lkml.org 
[lkml]   [2010]   [Nov]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [ANNOUNCE] New utility: 'trace'
(Please keep the initial Cc list)


On Wed, Nov 17, 2010 at 02:47:58PM +0200, Török Edwin wrote:
> I see --call-graph in the trace record -h output, but it doesn't seem to work on x86_64 for me.
> Do I need to pass some additional parameters to trace report, do I need additional CONFIG_ turned on?
> I have:
> CONFIG_STACKTRACE_SUPPORT=y
> CONFIG_HAVE_REGS_AND_STACK_ACCESS_API=y
> # CONFIG_CC_STACKPROTECTOR is not set
> CONFIG_STACKTRACE=y
> CONFIG_USER_STACKTRACE_SUPPORT=y
> # CONFIG_STACK_TRACER is not set
> # CONFIG_DEBUG_STACKOVERFLOW is not set
> # CONFIG_DEBUG_STACK_USAGE is not set
>
> Does it use the same code for callgraphs as ftrace, so I could choose userspace stacktrace?


So, this option comes from the fact this new tools wraps on perf record
command line. So yeah, your callchains are recorded, they are just not
reported by this new tool (yet).

Only perf report know how to deal with them for now.

You can play with that though, recording with "trace record -g" and look
at the result with "perf report", you'll get nice stats about your
callchains.

Now the tracing has yet to make use of it.

Ah and it doesn't use the same code than ftrace, but perf does kernel and
user callchain snapshot too. "-g" just record both.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2010-11-17 14:35    [W:0.137 / U:3.768 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site