Waiman Long
ddb20d1d3a
locking/rwsem: Optimize down_read_trylock()
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>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-c6x-dev@linux-c6x.org
Cc: linux-m68k@lists.linux-m68k.org
Cc: linux-riscv@lists.infradead.org
Cc: linux-um@lists.infradead.org
Cc: linux-xtensa@linux-xtensa.org
Cc: linuxppc-dev@lists.ozlabs.org
Cc: nios2-dev@lists.rocketboards.org
Cc: openrisc@lists.librecores.org
Cc: uclinux-h8-devel@lists.sourceforge.jp
Link: https://lkml.kernel.org/r/20190322143008.21313-4-longman@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-04-03 14:50:52 +02:00
..
2019-03-21 19:57:02 -07:00
2019-03-12 14:08:19 -07:00
2018-10-24 20:55:56 -04:00
2018-12-30 08:31:52 +00:00
2019-03-12 10:04:02 -07:00
2019-03-22 22:50:41 +01:00
2019-03-07 18:32:02 -08:00
2019-03-23 12:32:01 +01:00
2019-03-05 15:56:59 +01:00
2019-04-03 14:50:52 +02:00
2019-03-12 10:04:02 -07:00
2019-03-15 14:22:59 -07:00
2019-03-06 07:59:36 -08:00
2019-03-24 11:42:10 -07:00
2019-03-22 13:38:26 +01:00
2019-03-26 08:35:36 -04:00
2019-03-07 18:32:02 -08:00
2019-01-31 14:20:54 +01:00
2019-01-18 17:53:29 -05:00
2019-01-14 18:01:05 -05:00
2019-01-18 17:53:29 -05:00
2019-02-03 17:49:35 -05:00
2019-02-07 21:44:27 -05:00
2019-02-12 20:17:13 -05:00
2019-02-03 17:49:35 -05:00
2018-10-31 08:54:14 -07:00
2019-01-25 11:22:43 -08:00
2019-02-07 00:13:27 +01:00
2019-03-07 18:32:02 -08:00
2019-03-28 13:34:58 +01:00
2019-03-05 21:07:14 -08:00
2019-01-08 13:18:44 -08:00
2018-10-26 16:26:32 -07:00
2019-03-07 10:11:41 -08:00
2018-10-31 08:54:12 -07:00
2019-03-09 15:53:03 -08:00
2019-03-22 13:05:26 +01:00
2019-03-07 18:31:59 -08:00
2019-04-03 11:42:33 +02:00
2019-01-21 17:38:56 -03:00
2019-04-03 14:50:52 +02:00
2018-12-15 17:44:35 +09:00
2019-03-07 18:32:02 -08:00
2018-12-28 12:11:47 -08:00
2018-12-06 14:38:50 +00:00
2019-02-13 08:16:40 +01:00
2019-03-06 10:31:36 -08:00
2019-03-07 18:32:02 -08:00
2018-12-28 12:11:52 -08:00
2018-11-07 14:41:41 +01:00
2019-03-07 18:32:00 -08:00
2018-11-16 14:11:04 +08:00
2019-03-07 18:31:59 -08:00
2018-09-16 16:08:25 +02:00
2018-12-28 12:42:30 -08:00
2018-10-31 08:54:16 -07:00
2019-03-29 10:01:37 -07:00
2018-09-11 16:13:24 +01:00
2019-03-12 13:27:20 -07:00
2019-03-16 13:05:32 -07:00
2019-01-03 18:57:57 -08:00
2019-03-07 11:44:01 -08:00
2019-03-16 13:47:14 -07:00
2019-01-30 19:27:00 +01:00
2019-02-10 21:51:39 +01:00
2018-12-05 19:31:44 -08:00
2019-03-16 13:47:14 -07:00
2019-03-07 19:25:37 -08:00
2019-01-04 13:13:47 -08:00
2019-03-12 10:04:00 -07:00
2019-02-09 08:47:52 -08:00
2018-11-27 09:21:41 -08:00
2019-01-11 18:05:40 -08:00
2018-10-09 16:51:11 +02:00
2018-11-07 23:51:16 -06:00
2018-08-30 12:56:40 +02:00
2019-03-28 13:32:01 +01:00
2019-02-01 15:46:23 -08:00
2019-03-21 12:00:18 +01:00