Files
linux/kernel/Kconfig.locks
Dmitry Ilvokhin f7e2cb6d49 locking/qspinlock: Add contended_release tracepoint
Unlike mutex and rw_semaphore, qspinlock has no owner field, so "perf
lock contention --lock-owner" cannot attribute a contended spinlock to
its holder. The waiter-side contention_begin event records that a
spinlock is contended, but not by whom. Firing contended_release in the
holder's context at unlock is the only way to capture the holder of a
contended spinlock.

Combine the contention check, trace call and release in an out-of-line
queued_spin_release_traced() so the compiler need not preserve the lock
pointer in a callee-saved register across the call.

The check in queued_spin_unlock() is paid on every unlock, even while
the tracepoint is disabled: a static-branch NOP on x86_64, and a few
more instructions to manage a stack frame elsewhere. Gate it behind
CONFIG_QUEUED_SPINLOCKS_TRACE_CONTENDED_RELEASE (default n) so nobody
pays for a tracepoint they do not use. Sleeping locks fire
contended_release regardless.

On x86 this generic path is used only with PARAVIRT_SPINLOCKS=n (e.g.
defconfig). PARAVIRT_SPINLOCKS=y kernels keep the paravirt static_call
unlock and are wired up separately.

All below are with the QUEUED_SPINLOCKS_TRACE_CONTENDED_RELEASE option
enabled.

_raw_spin_unlock(), x86_64 defconfig, GCC 11, tracepoint compiled in but
disabled. The unlock is the single 'movb'. The only instruction added to
the executed path is the 2-byte static-branch NOP. The CALL to the
traced helper and the JMP back are emitted out of line and are reached
only once the static branch is patched on:

          endbr64                            ; 4 bytes
          xchg   %ax,%ax                     ; 2 static-branch NOP
                                             ;   (added)
          movb   $0x0,(%rdi)                 ; 3 unlock (single store)
       A: decl   %gs:__preempt_count         ; 7
          je     B                           ; 2
          jmp    __x86_return_thunk          ; 5
          call   queued_spin_release_traced  ; 5 out of line, reached
                                             ;   only when the
                                             ;   tracepoint is on
          jmp    A                           ; 2 (added)
       B: call   __SCT__preempt_schedule     ; 5
          jmp    __x86_return_thunk          ; 5

Baseline is the same stream without the NOP and the out-of-line
CALL/JMP: 31 bytes vs 40 (+9 bytes).

Binary size impact on x86_64, defconfig: +680 bytes (+0.00%), since all
standard configs out-of-line unlock. Architectures with inlined unlock
(s390 (always), csky and loongarch (both when !PREEMPTION)) will see a
bigger increase in binary size.

On the same path (x86_64, PARAVIRT_SPINLOCKS=n) with the tracepoint
disabled, a _raw_spin_unlock()-heavy nginx workload [1] shows no
measurable difference between baseline and patched kernels in
throughput, latency, cycles, instructions, IPC, or L1 instruction-cache
misses (kernel and total): all deltas stay within run-to-run noise.

Unlike x86, on arm64 the frame setup code (STP, MOV and LDP) lands on
the executed path in addition to static-branch NOP. Binary size impact
on arm64, defconfig: +932 bytes (+0.00%).

The _raw_spin_unlock()-heavy nginx workload reflects the larger hot
path: L1 instruction-cache misses rise ~1.4% (kernel and total) and
instruction count ~0.4%, consistent with the per-unlock frame.
cpu_cycles, throughput and latency show no measurable change and are
within run-to-run noise.

Architectures with fully custom qspinlock implementations (e.g.
PowerPC) are not covered by this change.

