RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

a.com>; intel-gvt- > d...@lists.freedesktop.org > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Thu, Jun 29, 2017 at 08:41:53AM +0200, Gerd Hoffmann wrote: > > Hi, > > > > > > Does gvt track the live

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

vger.kernel.org; Chen, Xiaoguang > ; Zhang, Tina ; Alex > Williamson ; Lv, Zhiyuan > ; Kirti Wankhede ; intel-gvt- > d...@lists.freedesktop.org > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Thu, Jun 29, 2017 at 08:41:53AM +0200, Gerd H

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Thu, Jun 29, 2017 at 08:41:53AM +0200, Gerd Hoffmann wrote: > Hi, > > > > Does gvt track the live cycle of all dma-bufs it has handed out? > > > > The V9 implementation does track the dma-bufs' live cycle. The > > original idea was that leaving the dma-bufs' live cycle management to > >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Thu, Jun 29, 2017 at 08:41:53AM +0200, Gerd Hoffmann wrote: > Hi, > > > > Does gvt track the live cycle of all dma-bufs it has handed out? > > > > The V9 implementation does track the dma-bufs' live cycle. The > > original idea was that leaving the dma-bufs' live cycle management to > >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > Does gvt track the live cycle of all dma-bufs it has handed out? > > The V9 implementation does track the dma-bufs' live cycle. The > original idea was that leaving the dma-bufs' live cycle management to > user mode. That is still the case, user space decides which dma-bufs it'll go

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > Does gvt track the live cycle of all dma-bufs it has handed out? > > The V9 implementation does track the dma-bufs' live cycle. The > original idea was that leaving the dma-bufs' live cycle management to > user mode. That is still the case, user space decides which dma-bufs it'll go

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

.org; Wang, Zhi A <zhi.a.w...@intel.com> > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > Hi, > > > Hmm, I don't like that interface.  Can you cite examples of other > > ioctls that behave this way?  It does

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

vger.kernel.org; Chen, Xiaoguang > ; Zhang, Tina ; Kirti > Wankhede ; Lv, Zhiyuan ; > intel-gvt-...@lists.freedesktop.org; Wang, Zhi A > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > Hi, > > > Hmm, I don't like that in

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > Hmm, I don't like that interface.  Can you cite examples of other > ioctls that behave this way?  It doesn't feel like an elegant user > interface; the user can get the dmabuf, but only after they query the > dmabuf, even though the get-dmabuf ioctl returns the same data as the >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > Hmm, I don't like that interface.  Can you cite examples of other > ioctls that behave this way?  It doesn't feel like an elegant user > interface; the user can get the dmabuf, but only after they query the > dmabuf, even though the get-dmabuf ioctl returns the same data as the >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Mon, 26 Jun 2017 08:39:17 +0200 Gerd Hoffmann wrote: > Hi, > > > > With the generation we can also do something different:  Pass in > > > plane_type and > > > generation, and have VFIO_DEVICE_GET_DMABUF_FD return an error in > > > case > > > the generation doesn't match. 

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Mon, 26 Jun 2017 08:39:17 +0200 Gerd Hoffmann wrote: > Hi, > > > > With the generation we can also do something different:  Pass in > > > plane_type and > > > generation, and have VFIO_DEVICE_GET_DMABUF_FD return an error in > > > case > > > the generation doesn't match.  In that case it

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > With the generation we can also do something different:  Pass in > > plane_type and > > generation, and have VFIO_DEVICE_GET_DMABUF_FD return an error in > > case > > the generation doesn't match.  In that case it doesn't make much > > sense any > > more to have a separate plane_info

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > With the generation we can also do something different:  Pass in > > plane_type and > > generation, and have VFIO_DEVICE_GET_DMABUF_FD return an error in > > case > > the generation doesn't match.  In that case it doesn't make much > > sense any > > more to have a separate plane_info

