This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Should we compile x86-64 ld.so with -mno-sse -mno-sse?
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Andreas Jaeger <aj at suse dot com>
- Cc: Richard Henderson <rth at twiddle dot net>, "Carlos O'Donell" <carlos at systemhalted dot org>, libc-alpha at sourceware dot org
- Date: Thu, 1 Nov 2012 16:17:34 -0700
- Subject: Re: Should we compile x86-64 ld.so with -mno-sse -mno-sse?
- References: <20121031132245.GA2117@gmail.com><2353366.qlHrN9GaAk@byrd><CAE2sS1gsk4rLVj9duAy7Ws_3a9ywnY7NdaYb8iGxGayXaju5Lw@mail.gmail.com><5091AD68.4020806@twiddle.net><CAMe9rOprLQuxXFMEgABYNRO=WKwb6MymPPVa5OpmdWm7wrMT8A@mail.gmail.com><50920612.30007@suse.com>
On Wed, Oct 31, 2012 at 10:18 PM, Andreas Jaeger <aj@suse.com> wrote:
> On 11/01/2012 12:18 AM, H.J. Lu wrote:
>>
>> On Wed, Oct 31, 2012 at 3:59 PM, Richard Henderson <rth@twiddle.net>
>> wrote:
>>>
>>> On 2012-11-01 03:10, Carlos O'Donell wrote:
>>>>
>>>> What use case is there for not wanting to touch SEE in ld.so?
>>>
>>>
>>> I can only imagine HJL is worried about AVX performance if the
>>> dynamic linker uses SSE insns.
>>>
>>
>> That is not the main reason. We make sure that ld.so string/memory
>> functions never touch SSE/AVX registers. It is not for performance,
>> but for correctness. Otherwise callee will get random values in
>> vector registers,
>
>
> And as you mention elsewhere, we test already for it - so is that test
> failing now for you? Under which condition?
>
We are depending on pure luck. Adding -mno-sse -mno-mmx should
ensure that ld.so doesn't use SSE nor MMX registers.
--
H.J.