svn commit: r191291 - in head: lib/libthr/thread libexec/rtld-elf/amd64 libexec/rtld-elf/arm libexec/rtld-elf/i386 libexec/rtld-elf/ia64 libexec/rtld-elf/mips libexec/rtld-elf/powerpc libexec/rtld-...

Ivan Voras ivoras at
Sun Apr 19 23:31:05 UTC 2009

2009/4/20 Robert Watson <rwatson at>:
> On Sun, 19 Apr 2009, Robert Watson wrote:
>>  Now that the kernel defines CACHE_LINE_SIZE in machine/param.h, use
>>  that definition in the custom locking code for the run-time linker
>>  rather than local definitions.
> This actually changes the line size used by the rtld code for pre-pthreads
> locking for several architectures.  I think this is an improvement, but if
> architecture maintainers could comment on that, that would be helpful.

Will there be infrastructure for creating per-CPU structures or is
using something like:

int mycounter[MAXCPU] __attribute__ ((aligned(CACHE_LINE_SIZE)));

considered enough?

More information about the svn-src-head mailing list