This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH 5/5] Add single-threaded path to _int_malloc
- From: DJ Delorie <dj at redhat dot com>
- To: Wilco Dijkstra <Wilco dot Dijkstra at arm dot com>
- Cc: libc-alpha at sourceware dot org, nd at arm dot com
- Date: Mon, 23 Oct 2017 14:50:34 -0400
- Subject: Re: [PATCH 5/5] Add single-threaded path to _int_malloc
- Authentication-results: sourceware.org; auth=none
- Authentication-results: ext-mx10.extmail.prod.ext.phx2.redhat.com; dmarc=none (p=none dis=none) header.from=redhat.com
- Authentication-results: ext-mx10.extmail.prod.ext.phx2.redhat.com; spf=fail smtp.mailfrom=dj at redhat dot com
- Dmarc-filter: OpenDMARC Filter v1.3.2 mx1.redhat.com 50DE06146F
Looks good to me.
Wilco Dijkstra <Wilco.Dijkstra@arm.com> writes:
> However moving items from the fast bin to the tcache could be done in
> a much more efficient way in the future.
I tested a couple different algorithms for that, but there's lots of
opportunity to "tweak" the cache filling.
>> Remember, "fast at any cost" is not the goal. "Fast while still
>> maintainable" is ;-)
>
> However those typically go together.
Sure, the interesting cases are when they don't ;-)
> We could use the ifunc mechanism to decide between different variants -
> including for example for the single-thread optimization.
I suspect ifuncs are chosen before we know if the program will spawn any
threads, though.