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > So maybe a "enum plane_state" (instead of "bool is_enabled")?  So > > we > > can clearly disturgish ENABLED, DISABLED, NOT_SUPPORTED cases? > > What's the difference between NOT_SUPPORTED and -ENOTTY on the ioctl? > Perhaps a bit in a flags field could specify EN/DIS-ABLED and leave >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > So maybe a "enum plane_state" (instead of "bool is_enabled")?  So > > we > > can clearly disturgish ENABLED, DISABLED, NOT_SUPPORTED cases? > > What's the difference between NOT_SUPPORTED and -ENOTTY on the ioctl? > Perhaps a bit in a flags field could specify EN/DIS-ABLED and leave >

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

r.kernel.org; Kirti > Wankhede <kwankh...@nvidia.com>; Chen, Xiaoguang > <xiaoguang.c...@intel.com>; intel-gvt-...@lists.freedesktop.org; Lv, Zhiyuan > <zhiyuan...@intel.com>; Wang, Zhi A <zhi.a.w...@intel.com>; Wang, Zhenyu > Z <zhenyu.z.w...@intel.com> &

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

aoguang > ; intel-gvt-...@lists.freedesktop.org; Lv, Zhiyuan > ; Wang, Zhi A ; Wang, Zhenyu > Z > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: > > Hi, > > > > Thanks for all the commen

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Fri, 23 Jun 2017 09:26:59 +0200 Gerd Hoffmann wrote: > Hi, > > > Is this only going to support accelerated driver output, not basic > > VGA > > modes for BIOS interaction? > > Right now there is no vgabios or uefi support for the vgpu. > > But even with that in place

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Fri, 23 Jun 2017 09:26:59 +0200 Gerd Hoffmann wrote: > Hi, > > > Is this only going to support accelerated driver output, not basic > > VGA > > modes for BIOS interaction? > > Right now there is no vgabios or uefi support for the vgpu. > > But even with that in place there still is the

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Fri, 23 Jun 2017 10:31:28 +0200 Gerd Hoffmann wrote: > On Fri, 2017-06-23 at 15:49 +0800, Zhi Wang wrote: > > Hi: > >  Thanks for the discussions! If the userspace application has  > > already maintained a LRU list, it looks like we don't need > > generation  > >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Fri, 23 Jun 2017 10:31:28 +0200 Gerd Hoffmann wrote: > On Fri, 2017-06-23 at 15:49 +0800, Zhi Wang wrote: > > Hi: > >  Thanks for the discussions! If the userspace application has  > > already maintained a LRU list, it looks like we don't need > > generation  > > anymore, > >

Re: [kra...@redhat.com: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations]

Hi Gred and Alex: Thanks for the reply! It would be better that kernel only provides functions instead of maintaining states from my point of view. If there is any existing async notification channel in vfio device fd? like reporting device events from vfio to QEMU? If so, It would be nice

Re: [kra...@redhat.com: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations]

Hi Gred and Alex: Thanks for the reply! It would be better that kernel only provides functions instead of maintaining states from my point of view. If there is any existing async notification channel in vfio device fd? like reporting device events from vfio to QEMU? If so, It would be nice

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Fri, 2017-06-23 at 15:49 +0800, Zhi Wang wrote: > Hi: >  Thanks for the discussions! If the userspace application has  > already maintained a LRU list, it looks like we don't need > generation  > anymore, generation isn't required, things are working just fine without that. It is just a

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Fri, 2017-06-23 at 15:49 +0800, Zhi Wang wrote: > Hi: >  Thanks for the discussions! If the userspace application has  > already maintained a LRU list, it looks like we don't need > generation  > anymore, generation isn't required, things are working just fine without that. It is just a

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi: Thanks for the discussions! If the userspace application has already maintained a LRU list, it looks like we don't need generation anymore, as userspace application will lookup the guest framebuffer from the LRU list by "offset". No matter how, it would know if this is a new guest

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi: Thanks for the discussions! If the userspace application has already maintained a LRU list, it looks like we don't need generation anymore, as userspace application will lookup the guest framebuffer from the LRU list by "offset". No matter how, it would know if this is a new guest

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > Is this only going to support accelerated driver output, not basic > VGA > modes for BIOS interaction? Right now there is no vgabios or uefi support for the vgpu. But even with that in place there still is the problem that the display device initialization happens before the guest runs

