-----Original Message-----
From: Jan Stancek [mailto:[email protected]] 
Sent: Tuesday, February 25, 2014 7:28 PM
To: [email protected]
Cc: Francesco RUNDO; [email protected]
Subject: Re: [LTP] [PATCH] madvise/madvise03.c: Changed test exit code





----- Original Message -----
> From: [email protected]
> To: "Francesco RUNDO" <[email protected]>
> Cc: "Jan Stancek" <[email protected]>, 
> [email protected]
> Sent: Tuesday, 25 February, 2014 6:59:23 PM
> Subject: Re: [LTP] [PATCH] madvise/madvise03.c: Changed test exit code
> 
> Hi!

Hi,

> > > > > Now, "do_fallocate()" needs fallocate() op. but on ext3 this 
> > > > > is not supported i.e. ENOTSUP.
> > > 
> > > >>Do you have kernel compiled with CONFIG_TMPFS=y?
> > > 
> > > No. I've just checked kernel with above option: the reported 
> > > madvise03 testcase worked fine (usage of shmem_fallocate()).
> > > So your analysis of my scenario is corret.
> > > Please , discard posted patch.
> > 
> > >>We can still handle this with some patch to skip first testcase if 
> > >>kernel is compiled without "CONFIG_TMPFS=y". I think that qualifies as 
> > >>TCONF.
> > >>I'll leave it up to you.
> > 
> > Hmmm....yes... as TCONF(and not TFAIL) is of course the right exit 
> > code in case of TMPFS not enabled on kernel side.
> > I let you know...
> 
> Generally LTP tests should handle gracefully when syscall returns 
> ENOSYS or ENOTSUP, in reality only cases that has been catched by 
> somebody are handled.
> 
> What about following patch?

For me your modified patch is ok.
It covers scenario in which TCONF is right exit code instead of TFAIL.

Thanks,
BR
FR

>>I was thinking about trying to mount tmpfs, but that's probably unnecessary 
>>complication.
>>Limiting it to MADV_REMOVE looks good and we have hint why it may happen:

>>Reviewed-by: Jan Stancek <[email protected]>

>># ./madvise03
>>madvise03    1  TCONF  :  madvise MADV_REMOVE returned ENOTSUP CONFIG_TMPFS=y 
>>not in kernel .config?
>>madvise03    2  TPASS  :  madvise test for MADV_DONTFORK PASSED
>>madvise03    3  TPASS  :  madvise test for MADV_DOFORK PASSED

> 
> madvise/madvise03: Return TCONF on ENOTSUP
> 
> If support for TMPFS (CONFIG_TMPFS) is not compiled in kernel madvise 
> will return ENOTSUP for these testcases. Return TCONF rather than 
> TFAIL int this case.
> 
> Signed-off-by: Francesco Rundo <[email protected]>
> Signed-off-by: Cyril Hrubis <[email protected]>
> ---
>  testcases/kernel/syscalls/madvise/madvise03.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/testcases/kernel/syscalls/madvise/madvise03.c
> b/testcases/kernel/syscalls/madvise/madvise03.c
> index aeae429..02b28d2 100644
> --- a/testcases/kernel/syscalls/madvise/madvise03.c
> +++ b/testcases/kernel/syscalls/madvise/madvise03.c
> @@ -157,6 +157,11 @@ static void cleanup(void)  static void 
> check_and_print(char *advice)  {
>       if (TEST_RETURN == -1) {
> +             if (TEST_ERRNO == ENOTSUP && !strcmp(advice, "MADV_REMOVE")) {
> +                     tst_resm(TCONF, "madvise MADV_REMOVE returned ENOTSUP"
> +                              " CONFIG_TMPFS=y not in kernel .config?");
> +                     return;
> +             }
>               tst_resm(TFAIL,
>                        "madvise test for %s failed with "
>                        "return = %ld, errno = %d : %s",
> 
> --
> Cyril Hrubis
> [email protected]
> 
------------------------------------------------------------------------------
Flow-based real-time traffic analytics software. Cisco certified tool.
Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer
Customize your own dashboards, set traffic alerts and generate reports.
Network behavioral analysis & security monitoring. All-in-one tool.
http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Reply via email to