This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Speaking on malloc internals at FOSS.in
I think at one point the MALLOPT environment variables weren't
properly honored.. Perhaps look into that?
On Mon, Nov 26, 2012 at 11:47 AM, Carlos O'Donell
<carlos@systemhalted.org> wrote:
> On Sun, Nov 25, 2012 at 11:06 PM, Siddhesh Poyarekar
> <siddhesh@redhat.com> wrote:
>> Hi,
>>
>> I'm doing a talk on malloc internals at a conference in Bangalore
>> called FOSS.in[1]. I have a slide on future plans or concerns, which
>> include the following right now:
>>
>> * A second look at dynamic thresholds since they worsen fragmentation
>> for some cases
>> * A look at measuring resource usage in the light of multiple arenas
>> * Smarter controls on arena creation, like looking at RLIMIT_AS while
>> deciding number of arenas.
>>
>> The above are tentative ideas based on my experience, which is not
>> even close to being as extensive as most people on this list. Hence,
>> I'd really appreciate feedback on this and also any additional ideas
>> that anyone may have. I'll obviously give credit for the ideas on my
>> slides.
>
> * Accepted concensus on opensource benchmarks to use for validating
> allocator changes.
> * Stable API that allows interaction with allocator running in an application.
> - Usable by a debugger, audit library etc.
> * Application hook to control arena creation.
> - Can use allocator API and application specific information to make
> decisions.
> * Systemtap probes
> - What do we measure? What additional probes do we need?
>
> That's just off the top of my head... hope it helps.
>
> Cheers,
> Carlos.