This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: [HELP] Systemtap fails to load kernel module
- From: fche at redhat dot com (Frank Ch. Eigler)
- To: Shuxin Yang <shuxinyang dot oss at gmail dot com>
- Cc: Jovi Zhangwei <jovi at cloudflare dot com>, systemtap at sourceware dot org
- Date: Sun, 10 Jan 2016 14:58:08 -0500
- Subject: Re: [HELP] Systemtap fails to load kernel module
- Authentication-results: sourceware.org; auth=none
- References: <CABPcSqLiPoC=d2zHsVXTPgb+DppYfNO8x0=a+Fzvoj637h7gmw at mail dot gmail dot com> <y0mr3hvivc0 dot fsf at fche dot csb> <CABPcSqKE+fPJmSxyd6-VbFF1HxGvtC0M-CU5c-LOvevuCSbYoA at mail dot gmail dot com> <20160107184316 dot GA26193 at redhat dot com> <5690DF48 dot 3060500 at gmail dot com>
Hi -
shuxinyang.oss wrote:
> [...] The problem and the only problem is that the merged resulting
> file is suspicious as the sample-count seems to be significantly
> less than the one we collected using non-bulk mode. [..]
You're right, that is suspicous. It may help to note that the I/O
mode (bulk vs. normal) in no way affects the probe placement &
execution, so there is no adjustment of the data gathering side that
should have an effect. Maybe it's another subtler buffer-exhaustion.
(If you're using a single bound-CPU process, then bulk mode probably
wouldn't have any advantage over the normal mode anyways.)
- FChE