[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [bug-idutils] [PATCH 1/9] tests: disable gnulib's get-rusage-as test
From: |
Jim Meyering |
Subject: |
Re: [bug-idutils] [PATCH 1/9] tests: disable gnulib's get-rusage-as test |
Date: |
Fri, 03 Feb 2012 09:17:52 +0100 |
Claudio Fontana wrote:
> On Thu, Feb 2, 2012 at 8:38 AM, Jim Meyering <address@hidden> wrote:
>> From: Jim Meyering <address@hidden>
>>
>> * bootstrap.conf (avoided_gnulib_modules): Disable the get-rusage-as
>> test: it's failing.
>
> Hi Jim,
>
> why are you disabling this test?
> When I was maintaining idutils some years back all tests worked (one
> did not and it was a bug that had to be fixed).
>
> Is this a new test? If not, maybe the test is just doing its job of
> reporting a real problem, which should be pursued and fixed?
Hi Claudio,
That test fails on Fedora 16 and I didn't have time to investigate.
Here are details, and I'm Cc'ing bug-gnulib:
$ ./test-get-rusage-as
../../gltests/test-get-rusage-as.c:56: assertion failed
zsh: abort (core dumped) ./test-get-rusage-as
I've marked line 56 below.
Each of the three variables value1, value2, value3 has the value 4218880.
Raising the 0x281237 to a value greater than 4218880 does not help.
int
main ()
{
uintptr_t value1, value2, value3;
value1 = get_rusage_as ();
malloc (0x88);
value2 = get_rusage_as ();
malloc (0x281237);
value3 = get_rusage_as ();
if (value1 == 0 && value2 == 0 && value3 == 0)
{
fprintf (stderr, "Skipping test: no way to determine address space
size\n");
return 77;
}
else
{
/* The address space size is positive. */
ASSERT (value1 > 0);
/* Allocating memory should not decrease the address space size. */
ASSERT (value2 >= value1);
ASSERT (value3 >= value2);
/* Allocating 2.5 MB of memory should increase the address space size. */
*** ASSERT (value3 > value1);
return 0;
}
}
- Re: [bug-idutils] [PATCH 1/9] tests: disable gnulib's get-rusage-as test,
Jim Meyering <=