[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250116101024.2d8040d0@gandalf.local.home>
Date: Thu, 16 Jan 2025 10:10:24 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: Linus Torvalds <torvalds@...ux-foundation.org>
Cc: LKML <linux-kernel@...r.kernel.org>, Masami Hiramatsu
<mhiramat@...nel.org>, Mark Rutland <mark.rutland@....com>, Mathieu
Desnoyers <mathieu.desnoyers@...icios.com>, Shrikanth Hegde
<sshegde@...ux.ibm.com>
Subject: Re: [GIT PULL] tracing: Fixes for v6.13-rc7
On Thu, 16 Jan 2025 09:53:33 -0500
Steven Rostedt <rostedt@...dmis.org> wrote:
> But the irqsoff and wakeup latency tracers, when the "display-graph"
> option was set, would use the function graph tracer to calculate the
> times of functions during the latency. The movement of the calltime
> calculation made the value zero for these tracers, and the output
> no longer showed the length of time of each tracer, but instead the
> absolute timestamp of when the function returned (rettime - calltime
> where calltime is now zero).
FYI,
I have a patch that removes the calltime and rettime fields from the
generic structure as they should not be used in the generic code. That is
how I found this bug. I noticed that the irqsoff and wakeup tracers used
that field as well, and when I tested those tracers with the display-graph
option enabled, that's when I noticed the regression.
But that change is a cleanup and not appropriate for an rc release. This
was more of a work around to fix the regression. The proper fix will come
in 6.14.
-- Steve
Powered by blists - more mailing lists