Re: [PATCH] drm/msm: Check for the GPU IOMMU during bind

[Jordan Crouse <jorcrous@amazon.com>]

On Fri, Mar 10, 2023 at 01:05:36AM +0200, Dmitry Baryshkov wrote:
> CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe.
> 
> 
> 
> On 10/03/2023 00:20, Jordan Crouse wrote:
> > While booting with amd,imageon on a headless target the GPU probe was
> > failing with -ENOSPC in get_pages() from msm_gem.c.
> > 
> > Investigation showed that the driver was using the default 16MB VRAM
> > carveout because msm_use_mmu() was returning false since headless devices
> > use a dummy parent device. Avoid this by extending the existing is_a2xx
> > priv member to check the GPU IOMMU state on all platforms and use that
> > check in msm_use_mmu().
> 
> I wonder if we can fix this by setting 'dummy_dev'->of_node to adreno's
> of_node. Did you check that possibility?

I said I would check and then never looped back around. This will stick
on my todo list for now and I'll check on the next cycle. If anybody
else wants to jump in the meantime then please go for it.

Jordan