This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Use single shared memory block pool for all pool allocators
- From: Andi Kleen <andi at firstfloor dot org>
- To: Mikhail Maltsev <maltsevm at gmail dot com>
- Cc: gcc-patches mailing list <gcc-patches at gcc dot gnu dot org>, mliska <mliska at suse dot cz>, Jeff Law <law at redhat dot com>
- Date: Sun, 26 Jul 2015 11:50:00 -0700
- Subject: Re: [PATCH] Use single shared memory block pool for all pool allocators
- Authentication-results: sourceware.org; auth=none
- References: <55B51674 dot 4080203 at gmail dot com>
Mikhail Maltsev <maltsevm@gmail.com> writes:
> Hi, all!
> Recently I did some profiling of GCC to find hotspots and areas of possible
> performance improvement among them. glibc malloc(3) is one of (perhaps
> known)
I've been compiling gcc with tcmalloc to do a similar speedup. It would be
interesting to compare that to your patch.
Another useful optimization is to adjust the allocation size to be >=
2MB. Then modern Linux kernels often can give you a large page,
which cuts down TLB overhead. I did similar changes some time
ago for the garbage collector.
BTW I saw big differences in larger LTO builds.
-Andi
--
ak@linux.intel.com -- Speaking for myself only