Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-26 Thread Yu, Zhang
On 9/26/2014 4:48 PM, Chris Wilson wrote: On Fri, Sep 26, 2014 at 04:26:20PM +0800, Zhang, Yu wrote: Hi Chris & Daniel, Thanks for your comments. Following are my understandings about the changes needed for this patch: 1> We do not need the guard page anymore between different VMs. For th

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-26 Thread Chris Wilson
On Fri, Sep 26, 2014 at 04:26:20PM +0800, Zhang, Yu wrote: > Hi Chris & Daniel, > > Thanks for your comments. Following are my understandings about > the changes needed for this patch: > > 1> We do not need the guard page anymore between different VMs. For > the very last physical GTT entry, le

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-26 Thread Zhang, Yu
Hi Chris & Daniel, Thanks for your comments. Following are my understandings about the changes needed for this patch: 1> We do not need the guard page anymore between different VMs. For the very last physical GTT entry, let's keep it pointing to a guard page. 2> To reserve the GMs in our

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-24 Thread Daniel Vetter
On Wed, Sep 24, 2014 at 08:35:50PM +0800, Zhang, Yu wrote: > Hi Daniel & Chris, > > Thank you very much for your comments, And sorry for my late reply.:) I > was focusing on other tasks previously. > See my questions below: > > On 9/23/2014 7:25 PM, Daniel Vetter wrote: > >On Tue, Sep 23, 201

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-24 Thread Chris Wilson
On Wed, Sep 24, 2014 at 08:35:50PM +0800, Zhang, Yu wrote: > Hi Daniel & Chris, > > Thank you very much for your comments, And sorry for my late > reply.:) I was focusing on other tasks previously. > See my questions below: > > On 9/23/2014 7:25 PM, Daniel Vetter wrote: > >On Tue, Sep 23, 201

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-24 Thread Zhang, Yu
Hi Daniel & Chris, Thank you very much for your comments, And sorry for my late reply.:) I was focusing on other tasks previously. See my questions below: On 9/23/2014 7:25 PM, Daniel Vetter wrote: On Tue, Sep 23, 2014 at 10:19:02AM +0100, Chris Wilson wrote: On Tue, Sep 23, 2014 at 10:2

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-23 Thread Daniel Vetter
On Tue, Sep 23, 2014 at 10:19:02AM +0100, Chris Wilson wrote: > On Tue, Sep 23, 2014 at 10:26:26AM +0200, Daniel Vetter wrote: > > On Fri, Sep 19, 2014 at 09:00:00PM +0100, Chris Wilson wrote: > > > On Fri, Sep 19, 2014 at 06:21:46PM +, Tian, Kevin wrote: > > > > > From: Chris Wilson > > > > >

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-23 Thread Chris Wilson
On Tue, Sep 23, 2014 at 10:26:26AM +0200, Daniel Vetter wrote: > On Fri, Sep 19, 2014 at 09:00:00PM +0100, Chris Wilson wrote: > > On Fri, Sep 19, 2014 at 06:21:46PM +, Tian, Kevin wrote: > > > > From: Chris Wilson > > > > The implementation also looks backwards. To work correctly with the GTT

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-23 Thread Daniel Vetter
On Fri, Sep 19, 2014 at 09:00:00PM +0100, Chris Wilson wrote: > On Fri, Sep 19, 2014 at 06:21:46PM +, Tian, Kevin wrote: > > > From: Chris Wilson > > > The implementation also looks backwards. To work correctly with the GTT > > > allocator, you need to preallocate the reserved space such that i

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-19 Thread Chris Wilson
On Fri, Sep 19, 2014 at 06:21:46PM +, Tian, Kevin wrote: > > From: Chris Wilson > > The implementation also looks backwards. To work correctly with the GTT > > allocator, you need to preallocate the reserved space such that it can > > only allocate from the allowed ranges. Similarly, it should

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-19 Thread Tian, Kevin
> From: Chris Wilson > Sent: Friday, September 19, 2014 1:06 AM > > On Sat, Sep 20, 2014 at 02:47:02AM +0800, Jike Song wrote: > > From: Yu Zhang > > > > In XenGT, the global graphic memory space is partitioned by multiple > > vgpu instances in different VMs. The ballooning code is added in > > i

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-19 Thread Daniel Vetter
On Fri, Sep 19, 2014 at 09:05:57AM +0100, Chris Wilson wrote: > On Sat, Sep 20, 2014 at 02:47:02AM +0800, Jike Song wrote: > > From: Yu Zhang > > If all entries may have prefetch issues, > > then this special guard page is necessary, to protect unexpected > > accesses into GTT entries partitioned

Re: [Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-19 Thread Chris Wilson
On Sat, Sep 20, 2014 at 02:47:02AM +0800, Jike Song wrote: > From: Yu Zhang > > In XenGT, the global graphic memory space is partitioned by multiple > vgpu instances in different VMs. The ballooning code is added in > i915_gem_setup_global_gtt(), utilizing the drm mm allocator APIs to > mark the

[Intel-gfx] [PATCH 2/8] drm/i915: Adds graphic address space ballooning logic

2014-09-18 Thread Jike Song
From: Yu Zhang In XenGT, the global graphic memory space is partitioned by multiple vgpu instances in different VMs. The ballooning code is added in i915_gem_setup_global_gtt(), utilizing the drm mm allocator APIs to mark the graphic address space which are partitioned out to other vgpus as reser