Using regular expression filter system call

[JingPiao Chen]

On Mon, Apr 10, 2017 at 11:04 AM, Zev Weiss <zev at bewilderbeest.net> wrote:
>On Mon, Apr 10, 2017 at 02:12:02AM +0900, Masatake YAMATO wrote:
>>> Recently, class %sched and %clock are added. Is it necessary to add
>>> regular expression to filter system call?
>>
>>Interesting feature.
>>
>>> I use '@' as the prefix of regular expression. I think following regular
>>> expressions may be used. @time, @[pg]id, @^wait @xattr, @^dup, @^send,
>>> @^recv, @^read, @^write, @^sem, @^shm, @^msg, @^inotify, @^pkey_
>>
>>How about '/' (slash)?
>>
>>vi and sed users may be familiar with the notation.
>>
>>Masatake YAMATO
>>
>
>Agreed, this sounds like it would be a very handy feature -- though I'd
>suggest /regex/ for the syntax (a slash at each end, rather than just at
>the start) to match sed, awk, perl (I think dtrace also has similar
>predicate syntax as well?) etc.

Yes, '/' is better, I updated the patch.

--
JingPiao Chen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20170411/95f9e73b/attachment.html>