linux-stable/include
Peter Zijlstra c427f69564 locking/mutex: Optimize __mutex_trylock_fast()
Use try_cmpxchg to avoid the pointless TEST instruction..
And add the (missing) atomic_long_try_cmpxchg*() wrappery.

On x86_64 this gives:

0000000000000710 <mutex_lock>:						0000000000000710 <mutex_lock>:
 710:   65 48 8b 14 25 00 00    mov    %gs:0x0,%rdx                      710:   65 48 8b 14 25 00 00    mov    %gs:0x0,%rdx
 717:   00 00                                                            717:   00 00
                        715: R_X86_64_32S       current_task                                    715: R_X86_64_32S       current_task
 719:   31 c0                   xor    %eax,%eax                         719:   31 c0                   xor    %eax,%eax
 71b:   f0 48 0f b1 17          lock cmpxchg %rdx,(%rdi)                 71b:   f0 48 0f b1 17          lock cmpxchg %rdx,(%rdi)
 720:   48 85 c0                test   %rax,%rax                         720:   75 02                   jne    724 <mutex_lock+0x14>
 723:   75 02                   jne    727 <mutex_lock+0x17>             722:   f3 c3                   repz retq
 725:   f3 c3                   repz retq                                724:   eb da                   jmp    700 <__mutex_lock_slowpath>
 727:   eb d7                   jmp    700 <__mutex_lock_slowpath>       726:   66 2e 0f 1f 84 00 00    nopw   %cs:0x0(%rax,%rax,1)
 729:   0f 1f 80 00 00 00 00    nopl   0x0(%rax)                         72d:   00 00 00

On ARM64 this gives:

000000000000638 <mutex_lock>:						0000000000000638 <mutex_lock>:
     638:       d5384101        mrs     x1, sp_el0                           638:       d5384101        mrs     x1, sp_el0
     63c:       d2800002        mov     x2, #0x0                             63c:       d2800002        mov     x2, #0x0
     640:       f9800011        prfm    pstl1strm, [x0]                      640:       f9800011        prfm    pstl1strm, [x0]
     644:       c85ffc03        ldaxr   x3, [x0]                             644:       c85ffc03        ldaxr   x3, [x0]
     648:       ca020064        eor     x4, x3, x2                           648:       ca020064        eor     x4, x3, x2
     64c:       b5000064        cbnz    x4, 658 <mutex_lock+0x20>            64c:       b5000064        cbnz    x4, 658 <mutex_lock+0x20>
     650:       c8047c01        stxr    w4, x1, [x0]                         650:       c8047c01        stxr    w4, x1, [x0]
     654:       35ffff84        cbnz    w4, 644 <mutex_lock+0xc>             654:       35ffff84        cbnz    w4, 644 <mutex_lock+0xc>
     658:       b40000c3        cbz     x3, 670 <mutex_lock+0x38>            658:       b5000043        cbnz    x3, 660 <mutex_lock+0x28>
     65c:       a9bf7bfd        stp     x29, x30, [sp,#-16]!                 65c:       d65f03c0        ret
     660:       910003fd        mov     x29, sp                              660:       a9bf7bfd        stp     x29, x30, [sp,#-16]!
     664:       97ffffef        bl      620 <__mutex_lock_slowpath>          664:       910003fd        mov     x29, sp
     668:       a8c17bfd        ldp     x29, x30, [sp],#16                   668:       97ffffee        bl      620 <__mutex_lock_slowpath>
     66c:       d65f03c0        ret                                          66c:       a8c17bfd        ldp     x29, x30, [sp],#16
     670:       d65f03c0        ret                                          670:       d65f03c0        ret

Reported-by: Matthew Wilcox <mawilcox@microsoft.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-05-04 10:02:39 +02:00
..
acpi xen: fixes for 4.17-rc1 2018-04-12 11:04:35 -07:00
asm-generic locking/mutex: Optimize __mutex_trylock_fast() 2018-05-04 10:02:39 +02:00
clocksource ARM: SoC platform updates for 4.17 2018-04-05 21:21:08 -07:00
crypto
drm drm: Fix HDCP downstream dev count read 2018-04-16 12:10:48 -04:00
dt-bindings The large diff this time around is from the addition of a new clk driver 2018-04-13 15:51:06 -07:00
keys
kvm
linux delayacct: Use raw_spinlocks 2018-04-27 14:34:51 +02:00
math-emu
media media updates for v4.17-rc1 2018-04-10 10:10:30 -07:00
memory
misc
net net: sched: ife: handle malformed tlv length 2018-04-22 21:12:00 -04:00
pcmcia
ras
rdma Merge candidates for 4.17 merge window 2018-04-06 17:35:43 -07:00
scsi scsi: core: remove reference to scsi_show_extd_sense() 2018-04-18 23:37:39 -04:00
soc ARM: SoC driver updates for 4.17 2018-04-05 21:29:35 -07:00
sound sound updates for 4.17-rc1 2018-04-05 10:42:07 -07:00
target
trace Following tracing fixes: 2018-04-26 16:22:47 -07:00
uapi virtio_balloon: add array of stat names 2018-04-24 21:44:01 +03:00
video
xen xen/sndif: Sync up with the canonical definition in Xen 2018-04-17 08:26:33 -04:00