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > Is this only going to support accelerated driver output, not basic > VGA > modes for BIOS interaction? Right now there is no vgabios or uefi support for the vgpu. But even with that in place there still is the problem that the display device initialization happens before the guest runs

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Thu, 22 Jun 2017 10:30:15 +0200 Gerd Hoffmann wrote: > Hi, > > > > VFIO_DEVICE_FLAGS_GFX_DMABUF? > > > > After proposing these, I'm kind of questioning their purpose.  In the > > case of a GFX region, the user is going to learn that this is > > supported > > as they

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Thu, 22 Jun 2017 10:30:15 +0200 Gerd Hoffmann wrote: > Hi, > > > > VFIO_DEVICE_FLAGS_GFX_DMABUF? > > > > After proposing these, I'm kind of questioning their purpose.  In the > > case of a GFX region, the user is going to learn that this is > > supported > > as they parse the region

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > VFIO_DEVICE_FLAGS_GFX_DMABUF? > > After proposing these, I'm kind of questioning their purpose.  In the > case of a GFX region, the user is going to learn that this is > supported > as they parse the region information and find the device specific > region identifying itself as a GFX

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > > VFIO_DEVICE_FLAGS_GFX_DMABUF? > > After proposing these, I'm kind of questioning their purpose.  In the > case of a GFX region, the user is going to learn that this is > supported > as they parse the region information and find the device specific > region identifying itself as a GFX

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

op.org; Wang, > Zhi A <zhi.a.w...@intel.com> > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Wed, 2017-06-21 at 09:20 +, Zhang, Tina wrote: > > Thanks for all the comments. I'm planning to cook the next version of >

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

esktop.org; linux-kernel@vger.kernel.org; Chen, Xiaoguang > ; Kirti Wankhede ; Lv, > Zhiyuan ; intel-gvt-...@lists.freedesktop.org; Wang, > Zhi A > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Wed, 2017-06-21 at 09:20 +, Zhang, Tin

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Wed, 21 Jun 2017 13:03:31 +0200 Gerd Hoffmann wrote: > On Wed, 2017-06-21 at 09:20 +, Zhang, Tina wrote: > > Thanks for all the comments. I'm planning to cook the next version of > > this patch set > > How about posting only this patch instead of the whole series

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Wed, 21 Jun 2017 13:03:31 +0200 Gerd Hoffmann wrote: > On Wed, 2017-06-21 at 09:20 +, Zhang, Tina wrote: > > Thanks for all the comments. I'm planning to cook the next version of > > this patch set > > How about posting only this patch instead of the whole series until > we've settled

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Wed, 2017-06-21 at 09:20 +, Zhang, Tina wrote: > Thanks for all the comments. I'm planning to cook the next version of > this patch set How about posting only this patch instead of the whole series until we've settled the interfaces? > Could the following two works? > #define

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Wed, 2017-06-21 at 09:20 +, Zhang, Tina wrote: > Thanks for all the comments. I'm planning to cook the next version of > this patch set How about posting only this patch instead of the whole series until we've settled the interfaces? > Could the following two works? > #define

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

gt;; intel-...@lists.freedesktop.org; > linux-kernel@vger.kernel.org; Kirti Wankhede <kwankh...@nvidia.com>; > Chen, Xiaoguang <xiaoguang.c...@intel.com>; intel-gvt- > d...@lists.freedesktop.org; Lv, Zhiyuan <zhiyuan...@intel.com>; Wang, Zhi A > <zhi.a.w...@

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

nel@vger.kernel.org; Kirti Wankhede ; > Chen, Xiaoguang ; intel-gvt- > d...@lists.freedesktop.org; Lv, Zhiyuan ; Wang, Zhi A > ; Wang, Zhenyu Z > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Tue, 20 Jun 2017 23:01:53 + >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > We already have VFIO_DEVICE_GET_INFO which returns: > > struct vfio_device_info { > __u32   argsz; > __u32   flags; > #define VFIO_DEVICE_FLAGS_RESET (1 << 0)/* Device supports > reset */ > #define VFIO_DEVICE_FLAGS_PCI   (1 << 1)/* vfio-pci device */ >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

