This is the mail archive of the
mailing list for the glibc project.
Re: Guard x86 elf/tst-xmmymm.out by fast-check?
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Thomas Schwinge <thomas at codesourcery dot com>
- Cc: GNU C Library <libc-alpha at sourceware dot org>
- Date: Fri, 20 Dec 2013 12:23:00 -0800
- Subject: Re: Guard x86 elf/tst-xmmymm.out by fast-check?
- Authentication-results: sourceware.org; auth=none
- References: <87a9fvguwq dot fsf at schwinge dot name> <CAMe9rOqMpZ2y55kGHaBgKr3Jy-8u+yEjTCWpnhPTRjspmngicQ at mail dot gmail dot com> <877gazgu0r dot fsf at schwinge dot name>
On Fri, Dec 20, 2013 at 12:13 PM, Thomas Schwinge
> On Fri, 20 Dec 2013 12:06:25 -0800, "H.J. Lu" <firstname.lastname@example.org> wrote:
>> On Fri, Dec 20, 2013 at 11:54 AM, Thomas Schwinge
>> <email@example.com> wrote:
>> > On my x86 GNU Hurd testing system, the x86 elf/tst-xmmymm.out test takes
>> > about 75 min of wall time, out of 130 min total, so more than half of
>> This sounds very odd. Do you have a very slow x86 machine?
> Yes, by today's standards: AMD Athlon XP with 1466 MHz. And the other
> decisive factor is that the Hurd's I/O system still is "a bit" slow.
So the problem is slow machine plus slow I/O. Why not
define slow I/O for Hurd and don't run it if I/O is slow?