This is the mail archive of the
mailing list for the binutils project.
Re: PATCH: Add a testcase for PR gas/14201
On Thu, Jun 07, 2012 at 07:43:25AM -0700, H.J. Lu wrote:
> On Thu, Jun 7, 2012 at 7:22 AM, Alan Modra <email@example.com> wrote:
> > On Thu, Jun 07, 2012 at 06:48:28AM -0700, H.J. Lu wrote:
> >> I checked in this patch to add a testcase for PR gas/14201.
> > Please revert. ?The testcase takes too long to run.
> I timed it on a 7 years old 32-bit dual-core machine. The total
> times of "make check -j4" are
make -j hides the real impact, and the testcase will likely fail on
machines without a large amount of memory.
> It is 1 more second. I will move it to i386 directory.
My changes to sb.c were really about removing an arbitrary limit on
string buffer size. It's not something that we need to test for,
surely! The test in the PR allocated a 1G string buffer. I see now
that your test is "only" 500M, but even so that is a ridiculous amount
of memory to require for an assembler test.
Australia Development Lab, IBM