This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] malloc: Use current (C11-style) atomics for fastbin access
- From: Wilco Dijkstra <Wilco dot Dijkstra at arm dot com>
- To: 'GNU C Library' <libc-alpha at sourceware dot org>, "anton at ozlabs dot org" <anton at ozlabs dot org>, Florian Weimer <fweimer at redhat dot com>
- Cc: nd <nd at arm dot com>, "tuliom at linux dot vnet dot ibm dot com" <tuliom at linux dot vnet dot ibm dot com>, "pc at us dot ibm dot com" <pc at us dot ibm dot com>, "wschmidt at us dot ibm dot com" <wschmidt at us dot ibm dot com>
- Date: Wed, 16 Jan 2019 11:52:37 +0000
- Subject: Re: [PATCH] malloc: Use current (C11-style) atomics for fastbin access
Hi,
Anton wrote:
> I see a 16% regression on ppc64le with a simple threaded malloc test
> case. I guess the C11 atomics aren't as good as what we have in glibc.
I see a 12% regression on Cortex-A72 with your test, so this slowdown
must affect all targets.
Florian wrote:
> Uh-oh. Would you please check if replacing the two atomic_load_acquire
> with atomic_load_relaxed restore the previous performance?
That improves things by about 5%, however it's still significantly slower.
The single-threaded case is also 10% slower than the previous release.
Is there a bug for this already?
Cheers,
Wilco