Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220712042258.293010-1-jstultz@google.com> <20220712042258.293010-2-jstultz@google.com>
 <6e490429-e34c-fa51-c6d9-3315c4639d9d@amd.com>
In-Reply-To: <6e490429-e34c-fa51-c6d9-3315c4639d9d@amd.com>
From:   John Stultz <jstultz@google.com>
Date:   Tue, 12 Jul 2022 09:50:44 -0700
Message-ID: <CANDhNCo1WAUC9DhGfT19Smq2Z5duruAAwOs6AWUWUvNAjZqgAg@mail.gmail.com>
Subject: Re: [RFC][PATCH 2/3] drm: vgem: Enable SYNCOBJ and SYNCOBJ_TIMELINE
 on vgem driver
To:     =?UTF-8?Q?Christian_K=C3=B6nig?= <christian.koenig@amd.com>
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
        Maxime Ripard <mripard@kernel.org>,
        Thomas Zimmermann <tzimmermann@suse.de>,
        Jason Ekstrand <jason@jlekstrand.net>,
        Lionel Landwerlin <lionel.g.landwerlin@intel.com>,
        David Airlie <airlied@linux.ie>,
        Daniel Vetter <daniel@ffwll.ch>,
        dri-devel@lists.freedesktop.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Jul 12, 2022 at 12:46 AM Christian K=C3=B6nig
<christian.koenig@amd.com> wrote:
>
> Am 12.07.22 um 06:22 schrieb John Stultz:
> > Allows for basic SYNCOBJ api testing, in environments
> > like VMs where there may not be a supported drm driver.
> >
> > Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> > Cc: Maxime Ripard <mripard@kernel.org>
> > Cc: Thomas Zimmermann <tzimmermann@suse.de>
> > Cc: Jason Ekstrand <jason@jlekstrand.net>
> > Cc: Christian K=C3=B6nig <christian.koenig@amd.com>
> > Cc: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
> > Cc: Chunming Zhou <david1.zhou@amd.com>
> > Cc: David Airlie <airlied@linux.ie>
> > Cc: Daniel Vetter <daniel@ffwll.ch>
> > Cc: dri-devel@lists.freedesktop.org
> > Signed-off-by: John Stultz <jstultz@google.com>
> > ---
> >   drivers/gpu/drm/vgem/vgem_drv.c | 3 ++-
> >   1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vge=
m_drv.c
> > index c5e3e5457737..e5427d7399da 100644
> > --- a/drivers/gpu/drm/vgem/vgem_drv.c
> > +++ b/drivers/gpu/drm/vgem/vgem_drv.c
> > @@ -109,7 +109,8 @@ static struct drm_gem_object *vgem_gem_create_objec=
t(struct drm_device *dev, siz
> >   }
> >
> >   static const struct drm_driver vgem_driver =3D {
> > -     .driver_features                =3D DRIVER_GEM | DRIVER_RENDER,
> > +     .driver_features                =3D DRIVER_GEM | DRIVER_RENDER |
> > +                                       DRIVER_SYNCOBJ | DRIVER_SYNCOBJ=
_TIMELINE,
>
> Well that's rather surprising. I'm not an export on VGEM, but AFAIK you
> need to adjust the CS interface to support that stuff as well.

Apologies, could you clarify a bit more what you mean here?  This was
just helpful to enable the generic userland ioctls for the example
test tool in this series.

Are you proposing to add interfaces so the vgem driver can
attach/signal syncobjs similar to the
DRM_IOCTL_VGEM_FENCE_ATTACH/DRM_IOCTL_VGEM_FENCE_SIGNAL calls?

thanks
-john