +} + +static inline void local_bh_enable_irq_enter(void) +{ + _local_bh_enable(); +} + +static inline void softirq_handle_begin(void) +{ + __local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET); +} + +static inline void softirq_handle_end(void) +{ + __local_bh_enable(SOFTIRQ_OFFSET); + WARN_ON_ONCE(in_interrupt()); +} + +static inline void ksoftirqd_run_begin(void) +{ + local_irq_disable(); +} + +static inline void ksoftirqd_run_end(void) +{ + local_irq_enable(); +} + +static inline void invoke_softirq(void) +{ + if (ksoftirqd_running(local_softirq_pending())) + return; + + if (!force_irqthreads) { +#ifdef CONFIG_HAVE_IRQ_EXIT_ON_IRQ_STACK + /* + * We can safely execute softirq on the current stack if + * it is the irq stack, because it should be near empty + * at this stage. + */ + __do_softirq(); +#else + /* + * Otherwise, irq_exit() is called on the task stack that can + * be potentially deep already. So call softirq in its own stack + * to prevent from any overrun. + */ + do_softirq_own_stack(); +#endif + } else { + wakeup_softirqd(); + } +} + +static inline bool should_wake_ksoftirqd(void) { return true; } + +#endif /* !CONFIG_PREEMPT_RT */ + /* * We restart softirq processing for at most MAX_SOFTIRQ_RESTART times, * but break the loop if need_resched() is set or after 2 ms. @@ -272,7 +539,7 @@ asmlinkage __visible void __softirq_entry __do_softirq(void) pending = local_softirq_pending(); account_irq_enter_time(current); - __local_bh_disable_ip(_RET_IP_, SOFTIRQ_OFFSET); + softirq_handle_begin(); in_hardirq = lockdep_softirq_start(); restart: @@ -307,8 +574,10 @@ asmlinkage __visible void __softirq_entry __do_softirq(void) pending >>= softirq_bit; } - if (__this_cpu_read(ksoftirqd) == current) + if (!IS_ENABLED(CONFIG_PREEMPT_RT) && + __this_cpu_read(ksoftirqd) == current) rcu_softirq_qs(); + local_irq_disable(); pending = local_softirq_pending(); @@ -322,11 +591,11 @@ asmlinkage __visible void __softirq_entry __do_softirq(void) lockdep_softirq_end(in_hardirq); account_irq_exit_time(current); - __local_bh_enable(SOFTIRQ_OFFSET); - WARN_ON_ONCE(in_interrupt()); + softirq_handle_end(); current_restore_flags(old_flags, PF_MEMALLOC); } +#ifndef CONFIG_PREEMPT_RT asmlinkage __visible void do_softirq(void) { __u32 pending; @@ -344,6 +613,7 @@ asmlinkage __visible void do_softirq(void) local_irq_restore(flags); } +#endif /** * irq_enter_rcu - Enter an interrupt context with RCU watching @@ -355,9 +625,9 @@ void irq_enter_rcu(void) * Prevent raise_softirq from needlessly waking up ksoftirqd * here, as softirq will be serviced on return from interrupt. */ - local_bh_disable(); + local_bh_disable_irq_enter(); tick_irq_enter(); - _local_bh_enable(); + local_bh_enable_irq_enter(); } __irq_enter(); } @@ -371,32 +641,6 @@ void irq_enter(void) irq_enter_rcu(); } -static inline void invoke_softirq(void) -{ - if (ksoftirqd_running(local_softirq_pending())) - return; - - if (!force_irqthreads) { -#ifdef CONFIG_HAVE_IRQ_EXIT_ON_IRQ_STACK - /* - * We can safely execute softirq on the current stack if - * it is the irq stack, because it should be near empty - * at this stage. - */ - __do_softirq(); -#else - /* - * Otherwise, irq_exit() is called on the task stack that can - * be potentially deep already. So call softirq in its own stack - * to prevent from any overrun. - */ - do_softirq_own_stack(); -#endif - } else { - wakeup_softirqd(); - } -} - static inline void tick_irq_exit(void) { #ifdef CONFIG_NO_HZ_COMMON @@ -466,7 +710,7 @@ inline void raise_softirq_irqoff(unsigned int nr) * Otherwise we wake up ksoftirqd to make sure we * schedule the softirq soon. */ - if (!in_interrupt()) + if (!in_interrupt() && should_wake_ksoftirqd()) wakeup_softirqd(); } @@ -606,6 +850,29 @@ void tasklet_init(struct tasklet_struct *t, } EXPORT_SYMBOL(tasklet_init); +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT) + +void tasklet_unlock_wait(struct tasklet_struct *t) +{ + while (test_bit(TASKLET_STATE_RUN, &(t)->state)) { + if (IS_ENABLED(CONFIG_PREEMPT_RT)) { + /* + * Prevent a live lock when current preempted soft + * interrupt processing or prevents ksoftirqd from + * running. If the tasklet runs on a different CPU + * then this has no effect other than doing the BH + * disable/enable dance for nothing. + */ + local_bh_disable(); + local_bh_enable(); + } else { + cpu_relax(); + } + } +} +EXPORT_SYMBOL(tasklet_unlock_wait); +#endif + void tasklet_kill(struct tasklet_struct *t) { if (in_interrupt()) @@ -613,7 +880,20 @@ void tasklet_kill(struct tasklet_struct *t) while (test_and_set_bit(TASKLET_STATE_SCHED, &t->state)) { do { - yield(); + if (IS_ENABLED(CONFIG_PREEMPT_RT)) { + /* + * Prevent a live lock when current + * preempted soft interrupt processing or + * prevents ksoftirqd from running. If the + * tasklet runs on a different CPU then + * this has no effect other than doing the + * BH disable/enable dance for nothing. + */ + local_bh_disable(); + local_bh_enable(); + } else { + yield(); + } } while (test_bit(TASKLET_STATE_SCHED, &t->state)); } tasklet_unlock_wait(t); @@ -643,18 +923,18 @@ static int ksoftirqd_should_run(unsigned int cpu) static void run_ksoftirqd(unsigned int cpu) { - local_irq_disable(); + ksoftirqd_run_begin(); if (local_softirq_pending()) { /* * We can safely run softirq on inline stack, as we are not deep * in the task stack here. */ __do_softirq(); - local_irq_enable(); + ksoftirqd_run_end(); cond_resched(); return; } - local_irq_enable(); + ksoftirqd_run_end(); } #ifdef CONFIG_HOTPLUG_CPU diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 865bb0228ab66..fb05a9293d007 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -42,11 +42,23 @@ struct cpu_stopper { struct list_head works; /* list of pending works */ struct cpu_stop_work stop_work; /* for stop_cpus */ + unsigned long caller; + cpu_stop_fn_t fn; }; static DEFINE_PER_CPU(struct cpu_stopper, cpu_stopper); static bool stop_machine_initialized = false; +void print_stop_info(const char *log_lvl, struct task_struct *task) +{ + struct cpu_stopper *stopper = this_cpu_ptr(&cpu_stopper); + + if (task != stopper->thread) + return; + + printk("%sStopper: %pS <- %pS\n", log_lvl, stopper->fn, (void *)stopper->caller); +} + /* static data for stop_cpus */ static DEFINE_MUTEX(stop_cpus_mutex); static bool stop_cpus_in_progress; @@ -123,7 +135,7 @@ static bool cpu_stop_queue_work(unsigned int cpu, struct cpu_stop_work *work) int stop_one_cpu(unsigned int cpu, cpu_stop_fn_t fn, void *arg) { struct cpu_stop_done done; - struct cpu_stop_work work = { .fn = fn, .arg = arg, .done = &done }; + struct cpu_stop_work work = { .fn = fn, .arg = arg, .done = &done, .caller = _RET_IP_ }; cpu_stop_init_done(&done, 1); if (!cpu_stop_queue_work(cpu, &work)) @@ -331,7 +343,8 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void * work1 = work2 = (struct cpu_stop_work){ .fn = multi_cpu_stop, .arg = &msdata, - .done = &done + .done = &done, + .caller = _RET_IP_, }; cpu_stop_init_done(&done, 2); @@ -367,7 +380,7 @@ int stop_two_cpus(unsigned int cpu1, unsigned int cpu2, cpu_stop_fn_t fn, void * bool stop_one_cpu_nowait(unsigned int cpu, cpu_stop_fn_t fn, void *arg, struct cpu_stop_work *work_buf) { - *work_buf = (struct cpu_stop_work){ .fn = fn, .arg = arg, }; + *work_buf = (struct cpu_stop_work){ .fn = fn, .arg = arg, .caller = _RET_IP_, }; return cpu_stop_queue_work(cpu, work_buf); } @@ -487,6 +500,8 @@ static void cpu_stopper_thread(unsigned int cpu) int ret; /* cpu stop callbacks must not sleep, make in_atomic() == T */ + stopper->caller = work->caller; + stopper->fn = fn; preempt_count_inc(); ret = fn(arg); if (done) { @@ -495,6 +510,8 @@ static void cpu_stopper_thread(unsigned int cpu) cpu_stop_signal_done(done); } preempt_count_dec(); + stopper->fn = NULL; + stopper->caller = 0; WARN_ONCE(preempt_count(), "cpu_stop: %ps(%p) leaked preempt count\n", fn, arg); goto repeat; diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 95b6a708b0406..822fa0c7f5dbe 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1828,7 +1828,7 @@ static void __hrtimer_init_sleeper(struct hrtimer_sleeper *sl, * expiry. */ if (IS_ENABLED(CONFIG_PREEMPT_RT)) { - if (task_is_realtime(current) && !(mode & HRTIMER_MODE_SOFT)) + if ((task_is_realtime(current) && !(mode & HRTIMER_MODE_SOFT)) || system_state != SYSTEM_RUNNING) mode |= HRTIMER_MODE_HARD; } @@ -1993,6 +1993,36 @@ SYSCALL_DEFINE2(nanosleep_time32, struct old_timespec32 __user *, rqtp, } #endif +#ifdef CONFIG_PREEMPT_RT +/* + * Sleep for 1 ms in hope whoever holds what we want will let it go. + */ +void cpu_chill(void) +{ + unsigned int freeze_flag = current->flags & PF_NOFREEZE; + struct task_struct *self = current; + ktime_t chill_time; + + raw_spin_lock_irq(&self->pi_lock); + self->saved_state = self->state; + __set_current_state_no_track(TASK_UNINTERRUPTIBLE); + raw_spin_unlock_irq(&self->pi_lock); + + chill_time = ktime_set(0, NSEC_PER_MSEC); + + current->flags |= PF_NOFREEZE; + schedule_hrtimeout(&chill_time, HRTIMER_MODE_REL_HARD); + if (!freeze_flag) + current->flags &= ~PF_NOFREEZE; + + raw_spin_lock_irq(&self->pi_lock); + __set_current_state_no_track(self->saved_state); + self->saved_state = TASK_RUNNING; + raw_spin_unlock_irq(&self->pi_lock); +} +EXPORT_SYMBOL(cpu_chill); +#endif + /* * Functions related to boot-time initialization: */ diff --git a/kernel/time/sched_clock.c b/kernel/time/sched_clock.c index 1c03eec6ca9b9..0642013dace49 100644 --- a/kernel/time/sched_clock.c +++ b/kernel/time/sched_clock.c @@ -35,7 +35,7 @@ * into a single 64-byte cache line. */ struct clock_data { - seqcount_t seq; + seqcount_latch_t seq; struct clock_read_data read_data[2]; ktime_t wrap_kt; unsigned long rate; @@ -76,7 +76,7 @@ struct clock_read_data *sched_clock_read_begin(unsigned int *seq) int sched_clock_read_retry(unsigned int seq) { - return read_seqcount_retry(&cd.seq, seq); + return read_seqcount_latch_retry(&cd.seq, seq); } unsigned long long notrace sched_clock(void) @@ -258,7 +258,7 @@ void __init generic_sched_clock_init(void) */ static u64 notrace suspended_sched_clock_read(void) { - unsigned int seq = raw_read_seqcount(&cd.seq); + unsigned int seq = raw_read_seqcount_latch(&cd.seq); return cd.read_data[seq & 1].epoch_cyc; } diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index f0199a4ba1add..1de4665dc52f4 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -925,7 +925,7 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts) if (unlikely(local_softirq_pending())) { static int ratelimit; - if (ratelimit < 10 && + if (ratelimit < 10 && !local_bh_blocked() && (local_softirq_pending() & SOFTIRQ_STOP_IDLE_MASK)) { pr_warn("NOHZ: local_softirq_pending %02x\n", (unsigned int) local_softirq_pending()); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 4c47f388a83f1..999c981ae766f 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -64,7 +64,7 @@ static struct timekeeper shadow_timekeeper; * See @update_fast_timekeeper() below. */ struct tk_fast { - seqcount_raw_spinlock_t seq; + seqcount_latch_t seq; struct tk_read_base base[2]; }; @@ -81,13 +81,13 @@ static struct clocksource dummy_clock = { }; static struct tk_fast tk_fast_mono ____cacheline_aligned = { - .seq = SEQCNT_RAW_SPINLOCK_ZERO(tk_fast_mono.seq, &timekeeper_lock), + .seq = SEQCNT_LATCH_ZERO(tk_fast_mono.seq), .base[0] = { .clock = &dummy_clock, }, .base[1] = { .clock = &dummy_clock, }, }; static struct tk_fast tk_fast_raw ____cacheline_aligned = { - .seq = SEQCNT_RAW_SPINLOCK_ZERO(tk_fast_raw.seq, &timekeeper_lock), + .seq = SEQCNT_LATCH_ZERO(tk_fast_raw.seq), .base[0] = { .clock = &dummy_clock, }, .base[1] = { .clock = &dummy_clock, }, }; @@ -467,7 +467,7 @@ static __always_inline u64 __ktime_get_fast_ns(struct tk_fast *tkf) tk_clock_read(tkr), tkr->cycle_last, tkr->mask)); - } while (read_seqcount_retry(&tkf->seq, seq)); + } while (read_seqcount_latch_retry(&tkf->seq, seq)); return now; } @@ -533,7 +533,7 @@ static __always_inline u64 __ktime_get_real_fast_ns(struct tk_fast *tkf) tk_clock_read(tkr), tkr->cycle_last, tkr->mask)); - } while (read_seqcount_retry(&tkf->seq, seq)); + } while (read_seqcount_latch_retry(&tkf->seq, seq)); return now; } diff --git a/kernel/time/timer.c b/kernel/time/timer.c index a50364df10543..e25cb9d7f09a1 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -1765,6 +1765,8 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h) { struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]); + irq_work_tick_soft(); + __run_timers(base); if (IS_ENABLED(CONFIG_NO_HZ_COMMON)) __run_timers(this_cpu_ptr(&timer_bases[BASE_DEF])); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d3e5de717df2f..21706f318a001 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2437,6 +2437,15 @@ enum print_line_t trace_handle_return(struct trace_seq *s) } EXPORT_SYMBOL_GPL(trace_handle_return); +static unsigned short migration_disable_value(struct task_struct *tsk) +{ +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + return tsk ? tsk->migration_disabled : 0; +#else + return 0; +#endif +} + void tracing_generic_entry_update(struct trace_entry *entry, unsigned short type, unsigned long flags, int pc) @@ -2444,6 +2453,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned short type, struct task_struct *tsk = current; entry->preempt_count = pc & 0xff; + entry->preempt_lazy_count = preempt_lazy_count(); entry->pid = (tsk) ? tsk->pid : 0; entry->type = type; entry->flags = @@ -2455,8 +2465,11 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned short type, ((pc & NMI_MASK ) ? TRACE_FLAG_NMI : 0) | ((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) | ((pc & SOFTIRQ_OFFSET) ? TRACE_FLAG_SOFTIRQ : 0) | - (tif_need_resched() ? TRACE_FLAG_NEED_RESCHED : 0) | + (tif_need_resched_now() ? TRACE_FLAG_NEED_RESCHED : 0) | + (need_resched_lazy() ? TRACE_FLAG_NEED_RESCHED_LAZY : 0) | (test_preempt_need_resched() ? TRACE_FLAG_PREEMPT_RESCHED : 0); + + entry->migrate_disable = migration_disable_value(tsk); } EXPORT_SYMBOL_GPL(tracing_generic_entry_update); @@ -3784,14 +3797,17 @@ unsigned long trace_total_entries(struct trace_array *tr) static void print_lat_help_header(struct seq_file *m) { - seq_puts(m, "# _------=> CPU# \n" - "# / _-----=> irqs-off \n" - "# | / _----=> need-resched \n" - "# || / _---=> hardirq/softirq \n" - "# ||| / _--=> preempt-depth \n" - "# |||| / delay \n" - "# cmd pid ||||| time | caller \n" - "# \\ / ||||| \\ | / \n"); + seq_puts(m, "# _--------=> CPU# \n" + "# / _-------=> irqs-off \n" + "# | / _------=> need-resched \n" + "# || / _-----=> need-resched-lazy\n" + "# ||| / _----=> hardirq/softirq \n" + "# |||| / _---=> preempt-depth \n" + "# ||||| / _--=> preempt-lazy-depth\n" + "# |||||| / _-=> migrate-disable \n" + "# ||||||| / delay \n" + "# cmd pid |||||||| time | caller \n" + "# \\ / |||||||| \\ | / \n"); } static void print_event_info(struct array_buffer *buf, struct seq_file *m) @@ -3825,13 +3841,16 @@ static void print_func_help_header_irq(struct array_buffer *buf, struct seq_file print_event_info(buf, m); - seq_printf(m, "# %.