Clean up the IFUNC implementations for powerpc in order to remove
unneeded macro definitions.
Tested on ppc64le with and without --disable-multi-arch flag.
* sysdeps/powerpc/powerpc64/multiarch/memchr-power7.S: Define the
implementation-specific function name and remove unneeded macros
definition.
* sysdeps/powerpc/powerpc64/multiarch/memrchr-power7.S: Likewise.
* sysdeps/powerpc/powerpc64/multiarch/rawmemchr-power7.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/memchr.S: Set a default
function name if not defined and pass as parameter to macros
accordingly.
* sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Likewise.
Current optimized powercp64/power7 memchr uses a strategy to check for
p versus align(p+n) (where 'p' is the input char pointer and n the
maximum size to check for the byte) without taking care for possible
overflow on the pointer addition in case of large 'n'.
It was triggered by 3038145ca2 where default rawmemchr (used to
created ppc64 rawmemchr in ifunc selection) now uses memchr (p, c, (size_t)-1)
on its implementation.
This patch fixes it by implement a satured addition where overflows
sets the maximum pointer size to UINTPTR_MAX.
Checked on powerpc64le-linux-gnu.
[BZ# 20971]
* sysdeps/powerpc/powerpc64/power7/memchr.S (__memchr): Avoid
overflow in pointer addition.
* string/test-memchr.c (do_test): Add an argument to pass as
the size on memchr.
(test_main): Add check for SIZE_MAX.
http://sourceware.org/ml/libc-alpha/2013-08/msg00105.html
Like strnlen, memchr and memrchr had a number of defects fixed by this
patch as well as adding little-endian support. The first one I
noticed was that the entry to the main loop needlessly checked for
"are we done yet?" when we know the size is large enough that we can't
be done. The second defect I noticed was that the main loop count was
wrong, which in turn meant that the small loop needed to handle an
extra word. Thirdly, there is nothing to say that the string can't
wrap around zero, except of course that we'd normally hit a segfault
on trying to read from address zero. Fixing that simplified a number
of places:
- /* Are we done already? */
- addi r9,r8,8
- cmpld r9,r7
- bge L(null)
becomes
+ cmpld r8,r7
+ beqlr
However, the exit gets an extra test because I test for being on the
last word then if so whether the byte offset is less than the end.
Overall, the change is a win.
Lastly, memrchr used the wrong cache hint.
* sysdeps/powerpc/powerpc64/power7/memchr.S: Replace rlwimi with
insrdi. Make better use of reg selection to speed exit slightly.
Schedule entry path a little better. Remove useless "are we done"
checks on entry to main loop. Handle wrapping around zero address.
Correct main loop count. Handle single left-over word from main
loop inline rather than by using loop_small. Remove extra word
case in loop_small caused by wrong loop count. Add little-endian
support.
* sysdeps/powerpc/powerpc32/power7/memchr.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/memrchr.S: Likewise. Use proper
cache hint.
* sysdeps/powerpc/powerpc32/power7/memrchr.S: Likewise.
* sysdeps/powerpc/powerpc64/power7/rawmemchr.S: Add little-endian
support. Avoid rlwimi.
* sysdeps/powerpc/powerpc32/power7/rawmemchr.S: Likewise.
Assorted tweaking, twisting and tuning to squeeze a few additional cycles
out of the memchr code. Changes include bypassing the shift pairs
(sld,srd) when they are not required, and unrolling the small_loop that
handles short and trailing strings.
Per scrollpipe data measuring aligned strings for 64-bit, these changes
save between five and eight cycles (9-13% overall) for short strings (<32),
Longer aligned strings see slight improvement of 1-3% due to bypassing the
shifts and the instruction rearranging.
Joseph Myers
Alan Modra
Wainer dos Santos Moschetta
Adhemerval Zanella
Allan McRae
Andreas Schwab
Will Schmidt
Paul Eggert
Luis Machado