| Message ID | 20190322143008.21313-4-longman@redhat.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
| Series | locking/rwsem: Rwsem rearchitecture part 0 | expand |
On Fri, Mar 22, 2019 at 7:30 AM Waiman Long <longman@redhat.com> wrote: > > Modify __down_read_trylock() to optimize for an unlocked rwsem and make > it generate slightly better code. Oh, that should teach me to read all patches in the series before starting to comment on them. So ignore my comment on #1. Linus
On Fri, Mar 22, 2019 at 10:30:08AM -0400, Waiman Long wrote: > Modify __down_read_trylock() to optimize for an unlocked rwsem and make > it generate slightly better code. > > Before this patch, down_read_trylock: > > 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> > 0x0000000000000005 <+5>: jmp 0x18 <down_read_trylock+24> > 0x0000000000000007 <+7>: lea 0x1(%rdx),%rcx > 0x000000000000000b <+11>: mov %rdx,%rax > 0x000000000000000e <+14>: lock cmpxchg %rcx,(%rdi) > 0x0000000000000013 <+19>: cmp %rax,%rdx > 0x0000000000000016 <+22>: je 0x23 <down_read_trylock+35> > 0x0000000000000018 <+24>: mov (%rdi),%rdx > 0x000000000000001b <+27>: test %rdx,%rdx > 0x000000000000001e <+30>: jns 0x7 <down_read_trylock+7> > 0x0000000000000020 <+32>: xor %eax,%eax > 0x0000000000000022 <+34>: retq > 0x0000000000000023 <+35>: mov %gs:0x0,%rax > 0x000000000000002c <+44>: or $0x3,%rax > 0x0000000000000030 <+48>: mov %rax,0x20(%rdi) > 0x0000000000000034 <+52>: mov $0x1,%eax > 0x0000000000000039 <+57>: retq > > After patch, down_read_trylock: > > 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> > 0x0000000000000005 <+5>: xor %eax,%eax > 0x0000000000000007 <+7>: lea 0x1(%rax),%rdx > 0x000000000000000b <+11>: lock cmpxchg %rdx,(%rdi) > 0x0000000000000010 <+16>: jne 0x29 <down_read_trylock+41> > 0x0000000000000012 <+18>: mov %gs:0x0,%rax > 0x000000000000001b <+27>: or $0x3,%rax > 0x000000000000001f <+31>: mov %rax,0x20(%rdi) > 0x0000000000000023 <+35>: mov $0x1,%eax > 0x0000000000000028 <+40>: retq > 0x0000000000000029 <+41>: test %rax,%rax > 0x000000000000002c <+44>: jns 0x7 <down_read_trylock+7> > 0x000000000000002e <+46>: xor %eax,%eax > 0x0000000000000030 <+48>: retq > > By using a rwsem microbenchmark, the down_read_trylock() rate (with a > load of 10 to lengthen the lock critical section) on a x86-64 system > before and after the patch were: > > Before Patch After Patch > # of Threads rlock rlock > ------------ ----- ----- > 1 14,496 14,716 > 2 8,644 8,453 > 4 6,799 6,983 > 8 5,664 7,190 > > On a ARM64 system, the performance results were: > > Before Patch After Patch > # of Threads rlock rlock > ------------ ----- ----- > 1 23,676 24,488 > 2 7,697 9,502 > 4 4,945 3,440 > 8 2,641 1,603 > > For the uncontended case (1 thread), the new down_read_trylock() is a > little bit faster. For the contended cases, the new down_read_trylock() > perform pretty well in x86-64, but performance degrades at high > contention level on ARM64. So, 70% for 4 threads, 61% for 4 threads - does this trend continue tailing off as the number of threads (and cores) increase?