Hi, > We already have VFIO_DEVICE_GET_INFO which returns: > > struct vfio_device_info { > __u32   argsz; > __u32   flags; > #define VFIO_DEVICE_FLAGS_RESET (1 << 0)/* Device supports > reset */ > #define VFIO_DEVICE_FLAGS_PCI   (1 << 1)/* vfio-pci device */ >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

gt; Lv, Zhiyuan <zhiyuan...@intel.com>; Wang, Zhi A <zhi.a.w...@intel.com>; > > Wang, Zhenyu Z <zhenyu.z.w...@intel.com> > > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > > operations > > > > On Tue, 20 Jun 20

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

sts.freedesktop.org; > > linux- > > ker...@vger.kernel.org; Kirti Wankhede ; Chen, > > Xiaoguang ; intel-gvt-...@lists.freedesktop.org; > > Lv, Zhiyuan ; Wang, Zhi A ; > > Wang, Zhenyu Z > > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > &g

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

.@vger.kernel.org; Kirti Wankhede <kwankh...@nvidia.com>; Chen, > Xiaoguang <xiaoguang.c...@intel.com>; intel-gvt-...@lists.freedesktop.org; > Lv, Zhiyuan <zhiyuan...@intel.com>; Wang, Zhi A <zhi.a.w...@intel.com>; > Wang, Zhenyu Z <zhenyu.z.w...@intel.com> &

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

t; Xiaoguang ; intel-gvt-...@lists.freedesktop.org; > Lv, Zhiyuan ; Wang, Zhi A ; > Wang, Zhenyu Z > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Tue, 20 Jun 2017 12:57:36 +0200 > Gerd Hoffmann wrote: > > > On Tue,

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On 6/20/2017 8:30 PM, Alex Williamson wrote: > On Tue, 20 Jun 2017 12:57:36 +0200 > Gerd Hoffmann wrote: > >> On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: >>> Hi, >>> >>> Thanks for all the comments. Here are the summaries: >>> >>> 1. Modify the structures to make

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On 6/20/2017 8:30 PM, Alex Williamson wrote: > On Tue, 20 Jun 2017 12:57:36 +0200 > Gerd Hoffmann wrote: > >> On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: >>> Hi, >>> >>> Thanks for all the comments. Here are the summaries: >>> >>> 1. Modify the structures to make it more general. >>>

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Tue, 20 Jun 2017 12:57:36 +0200 Gerd Hoffmann wrote: > On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: > > Hi, > > > > Thanks for all the comments. Here are the summaries: > > > > 1. Modify the structures to make it more general. > > struct

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Tue, 20 Jun 2017 12:57:36 +0200 Gerd Hoffmann wrote: > On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: > > Hi, > > > > Thanks for all the comments. Here are the summaries: > > > > 1. Modify the structures to make it more general. > > struct vfio_device_gfx_plane_info { > > __u64

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: > Hi, > > Thanks for all the comments. Here are the summaries: > > 1. Modify the structures to make it more general. > struct vfio_device_gfx_plane_info { > __u64 start; > __u64 drm_format_mod; > __u32 drm_format; >

Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

On Tue, 2017-06-20 at 08:41 +, Zhang, Tina wrote: > Hi, > > Thanks for all the comments. Here are the summaries: > > 1. Modify the structures to make it more general. > struct vfio_device_gfx_plane_info { > __u64 start; > __u64 drm_format_mod; > __u32 drm_format; >

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

kernel.org; Kirti > Wankhede <kwankh...@nvidia.com>; Chen, Xiaoguang > <xiaoguang.c...@intel.com>; intel-gvt-...@lists.freedesktop.org; Lv, Zhiyuan > <zhiyuan...@intel.com> > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations >

RE: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf operations

khede ; Chen, Xiaoguang > ; intel-gvt-...@lists.freedesktop.org; Lv, Zhiyuan > > Subject: Re: [Intel-gfx] [PATCH v9 5/7] vfio: Define vfio based dma-buf > operations > > On Mon, 19 Jun 2017 08:38:32 +0200 > Gerd Hoffmann wrote: > > > Hi, > > > > >