[LTP] Question about perf_event_open/Cap_bounds/su01 test cases

[Julio Cruz Barroso]

Hi Jan,

Thanks for your suggestion. 

I check the first patch (you already applied). Thanks for the changes. 

About the others two, I will send in separately.

Regards

Julio

-----Original Message-----
From: Jan Stancek [mailto:jstancek@redhat.com] 
Sent: Monday, March 14, 2016 8:11 PM
To: Julio Cruz Barroso
Cc: Cyril Hrubis; ltp@lists.linux.it
Subject: Re: [LTP] Question about perf_event_open/Cap_bounds/su01 test cases



----- Original Message -----
> From: "Julio Cruz Barroso" <julio.cruz@smartmatic.com>
> To: "Jan Stancek" <jstancek@redhat.com>
> Cc: "Cyril Hrubis" <chrubis@suse.cz>, ltp@lists.linux.it
> Sent: Sunday, 13 March, 2016 12:39:35 PM
> Subject: RE: [LTP] Question about perf_event_open/Cap_bounds/su01 test 
> cases The following test cases were fixed (attached patches for revision):

Hi,

Please post each patch separately. You're missing Signed-off-by in each and some introduce white-space errors. You can check your patches with ./scripts/checkpatch.pl in linux kernel tree. Also description could be a bit more verbose to explain "why" a patch is proposed.

> 
> - Cap_bounds: Change test to go only up to min(CAP_LAST_CAP,
> /proc/sys/kernel/cap_last_cap)

I made some tweaks here, fixed white-space errors, added check for proc file (older kernels don't have it), fixed an error, where CAP_LAST_CAP was used outside "#ifdef HAVE_LIBCAP"
and pushed.

> - getrusage03: Check available memory to run grandchild_maxrss, zombie 
> and sig_ign

"unsigned long availMem" should be fine if you first divide pagesize by 1024.
The "100" you picked as a threshold should be a define. Also it would be better if you moved it to other side of comparison. If someone has less than 100MB available memory then subtraction will underflow unsigned long.

> - mtest06_2: The default test case want to use 1G or more. Change GB 
> to MB for embedded devices (in mmap2.c).

This needs an explanation and also to answer if this testcase will still be valid for non-embedded devices.

Regards,
Jan