x86/uaccess: Optimize copy_user_enhanced_fast_string() for short strings
According to the Intel datasheet, the REP MOVSB instruction exposes a pretty heavy setup cost (50 ticks), which hurts short string copy operations. This change tries to avoid this cost by calling the explicit loop available in the unrolled code for strings shorter than 64 bytes. The 64 bytes cutoff value is arbitrary from the code logic point of view - it has been selected based on measurements, as the largest value that still ensures a measurable gain. Micro benchmarks of the __copy_from_user() function with lengths in the [0-63] range show this performance gain (shorter the string, larger the gain): - in the [55%-4%] range on Intel Xeon(R) CPU E5-2690 v4 - in the [72%-9%] range on Intel Core i7-4810MQ Other tested CPUs - namely Intel Atom S1260 and AMD Opteron 8216 - show no difference, because they do not expose the ERMS feature bit. Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alan Cox <gnomes@lxorguk.ukuu.org.uk> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Kees Cook <keescook@chromium.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/4533a1d101fd460f80e21329a34928fad521c1d4.1498744345.git.pabeni@redhat.com [ Clarified the changelog. ] Signed-off-by: Ingo Molnar <mingo@kernel.org>
This commit is contained in:
parent
4d8a991d46
commit
236222d393
|
@ -37,7 +37,7 @@ ENTRY(copy_user_generic_unrolled)
|
||||||
movl %edx,%ecx
|
movl %edx,%ecx
|
||||||
andl $63,%edx
|
andl $63,%edx
|
||||||
shrl $6,%ecx
|
shrl $6,%ecx
|
||||||
jz 17f
|
jz .L_copy_short_string
|
||||||
1: movq (%rsi),%r8
|
1: movq (%rsi),%r8
|
||||||
2: movq 1*8(%rsi),%r9
|
2: movq 1*8(%rsi),%r9
|
||||||
3: movq 2*8(%rsi),%r10
|
3: movq 2*8(%rsi),%r10
|
||||||
|
@ -58,7 +58,8 @@ ENTRY(copy_user_generic_unrolled)
|
||||||
leaq 64(%rdi),%rdi
|
leaq 64(%rdi),%rdi
|
||||||
decl %ecx
|
decl %ecx
|
||||||
jnz 1b
|
jnz 1b
|
||||||
17: movl %edx,%ecx
|
.L_copy_short_string:
|
||||||
|
movl %edx,%ecx
|
||||||
andl $7,%edx
|
andl $7,%edx
|
||||||
shrl $3,%ecx
|
shrl $3,%ecx
|
||||||
jz 20f
|
jz 20f
|
||||||
|
@ -174,6 +175,8 @@ EXPORT_SYMBOL(copy_user_generic_string)
|
||||||
*/
|
*/
|
||||||
ENTRY(copy_user_enhanced_fast_string)
|
ENTRY(copy_user_enhanced_fast_string)
|
||||||
ASM_STAC
|
ASM_STAC
|
||||||
|
cmpl $64,%edx
|
||||||
|
jb .L_copy_short_string /* less then 64 bytes, avoid the costly 'rep' */
|
||||||
movl %edx,%ecx
|
movl %edx,%ecx
|
||||||
1: rep
|
1: rep
|
||||||
movsb
|
movsb
|
||||||
|
|
Loading…
Reference in New Issue