On 03/22/2019 01:25 PM, Russell King - ARM Linux admin wrote: > On Fri, Mar 22, 2019 at 10:30:08AM -0400, Waiman Long wrote: >> Modify __down_read_trylock() to optimize for an unlocked rwsem and make >> it generate slightly better code. >> >> Before this patch, down_read_trylock: >> >> 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> >> 0x0000000000000005 <+5>: jmp 0x18 <down_read_trylock+24> >> 0x0000000000000007 <+7>: lea 0x1(%rdx),%rcx >> 0x000000000000000b <+11>: mov %rdx,%rax >> 0x000000000000000e <+14>: lock cmpxchg %rcx,(%rdi) >> 0x0000000000000013 <+19>: cmp %rax,%rdx >> 0x0000000000000016 <+22>: je 0x23 <down_read_trylock+35> >> 0x0000000000000018 <+24>: mov (%rdi),%rdx >> 0x000000000000001b <+27>: test %rdx,%rdx >> 0x000000000000001e <+30>: jns 0x7 <down_read_trylock+7> >> 0x0000000000000020 <+32>: xor %eax,%eax >> 0x0000000000000022 <+34>: retq >> 0x0000000000000023 <+35>: mov %gs:0x0,%rax >> 0x000000000000002c <+44>: or $0x3,%rax >> 0x0000000000000030 <+48>: mov %rax,0x20(%rdi) >> 0x0000000000000034 <+52>: mov $0x1,%eax >> 0x0000000000000039 <+57>: retq >> >> After patch, down_read_trylock: >> >> 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> >> 0x0000000000000005 <+5>: xor %eax,%eax >> 0x0000000000000007 <+7>: lea 0x1(%rax),%rdx >> 0x000000000000000b <+11>: lock cmpxchg %rdx,(%rdi) >> 0x0000000000000010 <+16>: jne 0x29 <down_read_trylock+41> >> 0x0000000000000012 <+18>: mov %gs:0x0,%rax >> 0x000000000000001b <+27>: or $0x3,%rax >> 0x000000000000001f <+31>: mov %rax,0x20(%rdi) >> 0x0000000000000023 <+35>: mov $0x1,%eax >> 0x0000000000000028 <+40>: retq >> 0x0000000000000029 <+41>: test %rax,%rax >> 0x000000000000002c <+44>: jns 0x7 <down_read_trylock+7> >> 0x000000000000002e <+46>: xor %eax,%eax >> 0x0000000000000030 <+48>: retq >> >> By using a rwsem microbenchmark, the down_read_trylock() rate (with a >> load of 10 to lengthen the lock critical section) on a x86-64 system >> before and after the patch were: >> >> Before Patch After Patch >> # of Threads rlock rlock >> ------------ ----- ----- >> 1 14,496 14,716 >> 2 8,644 8,453 >> 4 6,799 6,983 >> 8 5,664 7,190 >> >> On a ARM64 system, the performance results were: >> >> Before Patch After Patch >> # of Threads rlock rlock >> ------------ ----- ----- >> 1 23,676 24,488 >> 2 7,697 9,502 >> 4 4,945 3,440 >> 8 2,641 1,603 >> >> For the uncontended case (1 thread), the new down_read_trylock() is a >> little bit faster. For the contended cases, the new down_read_trylock() >> perform pretty well in x86-64, but performance degrades at high >> contention level on ARM64. > So, 70% for 4 threads, 61% for 4 threads - does this trend > continue tailing off as the number of threads (and cores) > increase? > I didn't try higher number of contending threads. I won't worry too much about contention as trylock is a one-off event. The chance of having more than one trylock happening simultaneously is very small. Cheers, Longman
