From: mathieu.desnoyers@efficios.com (Mathieu Desnoyers)
Subject: [lttng-dev] Support for large files
Date: Sat, 30 Jun 2012 14:13:13 -0400 [thread overview]
Message-ID: <20120630181313.GA30747@Krystal> (raw)
In-Reply-To: <524C960C5DFC794E82BE548D825F05CF28348826@EU-MBX-01.mgc.mentorg.com>
* Oestman, Fredrik (Fredrik_Oestman at mentor.com) wrote:
> Hi,
>
> We've run into a problem when attempting to produce large traces on a 32-bit machine.
>
> This is the console output:
>
> Session met-2012-06-15_16-09-25 created.
> Traces will be written in /home/nasir/lttng-traces/met-2012-06-15_16-09-25-20120615-160925
> UST channel met_tools enabled for session met-2012-06-15_16-09-25
> UST event met_func:* created in channel met_tools
> UST event met_call:* created in channel met_tools
> Tracing started for session met-2012-06-15_16-09-25
> Ackermann: Jun 15 2012 16:01:06
> Error: Error writing to tracefile
> Tracing stopped for session met-2012-06-15_16-09-25
> Session met-2012-06-15_16-09-25 destroyed
>
> This is the trace directory (from another, identical run):
>
> -rwxrwxrwx 1 nasir nasir 4096 2012-06-18 18:38 metadata*
> -rwxrwxrwx 1 nasir nasir 0 2012-06-18 18:38 met_tools_0*
> -rwxrwxrwx 1 nasir nasir 0 2012-06-18 18:38 met_tools_1*
> -rwxrwxrwx 1 nasir nasir 2147483647 2012-06-18 18:40 met_tools_2*
> -rwxrwxrwx 1 nasir nasir 0 2012-06-18 18:38 met_tools_3*
>
> That the application runs on one core only is expected. The file size is 2 GiB - 1, which
> is the limit for 32-bit file offset variables.
>
> On 64-bit machines, the problem doesn't occur.
>
> Is this problem known?
Good catch. I guess we did not test enough on 32-bit. It's fixed on the
lttng-tools master branch by commit:
commit c72b7d965eb65a9f49a5615cac731cec3082aa7f
Author: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
Date: Fri Jun 29 12:40:30 2012 +0200
Fix: support large files on 32-bit systems
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers at efficios.com>
it should make its way into stable-2.0 branch soon.
By the way, babeltrace has the same issue, but we need to switch from
fts.h back to ftw.h to fix that, since fts.h does not support LFS.
Thanks,
Mathieu
>
> We are using lttng-ust 2.0.1 and userspace-rcu 0.6.7.
>
>
> Cheers,
>
> Fredrik ?stman
>
>
> _______________________________________________
> lttng-dev mailing list
> lttng-dev at lists.lttng.org
> http://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
--
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
prev parent reply other threads:[~2012-06-30 18:13 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-28 8:12 Oestman, Fredrik
2012-06-30 18:13 ` Mathieu Desnoyers [this message]
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=20120630181313.GA30747@Krystal \
--to=mathieu.desnoyers@efficios.com \
/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