[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180117112556.7bc7fe0f@gandalf.local.home>
Date: Wed, 17 Jan 2018 11:25:56 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: Paul Menzel <pmenzel+linux-ftrace@...gen.mpg.de>
Cc: Ingo Molnar <mingo@...nel.org>, linux-kernel@...r.kernel.org,
regression@...mhuis.info,
Mario Limonciello <mario.limonciello@...l.com>,
Len Brown <len.brown@...el.com>,
"Rafael J. Wysocki" <rafael.j.wysocki@...el.com>
Subject: Re: Suspend/resume with FTRACE enabled doesn’t
work anymore
On Wed, 17 Jan 2018 11:19:43 -0500
Steven Rostedt <rostedt@...dmis.org> wrote:
> On Wed, 17 Jan 2018 11:10:31 -0500
> Steven Rostedt <rostedt@...dmis.org> wrote:
>
>
> > It's been fixed, and the the patch is in Linus's tree already:
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a0e3a18f4baf8e3754ac1e56f0ade924d0c0c721
>
> Although I just noticed that the revert, also reverted the required
> barriers. I'll go add them back in.
>
I take that back. Barriers are not needed, as interrupts do not need to
see the updates. And looking at the patch being reverted, I don't think
they were needed there either. The update to those bits only need to be
done for the current context to prevent recursion at that context. And
that was the case with the commit that was reverted.
-- Steve
Powered by blists - more mailing lists