Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1651811 > unrolled thread

[PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g

Started byXiaoguang Chen <xiaoguang.chen@intel.com>
First post2017-05-27 10:50 +0200
Last post2017-05-31 11:10 +0200
Articles 5 on this page of 25 — 7 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g Xiaoguang Chen <xiaoguang.chen@intel.com> - 2017-05-27 10:50 +0200
    [PATCH v6 2/6] drm/i915/gvt: OpRegion support for GVT-g Xiaoguang Chen <xiaoguang.chen@intel.com> - 2017-05-27 10:50 +0200
      Re: [PATCH v6 2/6] drm/i915/gvt: OpRegion support for GVT-g Zhenyu Wang <zhenyuw@linux.intel.com> - 2017-05-31 07:00 +0200
        RE: [PATCH v6 2/6] drm/i915/gvt: OpRegion support for GVT-g "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-05-31 08:30 +0200
          Re: [PATCH v6 2/6] drm/i915/gvt: OpRegion support for GVT-g Zhenyu Wang <zhenyuw@linux.intel.com> - 2017-05-31 08:40 +0200
            RE: [PATCH v6 2/6] drm/i915/gvt: OpRegion support for GVT-g "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-05-31 08:50 +0200
    [PATCH v6 5/6] drm/i915/gvt: Dmabuf support for GVT-g Xiaoguang Chen <xiaoguang.chen@intel.com> - 2017-05-27 10:50 +0200
      Re: [PATCH v6 5/6] drm/i915/gvt: Dmabuf support for GVT-g Gerd Hoffmann <kraxel@redhat.com> - 2017-05-31 14:10 +0200
        RE: [PATCH v6 5/6] drm/i915/gvt: Dmabuf support for GVT-g "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-06-01 05:10 +0200
      Re: [PATCH v6 5/6] drm/i915/gvt: Dmabuf support for GVT-g Chris Wilson <chris@chris-wilson.co.uk> - 2017-06-01 11:20 +0200
    [PATCH v6 1/6] drm/i915/gvt: Extend the GVT-g architecture to support vfio device region Xiaoguang Chen <xiaoguang.chen@intel.com> - 2017-05-27 10:50 +0200
    [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Xiaoguang Chen <xiaoguang.chen@intel.com> - 2017-05-27 10:50 +0200
      Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Gerd Hoffmann <kraxel@redhat.com> - 2017-05-29 09:30 +0200
        RE: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-05-31 08:20 +0200
          Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Kirti Wankhede <kwankhede@nvidia.com> - 2017-05-31 19:30 +0200
            RE: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-06-01 05:10 +0200
              Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf  operations Alex Williamson <alex.williamson@redhat.com> - 2017-06-01 18:40 +0200
                Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Kirti Wankhede <kwankhede@nvidia.com> - 2017-06-01 20:50 +0200
                  Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Gerd Hoffmann <kraxel@redhat.com> - 2017-06-02 10:40 +0200
                    Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Kirti Wankhede <kwankhede@nvidia.com> - 2017-06-05 10:30 +0200
                      Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations Gerd Hoffmann <kraxel@redhat.com> - 2017-06-06 10:00 +0200
    Re: [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g Gerd Hoffmann <kraxel@redhat.com> - 2017-05-30 12:30 +0200
      RE: [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-05-31 04:30 +0200
        Re: [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g Gerd Hoffmann <kraxel@redhat.com> - 2017-05-31 11:00 +0200
          RE: [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g "Chen, Xiaoguang" <xiaoguang.chen@intel.com> - 2017-05-31 11:10 +0200

Page 2 of 2 — ← Prev page 1 [2]


#1658466 — Re: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations

FromGerd Hoffmann <kraxel@redhat.com>
Date2017-06-06 10:00 +0200
SubjectRe: [PATCH v6 4/6] vfio: Define vfio based vgpu's dma-buf operations
Message-ID<tPhzb-5Qn-17@gated-at.bofh.it>
In reply to#1657383
On Mon, 2017-06-05 at 13:56 +0530, Kirti Wankhede wrote:
> 
> On 6/2/2017 2:08 PM, Gerd Hoffmann wrote:
> > 
> > > struct vfio_vgpu_surface_info {
> > >         __u64 start;
> > >         __u32 width;
> > >         __u32 height;
> > >         __u32 stride;
> > >         __u32 size;
> > >         __u32 x_pos;
> > >         __u32 y_pos;
> > >         __u32 padding;
> > >         /* Only used when VFIO_VGPU_SURFACE_DMABUF_* flags set */
> > >         __u64 drm_format_mod;
> > >         __u32 drm_format;
> > 
> > Why for dmabufs only?  Shouldn't the region specify the format
> > too? 
> > Even in case you are using a fixed one (say
> > DRM_FORMAT_XRGB8888) you
> > can explicitly say so in drm_format (and set drm_format_mod to
> > zero).
> > 
> 
> Definitions for PIXMAN formats and DRM formats are different. I think
> we need a flag to specify the format of surface that vendor driver is
> going to provide, PIXMAN or DRM.

No need to put that into the ioctl interface.  First, the kernel should
not worry about what userspace uses to process the data.  Second (most)
drm formats can trivially be mapped into pixman formats.

For example: PIXMAN_x8r8g8b8 (little endian) == DRM_FORMAT_XRGB8888

> If surface is provided through region in PIXMAN format, existing
> functions in QEMU can be used to get format from bpp value,
> qemu_default_pixman_format(). Similarly, display surface can be
> updated
> by QEMU using qemu_create_displaysurface_from() from mmaped region.

