Message ID | 20230905184533.959171-1-adrian.larumbe@collabora.com |
---|---|
Headers | show |
Series | Add fdinfo support to Panfrost | expand |
On Tue, 5 Sep 2023 19:45:23 +0100 Adrián Larumbe <adrian.larumbe@collabora.com> wrote: > BO's RSS is updated every time new pages are allocated on demand and mapped > for the object at GPU page fault's IRQ handler, but only for heap buffers. > The reason this is unnecessary for non-heap buffers is that they are mapped > onto the GPU's VA space and backed by physical memory in their entirety at > BO creation time. > > This calculation is unnecessary for imported PRIME objects, since heap > buffers cannot be exported by our driver, and the actual BO RSS size is the > one reported in its attached dmabuf structure. > > Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com> > --- > drivers/gpu/drm/panfrost/panfrost_gem.c | 14 ++++++++++++++ > drivers/gpu/drm/panfrost/panfrost_gem.h | 5 +++++ > drivers/gpu/drm/panfrost/panfrost_mmu.c | 12 ++++++++---- > 3 files changed, 27 insertions(+), 4 deletions(-) > > diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c > index 7d8f83d20539..cb92c0ed7615 100644 > --- a/drivers/gpu/drm/panfrost/panfrost_gem.c > +++ b/drivers/gpu/drm/panfrost/panfrost_gem.c > @@ -208,6 +208,19 @@ static enum drm_gem_object_status panfrost_gem_status(struct drm_gem_object *obj > return res; > } > > +static size_t panfrost_gem_rss(struct drm_gem_object *obj) > +{ > + struct panfrost_gem_object *bo = to_panfrost_bo(obj); > + > + if (bo->is_heap) > + return bo->heap_rss_size; > + else if (bo->base.pages) { > + WARN_ON(bo->heap_rss_size); > + return bo->base.base.size; > + } else > + return 0; Nit: please add brackets on all conditional blocks, even if only the second one needs it. > +} > + > static const struct drm_gem_object_funcs panfrost_gem_funcs = { > .free = panfrost_gem_free_object, > .open = panfrost_gem_open, > @@ -220,6 +233,7 @@ static const struct drm_gem_object_funcs panfrost_gem_funcs = { > .vunmap = drm_gem_shmem_object_vunmap, > .mmap = drm_gem_shmem_object_mmap, > .status = panfrost_gem_status, > + .rss = panfrost_gem_rss, > .vm_ops = &drm_gem_shmem_vm_ops, > }; > > diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.h b/drivers/gpu/drm/panfrost/panfrost_gem.h > index ad2877eeeccd..13c0a8149c3a 100644 > --- a/drivers/gpu/drm/panfrost/panfrost_gem.h > +++ b/drivers/gpu/drm/panfrost/panfrost_gem.h > @@ -36,6 +36,11 @@ struct panfrost_gem_object { > */ > atomic_t gpu_usecount; > > + /* > + * Object chunk size currently mapped onto physical memory > + */ > + size_t heap_rss_size; > + > bool noexec :1; > bool is_heap :1; > }; > diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c > index d54d4e7b2195..67c206124781 100644 > --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c > +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c > @@ -285,17 +285,19 @@ static void panfrost_mmu_flush_range(struct panfrost_device *pfdev, > pm_runtime_put_autosuspend(pfdev->dev); > } > > -static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, > +static size_t mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, > u64 iova, int prot, struct sg_table *sgt) > { > unsigned int count; > struct scatterlist *sgl; > struct io_pgtable_ops *ops = mmu->pgtbl_ops; > u64 start_iova = iova; > + size_t total = 0; > > for_each_sgtable_dma_sg(sgt, sgl, count) { > unsigned long paddr = sg_dma_address(sgl); > size_t len = sg_dma_len(sgl); > + total += len; > > dev_dbg(pfdev->dev, "map: as=%d, iova=%llx, paddr=%lx, len=%zx", mmu->as, iova, paddr, len); > > @@ -315,7 +317,7 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, > > panfrost_mmu_flush_range(pfdev, mmu, start_iova, iova - start_iova); > > - return 0; > + return total; > } > > int panfrost_mmu_map(struct panfrost_gem_mapping *mapping) > @@ -447,6 +449,7 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, > pgoff_t page_offset; > struct sg_table *sgt; > struct page **pages; > + size_t mapped_size; > > bomapping = addr_to_mapping(pfdev, as, addr); > if (!bomapping) > @@ -518,10 +521,11 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, > if (ret) > goto err_map; > > - mmu_map_sg(pfdev, bomapping->mmu, addr, > - IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); > + mapped_size = mmu_map_sg(pfdev, bomapping->mmu, addr, > + IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); > > bomapping->active = true; > + bo->heap_rss_size += mapped_size; The alloc-on-fault granularity is set static (2MB), so no need to make mmu_map_sg() return the mapped size, we can just do += SZ_2M if things worked. > > dev_dbg(pfdev->dev, "mapped page fault @ AS%d %llx", as, addr); >
