Impact: standardize IO on cached ops
On modern CPUs it is almost always a bad idea to use non-temporal stores,
as the regression in this commit has shown it:
30d697f: x86: fix performance regression in write() syscall
The kernel simply has no good information about whether using non-temporal
stores is a good idea or not - and trying to add heuristics only increases
complexity and inserts fragility.
The regression on cached write()s took very long to be found - over two
years. So dont take any chances and let the hardware decide how it makes
use of its caches.
The only exception is drivers/gpu/drm/i915/i915_gem.c: there were we are
absolutely sure that another entity (the GPU) will pick up the dirty
data immediately and that the CPU will not touch that data before the
GPU will.
Also, keep the _nocache() primitives to make it easier for people to
experiment with these details. There may be more clear-cut cases where
non-cached copies can be used, outside of filemap.c.
Cc: Salman Qazi <sqazi@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: make more types of copies non-temporal
This change makes the following simple fix:
30d697f: x86: fix performance regression in write() syscall
A bit more sophisticated: we check the 'total' number of bytes
written to decide whether to copy in a cached or a non-temporal
way.
This will for example cause the tail (modulo 4096 bytes) chunk
of a large write() to be non-temporal too - not just the page-sized
chunks.
Cc: Salman Qazi <sqazi@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup, enable future change
Add a 'total bytes copied' parameter to __copy_from_user_*nocache(),
and update all the callsites.
The parameter is not used yet - architecture code can use it to
more intelligently decide whether the copy should be cached or
non-temporal.
Cc: Salman Qazi <sqazi@google.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
While the introduction of __copy_from_user_nocache (see commit:
0812a579c9) may have been an improvement
for sufficiently large writes, there is evidence to show that it is
deterimental for small writes. Unixbench's fstime test gives the
following results for 256 byte writes with MAX_BLOCK of 2000:
2.6.29-rc6 ( 5 samples, each in KB/sec ):
283750, 295200, 294500, 293000, 293300
2.6.29-rc6 + this patch (5 samples, each in KB/sec):
313050, 3106750, 293350, 306300, 307900
2.6.18
395700, 342000, 399100, 366050, 359850
See w_test() in src/fstime.c in unixbench version 4.1.0. Basically, the above test
consists of counting how much we can write in this manner:
alarm(10);
while (!sigalarm) {
for (f_blocks = 0; f_blocks < 2000; ++f_blocks) {
write(f, buf, 256);
}
lseek(f, 0L, 0);
}
Note, there are other components to the write syscall regression
that are not addressed here.
Signed-off-by: Salman Qazi <sqazi@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
__copy_from_user() will return invalid value 16 when it fails to
access user space and the size is 10.
Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Change header guards named "ASM_X86__*" to "_ASM_X86_*" since:
a. the double underscore is ugly and pointless.
b. no leading underscore violates namespace constraints.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