[1]: https://lore.kernel.org/all/aiphFXe_TPNPxZ_n@shell.ilvokhin.com/

Signed-off-by: Dmitry Ilvokhin <d@ilvokhin.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juergen Gross <jgross@suse.com>
Link: https://patch.msgid.link/0d998e22a0c595f670cfc6725bb683323aced5cb.1785778551.git.d@ilvokhin.com
2026-08-07 17:58:10 +02:00

282 lines
5.7 KiB
Plaintext

# SPDX-License-Identifier: GPL-2.0-only
#
# The ARCH_INLINE foo is necessary because select ignores "depends on"
#
config ARCH_INLINE_SPIN_TRYLOCK
bool
config ARCH_INLINE_SPIN_TRYLOCK_BH
bool
config ARCH_INLINE_SPIN_LOCK
bool
config ARCH_INLINE_SPIN_LOCK_BH
bool
config ARCH_INLINE_SPIN_LOCK_IRQ
bool
config ARCH_INLINE_SPIN_LOCK_IRQSAVE
bool
config ARCH_INLINE_SPIN_UNLOCK
bool
config ARCH_INLINE_SPIN_UNLOCK_BH
bool
config ARCH_INLINE_SPIN_UNLOCK_IRQ
bool
config ARCH_INLINE_SPIN_UNLOCK_IRQRESTORE
bool
config ARCH_INLINE_READ_TRYLOCK
bool
config ARCH_INLINE_READ_LOCK
bool
config ARCH_INLINE_READ_LOCK_BH
bool
config ARCH_INLINE_READ_LOCK_IRQ
bool
config ARCH_INLINE_READ_LOCK_IRQSAVE
bool
config ARCH_INLINE_READ_UNLOCK
bool
config ARCH_INLINE_READ_UNLOCK_BH
bool
config ARCH_INLINE_READ_UNLOCK_IRQ
bool
config ARCH_INLINE_READ_UNLOCK_IRQRESTORE
bool
config ARCH_INLINE_WRITE_TRYLOCK
bool
config ARCH_INLINE_WRITE_LOCK
bool
config ARCH_INLINE_WRITE_LOCK_BH
bool
config ARCH_INLINE_WRITE_LOCK_IRQ
bool
config ARCH_INLINE_WRITE_LOCK_IRQSAVE
bool
config ARCH_INLINE_WRITE_UNLOCK
bool
config ARCH_INLINE_WRITE_UNLOCK_BH
bool
config ARCH_INLINE_WRITE_UNLOCK_IRQ
bool
config ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE
bool
config UNINLINE_SPIN_UNLOCK
bool
#
# lock_* functions are inlined when:
# - DEBUG_SPINLOCK=n and GENERIC_LOCKBREAK=n and ARCH_INLINE_*LOCK=y
#
# trylock_* functions are inlined when:
# - DEBUG_SPINLOCK=n and ARCH_INLINE_*LOCK=y
#
# unlock and unlock_irq functions are inlined when:
# - DEBUG_SPINLOCK=n and ARCH_INLINE_*LOCK=y
# or
# - DEBUG_SPINLOCK=n and PREEMPTION=n
#
# unlock_bh and unlock_irqrestore functions are inlined when:
# - DEBUG_SPINLOCK=n and ARCH_INLINE_*LOCK=y
#
if !DEBUG_SPINLOCK
config INLINE_SPIN_TRYLOCK
def_bool y
depends on ARCH_INLINE_SPIN_TRYLOCK
config INLINE_SPIN_TRYLOCK_BH
def_bool y
depends on ARCH_INLINE_SPIN_TRYLOCK_BH
config INLINE_SPIN_LOCK
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_SPIN_LOCK
config INLINE_SPIN_LOCK_BH
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_SPIN_LOCK_BH
config INLINE_SPIN_LOCK_IRQ
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_SPIN_LOCK_IRQ
config INLINE_SPIN_LOCK_IRQSAVE
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_SPIN_LOCK_IRQSAVE
config INLINE_SPIN_UNLOCK_BH
def_bool y
depends on ARCH_INLINE_SPIN_UNLOCK_BH
config INLINE_SPIN_UNLOCK_IRQ
def_bool y
depends on !PREEMPTION || ARCH_INLINE_SPIN_UNLOCK_IRQ
config INLINE_SPIN_UNLOCK_IRQRESTORE
def_bool y
depends on ARCH_INLINE_SPIN_UNLOCK_IRQRESTORE
config INLINE_READ_TRYLOCK
def_bool y
depends on ARCH_INLINE_READ_TRYLOCK
