Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20190716164221.15436-1-robdclark@gmail.com> <20190716164221.15436-2-robdclark@gmail.com>
 <156329635647.9436.7142001798245279241@skylake-alporthouse-com>
In-Reply-To: <156329635647.9436.7142001798245279241@skylake-alporthouse-com>
From:   Rob Clark <robdclark@gmail.com>
Date:   Tue, 16 Jul 2019 10:03:59 -0700
Message-ID: <CAF6AEGuSZ9sRdBituUWbWNuzEav=ZCuMqan_sgP8JcaW_2aOOw@mail.gmail.com>
Subject: Re: [PATCH 2/2] drm/vgem: use normal cached mmap'ings
To:     Chris Wilson <chris@chris-wilson.co.uk>
Cc:     dri-devel <dri-devel@lists.freedesktop.org>,
        Rob Clark <robdclark@chromium.org>,
        Deepak Sharma <deepak.sharma@amd.com>,
        Rodrigo Siqueira <rodrigosiqueiramelo@gmail.com>,
        Eric Biggers <ebiggers@google.com>,
        David Airlie <airlied@linux.ie>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Thomas Zimmermann <tzimmermann@suse.de>,
        Emil Velikov <emil.velikov@collabora.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Jul 16, 2019 at 10:01 AM Chris Wilson <chris@chris-wilson.co.uk> wrote:
>
> Quoting Rob Clark (2019-07-16 17:42:15)
> > From: Rob Clark <robdclark@chromium.org>
> >
> > Since there is no real device associated with vgem, it is impossible to
> > end up with appropriate dev->dma_ops, meaning that we have no way to
> > invalidate the shmem pages allocated by vgem.  So, at least on platforms
> > without drm_cflush_pages(), we end up with corruption when cache lines
> > from previous usage of vgem bo pages get evicted to memory.
> >
> > The only sane option is to use cached mappings.
> >
> > Signed-off-by: Rob Clark <robdclark@chromium.org>
> > ---
> > Possibly we could dma_sync_*_for_{device,cpu}() on dmabuf attach/detach,
> > although the ->gem_prime_{pin,unpin}() API isn't quite ideal for that as
> > it is.  And that doesn't really help for drivers that don't attach/
> > detach for each use.
> >
> > But AFAICT vgem is mainly used for dmabuf testing, so maybe we don't
> > need to care too much about use of cached mmap'ings.
>
> Sadly this regresses with i915 interop.
>
> Starting subtest: 4KiB-tiny-vgem-blt-early-read-child
> (gem_concurrent_blit:8309) CRITICAL: Test assertion failure function dmabuf_cmp_bo, file ../tests/i915/gem_concurrent_all.c:408:
> (gem_concurrent_blit:8309) CRITICAL: Failed assertion: v[((y)*(b->width) + (((y) + pass)%(b->width)))] == val
> (gem_concurrent_blit:8309) CRITICAL: error: 0 != 0xdeadbeef
>
> and igt/prime_vgem
>
> Can you please cc intel-gfx so CI can pick up these changes?
> -Chris

I suppose CI is actually reading the imported VGEM bo from GPU?  I can
try to wire up the attach/detach dma_sync, which might help..

BR,
-R