renegade-project/linux-next
2
0
Fork 0
mirror of https://github.com/edk2-porting/linux-next.git synced 2025-01-10 14:43:54 +08:00
Code
0ade34c370
linux-next/include/asm-generic
History
Clement Courbet 0ade34c370 lib: optimize cpumask_next_and() 
We've measured that we spend ~0.6% of sys cpu time in cpumask_next_and().
It's essentially a joined iteration in search for a non-zero bit, which is
currently implemented as a lookup join (find a nonzero bit on the lhs,
lookup the rhs to see if it's set there).

Implement a direct join (find a nonzero bit on the incrementally built
join).  Also add generic bitmap benchmarks in the new `test_find_bit`
module for new function (see `find_next_and_bit` in [2] and [3] below).

For cpumask_next_and, direct benchmarking shows that it's 1.17x to 14x
faster with a geometric mean of 2.1 on 32 CPUs [1].  No impact on memory
usage.  Note that on Arm, the new pure-C implementation still outperforms
the old one that uses a mix of C and asm (`find_next_bit`) [3].

[1] Approximate benchmark code:

```
  unsigned long src1p[nr_cpumask_longs] = {pattern1};
  unsigned long src2p[nr_cpumask_longs] = {pattern2};
  for (/*a bunch of repetitions*/) {
    for (int n = -1; n <= nr_cpu_ids; ++n) {
      asm volatile("" : "+rm"(src1p)); // prevent any optimization
      asm volatile("" : "+rm"(src2p));
      unsigned long result = cpumask_next_and(n, src1p, src2p);
      asm volatile("" : "+rm"(result));
    }
  }
```

Results:
pattern1    pattern2     time_before/time_after
0x0000ffff  0x0000ffff   1.65
0x0000ffff  0x00005555   2.24
0x0000ffff  0x00001111   2.94
0x0000ffff  0x00000000   14.0
0x00005555  0x0000ffff   1.67
0x00005555  0x00005555   1.71
0x00005555  0x00001111   1.90
0x00005555  0x00000000   6.58
0x00001111  0x0000ffff   1.46
0x00001111  0x00005555   1.49
0x00001111  0x00001111   1.45
0x00001111  0x00000000   3.10
0x00000000  0x0000ffff   1.18
0x00000000  0x00005555   1.18
0x00000000  0x00001111   1.17
0x00000000  0x00000000   1.25
-----------------------------
               geo.mean  2.06

[2] test_find_next_bit, X86 (skylake)

 [ 3913.477422] Start testing find_bit() with random-filled bitmap
 [ 3913.477847] find_next_bit: 160868 cycles, 16484 iterations
 [ 3913.477933] find_next_zero_bit: 169542 cycles, 16285 iterations
 [ 3913.478036] find_last_bit: 201638 cycles, 16483 iterations
 [ 3913.480214] find_first_bit: 4353244 cycles, 16484 iterations
 [ 3913.480216] Start testing find_next_and_bit() with random-filled
 bitmap
 [ 3913.481074] find_next_and_bit: 89604 cycles, 8216 iterations
 [ 3913.481075] Start testing find_bit() with sparse bitmap
 [ 3913.481078] find_next_bit: 2536 cycles, 66 iterations
 [ 3913.481252] find_next_zero_bit: 344404 cycles, 32703 iterations
 [ 3913.481255] find_last_bit: 2006 cycles, 66 iterations
 [ 3913.481265] find_first_bit: 17488 cycles, 66 iterations
 [ 3913.481266] Start testing find_next_and_bit() with sparse bitmap
 [ 3913.481272] find_next_and_bit: 764 cycles, 1 iterations

[3] test_find_next_bit, arm (v7 odroid XU3).

[  267.206928] Start testing find_bit() with random-filled bitmap
[  267.214752] find_next_bit: 4474 cycles, 16419 iterations
[  267.221850] find_next_zero_bit: 5976 cycles, 16350 iterations
[  267.229294] find_last_bit: 4209 cycles, 16419 iterations
[  267.279131] find_first_bit: 1032991 cycles, 16420 iterations
[  267.286265] Start testing find_next_and_bit() with random-filled
bitmap
[  267.302386] find_next_and_bit: 2290 cycles, 8140 iterations
[  267.309422] Start testing find_bit() with sparse bitmap
[  267.316054] find_next_bit: 191 cycles, 66 iterations
[  267.322726] find_next_zero_bit: 8758 cycles, 32703 iterations
[  267.329803] find_last_bit: 84 cycles, 66 iterations
[  267.336169] find_first_bit: 4118 cycles, 66 iterations
[  267.342627] Start testing find_next_and_bit() with sparse bitmap
[  267.356919] find_next_and_bit: 91 cycles, 1 iterations

