Re: [PATCH] builtin expansion of memcmp for powerpc

2016-09-24 Thread Aaron Sawdey
On Sat, 2016-09-24 at 09:32 -0400, David Edelsohn wrote: > This patch broke bootstrap on AIX. > > The ChangeLog entry was not committed to the gcc/ChangeLog file. Changelog entry added. Forgot to write that emacs buffer before commit. > TARGET_LITTLE_ENDIAN only is defined for PPC64 Linux.  This

Re: [PATCH] builtin expansion of memcmp for powerpc

2016-09-24 Thread David Edelsohn
This patch broke bootstrap on AIX. The ChangeLog entry was not committed to the gcc/ChangeLog file. TARGET_LITTLE_ENDIAN only is defined for PPC64 Linux. This patch should be using !BYTES_BIG_ENDIAN. Thanks, David

Re: [PATCH] builtin expansion of memcmp for powerpc

2016-09-22 Thread Segher Boessenkool
Hi Aaron, On Thu, Sep 22, 2016 at 03:10:24PM -0500, Aaron Sawdey wrote: > The powerpc target had a movmemsi pattern which supports memcpy() but > did not have anything for memcmp(). This adds support for builtin > expansion of memcmp() into inline code for modest constant lengths. > Performance on

[PATCH] builtin expansion of memcmp for powerpc

2016-09-22 Thread Aaron Sawdey
The powerpc target had a movmemsi pattern which supports memcpy() but did not have anything for memcmp(). This adds support for builtin expansion of memcmp() into inline code for modest constant lengths. Performance on power8 is in the range of 3-7x faster than calling memcmp() for lengths under 40