*s _-----=> irqs-off\n", prec, space); - seq_printf(m, "# %.*s / _----=> need-resched\n", prec, space); - seq_printf(m, "# %.*s| / _---=> hardirq/softirq\n", prec, space); - seq_printf(m, "# %.*s|| / _--=> preempt-depth\n", prec, space); - seq_printf(m, "# %.*s||| / delay\n", prec, space); - seq_printf(m, "# TASK-PID %.*s CPU# |||| TIMESTAMP FUNCTION\n", prec, " TGID "); - seq_printf(m, "# | | %.*s | |||| | |\n", prec, " | "); + seq_printf(m, "# %.*s _-------=> irqs-off\n", prec, space); + seq_printf(m, "# %.*s / _------=> need-resched\n", prec, space); + seq_printf(m, "# %.*s| / _-----=> need-resched-lazy\n", prec, space); + seq_printf(m, "# %.*s|| / _----=> hardirq/softirq\n", prec, space); + seq_printf(m, "# %.*s||| / _---=> preempt-depth\n", prec, space); + seq_printf(m, "# %.*s|||| / _--=> preempt-lazy-depth\n", prec, space); + seq_printf(m, "# %.*s||||| / _-=> migrate-disable\n", prec, space); + seq_printf(m, "# %.*s|||||| / delay\n", prec, space); + seq_printf(m, "# TASK-PID %.*s CPU# ||||||| TIMESTAMP FUNCTION\n", prec, " TGID "); + seq_printf(m, "# | | %.*s | ||||||| | |\n", prec, " | "); } void @@ -9249,7 +9268,6 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) tracing_off(); local_irq_save(flags); - printk_nmi_direct_enter(); /* Simulate the iterator */ trace_init_global_iter(&iter); @@ -9329,7 +9347,6 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) atomic_dec(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); } atomic_dec(&dump_running); - printk_nmi_direct_exit(); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(ftrace_dump); diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 610d21355526d..580f2355a8357 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -143,6 +143,7 @@ struct kretprobe_trace_entry_head { * NEED_RESCHED - reschedule is requested * HARDIRQ - inside an interrupt handler * SOFTIRQ - inside a softirq handler + * NEED_RESCHED_LAZY - lazy reschedule is requested */ enum trace_flag_type { TRACE_FLAG_IRQS_OFF = 0x01, @@ -152,6 +153,7 @@ enum trace_flag_type { TRACE_FLAG_SOFTIRQ = 0x10, TRACE_FLAG_PREEMPT_RESCHED = 0x20, TRACE_FLAG_NMI = 0x40, + TRACE_FLAG_NEED_RESCHED_LAZY = 0x80, }; #define TRACE_BUF_SIZE 1024 diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index a85effb2373bf..a88a78c240bbb 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -182,6 +182,8 @@ static int trace_define_common_fields(void) __common_field(unsigned char, flags); __common_field(unsigned char, preempt_count); __common_field(int, pid); + __common_field(unsigned char, migrate_disable); + __common_field(unsigned char, preempt_lazy_count); return ret; } diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index 000e9dc224c61..bc24ae8e3613a 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -441,6 +441,7 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry) { char hardsoft_irq; char need_resched; + char need_resched_lazy; char irqs_off; int hardirq; int softirq; @@ -471,6 +472,9 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry) break; } + need_resched_lazy = + (entry->flags & TRACE_FLAG_NEED_RESCHED_LAZY) ? 'L' : '.'; + hardsoft_irq = (nmi && hardirq) ? 'Z' : nmi ? 'z' : @@ -479,14 +483,25 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry) softirq ? 's' : '.' ; - trace_seq_printf(s, "%c%c%c", - irqs_off, need_resched, hardsoft_irq); + trace_seq_printf(s, "%c%c%c%c", + irqs_off, need_resched, need_resched_lazy, + hardsoft_irq); if (entry->preempt_count) trace_seq_printf(s, "%x", entry->preempt_count); else trace_seq_putc(s, '.'); + if (entry->preempt_lazy_count) + trace_seq_printf(s, "%x", entry->preempt_lazy_count); + else + trace_seq_putc(s, '.'); + + if (entry->migrate_disable) + trace_seq_printf(s, "%x", entry->migrate_disable); + else + trace_seq_putc(s, '.'); + return !trace_seq_has_overflowed(s); } diff --git a/kernel/workqueue.c b/kernel/workqueue.c index c41c3c17b86a3..bb8a84dbabb9d 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -4905,6 +4905,10 @@ static void unbind_workers(int cpu) pool->flags |= POOL_DISASSOCIATED; raw_spin_unlock_irq(&pool->lock); + + for_each_pool_worker(worker, pool) + WARN_ON_ONCE(set_cpus_allowed_ptr(worker->task, cpu_active_mask) < 0); + mutex_unlock(&wq_pool_attach_mutex); /* diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 0c781f912f9f0..16fcda68c2b6b 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1332,7 +1332,7 @@ config DEBUG_ATOMIC_SLEEP config DEBUG_LOCKING_API_SELFTESTS bool "Locking API boot-time self-tests" - depends on DEBUG_KERNEL + depends on DEBUG_KERNEL && !PREEMPT_RT help Say Y here if you want the kernel to run a short self-test during bootup. The self-test checks whether common types of locking bugs diff --git a/lib/cpumask.c b/lib/cpumask.c index 85da6ab4fbb5a..35924025097b9 100644 --- a/lib/cpumask.c +++ b/lib/cpumask.c @@ -267,3 +267,21 @@ int cpumask_any_and_distribute(const struct cpumask *src1p, return next; } EXPORT_SYMBOL(cpumask_any_and_distribute); + +int cpumask_any_distribute(const struct cpumask *srcp) +{ + int next, prev; + + /* NOTE: our first selection will skip 0. */ + prev = __this_cpu_read(distribute_cpu_mask_prev); + + next = cpumask_next(prev, srcp); + if (next >= nr_cpu_ids) + next = cpumask_first(srcp); + + if (next < nr_cpu_ids) + __this_cpu_write(distribute_cpu_mask_prev, next); + + return next; +} +EXPORT_SYMBOL(cpumask_any_distribute); diff --git a/lib/debugobjects.c b/lib/debugobjects.c index fe4557955d976..f8bfd257d0bb3 100644 --- a/lib/debugobjects.c +++ b/lib/debugobjects.c @@ -537,7 +537,10 @@ __debug_object_init(void *addr, struct debug_obj_descr *descr, int onstack) struct debug_obj *obj; unsigned long flags; - fill_pool(); +#ifdef CONFIG_PREEMPT_RT + if (preempt_count() == 0 && !irqs_disabled()) +#endif + fill_pool(); db = get_bucket((unsigned long) addr); diff --git a/lib/dump_stack.c b/lib/dump_stack.c index a00ee6eedc7c3..f5a33b6f773f7 100644 --- a/lib/dump_stack.c +++ b/lib/dump_stack.c @@ -12,6 +12,7 @@ #include #include #include +#include static char dump_stack_arch_desc_str[128]; @@ -57,6 +58,7 @@ void dump_stack_print_info(const char *log_lvl) log_lvl, dump_stack_arch_desc_str); print_worker_info(log_lvl, current); + print_stop_info(log_lvl, current); } /** diff --git a/lib/irq_poll.c b/lib/irq_poll.c index 2f17b488d58e1..7557bf7ecf1f1 100644 --- a/lib/irq_poll.c +++ b/lib/irq_poll.c @@ -37,6 +37,7 @@ void irq_poll_sched(struct irq_poll *iop) list_add_tail(&iop->list, this_cpu_ptr(&blk_cpu_iopoll)); raise_softirq_irqoff(IRQ_POLL_SOFTIRQ); local_irq_restore(flags); + preempt_check_resched_rt(); } EXPORT_SYMBOL(irq_poll_sched); @@ -72,6 +73,7 @@ void irq_poll_complete(struct irq_poll *iop) local_irq_save(flags); __irq_poll_complete(iop); local_irq_restore(flags); + preempt_check_resched_rt(); } EXPORT_SYMBOL(irq_poll_complete); @@ -96,6 +98,7 @@ static void __latent_entropy irq_poll_softirq(struct softirq_action *h) } local_irq_enable(); + preempt_check_resched_rt(); /* Even though interrupts have been re-enabled, this * access is safe because interrupts can only add new @@ -133,6 +136,7 @@ static void __latent_entropy irq_poll_softirq(struct softirq_action *h) __raise_softirq_irqoff(IRQ_POLL_SOFTIRQ); local_irq_enable(); + preempt_check_resched_rt(); } /** @@ -196,6 +200,7 @@ static int irq_poll_cpu_dead(unsigned int cpu) this_cpu_ptr(&blk_cpu_iopoll)); __raise_softirq_irqoff(IRQ_POLL_SOFTIRQ); local_irq_enable(); + preempt_check_resched_rt(); return 0; } diff --git a/lib/locking-selftest.c b/lib/locking-selftest.c index 14f44f59e7338..b110d073d2d2a 100644 --- a/lib/locking-selftest.c +++ b/lib/locking-selftest.c @@ -742,6 +742,8 @@ GENERATE_TESTCASE(init_held_rtmutex); #include "locking-selftest-spin-hardirq.h" GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_hard_spin) +#ifndef CONFIG_PREEMPT_RT + #include "locking-selftest-rlock-hardirq.h" GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_hard_rlock) @@ -757,9 +759,12 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_rlock) #include "locking-selftest-wlock-softirq.h" GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_wlock) +#endif + #undef E1 #undef E2 +#ifndef CONFIG_PREEMPT_RT /* * Enabling hardirqs with a softirq-safe lock held: */ @@ -792,6 +797,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2A_rlock) #undef E1 #undef E2 +#endif + /* * Enabling irqs with an irq-safe lock held: */ @@ -815,6 +822,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2A_rlock) #include "locking-selftest-spin-hardirq.h" GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_hard_spin) +#ifndef CONFIG_PREEMPT_RT + #include "locking-selftest-rlock-hardirq.h" GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_hard_rlock) @@ -830,6 +839,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_rlock) #include "locking-selftest-wlock-softirq.h" GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_wlock) +#endif + #undef E1 #undef E2 @@ -861,6 +872,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_wlock) #include "locking-selftest-spin-hardirq.h" GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_hard_spin) +#ifndef CONFIG_PREEMPT_RT + #include "locking-selftest-rlock-hardirq.h" GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_hard_rlock) @@ -876,6 +889,8 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_rlock) #include "locking-selftest-wlock-softirq.h" GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_wlock) +#endif + #undef E1 #undef E2 #undef E3 @@ -909,6 +924,8 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_wlock) #include "locking-selftest-spin-hardirq.h" GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_hard_spin) +#ifndef CONFIG_PREEMPT_RT + #include "locking-selftest-rlock-hardirq.h" GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_hard_rlock) @@ -924,10 +941,14 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_rlock) #include "locking-selftest-wlock-softirq.h" GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_wlock) +#endif + #undef E1 #undef E2 #undef E3 +#ifndef CONFIG_PREEMPT_RT + /* * read-lock / write-lock irq inversion. * @@ -990,6 +1011,10 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_soft_wlock) #undef E2 #undef E3 +#endif + +#ifndef CONFIG_PREEMPT_RT + /* * read-lock / write-lock recursion that is actually safe. */ @@ -1028,6 +1053,8 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion_soft) #undef E2 #undef E3 +#endif + /* * read-lock / write-lock recursion that is unsafe. */ @@ -2058,6 +2085,7 @@ void locking_selftest(void) printk(" --------------------------------------------------------------------------\n"); +#ifndef CONFIG_PREEMPT_RT /* * irq-context testcases: */ @@ -2070,6 +2098,28 @@ void locking_selftest(void) DO_TESTCASE_6x2("irq read-recursion", irq_read_recursion); // DO_TESTCASE_6x2B("irq read-recursion #2", irq_read_recursion2); +#else + /* On -rt, we only do hardirq context test for raw spinlock */ + DO_TESTCASE_1B("hard-irqs-on + irq-safe-A", irqsafe1_hard_spin, 12); + DO_TESTCASE_1B("hard-irqs-on + irq-safe-A", irqsafe1_hard_spin, 21); + + DO_TESTCASE_1B("hard-safe-A + irqs-on", irqsafe2B_hard_spin, 12); + DO_TESTCASE_1B("hard-safe-A + irqs-on", irqsafe2B_hard_spin, 21); + + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 123); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 132); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 213); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 231); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 312); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 321); + + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 123); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 132); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 213); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 231); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 312); + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 321); +#endif ww_tests(); diff --git a/lib/nmi_backtrace.c b/lib/nmi_backtrace.c index 15ca78e1c7d4e..77bf84987cda4 100644 --- a/lib/nmi_backtrace.c +++ b/lib/nmi_backtrace.c @@ -75,12 +75,6 @@ void nmi_trigger_cpumask_backtrace(const cpumask_t *mask, touch_softlockup_watchdog(); } - /* - * Force flush any remote buffers that might be stuck in IRQ context - * and therefore could not run their irq_work. - */ - printk_safe_flush(); - clear_bit_unlock(0, &backtrace_flag); put_cpu(); } diff --git a/lib/scatterlist.c b/lib/scatterlist.c index 5d63a8857f361..5569dac27afe8 100644 --- a/lib/scatterlist.c +++ b/lib/scatterlist.c @@ -811,7 +811,7 @@ void sg_miter_stop(struct sg_mapping_iter *miter) flush_kernel_dcache_page(miter->page); if (miter->__flags & SG_MITER_ATOMIC) { - WARN_ON_ONCE(preemptible()); + WARN_ON_ONCE(!pagefault_disabled()); kunmap_atomic(miter->addr); } else