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;
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 8BIT
In-Reply-To: <20201119130244.GN3121392@hirez.programming.kicks-ass.net>
References: <1185a97a-3780-3bce-d97d-ff9c2830e35d@gmail.com> <20201119130244.GN3121392@hirez.programming.kicks-ass.net>
Subject: Re: Deadlock cpuctx_mutex / pmus_lock / &mm->mmap_lock#2
From:   Chris Wilson <chris@chris-wilson.co.uk>
Cc:     Ingo Molnar <mingo@redhat.com>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
To:     Heiner Kallweit <hkallweit1@gmail.com>,
        Peter Zijlstra <peterz@infradead.org>
Date:   Thu, 19 Nov 2020 13:25:11 +0000
Message-ID: <160579231162.30679.12645532537616868812@build.alporthouse.com>
User-Agent: alot/0.9
Precedence: bulk

Quoting Peter Zijlstra (2020-11-19 13:02:44)
> 
> Chris, I suspect this is due to i915 calling stop machine with all sorts
> of locks held. Is there anything to be done about this? stop_machine()
> is really nasty to begin with.
> 
> What problem is it typing to solve?

If there is any concurrent access through a PCI bar (that is exported to
userspace via mmap) as the GTT is updated, results in undefined HW
behaviour (where that is not limited to users writing to other system
pages).

stop_machine() is the most foolproof method we know that works.

This particular cycle is easy to break by moving the copy_to_user to
after releasing perf_event_ctx_unlock in perf_read().
-Chris