----- Original Message -----
> From: chru...@suse.cz
> To: "Jan Stancek" <jstan...@redhat.com>
> Cc: ltp-list@lists.sourceforge.net
> Sent: Tuesday, 26 August, 2014 9:27:49 AM
> Subject: Re: [LTP] [PATCH] sbrk02: check for ret code in setup()
>
> Hi!
> > diff --git a/testcases/kernel/syscalls/sbrk/sbrk02.c
> > b/testcases/kernel/syscalls/sbrk/sbrk02.c
> > index caa8537..5e9d4b8 100644
> > --- a/testcases/kernel/syscalls/sbrk/sbrk02.c
> > +++ b/testcases/kernel/syscalls/sbrk/sbrk02.c
> > @@ -34,7 +34,7 @@ static void sbrk_verify(void);
> > static void cleanup(void);
> > static int exp_enos[] = { ENOMEM, 0 };
> >
> > -static long increment;
> > +static long increment = INC;
> >
> > int main(int argc, char *argv[])
> > {
> > @@ -60,12 +60,19 @@ int main(int argc, char *argv[])
> >
> > static void setup(void)
> > {
> > + void *ret = NULL;
> > +
> > tst_sig(NOFORK, DEF_HANDLER, cleanup);
> >
> > TEST_PAUSE;
> >
> > - for (increment = INC; errno == 0; increment+=INC)
> > - sbrk(increment);
> > + /* call sbrk until it fails or increment overflows */
> > + while (ret != (void *)-1 && increment > 0) {
> > + ret = sbrk(increment);
> > + increment += INC;
> > + }
> > + tst_resm(TINFO | TERRNO, "setup() bailing inc: %ld, ret: %p, sbrk: %p",
> > + increment, ret, sbrk(0));
> >
> > errno = 0;
>
> Ok, this fixes the inifinite loop, but doesn't the testcase still fail
> afterwards? The sbrk_verify() expects it to fail with ENOMEM and when
> the setup fails to prepare the condition it may not fail, or am I
> mistaken?
You are right, the testcase still fails, but that looks like genuine
problem in sbrk. Man page says: "On error, -1 is returned, and errno is set to
ENOMEM.".
Regards,
Jan
>
> --
> Cyril Hrubis
> chru...@suse.cz
>
------------------------------------------------------------------------------
Slashdot TV.
Video for Nerds. Stuff that matters.
http://tv.slashdot.org/
_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list