kunmap(miter->page); diff --git a/lib/smp_processor_id.c b/lib/smp_processor_id.c index 525222e4f4099..faaa927ac2c8b 100644 --- a/lib/smp_processor_id.c +++ b/lib/smp_processor_id.c @@ -26,6 +26,11 @@ unsigned int check_preemption_disabled(const char *what1, const char *what2) if (current->nr_cpus_allowed == 1) goto out; +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + if (current->migration_disabled) + goto out; +#endif + /* * It is valid to assume CPU-locality during early bootup: */ diff --git a/lib/test_lockup.c b/lib/test_lockup.c index f1a020bcc763e..864554e769738 100644 --- a/lib/test_lockup.c +++ b/lib/test_lockup.c @@ -480,6 +480,21 @@ static int __init test_lockup_init(void) return -EINVAL; #ifdef CONFIG_DEBUG_SPINLOCK +#ifdef CONFIG_PREEMPT_RT + if (test_magic(lock_spinlock_ptr, + offsetof(spinlock_t, lock.wait_lock.magic), + SPINLOCK_MAGIC) || + test_magic(lock_rwlock_ptr, + offsetof(rwlock_t, rtmutex.wait_lock.magic), + SPINLOCK_MAGIC) || + test_magic(lock_mutex_ptr, + offsetof(struct mutex, lock.wait_lock.magic), + SPINLOCK_MAGIC) || + test_magic(lock_rwsem_ptr, + offsetof(struct rw_semaphore, rtmutex.wait_lock.magic), + SPINLOCK_MAGIC)) + return -EINVAL; +#else if (test_magic(lock_spinlock_ptr, offsetof(spinlock_t, rlock.magic), SPINLOCK_MAGIC) || @@ -493,6 +508,7 @@ static int __init test_lockup_init(void) offsetof(struct rw_semaphore, wait_lock.magic), SPINLOCK_MAGIC)) return -EINVAL; +#endif #endif if ((wait_state != TASK_RUNNING || diff --git a/localversion-rt b/localversion-rt new file mode 100644 index 0000000000000..e095ab8197147 --- /dev/null +++ b/localversion-rt @@ -0,0 +1 @@ +-rt20 diff --git a/mm/Kconfig b/mm/Kconfig index 6c974888f86f9..056460878a2b4 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -387,7 +387,7 @@ config NOMMU_INITIAL_TRIM_EXCESS config TRANSPARENT_HUGEPAGE bool "Transparent Hugepage Support" - depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE + depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT select COMPACTION select XARRAY_MULTI help diff --git a/mm/highmem.c b/mm/highmem.c index 64d8dea47dd1a..7d3065719ce80 100644 --- a/mm/highmem.c +++ b/mm/highmem.c @@ -31,8 +31,11 @@ #include #include +#ifndef CONFIG_PREEMPT_RT #if defined(CONFIG_HIGHMEM) || defined(CONFIG_X86_32) DEFINE_PER_CPU(int, __kmap_atomic_idx); +EXPORT_PER_CPU_SYMBOL(__kmap_atomic_idx); +#endif #endif /* @@ -108,8 +111,6 @@ static inline wait_queue_head_t *get_pkmap_wait_queue_head(unsigned int color) atomic_long_t _totalhigh_pages __read_mostly; EXPORT_SYMBOL(_totalhigh_pages); -EXPORT_PER_CPU_SYMBOL(__kmap_atomic_idx); - unsigned int nr_free_highpages (void) { struct zone *zone; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6877c765b8d03..09fa87f7a8119 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -63,6 +63,7 @@ #include #include #include "slab.h" +#include #include @@ -90,6 +91,13 @@ bool cgroup_memory_noswap __read_mostly; static DECLARE_WAIT_QUEUE_HEAD(memcg_cgwb_frn_waitq); #endif +struct event_lock { + local_lock_t l; +}; +static DEFINE_PER_CPU(struct event_lock, event_lock) = { + .l = INIT_LOCAL_LOCK(l), +}; + /* Whether legacy memory+swap accounting is active */ static bool do_memsw_account(void) { @@ -821,6 +829,7 @@ void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx, pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec); memcg = pn->memcg; + preempt_disable_rt(); /* Update memcg */ __mod_memcg_state(memcg, idx, val); @@ -840,6 +849,7 @@ void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx, x = 0; } __this_cpu_write(pn->lruvec_stat_cpu->count[idx], x); + preempt_enable_rt(); } /** @@ -2154,6 +2164,7 @@ void unlock_page_memcg(struct page *page) EXPORT_SYMBOL(unlock_page_memcg); struct memcg_stock_pcp { + local_lock_t lock; struct mem_cgroup *cached; /* this never be root cgroup */ unsigned int nr_pages; @@ -2205,7 +2216,7 @@ static bool consume_stock(struct mem_cgroup *memcg, unsigned int nr_pages) if (nr_pages > MEMCG_CHARGE_BATCH) return ret; - local_irq_save(flags); + local_lock_irqsave(&memcg_stock.lock, flags); stock = this_cpu_ptr(&memcg_stock); if (memcg == stock->cached && stock->nr_pages >= nr_pages) { @@ -2213,7 +2224,7 @@ static bool consume_stock(struct mem_cgroup *memcg, unsigned int nr_pages) ret = true; } - local_irq_restore(flags); + local_unlock_irqrestore(&memcg_stock.lock, flags); return ret; } @@ -2248,14 +2259,14 @@ static void drain_local_stock(struct work_struct *dummy) * The only protection from memory hotplug vs. drain_stock races is * that we always operate on local CPU stock here with IRQ disabled */ - local_irq_save(flags); + local_lock_irqsave(&memcg_stock.lock, flags); stock = this_cpu_ptr(&memcg_stock); drain_obj_stock(stock); drain_stock(stock); clear_bit(FLUSHING_CACHED_CHARGE, &stock->flags); - local_irq_restore(flags); + local_unlock_irqrestore(&memcg_stock.lock, flags); } /* @@ -2267,7 +2278,7 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages) struct memcg_stock_pcp *stock; unsigned long flags; - local_irq_save(flags); + local_lock_irqsave(&memcg_stock.lock, flags); stock = this_cpu_ptr(&memcg_stock); if (stock->cached != memcg) { /* reset if necessary */ @@ -2280,7 +2291,7 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages) if (stock->nr_pages > MEMCG_CHARGE_BATCH) drain_stock(stock); - local_irq_restore(flags); + local_unlock_irqrestore(&memcg_stock.lock, flags); } /* @@ -2300,7 +2311,7 @@ static void drain_all_stock(struct mem_cgroup *root_memcg) * as well as workers from this path always operate on the local * per-cpu data. CPU up doesn't touch memcg_stock at all. */ - curcpu = get_cpu(); + curcpu = get_cpu_light(); for_each_online_cpu(cpu) { struct memcg_stock_pcp *stock = &per_cpu(memcg_stock, cpu); struct mem_cgroup *memcg; @@ -2323,7 +2334,7 @@ static void drain_all_stock(struct mem_cgroup *root_memcg) schedule_work_on(cpu, &stock->work); } } - put_cpu(); + put_cpu_light(); mutex_unlock(&percpu_charge_mutex); } @@ -3084,7 +3095,7 @@ static bool consume_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes) unsigned long flags; bool ret = false; - local_irq_save(flags); + local_lock_irqsave(&memcg_stock.lock, flags); stock = this_cpu_ptr(&memcg_stock); if (objcg == stock->cached_objcg && stock->nr_bytes >= nr_bytes) { @@ -3092,7 +3103,7 @@ static bool consume_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes) ret = true; } - local_irq_restore(flags); + local_unlock_irqrestore(&memcg_stock.lock, flags); return ret; } @@ -3151,7 +3162,7 @@ static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes) struct memcg_stock_pcp *stock; unsigned long flags; - local_irq_save(flags); + local_lock_irqsave(&memcg_stock.lock, flags); stock = this_cpu_ptr(&memcg_stock); if (stock->cached_objcg != objcg) { /* reset if necessary */ @@ -3165,7 +3176,7 @@ static void refill_obj_stock(struct obj_cgroup *objcg, unsigned int nr_bytes) if (stock->nr_bytes > PAGE_SIZE) drain_obj_stock(stock); - local_irq_restore(flags); + local_unlock_irqrestore(&memcg_stock.lock, flags); } int obj_cgroup_charge(struct obj_cgroup *objcg, gfp_t gfp, size_t size) @@ -5681,12 +5692,12 @@ static int mem_cgroup_move_account(struct page *page, ret = 0; - local_irq_disable(); + local_lock_irq(&event_lock.l); mem_cgroup_charge_statistics(to, page, nr_pages); memcg_check_events(to, page); mem_cgroup_charge_statistics(from, page, -nr_pages); memcg_check_events(from, page); - local_irq_enable(); + local_unlock_irq(&event_lock.l); out_unlock: unlock_page(page); out: @@ -6722,10 +6733,10 @@ int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask) css_get(&memcg->css); commit_charge(page, memcg); - local_irq_disable(); + local_lock_irq(&event_lock.l); mem_cgroup_charge_statistics(memcg, page, nr_pages); memcg_check_events(memcg, page); - local_irq_enable(); + local_unlock_irq(&event_lock.l); if (PageSwapCache(page)) { swp_entry_t entry = { .val = page_private(page) }; @@ -6769,11 +6780,11 @@ static void uncharge_batch(const struct uncharge_gather *ug) memcg_oom_recover(ug->memcg); } - local_irq_save(flags); + local_lock_irqsave(&event_lock.l, flags); __count_memcg_events(ug->memcg, PGPGOUT, ug->pgpgout); __this_cpu_add(ug->memcg->vmstats_percpu->nr_page_events, ug->nr_pages); memcg_check_events(ug->memcg, ug->dummy_page); - local_irq_restore(flags); + local_unlock_irqrestore(&event_lock.l, flags); /* drop reference from uncharge_page */ css_put(&ug->memcg->css); @@ -6927,10 +6938,10 @@ void mem_cgroup_migrate(struct page *oldpage, struct page *newpage) css_get(&memcg->css); commit_charge(newpage, memcg); - local_irq_save(flags); + local_lock_irqsave(&event_lock.l, flags); mem_cgroup_charge_statistics(memcg, newpage, nr_pages); memcg_check_events(memcg, newpage); - local_irq_restore(flags); + local_unlock_irqrestore(&event_lock.l, flags); } DEFINE_STATIC_KEY_FALSE(memcg_sockets_enabled_key); @@ -7050,9 +7061,13 @@ static int __init mem_cgroup_init(void) cpuhp_setup_state_nocalls(CPUHP_MM_MEMCQ_DEAD, "mm/memctrl:dead", NULL, memcg_hotplug_cpu_dead); - for_each_possible_cpu(cpu) - INIT_WORK(&per_cpu_ptr(&memcg_stock, cpu)->work, - drain_local_stock); + for_each_possible_cpu(cpu) { + struct memcg_stock_pcp *stock; + + stock = per_cpu_ptr(&memcg_stock, cpu); + INIT_WORK(&stock->work, drain_local_stock); + local_lock_init(&stock->lock); + } for_each_node(node) { struct mem_cgroup_tree_per_node *rtpn; @@ -7101,6 +7116,7 @@ void mem_cgroup_swapout(struct page *page, swp_entry_t entry) struct mem_cgroup *memcg, *swap_memcg; unsigned int nr_entries; unsigned short oldid; + unsigned long flags; VM_BUG_ON_PAGE(PageLRU(page), page); VM_BUG_ON_PAGE(page_count(page), page); @@ -7146,9 +7162,13 @@ void mem_cgroup_swapout(struct page *page, swp_entry_t entry) * important here to have the interrupts disabled because it is the * only synchronisation we have for updating the per-CPU variables. */ + local_lock_irqsave(&event_lock.l, flags); +#ifndef CONFIG_PREEMPT_RT VM_BUG_ON(!irqs_disabled()); +#endif mem_cgroup_charge_statistics(memcg, page, -nr_entries); memcg_check_events(memcg, page); + local_unlock_irqrestore(&event_lock.l, flags); css_put(&memcg->css); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 780c8f023b282..f581204e3f35a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -61,6 +61,7 @@ #include #include #include +#include #include #include #include @@ -357,6 +358,13 @@ EXPORT_SYMBOL(nr_node_ids); EXPORT_SYMBOL(nr_online_nodes); #endif +struct pa_lock { + local_lock_t l; +}; +static DEFINE_PER_CPU(struct pa_lock, pa_lock) = { + .l = INIT_LOCAL_LOCK(l), +}; + int page_group_by_mobility_disabled __read_mostly; #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT @@ -1283,7 +1291,7 @@ static inline void prefetch_buddy(struct page *page) } /* - * Frees a number of pages from the PCP lists + * Frees a number of pages which have been collected from the pcp lists. * Assumes all pages on list are in same zone, and of same order. * count is the number of pages to free. * @@ -1293,15 +1301,56 @@ static inline void prefetch_buddy(struct page *page) * And clear the zone's pages_scanned counter, to hold off the "all pages are * pinned" detection logic. */ -static void free_pcppages_bulk(struct zone *zone, int count, - struct per_cpu_pages *pcp) +static void free_pcppages_bulk(struct zone *zone, struct list_head *head, + bool zone_retry) +{ + bool isolated_pageblocks; + struct page *page, *tmp; + unsigned long flags; + + spin_lock_irqsave(&zone->lock, flags); + isolated_pageblocks = has_isolate_pageblock(zone); + + /* + * Use safe version since after __free_one_page(), + * page->lru.next will not point to original list. + */ + list_for_each_entry_safe(page, tmp, head, lru) { + int mt = get_pcppage_migratetype(page); + + if (page_zone(page) != zone) { + /* + * free_unref_page_list() sorts pages by zone. If we end + * up with pages from a different NUMA nodes belonging + * to the same ZONE index then we need to redo with the + * correct ZONE pointer. Skip the page for now, redo it + * on the next iteration. + */ + WARN_ON_ONCE(zone_retry == false); + if (zone_retry) + continue; + } + + /* MIGRATE_ISOLATE page should not go to pcplists */ + VM_BUG_ON_PAGE(is_migrate_isolate(mt), page); + /* Pageblock could have been isolated meanwhile */ + if (unlikely(isolated_pageblocks)) + mt = get_pageblock_migratetype(page); + + list_del(&page->lru); + __free_one_page(page, page_to_pfn(page), zone, 0, mt, true); + trace_mm_page_pcpu_drain(page, 0, mt); + } + spin_unlock_irqrestore(&zone->lock, flags); +} + +static void isolate_pcp_pages(int count, struct per_cpu_pages *pcp, + struct list_head *dst) { int migratetype = 0; int batch_free = 0; int prefetch_nr = 0; - bool isolated_pageblocks; - struct page *page, *tmp; - LIST_HEAD(head); + struct page *page; /* * Ensure proper count is passed which otherwise would stuck in the @@ -1338,7 +1387,7 @@ static void free_pcppages_bulk(struct zone *zone, int count, if (bulkfree_pcp_prepare(page)) continue; - list_add_tail(&page->lru, &head); + list_add_tail(&page->lru, dst); /* * We are going to put the page back to the global @@ -1353,26 +1402,6 @@ static void free_pcppages_bulk(struct zone *zone, int count, prefetch_buddy(page); } while (--count && --batch_free && !list_empty(list)); } - - spin_lock(&zone->lock); - isolated_pageblocks = has_isolate_pageblock(zone); - - /* - * Use safe version since after __free_one_page(), - * page->lru.next will not point to original list. - */ - list_for_each_entry_safe(page, tmp, &head, lru) { - int mt = get_pcppage_migratetype(page); - /* MIGRATE_ISOLATE page should not go to pcplists */ - VM_BUG_ON_PAGE(is_migrate_isolate(mt), page); - /* Pageblock could have been isolated meanwhile */ - if (unlikely(isolated_pageblocks)) - mt = get_pageblock_migratetype(page); - - __free_one_page(page, page_to_pfn(page), zone, 0, mt, true); - trace_mm_page_pcpu_drain(page, 