This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
RE: Controlling probe overhead
- From: "Stone, Joshua I" <joshua dot i dot stone at intel dot com>
- To: "Frank Ch. Eigler" <fche at redhat dot com>
- Cc: <systemtap at sources dot redhat dot com>
- Date: Mon, 12 Mar 2007 17:33:30 -0700
- Subject: RE: Controlling probe overhead
Frank Ch. Eigler wrote:
> On Mon, Mar 12, 2007 at 05:01:32PM -0700, Stone, Joshua I wrote:
>
>> [...] And since I like convoluted corner cases -- what do you do if
>> an end probe triggers an error?
>
> Why, my dear Watson, we allow sequence numbering also on the error
> probes. That way they logically interleave with end probes. Any end
> probes with a larger sequence number will not be run and error probes
> will. (An error that occurs during an error probe is actually easy to
> handle: that probe handler will unwind, but subsequent ones can run.)
Ok -- should there be a defined order for end/error probes with the same
sequence number? We could leave that ambiguous, but it would be just as
easy to impose one before the other.
Josh