[courbet@google.com: v6]
  Link: http://lkml.kernel.org/r/20171129095715.23430-1-courbet@google.com
[geert@linux-m68k.org: m68k/bitops: always include <asm-generic/bitops/find.h>]
  Link: http://lkml.kernel.org/r/1512556816-28627-1-git-send-email-geert@linux-m68k.org
Link: http://lkml.kernel.org/r/20171128131334.23491-1-courbet@google.com
Signed-off-by: Clement Courbet <courbet@google.com>
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Yury Norov <ynorov@caviumnetworks.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-02-06 18:32:44 -08:00
..
bitops lib: optimize cpumask_next_and() 2018-02-06 18:32:44 -08:00
4level-fixup.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
5level-fixup.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
asm-offsets.h asm-generic: Add common asm-offsets.h 2015-06-23 13:35:49 +09:00
asm-prototypes.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
atomic64.h locking/atomic: Fix atomic_set_release() for 'funny' architectures 2017-08-10 12:28:54 +02:00
atomic-long.h Merge branch 'linus' into locking/core, to resolve conflicts 2017-11-07 10:32:44 +01:00
atomic.h openrisc: add optimized atomic operations 2017-02-25 04:14:06 +09:00
audit_change_attr.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
audit_dir_write.h audit/stable-4.15 PR 20171113 2017-11-15 13:28:48 -08:00
audit_read.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
audit_signal.h
audit_write.h audit/stable-4.15 PR 20171113 2017-11-15 13:28:48 -08:00
barrier.h locking/spinlock, arch: Update and fix spin_unlock_wait() implementations 2016-06-14 11:55:15 +02:00
bitops.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
bitsperlong.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
bug.h bug: fix "cut here" location for __WARN_TAINT architectures 2017-11-17 16:10:01 -08:00
bugs.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
cache.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
cacheflush.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
checksum.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
cmpxchg-local.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
cmpxchg.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
current.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
delay.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
device.h
div64.h A relatively calm cycle for the docs tree again. 2017-11-13 08:25:06 -08:00
dma-contiguous.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
dma-mapping.h dma-direct: rename dma_noop to dma_direct 2018-01-15 09:35:06 +01:00
dma.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
early_ioremap.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
emergency-restart.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
error-injection.h error-injection: Add injectable error types 2018-01-12 17:33:38 -08:00
exec.h Split arch_align_stack() out from asm-generic/system.h 2012-03-28 18:30:03 +01:00
export.h modversions: treat symbol CRCs as 32 bit quantities 2017-02-03 08:28:25 -08:00
extable.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
fb.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
fixmap.h asm-generic: Fix local variable shadow in __set_fixmap_offset 2016-02-16 15:10:44 +00:00
ftrace.h
futex.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
getorder.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
gpio.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
hardirq.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
hugetlb.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
hw_irq.h
ide_iops.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
int-ll64.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
io.h x86/io: Remove xlate_dev_kmem_ptr() duplication 2017-07-24 11:18:21 +02:00
ioctl.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
iomap.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
irq_regs.h
irq_work.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
irq.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
irqflags.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
kdebug.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
kmap_types.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
kprobes.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
kvm_para.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
linkage.h
local64.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
local.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
mcs_spinlock.h locking/mcs: Allow architecture specific asm files to be used for contended case 2014-02-09 21:18:52 +01:00
memory_model.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
mm_hooks.h arch, mm: Allow arch_dup_mmap() to fail 2017-12-22 20:13:01 +01:00
mm-arch-hooks.h mm: clean up per architecture MM hook header files 2015-07-17 16:39:53 -07:00
mmu_context.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
mmu.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
module.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
msi.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
page.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
param.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
parport.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pci_iomap.h PCI: Add pci_iomap_wc() variants 2015-08-25 09:59:45 +02:00
pci.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
percpu.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pgalloc.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pgtable-nop4d-hack.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pgtable-nop4d.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pgtable-nopmd.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pgtable-nopud.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
pgtable.h mm/thp: remove pmd_huge_split_prepare() 2018-01-31 17:18:38 -08:00
preempt.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
ptrace.h
qrwlock_types.h Merge branch 'linus' into locking/core, to resolve conflicts 2017-11-07 10:32:44 +01:00
qrwlock.h locking/qrwlock: Prevent slowpath writers getting held up by fastpath 2017-10-25 10:57:25 +02:00
qspinlock_types.h locking/qspinlock: Move __ARCH_SPIN_LOCK_UNLOCKED to qspinlock_types.h 2016-02-29 10:02:43 +01:00
qspinlock.h locking/arch: Remove dummy arch_{read,spin,write}_lock_flags() implementations 2017-10-10 11:50:19 +02:00
resource.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
rwsem.h Merge branch 'linus' into locking/core, to resolve conflicts 2017-11-07 10:32:44 +01:00
seccomp.h seccomp: Get compat syscalls from asm-generic header 2016-05-13 14:02:00 +02:00
sections.h sections: split dereference_function_descriptor() 2018-01-09 10:45:37 +01:00
segment.h
serial.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
set_memory.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
signal.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
simd.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
sizes.h ARM: 7430/1: sizes.h: move from asm-generic to <linux/sizes.h> 2012-06-28 17:14:34 +01:00
spinlock.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
statfs.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
string.h
switch_to.h Split the switch_to() wrapper out of asm-generic/system.h 2012-03-28 18:30:03 +01:00
syscall.h syscall.h: fix doc text for syscall_get_arch() 2014-09-23 16:20:00 -04:00
syscalls.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
termios-base.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
termios.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
timex.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
tlb.h mm: fix MADV_[FREE|DONTNEED] TLB flush miss problem 2017-08-10 15:54:07 -07:00
tlbflush.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
topology.h include/asm-generic/topology.h: remove unused parent_node() macro 2017-11-17 16:10:05 -08:00
trace_clock.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
uaccess.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
unaligned.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
unistd.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
user.h
vga.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
vmlinux.lds.h Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2018-01-31 14:31:10 -08:00
vtime.h include/asm-generic/vtime.h: avoid zero-length file 2013-09-30 14:31:02 -07:00
word-at-a-time.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
xor.h asm-generic: xor: mark static functions as __maybe_unused 2012-10-03 21:21:06 +02:00