Implemented -e trace=%clock option

[Dmitry V. Levin]

Hi,

On Tue, Mar 14, 2017 at 10:18:27PM +0530, Rishi Bhatt wrote:
> Hey,
> I have implemented clock.test test case,and ya clock_nanosleep's test
> has setitimer as one of the option,so i did the same and and included
> setitimer in my test case so that $LOG and $EXP  have the same out
> put. But well i am still in doubt that why printing setitimer is
> necessary in clock_nanosleep.c?

No, I don't think that printing setitimer is necessary in clock_nanosleep.test.

> Well i am not sure about the changeLog thing but what i have
> understood in README-hacking i have implemented that.
> 
> I have used TRACE_SCHED 00030,well Eugene said that i have to use a
> code which doesn't collide with other TRACE_*  so what collision means
> here?

Well, collision is a quite simple concept, isn't it?  Try
$ ./strace -e%clock tests/rt_tgsigqueueinfo
- I bet you'd be surprised with the output.

> Also i have fixed the other issues that Eugene had pointed out.

Not really, at least the following issues are still there:

Applying: Implemented -e trace=%clock option.
.git/rebase-apply/patch:883: new blank line at EOF.

timeout: failed to run command '../../tests-m32/clock.test': Permission denied

BTW, the patch itself was not inlined correctly, just try to "git am" it
and you'll see what went wrong.


-- 
ldv
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 801 bytes
Desc: not available
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20170314/cda9cfa1/attachment.bin>