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

[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, 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 +0000, 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 evict any
> >>>>>>conflicting nodes when deballooning.
> >>>>>
> >>>>>Could you elaborate a bit for above suggestion?
> >>>>
> >>>>My expectation was that the dev_priv->gtt.base.vm would contain exactly
> >>>>two holes after setup (in the mappable and non-mappable range). To do
> >>>>that you would explicitly reserve everything barred from this client
> >>>>using a set of drm_mm_reserve_node()
> >>>
> >>>Essentially a reserve_node implements what you open-code with
> >>>insert_node_range right now.
> >>
> >>Heh, there is a big difference. One inserts exactly where you ask and
> >>fails if it conflicts, the other inserts where it feels like within that
> >>range.
> 
> Do you mean drm_mm_search_free_in_range_generic() may not get
> reserve the exact range we are expecting to? Is this why you'd
> prefer the drm_mm_reserve_node()?
> 
> Besides, the ggtt_vm->mm is just initialized right before the
> ballooning code in routine i915_gem_setup_global_gtt(), so is there
> any chance the range to be partitioned out is already reserved by
> someone else?

No. It is just that drm_mm_reserve_node() was added to have the precise
semantics expected here with the strict checks. And should work better
with dynamic ballooning...
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre