| Message ID | 20180104163842.11635-1-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 4 January 2018 at 16:38, Chris Wilson <chris@chris-wilson.co.uk> wrote: > In some iommu, e.g. swiotlb, the available space can be quite limited. > So we employ a trial-and-error approach to seeing if our large > contiguous chunks can fit, and if that fails we try again with smaller > chunks after trying to free our own lazily allocated blobs. As we use a > trial-and-error approach, we do not want dma_map_sg() to emit a WARN of > its own accord, we want to gracefully report the error back to the caller > instead. > > Note that our noisy culprit, swiotlb, doesn't honour the flag, yet. > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> > Cc: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Quoting Matthew Auld (2018-01-04 17:20:16) > On 4 January 2018 at 16:38, Chris Wilson <chris@chris-wilson.co.uk> wrote: > > In some iommu, e.g. swiotlb, the available space can be quite limited. > > So we employ a trial-and-error approach to seeing if our large > > contiguous chunks can fit, and if that fails we try again with smaller > > chunks after trying to free our own lazily allocated blobs. As we use a > > trial-and-error approach, we do not want dma_map_sg() to emit a WARN of > > its own accord, we want to gracefully report the error back to the caller > > instead. > > > > Note that our noisy culprit, swiotlb, doesn't honour the flag, yet. > > > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> > > Cc: Matthew Auld <matthew.auld@intel.com> > Reviewed-by: Matthew Auld <matthew.auld@intel.com> Thanks, pushed the small tweak for a silent future. -Chris
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index c5f393870532..f2a0f556da21 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -2335,9 +2335,10 @@ int i915_gem_gtt_prepare_pages(struct drm_i915_gem_object *obj, struct sg_table *pages) { do { - if (dma_map_sg(&obj->base.dev->pdev->dev, - pages->sgl, pages->nents, - PCI_DMA_BIDIRECTIONAL)) + if (dma_map_sg_attrs(&obj->base.dev->pdev->dev, + pages->sgl, pages->nents, + PCI_DMA_BIDIRECTIONAL, + DMA_ATTR_NO_WARN)) return 0; /* If the DMA remap fails, one cause can be that we have
In some iommu, e.g. swiotlb, the available space can be quite limited. So we employ a trial-and-error approach to seeing if our large contiguous chunks can fit, and if that fails we try again with smaller chunks after trying to free our own lazily allocated blobs. As we use a trial-and-error approach, we do not want dma_map_sg() to emit a WARN of its own accord, we want to gracefully report the error back to the caller instead. Note that our noisy culprit, swiotlb, doesn't honour the flag, yet. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Matthew Auld <matthew.auld@intel.com> --- drivers/gpu/drm/i915/i915_gem_gtt.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)