[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240207061429.3e29afc8@rorschach.local.home>
Date: Wed, 7 Feb 2024 06:14:29 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: Joel Fernandes <joel@...lfernandes.org>
Cc: Huang Yiwei <quic_hyiwei@...cinc.com>, mhiramat@...nel.org,
mark.rutland@....com, mcgrof@...nel.org, keescook@...omium.org,
j.granados@...sung.com, mathieu.desnoyers@...icios.com, corbet@....net,
linux-kernel@...r.kernel.org, linux-trace-kernel@...r.kernel.org,
linux-fsdevel@...r.kernel.org, linux-doc@...r.kernel.org,
quic_bjorande@...cinc.com, quic_tsoni@...cinc.com, quic_satyap@...cinc.com,
quic_aiquny@...cinc.com, kernel@...cinc.com, Ross Zwisler
<zwisler@...gle.com>
Subject: Re: [PATCH v4] tracing: Support to dump instance traces by
ftrace_dump_on_oops
On Wed, 7 Feb 2024 05:24:58 -0500
Joel Fernandes <joel@...lfernandes.org> wrote:
> Btw, hopefully the "trace off on warning" and related boot parameters also apply
> to instances, I haven't personally checked but I often couple those with the
> dump-on-oops ones.
Currently they do not. It would require an updated interface to do so,
as sometimes instances can be used to continue tracing after a warning,
so I don't want to make it for all instances.
Perhaps we need an option for these too, and have all options be
updated via the command line. That way we don't need to make special
boot line parameters for this. If we move these to options (keeping the
proc interface for backward compatibility) it would make most features
available to all with one change.
-- Steve
Powered by blists - more mailing lists