Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 4 May 2020 16:43:55 -0400
From:   Steven Rostedt <rostedt@goodmis.org>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Joerg Roedel <jroedel@suse.de>,
        Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        Ingo Molnar <mingo@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Borislav Petkov <bp@alien8.de>,
        Andrew Morton <akpm@linux-foundation.org>,
        Shile Zhang <shile.zhang@linux.alibaba.com>,
        Andy Lutomirski <luto@amacapital.net>,
        "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Tzvetomir Stoyanov <tz.stoyanov@gmail.com>
Subject: Re: [PATCH] percpu: Sync vmalloc mappings in pcpu_alloc() and
 free_percpu()
Message-ID: <20200504164355.7551de82@gandalf.local.home>
In-Reply-To: <20200504202540.GD5298@hirez.programming.kicks-ass.net>
References: <20200429105941.GQ30814@suse.de>
        <20200429082854.6e1796b5@oasis.local.home>
        <20200429100731.201312a9@gandalf.local.home>
        <20200430141120.GA8135@suse.de>
        <20200430121136.6d7aeb22@gandalf.local.home>
        <20200430191434.GC8135@suse.de>
        <20200430211308.74a994dc@oasis.local.home>
        <1902703609.78863.1588300015661.JavaMail.zimbra@efficios.com>
        <20200430223919.50861011@gandalf.local.home>
        <20200504151236.GI8135@suse.de>
        <20200504202540.GD5298@hirez.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, 4 May 2020 22:25:40 +0200
Peter Zijlstra <peterz@infradead.org> wrote:

> I'm only following this with half an eye atm, but one solution is to
> pull the vmalloc fault out of the trace path entirely.
> 
> With the current code that is 'interesting', but with the pile of
> patches Thomas is sitting on that would be quite feasible.

As I assume that's not something that would be backported to the stable
kernels, I'm still going to add the workaround.

But getting rid of vmalloc_sync_mappings() altogether is still a good idea
IMO.

-- Steve