0, mt); - } - spin_unlock(&zone->lock); } static void free_one_page(struct zone *zone, @@ -1473,10 +1502,10 @@ static void __free_pages_ok(struct page *page, unsigned int order) return; migratetype = get_pfnblock_migratetype(page, pfn); - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); __count_vm_events(PGFREE, 1 << order); free_one_page(page_zone(page), page, pfn, order, migratetype); - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); } void __free_pages_core(struct page *page, unsigned int order) @@ -2877,13 +2906,18 @@ void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp) { unsigned long flags; int to_drain, batch; + LIST_HEAD(dst); - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); batch = READ_ONCE(pcp->batch); to_drain = min(pcp->count, batch); if (to_drain > 0) - free_pcppages_bulk(zone, to_drain, pcp); - local_irq_restore(flags); + isolate_pcp_pages(to_drain, pcp, &dst); + + local_unlock_irqrestore(&pa_lock.l, flags); + + if (to_drain > 0) + free_pcppages_bulk(zone, &dst, false); } #endif @@ -2899,14 +2933,21 @@ static void drain_pages_zone(unsigned int cpu, struct zone *zone) unsigned long flags; struct per_cpu_pageset *pset; struct per_cpu_pages *pcp; + LIST_HEAD(dst); + int count; - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); pset = per_cpu_ptr(zone->pageset, cpu); pcp = &pset->pcp; - if (pcp->count) - free_pcppages_bulk(zone, pcp->count, pcp); - local_irq_restore(flags); + count = pcp->count; + if (count) + isolate_pcp_pages(count, pcp, &dst); + + local_unlock_irqrestore(&pa_lock.l, flags); + + if (count) + free_pcppages_bulk(zone, &dst, false); } /* @@ -2954,9 +2995,9 @@ static void drain_local_pages_wq(struct work_struct *work) * cpu which is allright but we also have to make sure to not move to * a different one. */ - preempt_disable(); + migrate_disable(); drain_local_pages(drain->zone); - preempt_enable(); + migrate_enable(); } /* @@ -3105,7 +3146,8 @@ static bool free_unref_page_prepare(struct page *page, unsigned long pfn) return true; } -static void free_unref_page_commit(struct page *page, unsigned long pfn) +static void free_unref_page_commit(struct page *page, unsigned long pfn, + struct list_head *dst) { struct zone *zone = page_zone(page); struct per_cpu_pages *pcp; @@ -3134,7 +3176,8 @@ static void free_unref_page_commit(struct page *page, unsigned long pfn) pcp->count++; if (pcp->count >= pcp->high) { unsigned long batch = READ_ONCE(pcp->batch); - free_pcppages_bulk(zone, batch, pcp); + + isolate_pcp_pages(batch, pcp, dst); } } @@ -3145,13 +3188,17 @@ void free_unref_page(struct page *page) { unsigned long flags; unsigned long pfn = page_to_pfn(page); + struct zone *zone = page_zone(page); + LIST_HEAD(dst); if (!free_unref_page_prepare(page, pfn)) return; - local_irq_save(flags); - free_unref_page_commit(page, pfn); - local_irq_restore(flags); + local_lock_irqsave(&pa_lock.l, flags); + free_unref_page_commit(page, pfn, &dst); + local_unlock_irqrestore(&pa_lock.l, flags); + if (!list_empty(&dst)) + free_pcppages_bulk(zone, &dst, false); } /* @@ -3162,6 +3209,11 @@ void free_unref_page_list(struct list_head *list) struct page *page, *next; unsigned long flags, pfn; int batch_count = 0; + struct list_head dsts[__MAX_NR_ZONES]; + int i; + + for (i = 0; i < __MAX_NR_ZONES; i++) + INIT_LIST_HEAD(&dsts[i]); /* Prepare pages for freeing */ list_for_each_entry_safe(page, next, list, lru) { @@ -3171,25 +3223,42 @@ void free_unref_page_list(struct list_head *list) set_page_private(page, pfn); } - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); list_for_each_entry_safe(page, next, list, lru) { unsigned long pfn = page_private(page); + enum zone_type type; set_page_private(page, 0); trace_mm_page_free_batched(page); - free_unref_page_commit(page, pfn); + type = page_zonenum(page); + free_unref_page_commit(page, pfn, &dsts[type]); /* * Guard against excessive IRQ disabled times when we get * a large list of pages to free. */ if (++batch_count == SWAP_CLUSTER_MAX) { - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); batch_count = 0; - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); } } - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); + + for (i = 0; i < __MAX_NR_ZONES; ) { + struct page *page; + struct zone *zone; + + if (list_empty(&dsts[i])) { + i++; + continue; + } + + page = list_first_entry(&dsts[i], struct page, lru); + zone = page_zone(page); + + free_pcppages_bulk(zone, &dsts[i], true); + } } /* @@ -3343,7 +3412,7 @@ static struct page *rmqueue_pcplist(struct zone *preferred_zone, struct page *page; unsigned long flags; - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); pcp = &this_cpu_ptr(zone->pageset)->pcp; list = &pcp->lists[migratetype]; page = __rmqueue_pcplist(zone, migratetype, alloc_flags, pcp, list); @@ -3351,7 +3420,7 @@ static struct page *rmqueue_pcplist(struct zone *preferred_zone, __count_zid_vm_events(PGALLOC, page_zonenum(page), 1); zone_statistics(preferred_zone, zone); } - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); return page; } @@ -3385,7 +3454,8 @@ struct page *rmqueue(struct zone *preferred_zone, * allocate greater than order-1 page units with __GFP_NOFAIL. */ WARN_ON_ONCE((gfp_flags & __GFP_NOFAIL) && (order > 1)); - spin_lock_irqsave(&zone->lock, flags); + local_lock_irqsave(&pa_lock.l, flags); + spin_lock(&zone->lock); do { page = NULL; @@ -3411,7 +3481,7 @@ struct page *rmqueue(struct zone *preferred_zone, __count_zid_vm_events(PGALLOC, page_zonenum(page), 1 << order); zone_statistics(preferred_zone, zone); - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); out: /* Separate test+clear to avoid unnecessary atomics */ @@ -3424,7 +3494,7 @@ struct page *rmqueue(struct zone *preferred_zone, return page; failed: - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); return NULL; } @@ -8697,7 +8767,7 @@ void zone_pcp_reset(struct zone *zone) struct per_cpu_pageset *pset; /* avoid races with drain_pages() */ - local_irq_save(flags); + local_lock_irqsave(&pa_lock.l, flags); if (zone->pageset != &boot_pageset) { for_each_online_cpu(cpu) { pset = per_cpu_ptr(zone->pageset, cpu); @@ -8706,7 +8776,7 @@ void zone_pcp_reset(struct zone *zone) free_percpu(zone->pageset); zone->pageset = &boot_pageset; } - local_irq_restore(flags); + local_unlock_irqrestore(&pa_lock.l, flags); } #ifdef CONFIG_MEMORY_HOTREMOVE diff --git a/mm/shmem.c b/mm/shmem.c index 8e2b35ba93ad1..e029b943ebed8 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -278,10 +278,10 @@ static int shmem_reserve_inode(struct super_block *sb, ino_t *inop) ino_t ino; if (!(sb->s_flags & SB_KERNMOUNT)) { - spin_lock(&sbinfo->stat_lock); + raw_spin_lock(&sbinfo->stat_lock); if (sbinfo->max_inodes) { if (!sbinfo->free_inodes) { - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); return -ENOSPC; } sbinfo->free_inodes--; @@ -304,7 +304,7 @@ static int shmem_reserve_inode(struct super_block *sb, ino_t *inop) } *inop = ino; } - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); } else if (inop) { /* * __shmem_file_setup, one of our callers, is lock-free: it @@ -319,13 +319,14 @@ static int shmem_reserve_inode(struct super_block *sb, ino_t *inop) * to worry about things like glibc compatibility. */ ino_t *next_ino; + next_ino = per_cpu_ptr(sbinfo->ino_batch, get_cpu()); ino = *next_ino; if (unlikely(ino % SHMEM_INO_BATCH == 0)) { - spin_lock(&sbinfo->stat_lock); + raw_spin_lock(&sbinfo->stat_lock); ino = sbinfo->next_ino; sbinfo->next_ino += SHMEM_INO_BATCH; - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); if (unlikely(is_zero_ino(ino))) ino++; } @@ -341,9 +342,9 @@ static void shmem_free_inode(struct super_block *sb) { struct shmem_sb_info *sbinfo = SHMEM_SB(sb); if (sbinfo->max_inodes) { - spin_lock(&sbinfo->stat_lock); + raw_spin_lock(&sbinfo->stat_lock); sbinfo->free_inodes++; - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); } } @@ -1479,10 +1480,10 @@ static struct mempolicy *shmem_get_sbmpol(struct shmem_sb_info *sbinfo) { struct mempolicy *mpol = NULL; if (sbinfo->mpol) { - spin_lock(&sbinfo->stat_lock); /* prevent replace/use races */ + raw_spin_lock(&sbinfo->stat_lock); /* prevent replace/use races */ mpol = sbinfo->mpol; mpol_get(mpol); - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); } return mpol; } @@ -3582,9 +3583,10 @@ static int shmem_reconfigure(struct fs_context *fc) struct shmem_options *ctx = fc->fs_private; struct shmem_sb_info *sbinfo = SHMEM_SB(fc->root->d_sb); unsigned long inodes; + struct mempolicy *mpol = NULL; const char *err; - spin_lock(&sbinfo->stat_lock); + raw_spin_lock(&sbinfo->stat_lock); inodes = sbinfo->max_inodes - sbinfo->free_inodes; if ((ctx->seen & SHMEM_SEEN_BLOCKS) && ctx->blocks) { if (!sbinfo->max_blocks) { @@ -3629,14 +3631,15 @@ static int shmem_reconfigure(struct fs_context *fc) * Preserve previous mempolicy unless mpol remount option was specified. */ if (ctx->mpol) { - mpol_put(sbinfo->mpol); + mpol = sbinfo->mpol; sbinfo->mpol = ctx->mpol; /* transfers initial ref */ ctx->mpol = NULL; } - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); + mpol_put(mpol); return 0; out: - spin_unlock(&sbinfo->stat_lock); + raw_spin_unlock(&sbinfo->stat_lock); return invalfc(fc, "%s", err); } @@ -3753,7 +3756,7 @@ static int shmem_fill_super(struct super_block *sb, struct fs_context *fc) sbinfo->mpol = ctx->mpol; ctx->mpol = NULL; - spin_lock_init(&sbinfo->stat_lock); + raw_spin_lock_init(&sbinfo->stat_lock); if (percpu_counter_init(&sbinfo->used_blocks, 0, GFP_KERNEL)) goto failed; spin_lock_init(&sbinfo->shrinklist_lock); diff --git a/mm/slab.c b/mm/slab.c index f658e86ec8cee..3dbddaad8a32b 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -233,7 +233,7 @@ static void kmem_cache_node_init(struct kmem_cache_node *parent) parent->shared = NULL; parent->alien = NULL; parent->colour_next = 0; - spin_lock_init(&parent->list_lock); + raw_spin_lock_init(&parent->list_lock); parent->free_objects = 0; parent->free_touched = 0; } @@ -558,9 +558,9 @@ static noinline void cache_free_pfmemalloc(struct kmem_cache *cachep, page_node = page_to_nid(page); n = get_node(cachep, page_node); - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); free_block(cachep, &objp, 1, page_node, &list); - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); slabs_destroy(cachep, &list); } @@ -698,7 +698,7 @@ static void __drain_alien_cache(struct kmem_cache *cachep, struct kmem_cache_node *n = get_node(cachep, node); if (ac->avail) { - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); /* * Stuff objects into the remote nodes shared array first. * That way we could avoid the overhead of putting the objects @@ -709,7 +709,7 @@ static void __drain_alien_cache(struct kmem_cache *cachep, free_block(cachep, ac->entry, ac->avail, node, list); ac->avail = 0; - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); } } @@ -782,9 +782,9 @@ static int __cache_free_alien(struct kmem_cache *cachep, void *objp, slabs_destroy(cachep, &list); } else { n = get_node(cachep, page_node); - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); free_block(cachep, &objp, 1, page_node, &list); - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); slabs_destroy(cachep, &list); } return 1; @@ -825,10 +825,10 @@ static int init_cache_node(struct kmem_cache *cachep, int node, gfp_t gfp) */ n = get_node(cachep, node); if (n) { - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); n->free_limit = (1 + nr_cpus_node(node)) * cachep->batchcount + cachep->num; - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); return 0; } @@ -907,7 +907,7 @@ static int setup_kmem_cache_node(struct kmem_cache *cachep, goto fail; n = get_node(cachep, node); - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); if (n->shared && force_change) { free_block(cachep, n->shared->entry, n->shared->avail, node, &list); @@ -925,7 +925,7 @@ static int setup_kmem_cache_node(struct kmem_cache *cachep, new_alien = NULL; } - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); slabs_destroy(cachep, &list); /* @@ -964,7 +964,7 @@ static void cpuup_canceled(long cpu) if (!n) continue; - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); /* Free limit for this kmem_cache_node */ n->free_limit -= cachep->batchcount; @@ -975,7 +975,7 @@ static void cpuup_canceled(long cpu) nc->avail = 0; if (!cpumask_empty(mask)) { - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); goto free_slab; } @@ -989,7 +989,7 @@ static void cpuup_canceled(long cpu) alien = n->alien; n->alien = NULL; - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); kfree(shared); if (alien) { @@ -1173,7 +1173,7 @@ static void __init init_list(struct kmem_cache *cachep, struct kmem_cache_node * /* * Do not assume that spinlocks can be initialized via memcpy: */ - spin_lock_init(&ptr->list_lock); + raw_spin_lock_init(&ptr->list_lock); MAKE_ALL_LISTS(cachep, ptr, nodeid); cachep->node[nodeid] = ptr; @@ -1344,11 +1344,11 @@ slab_out_of_memory(struct kmem_cache *cachep, gfp_t gfpflags, int nodeid) for_each_kmem_cache_node(cachep, node, n) { unsigned long total_slabs, free_slabs, free_objs; - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); total_slabs = n->total_slabs; free_slabs = n->free_slabs; free_objs = n->free_objects; - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); pr_warn(" node %d: slabs: %ld/%ld, objs: %ld/%ld\n", node, total_slabs - free_slabs, total_slabs, @@ -2106,7 +2106,7 @@ static void check_spinlock_acquired(struct kmem_cache *cachep) { #ifdef CONFIG_SMP check_irq_off(); - assert_spin_locked(&get_node(cachep, numa_mem_id())->list_lock); + assert_raw_spin_locked(&get_node(cachep, numa_mem_id())->list_lock); #endif } @@ -2114,7 +2114,7 @@ static void check_spinlock_acquired_node(struct kmem_cache *cachep, int node) { #ifdef CONFIG_SMP check_irq_off(); - assert_spin_locked(&get_node(cachep, node)->list_lock); + assert_raw_spin_locked(&get_node(cachep, node)->list_lock); #endif } @@ -2154,9 +2154,9 @@ static void do_drain(void *arg) check_irq_off(); ac = cpu_cache_get(cachep); n = get_node(cachep, node); - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); free_block(cachep, ac->entry, ac->avail, node, &list); - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); ac->avail = 0; slabs_destroy(cachep, &list); } @@ -2174,9 +2174,9 @@ static void drain_cpu_caches(struct kmem_cache *cachep) drain_alien_cache(cachep, n->alien); for_each_kmem_cache_node(cachep, node, n) { - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); drain_array_locked(cachep, n->shared, node, true, &list); - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); slabs_destroy(cachep, &list); } @@ -2198,10 +2198,10 @@ static int drain_freelist(struct kmem_cache *cache, nr_freed = 0; while (nr_freed < tofree && !list_empty(&n->slabs_free)) { - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); p = n->slabs_free.prev; if (p == &n->slabs_free) { - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); goto out; } @@ -2214,7 +2214,7 @@ static int drain_freelist(struct kmem_cache *cache, * to the cache. */ n->free_objects -= cache->num; - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); slab_destroy(cache, page); nr_freed++; } @@ -2652,7 +2652,7 @@ static void cache_grow_end(struct kmem_cache *cachep, struct page *page) INIT_LIST_HEAD(&page->slab_list); n = get_node(cachep, page_to_nid(page)); - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); n->total_slabs++; if (!page->active) { list_add_tail(&page->slab_list, &n->slabs_free); @@ -2662,7 +2662,7 @@ static void cache_grow_end(struct kmem_cache *cachep, struct page *page) STATS_INC_GROWN(cachep); n->free_objects += cachep->num - page->active; - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); fixup_objfreelist_debug(cachep, &list); } @@ -2828,7 +2828,7 @@ static struct page *get_first_slab(struct kmem_cache_node *n, bool pfmemalloc) { struct page *page; - assert_spin_locked(&n->list_lock); + assert_raw_spin_locked(&n->list_lock); page = list_first_entry_or_null(&n->slabs_partial, struct page, slab_list); if (!page) { @@ -2855,10 +2855,10 @@ static noinline void *cache_alloc_pfmemalloc(struct kmem_cache *cachep, if (!gfp_pfmemalloc_allowed(flags)) return NULL; - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); page = get_first_slab(n, true); if (!page) { - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); return NULL; } @@ -2867,7 +2867,7 @@ static noinline void *cache_alloc_pfmemalloc(struct kmem_cache *cachep, fixup_slab_list(cachep, n, page, &list); - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); fixup_objfreelist_debug(cachep, &list); return obj; @@ -2926,7 +2926,7 @@ static void *cache_alloc_refill(struct kmem_cache *cachep, gfp_t flags) if (!n->free_objects && (!shared || !shared->avail)) goto direct_grow; - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); shared = READ_ONCE(n->shared); /* See if we can refill from the shared array */ @@ -2950,7 +2950,7 @@ static void *cache_alloc_refill(struct kmem_cache *cachep, gfp_t flags) must_grow: n->free_objects -= ac->avail; alloc_done: - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); fixup_objfreelist_debug(cachep, &list); direct_grow: @@ -3175,7 +3175,7 @@ static void *____cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, BUG_ON(!n); check_irq_off(); - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); page = get_first_slab(n, false); if (!page) goto must_grow; @@ -3193,12 +3193,12 @@ static void *____cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, fixup_slab_list(cachep, n, page, &list); - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); fixup_objfreelist_debug(cachep, &list); return obj; must_grow: - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); page = cache_grow_begin(cachep, gfp_exact_node(flags), nodeid); if (page) { /* This slab isn't counted yet so don't update free_objects */ @@ -3376,7 +3376,7 @@ static void cache_flusharray(struct kmem_cache *cachep, struct array_cache *ac) check_irq_off(); n = get_node(cachep, node); - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); if (n->shared) { struct array_cache *shared_array = n->shared; int max = shared_array->limit - shared_array->avail; @@ -3405,7 +3405,7 @@ static void cache_flusharray(struct kmem_cache *cachep, struct array_cache *ac) STATS_SET_FREEABLE(cachep, i); } #endif - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); ac->avail -= batchcount; memmove(ac->entry, &(ac->entry[batchcount]), sizeof(void *)*ac->avail); slabs_destroy(cachep, &list); @@ -3834,9 +3834,9 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit, node = cpu_to_mem(cpu); n = get_node(cachep, node); - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); free_block(cachep, ac->entry, ac->avail, node, &list); - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); slabs_destroy(cachep, &list); } free_percpu(prev); @@ -3931,9 +3931,9 @@ static void drain_array(struct kmem_cache *cachep, struct kmem_cache_node *n, return; } - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); drain_array_locked(cachep, ac, node, false, &list); - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); slabs_destroy(cachep, &list); } @@ -4017,7 +4017,7 @@ void get_slabinfo(struct kmem_cache *cachep, struct slabinfo *sinfo) for_each_kmem_cache_node(cachep, node, n) { check_irq_on(); - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); total_slabs += n->total_slabs; free_slabs += n->free_slabs; @@ -4026,7 +4026,7 @@ void get_slabinfo(struct kmem_cache *cachep, struct slabinfo *sinfo) if (n->shared) shared_avail += n->shared->avail; - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); } num_objs = total_slabs * cachep->num; active_slabs = total_slabs - free_slabs; diff --git a/mm/slab.h b/mm/slab.h index 6cc323f1313af..089bcef627e62 100644 --- a/mm/slab.h +++ b/mm/slab.h @@ -530,7 +530,7 @@ static inline void slab_post_alloc_hook(struct kmem_cache *s, * The slab lists for all objects. */ struct kmem_cache_node { - spinlock_t list_lock; + raw_spinlock_t list_lock; #ifdef CONFIG_SLAB struct list_head slabs_partial; /* partial list first, better asm code */ diff --git a/mm/slub.c b/mm/slub.c index 6d3574013b2f8..795b9a3488df3 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -434,7 +434,7 @@ static inline bool cmpxchg_double_slab(struct kmem_cache *s, struct page *page, #ifdef CONFIG_SLUB_DEBUG static unsigned long object_map[BITS_TO_LONGS(MAX_OBJS_PER_PAGE)]; -static DEFINE_SPINLOCK(object_map_lock); +static DEFINE_RAW_SPINLOCK(object_map_lock); /* * Determine a map of object in use on a page. @@ -450,7 +450,7 @@ static unsigned long *get_map(struct kmem_cache *s, struct page *page) VM_BUG_ON(!irqs_disabled()); - spin_lock(&object_map_lock); + raw_spin_lock(&object_map_lock); bitmap_zero(object_map, page->objects); @@ -463,7 +463,7 @@ static unsigned long *get_map(struct kmem_cache *s, struct page *page) static void put_map(unsigned long *map) __releases(&object_map_lock) { VM_BUG_ON(map != object_map); - spin_unlock(&object_map_lock); + raw_spin_unlock(&object_map_lock); } static inline unsigned int size_from_object(struct kmem_cache *s) @@ -1213,7 +1213,7 @@ static noinline int free_debug_processing( unsigned long flags; int ret = 0; - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); slab_lock(page); if (s->flags & SLAB_CONSISTENCY_CHECKS) { @@ -1248,7 +1248,7 @@ static noinline int free_debug_processing( bulk_cnt, cnt); slab_unlock(page); - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); if (!ret) slab_fix(s, "Object at 0x%p not freed", object); return ret; @@ -1496,6 +1496,12 @@ static bool freelist_corrupted(struct kmem_cache *s, struct page *page, } #endif /* CONFIG_SLUB_DEBUG */ +struct slub_free_list { + raw_spinlock_t lock; + struct list_head list; +}; +static DEFINE_PER_CPU(struct slub_free_list, slub_free_list); + /* * Hooks for other subsystems that check memory allocations. In a typical * production configuration these hooks all should produce no code at all. @@ -1739,10 +1745,18 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node) void *start, *p, *next; int idx; bool shuffle; + bool enableirqs = false; flags &= gfp_allowed_mask; if (gfpflags_allow_blocking(flags)) + enableirqs = true; + +#ifdef CONFIG_PREEMPT_RT + if (system_state > SYSTEM_BOOTING && system_state < SYSTEM_SUSPEND) + enableirqs = true; +#endif + if (enableirqs) local_irq_enable(); flags |= s->allocflags; @@ -1801,7 +1815,7 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node) page->frozen = 1; out: - if (gfpflags_allow_blocking(flags)) + if (enableirqs) local_irq_disable(); if (!page) return NULL; @@ -1844,6 +1858,16 @@ static void __free_slab(struct kmem_cache *s, struct page *page) __free_pages(page, order); } +static void free_delayed(struct list_head *h) +{ + while (!list_empty(h)) { + struct page *page = list_first_entry(h, struct page, lru); + + list_del(&page->lru); + __free_slab(page->slab_cache, page); + } +} + static void rcu_free_slab(struct rcu_head *h) { struct page *page = container_of(h, struct page, rcu_head); @@ -1855,6 +1879,12 @@ static void free_slab(struct kmem_cache *s, struct page *page) { if (unlikely(s->flags & SLAB_TYPESAFE_BY_RCU)) { call_rcu(&page->rcu_head, rcu_free_slab); + } else if (irqs_disabled()) { + struct slub_free_list *f = this_cpu_ptr(&slub_free_list); + + raw_spin_lock(&f->lock); + list_add(&page->lru, &f->list); + raw_spin_unlock(&f->lock); } else __free_slab(s, page); } @@ -1962,7 +1992,7 @@ static void *get_partial_node(struct kmem_cache *s, struct kmem_cache_node *n, if (!n || !n->nr_partial) return NULL; - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); list_for_each_entry_safe(page, page2, &n->partial, slab_list) { void *t; @@ -1987,7 +2017,7 @@ static void *get_partial_node(struct kmem_cache *s, struct kmem_cache_node *n, break; } - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); return object; } @@ -2241,7 +2271,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page, * that acquire_slab() will see a slab page that * is frozen */ - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); } } else { m = M_FULL; @@ -2252,7 +2282,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page, * slabs from diagnostic functions will not see * any frozen slabs. */ - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); } } @@ -2276,7 +2306,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page, goto redo; if (lock) - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); if (m == M_PARTIAL) stat(s, tail); @@ -2315,10 +2345,10 @@ static void unfreeze_partials(struct kmem_cache *s, n2 = get_node(s, page_to_nid(page)); if (n != n2) { if (n) - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); n = n2; - spin_lock(&n->list_lock); + raw_spin_lock(&n->list_lock); } do { @@ -2347,7 +2377,7 @@ static void unfreeze_partials(struct kmem_cache *s, } if (n) - spin_unlock(&n->list_lock); + raw_spin_unlock(&n->list_lock); while (discard_page) { page = discard_page; @@ -2384,14 +2414,21 @@ static void put_cpu_partial(struct kmem_cache *s, struct page *page, int drain) pobjects = oldpage->pobjects; pages = oldpage->pages; if (drain && pobjects > slub_cpu_partial(s)) { + struct slub_free_list *f; unsigned long flags; + LIST_HEAD(tofree); /* * partial array is full. Move the existing * set to the per node partial list. */ local_irq_save(flags); unfreeze_partials(s, this_cpu_ptr(s->cpu_slab)); + f = this_cpu_ptr(&slub_free_list); + raw_spin_lock(&f->lock); + list_splice_init(&f->list, &tofree); + raw_spin_unlock(&f->lock); local_irq_restore(flags); + free_delayed(&tofree); oldpage = NULL; pobjects = 0; pages = 0; @@ -2459,7 +2496,19 @@ static bool has_cpu_slab(int cpu, void *info) static void flush_all(struct kmem_cache *s) { + LIST_HEAD(tofree); + int cpu; + on_each_cpu_cond(has_cpu_slab, flush_cpu_slab, s, 1); + for_each_online_cpu(cpu) { + struct slub_free_list *f; + + f = &per_cpu(slub_free_list, cpu); + raw_spin_lock_irq(&f->lock); + list_splice_init(&f->list, &tofree); + raw_spin_unlock_irq(&f->lock); + free_delayed(&tofree); + } } /* @@ -2514,10 +2563,10 @@ static unsigned long count_partial(struct kmem_cache_node *n, unsigned long x = 0; struct page *page; - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); list_for_each_entry(page, &n->partial, slab_list) x += get_count(page); - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); return x; } #endif /* CONFIG_SLUB_DEBUG || CONFIG_SYSFS */ @@ -2656,8 +2705,10 @@ static inline void *get_freelist(struct kmem_cache *s, struct page *page) * already disabled (which is the case for bulk allocation). */ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, - unsigned long addr, struct kmem_cache_cpu *c) + unsigned long addr, struct kmem_cache_cpu *c, + struct list_head *to_free) { + struct slub_free_list *f; void *freelist; struct page *page; @@ -2723,6 +2774,13 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, VM_BUG_ON(!c->page->frozen); c->freelist = get_freepointer(s, freelist); c->tid = next_tid(c->tid); + +out: + f = this_cpu_ptr(&slub_free_list); + raw_spin_lock(&f->lock); + list_splice_init(&f->list, to_free); + raw_spin_unlock(&f->lock); + return freelist; new_slab: @@ -2738,7 +2796,7 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, if (unlikely(!freelist)) { slab_out_of_memory(s, gfpflags, node); - return NULL; + goto out; } page = c->page; @@ -2751,7 +2809,7 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, goto new_slab; /* Slab failed checks. Next slab needed */ deactivate_slab(s, page, get_freepointer(s, freelist), c); - return freelist; + goto out; } /* @@ -2763,6 +2821,7 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, { void *p; unsigned long flags; + LIST_HEAD(tofree); local_irq_save(flags); #ifdef CONFIG_PREEMPTION @@ -2774,8 +2833,9 