From: francis.giraldeau@usherbrooke.ca (Francis Giraldeau)
Subject: [ltt-dev] No page alloc events
Date: Thu, 30 Jun 2011 09:12:53 -0400 [thread overview]
Message-ID: <1309439573.2299.16.camel@berta> (raw)
Hi,
I got a really strange behavior with mm events. When the process exits,
there are a bunch of mm.page_free that occurs, but mm.page_alloc never
occur before them which is weird. I was expecting the number of
page_free to equal the number of page_alloc, only page_free fires.
If I understand correctly, if a program does a malloc, libc internally
calls sys_brk and heap is increased. But, physical memory pages are only
allocated by the page fault handler when they are accessed. So, I should
have seen the page_alloc when writing or reading the memory, isn't?
Well, page fauts are present, but then again, no page_alloc.
The trace_page_alloc tracepoint is in __alloc_pages_slowpath, and is
normally called by __do_fault. It seems that this is never called, so
maybe the tracepoint should belong to another function?
Using kernel 2.6.38.6-lttng-0.249 and lttng-modules v0.19-stable.
Have a nice day,
Francis
reply other threads:[~2011-06-30 13:12 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1309439573.2299.16.camel@berta \
--to=francis.giraldeau@usherbrooke.ca \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox