[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200210221521.59928416@rorschach.local.home>
Date: Mon, 10 Feb 2020 22:15:21 -0500
From: Steven Rostedt <rostedt@...dmis.org>
To: Yiwei Zhang <zzyiwei@...gle.com>
Cc: mingo@...hat.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] Add gpu memory tracepoints
On Mon, 10 Feb 2020 19:05:35 -0800
Yiwei Zhang <zzyiwei@...gle.com> wrote:
> Thanks for the prompt reply!
>
> The tracepoint proposed here is for tracking global gpu memory usage
> total counter and per-process gpu memory usage total counter. The
> tracepoint is for gfx drivers who have implemented gpu memory tracking
> system. The tracepoint expects the de-duplication of the shared memory
> is done inside the tracking system.
>
> On Android, the graphics driver has implemented gpu memory tracking.
> First, we'd like to profiler GPU memory with this tracepoint. Second,
> we implement eBPF programs and attach to this tracepoint for tracking
> GPU memory at runtime on production devices. However, the tracepoint +
> eBPF approach requires the tracepoint to be upstreamed so that it's
> considered a stable interface which Android common kernel can carry it
> forever.
Then it needs to live in the drivers/gpu directory. kernel/trace is for
tracing infrastructure and not for adding trace points.
-- Steve
Powered by blists - more mailing lists