Hi, Could you share the microbenchmark you are using ? I'd like to test the series on powerpc. Thanks Christophe Le 22/03/2019 à 15:30, Waiman Long a écrit : > Modify __down_read_trylock() to optimize for an unlocked rwsem and make > it generate slightly better code. > > Before this patch, down_read_trylock: > > 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> > 0x0000000000000005 <+5>: jmp 0x18 <down_read_trylock+24> > 0x0000000000000007 <+7>: lea 0x1(%rdx),%rcx > 0x000000000000000b <+11>: mov %rdx,%rax > 0x000000000000000e <+14>: lock cmpxchg %rcx,(%rdi) > 0x0000000000000013 <+19>: cmp %rax,%rdx > 0x0000000000000016 <+22>: je 0x23 <down_read_trylock+35> > 0x0000000000000018 <+24>: mov (%rdi),%rdx > 0x000000000000001b <+27>: test %rdx,%rdx > 0x000000000000001e <+30>: jns 0x7 <down_read_trylock+7> > 0x0000000000000020 <+32>: xor %eax,%eax > 0x0000000000000022 <+34>: retq > 0x0000000000000023 <+35>: mov %gs:0x0,%rax > 0x000000000000002c <+44>: or $0x3,%rax > 0x0000000000000030 <+48>: mov %rax,0x20(%rdi) > 0x0000000000000034 <+52>: mov $0x1,%eax > 0x0000000000000039 <+57>: retq > > After patch, down_read_trylock: > > 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> > 0x0000000000000005 <+5>: xor %eax,%eax > 0x0000000000000007 <+7>: lea 0x1(%rax),%rdx > 0x000000000000000b <+11>: lock cmpxchg %rdx,(%rdi) > 0x0000000000000010 <+16>: jne 0x29 <down_read_trylock+41> > 0x0000000000000012 <+18>: mov %gs:0x0,%rax > 0x000000000000001b <+27>: or $0x3,%rax > 0x000000000000001f <+31>: mov %rax,0x20(%rdi) > 0x0000000000000023 <+35>: mov $0x1,%eax > 0x0000000000000028 <+40>: retq > 0x0000000000000029 <+41>: test %rax,%rax > 0x000000000000002c <+44>: jns 0x7 <down_read_trylock+7> > 0x000000000000002e <+46>: xor %eax,%eax > 0x0000000000000030 <+48>: retq > > By using a rwsem microbenchmark, the down_read_trylock() rate (with a > load of 10 to lengthen the lock critical section) on a x86-64 system > before and after the patch were: > > Before Patch After Patch > # of Threads rlock rlock > ------------ ----- ----- > 1 14,496 14,716 > 2 8,644 8,453 > 4 6,799 6,983 > 8 5,664 7,190 > > On a ARM64 system, the performance results were: > > Before Patch After Patch > # of Threads rlock rlock > ------------ ----- ----- > 1 23,676 24,488 > 2 7,697 9,502 > 4 4,945 3,440 > 8 2,641 1,603 > > For the uncontended case (1 thread), the new down_read_trylock() is a > little bit faster. For the contended cases, the new down_read_trylock() > perform pretty well in x86-64, but performance degrades at high > contention level on ARM64. > > Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> > Signed-off-by: Waiman Long <longman@redhat.com> > --- > kernel/locking/rwsem.h | 13 ++++++++----- > 1 file changed, 8 insertions(+), 5 deletions(-) > > diff --git a/kernel/locking/rwsem.h b/kernel/locking/rwsem.h > index 45ee00236e03..1f5775aa6a1d 100644 > --- a/kernel/locking/rwsem.h > +++ b/kernel/locking/rwsem.h > @@ -174,14 +174,17 @@ static inline int __down_read_killable(struct rw_semaphore *sem) > > static inline int __down_read_trylock(struct rw_semaphore *sem) > { > - long tmp; > + /* > + * Optimize for the case when the rwsem is not locked at all. > + */ > + long tmp = RWSEM_UNLOCKED_VALUE; > > - while ((tmp = atomic_long_read(&sem->count)) >= 0) { > - if (tmp == atomic_long_cmpxchg_acquire(&sem->count, tmp, > - tmp + RWSEM_ACTIVE_READ_BIAS)) { > + do { > + if (atomic_long_try_cmpxchg_acquire(&sem->count, &tmp, > + tmp + RWSEM_ACTIVE_READ_BIAS)) { > return 1; > } > - } > + } while (tmp >= 0); > return 0; > } > >