config INLINE_READ_LOCK
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_READ_LOCK
config INLINE_READ_LOCK_BH
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_READ_LOCK_BH
config INLINE_READ_LOCK_IRQ
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_READ_LOCK_IRQ
config INLINE_READ_LOCK_IRQSAVE
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_READ_LOCK_IRQSAVE
config INLINE_READ_UNLOCK
def_bool y
depends on !PREEMPTION || ARCH_INLINE_READ_UNLOCK
config INLINE_READ_UNLOCK_BH
def_bool y
depends on ARCH_INLINE_READ_UNLOCK_BH
config INLINE_READ_UNLOCK_IRQ
def_bool y
depends on !PREEMPTION || ARCH_INLINE_READ_UNLOCK_IRQ
config INLINE_READ_UNLOCK_IRQRESTORE
def_bool y
depends on ARCH_INLINE_READ_UNLOCK_IRQRESTORE
config INLINE_WRITE_TRYLOCK
def_bool y
depends on ARCH_INLINE_WRITE_TRYLOCK
config INLINE_WRITE_LOCK
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_WRITE_LOCK
config INLINE_WRITE_LOCK_BH
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_WRITE_LOCK_BH
config INLINE_WRITE_LOCK_IRQ
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_WRITE_LOCK_IRQ
config INLINE_WRITE_LOCK_IRQSAVE
def_bool y
depends on !GENERIC_LOCKBREAK && ARCH_INLINE_WRITE_LOCK_IRQSAVE
config INLINE_WRITE_UNLOCK
def_bool y
depends on !PREEMPTION || ARCH_INLINE_WRITE_UNLOCK
config INLINE_WRITE_UNLOCK_BH
def_bool y
depends on ARCH_INLINE_WRITE_UNLOCK_BH
config INLINE_WRITE_UNLOCK_IRQ
def_bool y
depends on !PREEMPTION || ARCH_INLINE_WRITE_UNLOCK_IRQ
config INLINE_WRITE_UNLOCK_IRQRESTORE
def_bool y
depends on ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE
endif
config ARCH_SUPPORTS_ATOMIC_RMW
bool
config MUTEX_SPIN_ON_OWNER
def_bool y
depends on SMP && ARCH_SUPPORTS_ATOMIC_RMW
config RWSEM_SPIN_ON_OWNER
def_bool y
depends on SMP && ARCH_SUPPORTS_ATOMIC_RMW
config LOCK_SPIN_ON_OWNER
def_bool y
depends on MUTEX_SPIN_ON_OWNER || RWSEM_SPIN_ON_OWNER
config ARCH_USE_QUEUED_SPINLOCKS
bool
config QUEUED_SPINLOCKS
def_bool y if ARCH_USE_QUEUED_SPINLOCKS
depends on SMP
config QUEUED_SPINLOCKS_TRACE_CONTENDED_RELEASE
bool "Trace contended_release on queued spinlocks"
depends on QUEUED_SPINLOCKS && TRACEPOINTS
help
Fire the lock:contended_release tracepoint when a contended queued
spinlock is released, so it is possible to attribute a contended
spinlock to its holder.
Architectures that can patch the unlock site do this at no cost and
do not need this option.
Everywhere else the check is compiled into queued_spin_unlock() and
a small cost is paid on every unlock even when the tracepoint is
disabled: a static-branch NOP and possibly a few more instructions
to manage a stack frame.
Sleeping locks fire lock:contended_release regardless of this option.
If unsure, say N.
config BPF_ARCH_SPINLOCK
bool
config ARCH_USE_QUEUED_RWLOCKS
bool
config QUEUED_RWLOCKS
def_bool y if ARCH_USE_QUEUED_RWLOCKS
depends on SMP && !PREEMPT_RT
config ARCH_HAS_MMIOWB
bool
config MMIOWB
def_bool y if ARCH_HAS_MMIOWB
depends on SMP