search for: use_dma32

...> > > - ?????? adev->need_swiotlb, > > > - ?????? dma_addressing_limited(adev- > > > >dev)); > > > + ?????? (struct ttm_device_init_flags){ > > > + ?????? .use_dma_alloc = adev- > > > > need_swiotlb, > > > + ?????? .use_dma32 = > > > dma_addressing_limited(adev->dev) > > > + ?????? }); > > > ?? if (r) { > > > ?? DRM_ERROR("failed initializing buffer object > > > driver(%d).\n", r); > > > ?? return r; > > > diff --git a/drivers/gpu/drm/drm_ge...

...->need_swiotlb, > > > > - ?????? dma_addressing_limited(adev- > > > > >dev)); > > > > + ?????? (struct ttm_device_init_flags){ > > > > + ?????? .use_dma_alloc = adev- > > > > > need_swiotlb, > > > > + ?????? .use_dma32 = > > > > dma_addressing_limited(adev->dev) > > > > + ?????? }); I can't say I like declaring the ttm_device_init_flags as part of the function arguments. I'd prefer this a local variable but this just an opinion. Not blocker but if other agree with me, maybe c...

Some graphics APIs differentiate between out-of-graphics-memory and out-of-host-memory (system memory). Add a device init flag to have -ENOSPC propagated from the resource managers instead of being converted to -ENOMEM, to aid driver stacks in determining what error code to return or whether corrective action can be taken at the driver level. The first patch deals with a ttm_device_init()