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node, c = this_cpu_ptr(s->cpu_slab); #endif - p = ___slab_alloc(s, gfpflags, node, addr, c); + p = ___slab_alloc(s, gfpflags, node, addr, c, &tofree); local_irq_restore(flags); + free_delayed(&tofree); return p; } @@ -2809,6 +2869,10 @@ static __always_inline void *slab_alloc_node(struct kmem_cache *s, unsigned long tid; struct obj_cgroup *objcg = NULL; + if (IS_ENABLED(CONFIG_PREEMPT_RT) && IS_ENABLED(CONFIG_DEBUG_ATOMIC_SLEEP)) + WARN_ON_ONCE(!preemptible() && + (system_state > SYSTEM_BOOTING && system_state < SYSTEM_SUSPEND)); + s = slab_pre_alloc_hook(s, &objcg, 1, gfpflags); if (!s) return NULL; @@ -2975,7 +3039,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, do { if (unlikely(n)) { - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); n = NULL; } prior = page->freelist; @@ -3007,7 +3071,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, * Otherwise the list_lock will synchronize with * other processors updating the list of slabs. */ - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); } } @@ -3048,7 +3112,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, add_partial(n, page, DEACTIVATE_TO_TAIL); stat(s, FREE_ADD_PARTIAL); } - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); return; slab_empty: @@ -3063,7 +3127,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page, remove_full(s, n, page); } - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); stat(s, FREE_SLAB); discard_slab(s, page); } @@ -3270,9 +3334,14 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size, void **p) { struct kmem_cache_cpu *c; + LIST_HEAD(to_free); int i; struct obj_cgroup *objcg = NULL; + if (IS_ENABLED(CONFIG_PREEMPT_RT) && IS_ENABLED(CONFIG_DEBUG_ATOMIC_SLEEP)) + WARN_ON_ONCE(!preemptible() && + (system_state > SYSTEM_BOOTING && system_state < SYSTEM_SUSPEND)); + /* memcg and kmem_cache debug support */ s = slab_pre_alloc_hook(s, &objcg, size, flags); if (unlikely(!s)) @@ -3303,7 +3372,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size, * of re-populating per CPU c->freelist */ p[i] = ___slab_alloc(s, flags, NUMA_NO_NODE, - _RET_IP_, c); + _RET_IP_, c, &to_free); if (unlikely(!p[i])) goto error; @@ -3318,6 +3387,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size, } c->tid = next_tid(c->tid); local_irq_enable(); + free_delayed(&to_free); /* Clear memory outside IRQ disabled fastpath loop */ if (unlikely(slab_want_init_on_alloc(flags, s))) { @@ -3332,6 +3402,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size, return i; error: local_irq_enable(); + free_delayed(&to_free); slab_post_alloc_hook(s, objcg, flags, i, p); __kmem_cache_free_bulk(s, i, p); return 0; @@ -3467,7 +3538,7 @@ static void init_kmem_cache_node(struct kmem_cache_node *n) { n->nr_partial = 0; - spin_lock_init(&n->list_lock); + raw_spin_lock_init(&n->list_lock); INIT_LIST_HEAD(&n->partial); #ifdef CONFIG_SLUB_DEBUG atomic_long_set(&n->nr_slabs, 0); @@ -3868,7 +3939,7 @@ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n) struct page *page, *h; BUG_ON(irqs_disabled()); - spin_lock_irq(&n->list_lock); + raw_spin_lock_irq(&n->list_lock); list_for_each_entry_safe(page, h, &n->partial, slab_list) { if (!page->inuse) { remove_partial(n, page); @@ -3878,7 +3949,7 @@ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n) "Objects remaining in %s on __kmem_cache_shutdown()"); } } - spin_unlock_irq(&n->list_lock); + raw_spin_unlock_irq(&n->list_lock); list_for_each_entry_safe(page, h, &discard, slab_list) discard_slab(s, page); @@ -4149,7 +4220,7 @@ int __kmem_cache_shrink(struct kmem_cache *s) for (i = 0; i < SHRINK_PROMOTE_MAX; i++) INIT_LIST_HEAD(promote + i); - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); /* * Build lists of slabs to discard or promote. @@ -4180,7 +4251,7 @@ int __kmem_cache_shrink(struct kmem_cache *s) for (i = SHRINK_PROMOTE_MAX - 1; i >= 0; i--) list_splice(promote + i, &n->partial); - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); /* Release empty slabs */ list_for_each_entry_safe(page, t, &discard, slab_list) @@ -4355,6 +4426,12 @@ void __init kmem_cache_init(void) { static __initdata struct kmem_cache boot_kmem_cache, boot_kmem_cache_node; + int cpu; + + for_each_possible_cpu(cpu) { + raw_spin_lock_init(&per_cpu(slub_free_list, cpu).lock); + INIT_LIST_HEAD(&per_cpu(slub_free_list, cpu).list); + } if (debug_guardpage_minorder()) slub_max_order = 0; @@ -4542,7 +4619,7 @@ static int validate_slab_node(struct kmem_cache *s, struct page *page; unsigned long flags; - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); list_for_each_entry(page, &n->partial, slab_list) { validate_slab(s, page); @@ -4564,7 +4641,7 @@ static int validate_slab_node(struct kmem_cache *s, s->name, count, atomic_long_read(&n->nr_slabs)); out: - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); return count; } @@ -4743,12 +4820,12 @@ static int list_locations(struct kmem_cache *s, char *buf, if (!atomic_long_read(&n->nr_slabs)) continue; - spin_lock_irqsave(&n->list_lock, flags); + raw_spin_lock_irqsave(&n->list_lock, flags); list_for_each_entry(page, &n->partial, slab_list) process_slab(&t, s, page, alloc); list_for_each_entry(page, &n->full, slab_list) process_slab(&t, s, page, alloc); - spin_unlock_irqrestore(&n->list_lock, flags); + raw_spin_unlock_irqrestore(&n->list_lock, flags); } for (i = 0; i < t.count; i++) { diff --git a/mm/swap.c b/mm/swap.c index e7bdf094f76a0..65ef7e3525bfa 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -763,10 +763,20 @@ static void lru_add_drain_per_cpu(struct work_struct *dummy) */ void lru_add_drain_all(void) { - static seqcount_t seqcount = SEQCNT_ZERO(seqcount); - static DEFINE_MUTEX(lock); + /* + * lru_drain_gen - Global pages generation number + * + * (A) Definition: global lru_drain_gen = x implies that all generations + * 0 < n <= x are already *scheduled* for draining. + * + * This is an optimization for the highly-contended use case where a + * user space workload keeps constantly generating a flow of pages for + * each CPU. + */ + static unsigned int lru_drain_gen; static struct cpumask has_work; - int cpu, seq; + static DEFINE_MUTEX(lock); + unsigned cpu, this_gen; /* * Make sure nobody triggers this path before mm_percpu_wq is fully @@ -775,21 +785,54 @@ void lru_add_drain_all(void) if (WARN_ON(!mm_percpu_wq)) return; - seq = raw_read_seqcount_latch(&seqcount); + /* + * Guarantee pagevec counter stores visible by this CPU are visible to + * other CPUs before loading the current drain generation. + */ + smp_mb(); + + /* + * (B) Locally cache global LRU draining generation number + * + * The read barrier ensures that the counter is loaded before the mutex + * is taken. It pairs with smp_mb() inside the mutex critical section + * at (D). + */ + this_gen = smp_load_acquire(&lru_drain_gen); mutex_lock(&lock); /* - * Piggyback on drain started and finished while we waited for lock: - * all pages pended at the time of our enter were drained from vectors. + * (C) Exit the draining operation if a newer generation, from another + * lru_add_drain_all(), was already scheduled for draining. Check (A). */ - if (__read_seqcount_retry(&seqcount, seq)) + if (unlikely(this_gen != lru_drain_gen)) goto done; - raw_write_seqcount_latch(&seqcount); + /* + * (D) Increment global generation number + * + * Pairs with smp_load_acquire() at (B), outside of the critical + * section. Use a full memory barrier to guarantee that the new global + * drain generation number is stored before loading pagevec counters. + * + * This pairing must be done here, before the for_each_online_cpu loop + * below which drains the page vectors. + * + * Let x, y, and z represent some system CPU numbers, where x < y < z. + * Assume CPU #z is is in the middle of the for_each_online_cpu loop + * below and has already reached CPU #y's per-cpu data. CPU #x comes + * along, adds some pages to its per-cpu vectors, then calls + * lru_add_drain_all(). + * + * If the paired barrier is done at any later step, e.g. after the + * loop, CPU #x will just exit at (C) and miss flushing out all of its + * added pages. + */ + WRITE_ONCE(lru_drain_gen, lru_drain_gen + 1); + smp_mb(); cpumask_clear(&has_work); - for_each_online_cpu(cpu) { struct work_struct *work = &per_cpu(lru_add_drain_work, cpu); @@ -801,7 +844,7 @@ void lru_add_drain_all(void) need_activate_page_drain(cpu)) { INIT_WORK(work, lru_add_drain_per_cpu); queue_work_on(cpu, mm_percpu_wq, work); - cpumask_set_cpu(cpu, &has_work); + __cpumask_set_cpu(cpu, &has_work); } } @@ -816,7 +859,7 @@ void lru_add_drain_all(void) { lru_add_drain(); } -#endif +#endif /* CONFIG_SMP */ /** * release_pages - batched put_page() diff --git a/mm/vmalloc.c b/mm/vmalloc.c index be4724b916b3e..994acb959d00d 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -1544,7 +1544,7 @@ static void *new_vmap_block(unsigned int order, gfp_t gfp_mask) struct vmap_block *vb; struct vmap_area *va; unsigned long vb_idx; - int node, err; + int node, err, cpu; void *vaddr; node = numa_node_id(); @@ -1581,11 +1581,12 @@ static void *new_vmap_block(unsigned int order, gfp_t gfp_mask) return ERR_PTR(err); } - vbq = &get_cpu_var(vmap_block_queue); + cpu = get_cpu_light(); + vbq = this_cpu_ptr(&vmap_block_queue); spin_lock(&vbq->lock); list_add_tail_rcu(&vb->free_list, &vbq->free); spin_unlock(&vbq->lock); - put_cpu_var(vmap_block_queue); + put_cpu_light(); return vaddr; } @@ -1650,6 +1651,7 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask) struct vmap_block *vb; void *vaddr = NULL; unsigned int order; + int cpu; BUG_ON(offset_in_page(size)); BUG_ON(size > PAGE_SIZE*VMAP_MAX_ALLOC); @@ -1664,7 +1666,8 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask) order = get_order(size); rcu_read_lock(); - vbq = &get_cpu_var(vmap_block_queue); + cpu = get_cpu_light(); + vbq = this_cpu_ptr(&vmap_block_queue); list_for_each_entry_rcu(vb, &vbq->free, free_list) { unsigned long pages_off; @@ -1687,7 +1690,7 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask) break; } - put_cpu_var(vmap_block_queue); + put_cpu_light(); rcu_read_unlock(); /* Allocate new block if nothing was found */ diff --git a/mm/vmstat.c b/mm/vmstat.c index 4f7b4ee6aa123..0fc677378d3dd 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -321,6 +321,7 @@ void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item, long x; long t; + preempt_disable_rt(); x = delta + __this_cpu_read(*p); t = __this_cpu_read(pcp->stat_threshold); @@ -330,6 +331,7 @@ void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item, x = 0; } __this_cpu_write(*p, x); + preempt_enable_rt(); } EXPORT_SYMBOL(__mod_zone_page_state); @@ -346,6 +348,7 @@ void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, delta >>= PAGE_SHIFT; } + preempt_disable_rt(); x = delta + __this_cpu_read(*p); t = __this_cpu_read(pcp->stat_threshold); @@ -355,6 +358,7 @@ void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item, x = 0; } __this_cpu_write(*p, x); + preempt_enable_rt(); } EXPORT_SYMBOL(__mod_node_page_state); @@ -387,6 +391,7 @@ void __inc_zone_state(struct zone *zone, enum zone_stat_item item) s8 __percpu *p = pcp->vm_stat_diff + item; s8 v, t; + preempt_disable_rt(); v = __this_cpu_inc_return(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v > t)) { @@ -395,6 +400,7 @@ void __inc_zone_state(struct zone *zone, enum zone_stat_item item) zone_page_state_add(v + overstep, zone, item); __this_cpu_write(*p, -overstep); } + preempt_enable_rt(); } void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item) @@ -405,6 +411,7 @@ void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item) VM_WARN_ON_ONCE(vmstat_item_in_bytes(item)); + preempt_disable_rt(); v = __this_cpu_inc_return(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v > t)) { @@ -413,6 +420,7 @@ void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item) node_page_state_add(v + overstep, pgdat, item); __this_cpu_write(*p, -overstep); } + preempt_enable_rt(); } void __inc_zone_page_state(struct page *page, enum zone_stat_item item) @@ -433,6 +441,7 @@ void __dec_zone_state(struct zone *zone, enum zone_stat_item item) s8 __percpu *p = pcp->vm_stat_diff + item; s8 v, t; + preempt_disable_rt(); v = __this_cpu_dec_return(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v < - t)) { @@ -441,6 +450,7 @@ void __dec_zone_state(struct zone *zone, enum zone_stat_item item) zone_page_state_add(v - overstep, zone, item); __this_cpu_write(*p, overstep); } + preempt_enable_rt(); } void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item) @@ -451,6 +461,7 @@ void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item) VM_WARN_ON_ONCE(vmstat_item_in_bytes(item)); + preempt_disable_rt(); v = __this_cpu_dec_return(*p); t = __this_cpu_read(pcp->stat_threshold); if (unlikely(v < - t)) { @@ -459,6 +470,7 @@ void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item) node_page_state_add(v - overstep, pgdat, item); __this_cpu_write(*p, overstep); } + preempt_enable_rt(); } void __dec_zone_page_state(struct page *page, enum zone_stat_item item) diff --git a/mm/workingset.c b/mm/workingset.c index 92e66113a5779..29609352552fc 100644 --- a/mm/workingset.c +++ b/mm/workingset.c @@ -432,6 +432,8 @@ static struct list_lru shadow_nodes; void workingset_update_node(struct xa_node *node) { + struct address_space *mapping; + /* * Track non-empty nodes that contain only shadow entries; * unlink those that contain pages or are being freed. @@ -440,7 +442,8 @@ void workingset_update_node(struct xa_node *node) * already where they should be. The list_empty() test is safe * as node->private_list is protected by the i_pages lock. */ - VM_WARN_ON_ONCE(!irqs_disabled()); /* For __inc_lruvec_page_state */ + mapping = container_of(node->array, struct address_space, i_pages); + lockdep_assert_held(&mapping->i_pages.xa_lock); if (node->count && node->count == node->nr_values) { if (list_empty(&node->private_list)) { diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index c36fdff9a3713..2cc22ee7b8945 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -57,6 +57,7 @@ #include #include #include +#include #define ZSPAGE_MAGIC 0x58 @@ -77,6 +78,20 @@ #define ZS_HANDLE_SIZE (sizeof(unsigned long)) +#ifdef CONFIG_PREEMPT_RT + +struct zsmalloc_handle { + unsigned long addr; + struct mutex lock; +}; + +#define ZS_HANDLE_ALLOC_SIZE (sizeof(struct zsmalloc_handle)) + +#else + +#define ZS_HANDLE_ALLOC_SIZE (sizeof(unsigned long)) +#endif + /* * Object location (, ) is encoded as * a single (unsigned long) handle value. @@ -293,6 +308,7 @@ struct zspage { }; struct mapping_area { + local_lock_t lock; #ifdef CONFIG_ZSMALLOC_PGTABLE_MAPPING struct vm_struct *vm; /* vm area for mapping object that span pages */ #else @@ -326,7 +342,7 @@ static void SetZsPageMovable(struct zs_pool *pool, struct zspage *zspage) {} static int create_cache(struct zs_pool *pool) { - pool->handle_cachep = kmem_cache_create("zs_handle", ZS_HANDLE_SIZE, + pool->handle_cachep = kmem_cache_create("zs_handle", ZS_HANDLE_ALLOC_SIZE, 0, 0, NULL); if (!pool->handle_cachep) return 1; @@ -350,10 +366,27 @@ static void destroy_cache(struct zs_pool *pool) static unsigned long cache_alloc_handle(struct zs_pool *pool, gfp_t gfp) { - return (unsigned long)kmem_cache_alloc(pool->handle_cachep, - gfp & ~(__GFP_HIGHMEM|__GFP_MOVABLE)); + void *p; + + p = kmem_cache_alloc(pool->handle_cachep, + gfp & ~(__GFP_HIGHMEM|__GFP_MOVABLE)); +#ifdef CONFIG_PREEMPT_RT + if (p) { + struct zsmalloc_handle *zh = p; + + mutex_init(&zh->lock); + } +#endif + return (unsigned long)p; } +#ifdef CONFIG_PREEMPT_RT +static struct zsmalloc_handle *zs_get_pure_handle(unsigned long handle) +{ + return (void *)(handle &~((1 << OBJ_TAG_BITS) - 1)); +} +#endif + static void cache_free_handle(struct zs_pool *pool, unsigned long handle) { kmem_cache_free(pool->handle_cachep, (void *)handle); @@ -372,12 +405,18 @@ static void cache_free_zspage(struct zs_pool *pool, struct zspage *zspage) static void record_obj(unsigned long handle, unsigned long obj) { +#ifdef CONFIG_PREEMPT_RT + struct zsmalloc_handle *zh = zs_get_pure_handle(handle); + + WRITE_ONCE(zh->addr, obj); +#else /* * lsb of @obj represents handle lock while other bits * represent object value the handle is pointing so * updating shouldn't do store tearing. */ WRITE_ONCE(*(unsigned long *)handle, obj); +#endif } /* zpool driver */ @@ -459,7 +498,10 @@ MODULE_ALIAS("zpool-zsmalloc"); #endif /* CONFIG_ZPOOL */ /* per-cpu VM mapping areas for zspage accesses that cross page boundaries */ -static DEFINE_PER_CPU(struct mapping_area, zs_map_area); +static DEFINE_PER_CPU(struct mapping_area, zs_map_area) = { + /* XXX remove this and use a spin_lock_t in pin_tag() */ + .lock = INIT_LOCAL_LOCK(lock), +}; static bool is_zspage_isolated(struct zspage *zspage) { @@ -869,7 +911,13 @@ static unsigned long location_to_obj(struct page *page, unsigned int obj_idx) static unsigned long handle_to_obj(unsigned long handle) { +#ifdef CONFIG_PREEMPT_RT + struct zsmalloc_handle *zh = zs_get_pure_handle(handle); + + return zh->addr; +#else return *(unsigned long *)handle; +#endif } static unsigned long obj_to_head(struct page *page, void *obj) @@ -883,22 +931,46 @@ static unsigned long obj_to_head(struct page *page, void *obj) static inline int testpin_tag(unsigned long handle) { +#ifdef CONFIG_PREEMPT_RT + struct zsmalloc_handle *zh = zs_get_pure_handle(handle); + + return mutex_is_locked(&zh->lock); +#else return bit_spin_is_locked(HANDLE_PIN_BIT, (unsigned long *)handle); +#endif } static inline int trypin_tag(unsigned long handle) { +#ifdef CONFIG_PREEMPT_RT + struct zsmalloc_handle *zh = zs_get_pure_handle(handle); + + return mutex_trylock(&zh->lock); +#else return bit_spin_trylock(HANDLE_PIN_BIT, (unsigned long *)handle); +#endif } static void pin_tag(unsigned long handle) __acquires(bitlock) { +#ifdef CONFIG_PREEMPT_RT + struct zsmalloc_handle *zh = zs_get_pure_handle(handle); + + return mutex_lock(&zh->lock); +#else bit_spin_lock(HANDLE_PIN_BIT, (unsigned long *)handle); +#endif } static void unpin_tag(unsigned long handle) __releases(bitlock) { +#ifdef CONFIG_PREEMPT_RT + struct zsmalloc_handle *zh = zs_get_pure_handle(handle); + + return mutex_unlock(&zh->lock); +#else bit_spin_unlock(HANDLE_PIN_BIT, (unsigned long *)handle); +#endif } static void reset_page(struct page *page) @@ -1326,7 +1398,8 @@ void *zs_map_object(struct zs_pool *pool, unsigned long handle, class = pool->size_class[class_idx]; off = (class->size * obj_idx) & ~PAGE_MASK; - area = &get_cpu_var(zs_map_area); + local_lock(&zs_map_area.lock); + area = this_cpu_ptr(&zs_map_area); area->vm_mm = mm; if (off + class->size <= PAGE_SIZE) { /* this object is contained entirely within a page */ @@ -1380,7 +1453,7 @@ void zs_unmap_object(struct zs_pool *pool, unsigned long handle) __zs_unmap_object(area, pages, off, class->size); } - put_cpu_var(zs_map_area); + local_unlock(&zs_map_area.lock); migrate_read_unlock(zspage); unpin_tag(handle); diff --git a/mm/zswap.c b/mm/zswap.c index fbb782924ccc5..78a20f7b00f2c 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -387,27 +388,35 @@ static struct zswap_entry *zswap_entry_find_get(struct rb_root *root, /********************************* * per-cpu code **********************************/ -static DEFINE_PER_CPU(u8 *, zswap_dstmem); +struct zswap_comp { + /* Used for per-CPU dstmem and tfm */ + local_lock_t lock; + u8 *dstmem; +}; + +static DEFINE_PER_CPU(struct zswap_comp, zswap_comp); static int zswap_dstmem_prepare(unsigned int cpu) { + struct zswap_comp *zcomp; u8 *dst; dst = kmalloc_node(PAGE_SIZE * 2, GFP_KERNEL, cpu_to_node(cpu)); if (!dst) return -ENOMEM; - per_cpu(zswap_dstmem, cpu) = dst; + zcomp = per_cpu_ptr(&zswap_comp, cpu); + zcomp->dstmem = dst; return 0; } static int zswap_dstmem_dead(unsigned int cpu) { - u8 *dst; + struct zswap_comp *zcomp; - dst = per_cpu(zswap_dstmem, cpu); - kfree(dst); - per_cpu(zswap_dstmem, cpu) = NULL; + zcomp = per_cpu_ptr(&zswap_comp, cpu); + kfree(zcomp->dstmem); + zcomp->dstmem = NULL; return 0; } @@ -919,10 +928,11 @@ static int zswap_writeback_entry(struct zpool *pool, unsigned long handle) dlen = PAGE_SIZE; src = (u8 *)zhdr + sizeof(struct zswap_header); dst = kmap_atomic(page); - tfm = *get_cpu_ptr(entry->pool->tfm); + local_lock(&zswap_comp.lock); + tfm = *this_cpu_ptr(entry->pool->tfm); ret = crypto_comp_decompress(tfm, src, entry->length, dst, &dlen); - put_cpu_ptr(entry->pool->tfm); + local_unlock(&zswap_comp.lock); kunmap_atomic(dst); BUG_ON(ret); BUG_ON(dlen != PAGE_SIZE); @@ -1074,12 +1084,12 @@ static int zswap_frontswap_store(unsigned type, pgoff_t offset, } /* compress */ - dst = get_cpu_var(zswap_dstmem); - tfm = *get_cpu_ptr(entry->pool->tfm); + local_lock(&zswap_comp.lock); + dst = *this_cpu_ptr(&zswap_comp.dstmem); + tfm = *this_cpu_ptr(entry->pool->tfm); src = kmap_atomic(page); ret = crypto_comp_compress(tfm, src, PAGE_SIZE, dst, &dlen); kunmap_atomic(src); - put_cpu_ptr(entry->pool->tfm); if (ret) { ret = -EINVAL; goto put_dstmem; @@ -1103,7 +1113,7 @@ static int zswap_frontswap_store(unsigned type, pgoff_t offset, memcpy(buf, &zhdr, hlen); memcpy(buf + hlen, dst, dlen); zpool_unmap_handle(entry->pool->zpool, handle); - put_cpu_var(zswap_dstmem); + local_unlock(&zswap_comp.lock); /* populate entry */ entry->offset = offset; @@ -1131,7 +1141,7 @@ static int zswap_frontswap_store(unsigned type, pgoff_t offset, return 0; put_dstmem: - put_cpu_var(zswap_dstmem); + local_unlock(&zswap_comp.lock); zswap_pool_put(entry->pool); freepage: zswap_entry_cache_free(entry); @@ -1176,9 +1186,10 @@ static int zswap_frontswap_load(unsigned type, pgoff_t offset, if (zpool_evictable(entry->pool->zpool)) src += sizeof(struct zswap_header); dst = kmap_atomic(page); - tfm = *get_cpu_ptr(entry->pool->tfm); + local_lock(&zswap_comp.lock); + tfm = *this_cpu_ptr(entry->pool->tfm); ret = crypto_comp_decompress(tfm, src, entry->length, dst, &dlen); - put_cpu_ptr(entry->pool->tfm); + local_unlock(&zswap_comp.lock); kunmap_atomic(dst); zpool_unmap_handle(entry->pool->zpool, entry->handle); BUG_ON(ret); diff --git a/net/Kconfig b/net/Kconfig index 3831206977a17..81ae878ae553d 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -282,7 +282,7 @@ config CGROUP_NET_CLASSID config NET_RX_BUSY_POLL bool - default y + default y if !PREEMPT_RT config BQL bool diff --git a/net/core/dev.c b/net/core/dev.c index 4906b44af8506..3063e57529c6d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -219,14 +219,14 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex) static inline void rps_lock(struct softnet_data *sd) { #ifdef CONFIG_RPS - spin_lock(&sd->input_pkt_queue.lock); + raw_spin_lock(&sd->input_pkt_queue.raw_lock); #endif } static inline void rps_unlock(struct softnet_data *sd) { #ifdef CONFIG_RPS - spin_unlock(&sd->input_pkt_queue.lock); + raw_spin_unlock(&sd->input_pkt_queue.raw_lock); #endif } @@ -3034,6 +3034,7 @@ static void __netif_reschedule(struct Qdisc *q) sd->output_queue_tailp = &q->next_sched; raise_softirq_irqoff(NET_TX_SOFTIRQ); local_irq_restore(flags); + preempt_check_resched_rt(); } void __netif_schedule(struct Qdisc *q) @@ -3096,6 +3097,7 @@ void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason) __this_cpu_write(softnet_data.completion_queue, skb); raise_softirq_irqoff(NET_TX_SOFTIRQ); local_irq_restore(flags); + preempt_check_resched_rt(); } EXPORT_SYMBOL(__dev_kfree_skb_irq); @@ -3762,7 +3764,11 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, * This permits qdisc->running owner to get the lock more * often and dequeue packets faster. */ +#ifdef CONFIG_PREEMPT_RT + contended = true; +#else contended = qdisc_is_running(q); +#endif if (unlikely(contended)) spin_lock(&q->busylock); @@ -4558,6 +4564,7 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu, rps_unlock(sd); local_irq_restore(flags); + preempt_check_resched_rt(); atomic_long_inc(&skb->dev->rx_dropped); kfree_skb(skb); @@ -4773,7 +4780,7 @@ static int netif_rx_internal(struct sk_buff *skb) struct rps_dev_flow voidflow, *rflow = &voidflow; int cpu; - preempt_disable(); + migrate_disable(); rcu_read_lock(); cpu = get_rps_cpu(skb->dev, skb, &rflow); @@ -4783,14 +4790,14 @@ static int netif_rx_internal(struct sk_buff *skb) ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail); rcu_read_unlock(); - preempt_enable(); + migrate_enable(); } else #endif { unsigned int qtail; - ret = enqueue_to_backlog(skb, get_cpu(), &qtail); - put_cpu(); + ret = enqueue_to_backlog(skb, get_cpu_light(), &qtail); + put_cpu_light(); } return ret; } @@ -4829,11 +4836,9 @@ int netif_rx_ni(struct sk_buff *skb) trace_netif_rx_ni_entry(skb); - preempt_disable(); + local_bh_disable(); err = netif_rx_internal(skb); - if (local_softirq_pending()) - do_softirq(); - preempt_enable(); + local_bh_enable(); trace_netif_rx_ni_exit(err); return err; @@ -6202,12 +6207,14 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd) sd->rps_ipi_list = NULL; local_irq_enable(); + preempt_check_resched_rt(); /* Send pending IPI's to kick RPS processing on remote cpus. */ net_rps_send_ipi(remsd); } else #endif local_irq_enable(); + preempt_check_resched_rt(); } static bool sd_has_rps_ipi_waiting(struct softnet_data *sd) @@ -6285,6 +6292,7 @@ void __napi_schedule(struct napi_struct *n) local_irq_save(flags); ____napi_schedule(this_cpu_ptr(&softnet_data), n); local_irq_restore(flags); + preempt_check_resched_rt(); } EXPORT_SYMBOL(__napi_schedule); @@ -10711,6 +10719,7 @@ static int dev_cpu_dead(unsigned int oldcpu) raise_softirq_irqoff(NET_TX_SOFTIRQ); local_irq_enable(); + preempt_check_resched_rt(); #ifdef CONFIG_RPS remsd = oldsd->rps_ipi_list; @@ -10724,7 +10733,7 @@ static int dev_cpu_dead(unsigned int oldcpu) netif_rx_ni(skb); input_queue_head_incr(oldsd); } - while ((skb = skb_dequeue(&oldsd->input_pkt_queue))) { + while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) { netif_rx_ni(skb); input_queue_head_incr(oldsd); } @@ -11040,7 +11049,7 @@ static int __init net_dev_init(void) INIT_WORK(flush, flush_backlog); - skb_queue_head_init(&sd->input_pkt_queue); + skb_queue_head_init_raw(&sd->input_pkt_queue); skb_queue_head_init(&sd->process_queue); #ifdef CONFIG_XFRM_OFFLOAD skb_queue_head_init(&sd->xfrm_backlog); diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 80dbf2f4016e2..698b02dfeaaf7 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -42,7 +42,7 @@ struct net_rate_estimator { struct gnet_stats_basic_packed *bstats; spinlock_t *stats_lock; - seqcount_t *running; + net_seqlock_t *running; struct gnet_stats_basic_cpu __percpu *cpu_bstats; u8 ewma_log; u8 intvl_log; /* period : (250ms << intvl_log) */ @@ -125,7 +125,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct net_rate_estimator __rcu **rate_est, spinlock_t *lock, - seqcount_t *running, + net_seqlock_t *running, struct nlattr *opt) { struct gnet_estimator *parm = nla_data(opt); @@ -223,7 +223,7 @@ int gen_replace_estimator(struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu_bstats, struct net_rate_estimator __rcu **rate_est, spinlock_t *lock, - seqcount_t *running, struct nlattr *opt) + net_seqlock_t *running, struct nlattr *opt) { return gen_new_estimator(bstats, cpu_bstats, rate_est, lock, running, opt); diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c index e491b083b3485..ef432cea2e105 100644 --- a/net/core/gen_stats.c +++ b/net/core/gen_stats.c @@ -137,7 +137,7 @@ __gnet_stats_copy_basic_cpu(struct gnet_stats_basic_packed *bstats, } void -__gnet_stats_copy_basic(const seqcount_t *running, +__gnet_stats_copy_basic(net_seqlock_t *running, struct gnet_stats_basic_packed *bstats, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b) @@ -150,15 +150,15 @@ __gnet_stats_copy_basic(const seqcount_t *running, } do { if (running) - seq = read_seqcount_begin(running); + seq = net_seq_begin(running); bstats->bytes = b->bytes; bstats->packets = b->packets; - } while (running && read_seqcount_retry(running, seq)); + } while (running && net_seq_retry(running, seq)); } EXPORT_SYMBOL(__gnet_stats_copy_basic); static int -___gnet_stats_copy_basic(const seqcount_t *running, +___gnet_stats_copy_basic(net_seqlock_t *running, struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b, @@ -204,7 +204,7 @@ ___gnet_stats_copy_basic(const seqcount_t *running, * if the room in the socket buffer was not sufficient. */ int -gnet_stats_copy_basic(const seqcount_t *running, +gnet_stats_copy_basic(net_seqlock_t *running, struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b) @@ -228,7 +228,7 @@ EXPORT_SYMBOL(gnet_stats_copy_basic); * if the room in the socket buffer was not sufficient. */ int -gnet_stats_copy_basic_hw(const seqcount_t *running, +gnet_stats_copy_basic_hw(net_seqlock_t *running, struct gnet_dump *d, struct gnet_stats_basic_cpu __percpu *cpu, struct gnet_stats_basic_packed *b) diff --git a/net/core/sock.c b/net/core/sock.c index 6c5c6b18eff4c..dc252f9aaf7e0 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3049,12 +3049,11 @@ void lock_sock_nested(struct sock *sk, int subclass) if (sk->sk_lock.owned) __lock_sock(sk); sk->sk_lock.owned = 1; - spin_unlock(&sk->sk_lock.slock); + spin_unlock_bh(&sk->sk_lock.slock); /* * The sk_lock has mutex_lock() semantics here: */ mutex_acquire(&sk->sk_lock.dep_map, subclass, 0, _RET_IP_); - local_bh_enable(); } EXPORT_SYMBOL(lock_sock_nested); @@ -3103,12 +3102,11 @@ bool lock_sock_fast(struct sock *sk) __lock_sock(sk); sk->sk_lock.owned = 1; - spin_unlock(&sk->sk_lock.slock); + spin_unlock_bh(&sk->sk_lock.slock); /* * The sk_lock has mutex_lock() semantics here: */ mutex_acquire(&sk->sk_lock.dep_map, 0, 0, _RET_IP_); - local_bh_enable(); return true; } EXPORT_SYMBOL(lock_sock_fast); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 239e54474b653..fcb105cbb5465 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -585,7 +585,9 @@ int __inet_hash(struct sock *sk, struct sock *osk) int err = 0; if (sk->sk_state != TCP_LISTEN) { + local_bh_disable(); inet_ehash_nolisten(sk, osk); + local_bh_enable(); return 0; } WARN_ON(!sk_unhashed(sk)); @@ -617,11 +619,8 @@ int inet_hash(struct sock *sk) { int err = 0; - if (sk->sk_state != TCP_CLOSE) { - local_bh_disable(); + if (sk->sk_state != TCP_CLOSE) err = __inet_hash(sk, NULL); - local_bh_enable(); - } return err; } @@ -632,17 +631,20 @@ void inet_unhash(struct sock *sk) struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; struct inet_listen_hashbucket *ilb = NULL; spinlock_t *lock; + bool state_listen; if (sk_unhashed(sk)) return; if (sk->sk_state == TCP_LISTEN) { + state_listen = true; ilb = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)]; - lock = &ilb->lock; + spin_lock(&ilb->lock); } else { + state_listen = false; lock = inet_ehash_lockp(hashinfo, sk->sk_hash); + spin_lock_bh(lock); } - spin_lock_bh(lock); if (sk_unhashed(sk)) goto unlock; @@ -655,7 +657,10 @@ void inet_unhash(struct sock *sk) __sk_nulls_del_node_init_rcu(sk); sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); unlock: - spin_unlock_bh(lock); + if (state_listen) + spin_unlock(&ilb->lock); + else + spin_unlock_bh(lock); } EXPORT_SYMBOL_GPL(inet_unhash); diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 2d3add9e61162..50fd17cbf3ec7 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -335,11 +335,8 @@ int inet6_hash(struct sock *sk) { int err = 0; - if (sk->sk_state != TCP_CLOSE) { - local_bh_disable(); + if (sk->sk_state != TCP_CLOSE) err = __inet_hash(sk, NULL); - local_bh_enable(); - } return err; } diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 2a76a2f5ed88c..1542f1a5a31c8 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1257,7 +1257,7 @@ static struct Qdisc *qdisc_create(struct net_device *dev, rcu_assign_pointer(sch->stab, stab); } if (tca[TCA_RATE]) { - seqcount_t *running; + net_seqlock_t *running; err = -EOPNOTSUPP; if (sch->flags & TCQ_F_MQROOT) { diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 54c417244642a..7ce1abfd68a69 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -553,7 +553,11 @@ struct Qdisc noop_qdisc = { .ops = &noop_qdisc_ops, .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), .dev_queue = &noop_netdev_queue, +#ifdef CONFIG_PREEMPT_RT + .running = __SEQLOCK_UNLOCKED(noop_qdisc.running), +#else .running = SEQCNT_ZERO(noop_qdisc.running), +#endif .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock), .gso_skb = { .next = (struct sk_buff *)&noop_qdisc.gso_skb, @@ -858,9 +862,15 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, lockdep_set_class(&sch->busylock, dev->qdisc_tx_busylock ?: &qdisc_tx_busylock); +#ifdef CONFIG_PREEMPT_RT + seqlock_init(&sch->running); + lockdep_set_class(&sch->running.lock, + dev->qdisc_running_key ?: &qdisc_running_key); +#else seqcount_init(&sch->running); lockdep_set_class(&sch->running, dev->qdisc_running_key ?: &qdisc_running_key); +#endif sch->ops = ops; sch->flags = ops->static_flags; diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index 43cf8dbde898b..d5516102491b7 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -422,7 +422,7 @@ void svc_xprt_do_enqueue(struct svc_xprt *xprt) if (test_and_set_bit(XPT_BUSY, &xprt->xpt_flags)) return; - cpu = get_cpu(); + cpu = get_cpu_light(); pool = svc_pool_for_cpu(xprt->xpt_server, cpu); atomic_long_inc(&pool->sp_stats.packets); @@ -446,7 +446,7 @@ void svc_xprt_do_enqueue(struct svc_xprt *xprt) rqstp = NULL; out_unlock: rcu_read_unlock(); - put_cpu(); + put_cpu_light(); trace_svc_xprt_do_enqueue(xprt, rqstp); } EXPORT_SYMBOL_GPL(svc_xprt_do_enqueue); diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index efc89a92961df..4e1216d04441d 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -44,7 +44,7 @@ static void xfrm_state_gc_task(struct work_struct *work); */ static unsigned int xfrm_state_hashmax __read_mostly = 1 * 1024 * 1024; -static __read_mostly seqcount_t xfrm_state_hash_generation = SEQCNT_ZERO(xfrm_state_hash_generation); +static __read_mostly seqcount_spinlock_t xfrm_state_hash_generation; static struct kmem_cache *xfrm_state_cache __ro_after_init; static DECLARE_WORK(xfrm_state_gc_work, xfrm_state_gc_task); @@ -139,6 +139,11 @@ static void xfrm_hash_resize(struct work_struct *work) return; } + /* XXX - the locking which protects the sequence counter appears + * to be broken here. The sequence counter is global, but the + * spinlock used for the sequence counter write serialization is + * per network namespace... + */ spin_lock_bh(&net->xfrm.xfrm_state_lock); write_seqcount_begin(&xfrm_state_hash_generation); @@ -2589,6 +2594,8 @@ int __net_init xfrm_state_init(struct net *net) net->xfrm.state_num = 0; INIT_WORK(&net->xfrm.state_hash_work, xfrm_hash_resize); spin_lock_init(&net->xfrm.xfrm_state_lock); + seqcount_spinlock_init(&xfrm_state_hash_generation, + &net->xfrm.xfrm_state_lock); return 0; out_byspi: diff --git a/scripts/gdb/linux/dmesg.py b/scripts/gdb/linux/dmesg.py index 2fa7bb83885f0..a92c55bd8de54 100644 --- a/scripts/gdb/linux/dmesg.py +++ b/scripts/gdb/linux/dmesg.py @@ -16,8 +16,13 @@ import sys from linux import utils -printk_log_type = utils.CachedType("struct printk_log") - +printk_info_type = utils.CachedType("struct printk_info") +prb_data_blk_lpos_type = utils.CachedType("struct prb_data_blk_lpos") +prb_desc_type = utils.CachedType("struct prb_desc") +prb_desc_ring_type = utils.CachedType("struct prb_desc_ring") +prb_data_ring_type = utils.CachedType("struct prb_data_ring") +printk_ringbuffer_type = utils.CachedType("struct printk_ringbuffer") +atomic_long_type = utils.CachedType("atomic_long_t") class LxDmesg(gdb.Command): """Print Linux kernel log buffer.""" @@ -26,44 +31,110 @@ printk_log_type = utils.CachedType("struct printk_log") super(LxDmesg, self).__init__("lx-dmesg", gdb.COMMAND_DATA) def invoke(self, arg, from_tty): - log_buf_addr = int(str(gdb.parse_and_eval( - "(void *)'printk.c'::log_buf")).split()[0], 16) - log_first_idx = int(gdb.parse_and_eval("'printk.c'::log_first_idx")) - log_next_idx = int(gdb.parse_and_eval("'printk.c'::log_next_idx")) - log_buf_len = int(gdb.parse_and_eval("'printk.c'::log_buf_len")) - inf = gdb.inferiors()[0] - start = log_buf_addr + log_first_idx - if log_first_idx < log_next_idx: - log_buf_2nd_half = -1 - length = log_next_idx - log_first_idx - log_buf = utils.read_memoryview(inf, start, length).tobytes() - else: - log_buf_2nd_half = log_buf_len - log_first_idx - a = utils.read_memoryview(inf, start, log_buf_2nd_half) - b = utils.read_memoryview(inf, log_buf_addr, log_next_idx) - log_buf = a.tobytes() + b.tobytes() - length_offset = printk_log_type.get_type()['len'].bitpos // 8 - text_len_offset = printk_log_type.get_type()['text_len'].bitpos // 8 - time_stamp_offset = printk_log_type.get_type()['ts_nsec'].bitpos // 8 - text_offset = printk_log_type.get_type().sizeof + # read in prb structure + prb_addr = int(str(gdb.parse_and_eval("(void *)'printk.c'::prb")).split()[0], 16) + sz = printk_ringbuffer_type.get_type().sizeof + prb = utils.read_memoryview(inf, prb_addr, sz).tobytes() - pos = 0 - while pos < log_buf.__len__(): - length = utils.read_u16(log_buf, pos + length_offset) - if length == 0: - if log_buf_2nd_half == -1: - gdb.write("Corrupted log buffer!\n") + # read in descriptor ring structure + off = printk_ringbuffer_type.get_type()['desc_ring'].bitpos // 8 + addr = prb_addr + off + sz = prb_desc_ring_type.get_type().sizeof + desc_ring = utils.read_memoryview(inf, addr, sz).tobytes() + + # read in descriptor array + off = prb_desc_ring_type.get_type()['count_bits'].bitpos // 8 + desc_ring_count = 1 << utils.read_u32(desc_ring, off) + desc_sz = prb_desc_type.get_type().sizeof + off = prb_desc_ring_type.get_type()['descs'].bitpos // 8 + addr = utils.read_ulong(desc_ring, off) + descs = utils.read_memoryview(inf, addr, desc_sz * desc_ring_count).tobytes() + + # read in info array + info_sz = printk_info_type.get_type().sizeof + off = prb_desc_ring_type.get_type()['infos'].bitpos // 8 + addr = utils.read_ulong(desc_ring, off) + infos = utils.read_memoryview(inf, addr, info_sz * desc_ring_count).tobytes() + + # read in text data ring structure + off = printk_ringbuffer_type.get_type()['text_data_ring'].bitpos // 8 + addr = prb_addr + off + sz = prb_data_ring_type.get_type().sizeof + text_data_ring = utils.read_memoryview(inf, addr, sz).tobytes() + + # read in text data + off = prb_data_ring_type.get_type()['size_bits'].bitpos // 8 + text_data_sz = 1 << utils.read_u32(text_data_ring, off) + off = prb_data_ring_type.get_type()['data'].bitpos // 8 + addr = utils.read_ulong(text_data_ring, off) + text_data = utils.read_memoryview(inf, addr, text_data_sz).tobytes() + + counter_off = atomic_long_type.get_type()['counter'].bitpos // 8 + + sv_off = prb_desc_type.get_type()['state_var'].bitpos // 8 + + off = prb_desc_type.get_type()['text_blk_lpos'].bitpos // 8 + begin_off = off + (prb_data_blk_lpos_type.get_type()['begin'].bitpos // 8) + next_off = off + (prb_data_blk_lpos_type.get_type()['next'].bitpos // 8) + + ts_off = printk_info_type.get_type()['ts_nsec'].bitpos // 8 + len_off = printk_info_type.get_type()['text_len'].bitpos // 8 + + # definitions from kernel/printk/printk_ringbuffer.h + desc_committed = 1 + desc_finalized = 2 + desc_sv_bits = utils.get_long_type().sizeof * 8 + desc_flags_shift = desc_sv_bits - 2 + desc_flags_mask = 3 << desc_flags_shift + desc_id_mask = ~desc_flags_mask + + # read in tail and head descriptor ids + off = prb_desc_ring_type.get_type()['tail_id'].bitpos // 8 + tail_id = utils.read_u64(desc_ring, off + counter_off) + off = prb_desc_ring_type.get_type()['head_id'].bitpos // 8 + head_id = utils.read_u64(desc_ring, off + counter_off) + + did = tail_id + while True: + ind = did % desc_ring_count + desc_off = desc_sz * ind + info_off = info_sz * ind + + # skip non-committed record + state = 3 & (utils.read_u64(descs, desc_off + sv_off + + counter_off) >> desc_flags_shift) + if state != desc_committed and state != desc_finalized: + if did == head_id: break - pos = log_buf_2nd_half + did = (did + 1) & desc_id_mask continue - text_len = utils.read_u16(log_buf, pos + text_len_offset) - text_start = pos + text_offset - text = log_buf[text_start:text_start + text_len].decode( - encoding='utf8', errors='replace') - time_stamp = utils.read_u64(log_buf, pos + time_stamp_offset) + begin = utils.read_ulong(descs, desc_off + begin_off) % text_data_sz + end = utils.read_ulong(descs, desc_off + next_off) % text_data_sz + + # handle data-less record + if begin & 1 == 1: + text = "" + else: + # handle wrapping data block + if begin > end: + begin = 0 + + # skip over descriptor id + text_start = begin + utils.get_long_type().sizeof + + text_len = utils.read_u16(infos, info_off + len_off) + + # handle truncated message + if end - text_start < text_len: + text_len = end - text_start + + text = text_data[text_start:text_start + text_len].decode( + encoding='utf8', errors='replace') + + time_stamp = utils.read_u64(infos, info_off + ts_off) for line in text.splitlines(): msg = u"[{time:12.6f}] {line}\n".format( @@ -75,7 +146,9 @@ printk_log_type = utils.CachedType("struct printk_log") msg = msg.encode(encoding='utf8', errors='replace') gdb.write(msg) - pos += length + if did == head_id: + break + did = (did + 1) & desc_id_mask LxDmesg() diff --git a/scripts/gdb/linux/utils.py b/scripts/gdb/linux/utils.py index ea94221dbd392..ff7c1799d588f 100644 --- a/scripts/gdb/linux/utils.py +++ b/scripts/gdb/linux/utils.py @@ -123,6 +123,13 @@ target_endianness = None return read_u32(buffer, offset + 4) + (read_u32(buffer, offset) << 32) +def read_ulong(buffer, offset): + if get_long_type().sizeof == 8: + return read_u64(buffer, offset) + else: + return read_u32(buffer, offset) + + target_arch = None