A thin wrapper which accepts a struct vfio_vgpu_surface_info and
translates that into a qemu_create_displaysurface_from() call should be
pretty small.

cheers,
  Gerd

[toc] | [prev] | [next] | [standalone]


#1653085

FromGerd Hoffmann <kraxel@redhat.com>
Date2017-05-30 12:30 +0200
Message-ID<tMMzw-677-17@gated-at.bofh.it>
In reply to#1651811
  Hi,

> This patch set adds the dma-buf support for intel GVT-g.
> dma-buf is a uniform mechanism to share DMA buffers across different
> devices and sub-systems.
> dma-buf for intel GVT-g is mainly used to share the vgpu's
> framebuffer
> to other users or sub-systems so they can use the dma-buf to show the
> desktop of a vm which uses intel vgpu.

Which branch is this based on?
Applying to gvt-stable-4.11 doesn't work.
Applying to drm-intel-next doesn't work either ...

cheers,
  Gerd

[toc] | [prev] | [next] | [standalone]


#1653787

From"Chen, Xiaoguang" <xiaoguang.chen@intel.com>
Date2017-05-31 04:30 +0200
Message-ID<tN1yx-76t-7@gated-at.bofh.it>
In reply to#1653085
Hi Gerd,

It is based on 4.12.0-rc1

>-----Original Message-----
>From: Gerd Hoffmann [mailto:kraxel@redhat.com]
>Sent: Tuesday, May 30, 2017 6:24 PM
>To: Chen, Xiaoguang <xiaoguang.chen@intel.com>;
>alex.williamson@redhat.com; chris@chris-wilson.co.uk; intel-
>gfx@lists.freedesktop.org; linux-kernel@vger.kernel.org;
>zhenyuw@linux.intel.com; Lv, Zhiyuan <zhiyuan.lv@intel.com>; intel-gvt-
>dev@lists.freedesktop.org; Wang, Zhi A <zhi.a.wang@intel.com>; Tian, Kevin
><kevin.tian@intel.com>
>Subject: Re: [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g
>
>  Hi,
>
>> This patch set adds the dma-buf support for intel GVT-g.
>> dma-buf is a uniform mechanism to share DMA buffers across different
>> devices and sub-systems.
>> dma-buf for intel GVT-g is mainly used to share the vgpu's framebuffer
>> to other users or sub-systems so they can use the dma-buf to show the
>> desktop of a vm which uses intel vgpu.
>
>Which branch is this based on?
>Applying to gvt-stable-4.11 doesn't work.
>Applying to drm-intel-next doesn't work either ...
>
>cheers,
>  Gerd

[toc] | [prev] | [next] | [standalone]


#1654019

FromGerd Hoffmann <kraxel@redhat.com>
Date2017-05-31 11:00 +0200
Message-ID<tN7DY-2u7-7@gated-at.bofh.it>
In reply to#1653787
On Wed, 2017-05-31 at 02:29 +0000, Chen, Xiaoguang wrote:
> Hi Gerd,
> 
> It is based on 4.12.0-rc1

Applies, good.
But then fails to build:

error: ‘struct vfio_vgpu_dmabuf_info’ has no member named ‘resv’

gvt/kvmgt.c:611:11: note: in expansion of macro ‘offsetofend’
   minsz = offsetofend(struct vfio_vgpu_dmabuf_info, resv);

/me wonders how this was tested ...

cheers,
  Gerd

[toc] | [prev] | [next] | [standalone]


#1654028

From"Chen, Xiaoguang" <xiaoguang.chen@intel.com>
Date2017-05-31 11:10 +0200
Message-ID<tN7NE-2Od-31@gated-at.bofh.it>
In reply to#1654019
Hi Gerd,

I found this problem once I sent the patches :(

I checked the uapi definitions and found it is usually called pad to do the aligning. So I changed the 'resv' to 'pad' in the patch but forgot to update it in the last patch and did not test after the "small" change. Next time I will test even when the change is very small.

I will change this in the next version.
Sorry for the mistake.

>-----Original Message-----
>From: Gerd Hoffmann [mailto:kraxel@redhat.com]
>Sent: Wednesday, May 31, 2017 4:59 PM
>To: Chen, Xiaoguang <xiaoguang.chen@intel.com>;
>alex.williamson@redhat.com; chris@chris-wilson.co.uk; intel-
>gfx@lists.freedesktop.org; linux-kernel@vger.kernel.org;
>zhenyuw@linux.intel.com; Lv, Zhiyuan <zhiyuan.lv@intel.com>; intel-gvt-
>dev@lists.freedesktop.org; Wang, Zhi A <zhi.a.wang@intel.com>; Tian, Kevin
><kevin.tian@intel.com>
>Subject: Re: [PATCH v6 0/6] drm/i915/gvt: Dma-buf support for GVT-g
>
>On Wed, 2017-05-31 at 02:29 +0000, Chen, Xiaoguang wrote:
>> Hi Gerd,
>>
>> It is based on 4.12.0-rc1
>
>Applies, good.
>But then fails to build:
>
>error: ‘struct vfio_vgpu_dmabuf_info’ has no member named ‘resv’
>
>gvt/kvmgt.c:611:11: note: in expansion of macro ‘offsetofend’
>   minsz = offsetofend(struct vfio_vgpu_dmabuf_info, resv);
>
>/me wonders how this was tested ...
>
>cheers,
>  Gerd

[toc] | [prev] | [standalone]


Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | linux.kernel


csiph-web