This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC] Tunable elision patch for siddhesh/tunables
- From: "Paul E. Murphy" <murphyp at linux dot vnet dot ibm dot com>
- To: Andi Kleen <andi at firstfloor dot org>
- Cc: Siddhesh Poyarekar <siddhesh at redhat dot com>, "Carlos O'Donell" <carlos at redhat dot com>, Steve Munroe <sjmunroe at us dot ibm dot com>, Tulio Magno Quites Machado Filho <tuliom at linux dot vnet dot ibm dot com>, "libc-alpha at sourceware dot org" <libc-alpha at sourceware dot org>, Roland McGrath <roland at hack dot frob dot com>
- Date: Tue, 13 Oct 2015 09:02:22 -0500
- Subject: Re: [RFC] Tunable elision patch for siddhesh/tunables
- Authentication-results: sourceware.org; auth=none
- References: <5612A237 dot 7040409 at linux dot vnet dot ibm dot com> <20151013002755 dot GC3533 at two dot firstfloor dot org>
On 10/12/2015 07:27 PM, Andi Kleen wrote:
> On Mon, Oct 05, 2015 at 11:15:51AM -0500, Paul E. Murphy wrote:
>> Siddhesh et al,
>>
>> In continuing the glibc tunables discussion, I've made some
>> additional patches which I think would help our work on the
>> siddhesh/tunables branch.
>>
>> I've split this into two patches:
>>
>> * The first should enable tunables to initialize without
>> calling malloc(), and prior to __environ being set.
>>
>> * The second adds a common tunables initialization function
>> for elision, and enables it for supported archs.
>>
>> I have tested this on PPC64. I need help testing for
>> supported s390 and x86 platforms.
On the S390 side of things, I made a refactoring mistake.
"ELISION_SHOULD_ENABLE" should be "ELISION_CAN_ENABLE" if
anyone attempts to test.
>
> Sorry for the delay. I built the patch on x86_64. So far
> it seems to completely disable lock elision, even when
> explicitly enabling it with GNU_LIBC_TUNABLES=pthread.elision_enable=1
No worries. Did you try that exact string? It should be the
yet more verbose "glibc.pthread.elision_enable=1" or 0.
Siddhesh, is it necessary to prefix each tunable with "glibc."?
It is implied by the name of the envvar.
>
> I'll debug later to see why that is.
Much appreciated.
>
> -Andi
>