diff --git a/kernel/locking/rwsem.h b/kernel/locking/rwsem.h index 45ee00236e03..1f5775aa6a1d 100644 --- a/kernel/locking/rwsem.h +++ b/kernel/locking/rwsem.h @@ -174,14 +174,17 @@ static inline int __down_read_killable(struct rw_semaphore *sem) static inline int __down_read_trylock(struct rw_semaphore *sem) { - long tmp; + /* + * Optimize for the case when the rwsem is not locked at all. + */ + long tmp = RWSEM_UNLOCKED_VALUE; - while ((tmp = atomic_long_read(&sem->count)) >= 0) { - if (tmp == atomic_long_cmpxchg_acquire(&sem->count, tmp, - tmp + RWSEM_ACTIVE_READ_BIAS)) { + do { + if (atomic_long_try_cmpxchg_acquire(&sem->count, &tmp, + tmp + RWSEM_ACTIVE_READ_BIAS)) { return 1; } - } + } while (tmp >= 0); return 0; }
Modify __down_read_trylock() to optimize for an unlocked rwsem and make it generate slightly better code. Before this patch, down_read_trylock: 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> 0x0000000000000005 <+5>: jmp 0x18 <down_read_trylock+24> 0x0000000000000007 <+7>: lea 0x1(%rdx),%rcx 0x000000000000000b <+11>: mov %rdx,%rax 0x000000000000000e <+14>: lock cmpxchg %rcx,(%rdi) 0x0000000000000013 <+19>: cmp %rax,%rdx 0x0000000000000016 <+22>: je 0x23 <down_read_trylock+35> 0x0000000000000018 <+24>: mov (%rdi),%rdx 0x000000000000001b <+27>: test %rdx,%rdx 0x000000000000001e <+30>: jns 0x7 <down_read_trylock+7> 0x0000000000000020 <+32>: xor %eax,%eax 0x0000000000000022 <+34>: retq 0x0000000000000023 <+35>: mov %gs:0x0,%rax 0x000000000000002c <+44>: or $0x3,%rax 0x0000000000000030 <+48>: mov %rax,0x20(%rdi) 0x0000000000000034 <+52>: mov $0x1,%eax 0x0000000000000039 <+57>: retq After patch, down_read_trylock: 0x0000000000000000 <+0>: callq 0x5 <down_read_trylock+5> 0x0000000000000005 <+5>: xor %eax,%eax 0x0000000000000007 <+7>: lea 0x1(%rax),%rdx 0x000000000000000b <+11>: lock cmpxchg %rdx,(%rdi) 0x0000000000000010 <+16>: jne 0x29 <down_read_trylock+41> 0x0000000000000012 <+18>: mov %gs:0x0,%rax 0x000000000000001b <+27>: or $0x3,%rax 0x000000000000001f <+31>: mov %rax,0x20(%rdi) 0x0000000000000023 <+35>: mov $0x1,%eax 0x0000000000000028 <+40>: retq 0x0000000000000029 <+41>: test %rax,%rax 0x000000000000002c <+44>: jns 0x7 <down_read_trylock+7> 0x000000000000002e <+46>: xor %eax,%eax 0x0000000000000030 <+48>: retq By using a rwsem microbenchmark, the down_read_trylock() rate (with a load of 10 to lengthen the lock critical section) on a x86-64 system before and after the patch were: Before Patch After Patch # of Threads rlock rlock ------------ ----- ----- 1 14,496 14,716 2 8,644 8,453 4 6,799 6,983 8 5,664 7,190 On a ARM64 system, the performance results were: Before Patch After Patch # of Threads rlock rlock ------------ ----- ----- 1 23,676 24,488 2 7,697 9,502 4 4,945 3,440 8 2,641 1,603 For the uncontended case (1 thread), the new down_read_trylock() is a little bit faster. For the contended cases, the new down_read_trylock() perform pretty well in x86-64, but performance degrades at high contention level on ARM64. Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Waiman Long <longman@redhat.com> --- kernel/locking/rwsem.h | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)