On 06.09.2023 10:01, Boris Brezillon wrote: >On Tue, 5 Sep 2023 19:45:23 +0100 >Adrián Larumbe <adrian.larumbe@collabora.com> wrote: > >> BO's RSS is updated every time new pages are allocated on demand and mapped >> for the object at GPU page fault's IRQ handler, but only for heap buffers. >> The reason this is unnecessary for non-heap buffers is that they are mapped >> onto the GPU's VA space and backed by physical memory in their entirety at >> BO creation time. >> >> This calculation is unnecessary for imported PRIME objects, since heap >> buffers cannot be exported by our driver, and the actual BO RSS size is the >> one reported in its attached dmabuf structure. >> >> Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com> >> --- >> drivers/gpu/drm/panfrost/panfrost_gem.c | 14 ++++++++++++++ >> drivers/gpu/drm/panfrost/panfrost_gem.h | 5 +++++ >> drivers/gpu/drm/panfrost/panfrost_mmu.c | 12 ++++++++---- >> 3 files changed, 27 insertions(+), 4 deletions(-) >> >> diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c >> index 7d8f83d20539..cb92c0ed7615 100644 >> --- a/drivers/gpu/drm/panfrost/panfrost_gem.c >> +++ b/drivers/gpu/drm/panfrost/panfrost_gem.c >> @@ -208,6 +208,19 @@ static enum drm_gem_object_status panfrost_gem_status(struct drm_gem_object *obj >> return res; >> } >> >> +static size_t panfrost_gem_rss(struct drm_gem_object *obj) >> +{ >> + struct panfrost_gem_object *bo = to_panfrost_bo(obj); >> + >> + if (bo->is_heap) >> + return bo->heap_rss_size; >> + else if (bo->base.pages) { >> + WARN_ON(bo->heap_rss_size); >> + return bo->base.base.size; >> + } else >> + return 0; > >Nit: please add brackets on all conditional blocks, even if only the >second one needs it. > >> +} >> + >> static const struct drm_gem_object_funcs panfrost_gem_funcs = { >> .free = panfrost_gem_free_object, >> .open = panfrost_gem_open, >> @@ -220,6 +233,7 @@ static const struct drm_gem_object_funcs panfrost_gem_funcs = { >> .vunmap = drm_gem_shmem_object_vunmap, >> .mmap = drm_gem_shmem_object_mmap, >> .status = panfrost_gem_status, >> + .rss = panfrost_gem_rss, >> .vm_ops = &drm_gem_shmem_vm_ops, >> }; >> >> diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.h b/drivers/gpu/drm/panfrost/panfrost_gem.h >> index ad2877eeeccd..13c0a8149c3a 100644 >> --- a/drivers/gpu/drm/panfrost/panfrost_gem.h >> +++ b/drivers/gpu/drm/panfrost/panfrost_gem.h >> @@ -36,6 +36,11 @@ struct panfrost_gem_object { >> */ >> atomic_t gpu_usecount; >> >> + /* >> + * Object chunk size currently mapped onto physical memory >> + */ >> + size_t heap_rss_size; >> + >> bool noexec :1; >> bool is_heap :1; >> }; >> diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c >> index d54d4e7b2195..67c206124781 100644 >> --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c >> +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c >> @@ -285,17 +285,19 @@ static void panfrost_mmu_flush_range(struct panfrost_device *pfdev, >> pm_runtime_put_autosuspend(pfdev->dev); >> } >> >> -static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, >> +static size_t mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, >> u64 iova, int prot, struct sg_table *sgt) >> { >> unsigned int count; >> struct scatterlist *sgl; >> struct io_pgtable_ops *ops = mmu->pgtbl_ops; >> u64 start_iova = iova; >> + size_t total = 0; >> >> for_each_sgtable_dma_sg(sgt, sgl, count) { >> unsigned long paddr = sg_dma_address(sgl); >> size_t len = sg_dma_len(sgl); >> + total += len; >> >> dev_dbg(pfdev->dev, "map: as=%d, iova=%llx, paddr=%lx, len=%zx", mmu->as, iova, paddr, len); >> >> @@ -315,7 +317,7 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, >> >> panfrost_mmu_flush_range(pfdev, mmu, start_iova, iova - start_iova); >> >> - return 0; >> + return total; >> } >> >> int panfrost_mmu_map(struct panfrost_gem_mapping *mapping) >> @@ -447,6 +449,7 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, >> pgoff_t page_offset; >> struct sg_table *sgt; >> struct page **pages; >> + size_t mapped_size; >> >> bomapping = addr_to_mapping(pfdev, as, addr); >> if (!bomapping) >> @@ -518,10 +521,11 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, >> if (ret) >> goto err_map; >> >> - mmu_map_sg(pfdev, bomapping->mmu, addr, >> - IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); >> + mapped_size = mmu_map_sg(pfdev, bomapping->mmu, addr, >> + IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); >> >> bomapping->active = true; >> + bo->heap_rss_size += mapped_size; > >The alloc-on-fault granularity is set static (2MB), so no need to >make mmu_map_sg() return the mapped size, we can just do += SZ_2M if >things worked. At the moment mmu_map_sg is treated as though it always succeeds in mapping the page. Would it be alright if I changed it so that we take into account the unlikely case that ops->map_pages might fail? Something like this: https://gitlab.collabora.com/-/snippets/323 >> >> dev_dbg(pfdev->dev, "mapped page fault @ AS%d %llx", as, addr); >>
On Sat, 9 Sep 2023 17:42:02 +0100 Adrián Larumbe <adrian.larumbe@collabora.com> wrote: > On 06.09.2023 10:01, Boris Brezillon wrote: > >On Tue, 5 Sep 2023 19:45:23 +0100 > >Adrián Larumbe <adrian.larumbe@collabora.com> wrote: > > > >> BO's RSS is updated every time new pages are allocated on demand and mapped > >> for the object at GPU page fault's IRQ handler, but only for heap buffers. > >> The reason this is unnecessary for non-heap buffers is that they are mapped > >> onto the GPU's VA space and backed by physical memory in their entirety at > >> BO creation time. > >> > >> This calculation is unnecessary for imported PRIME objects, since heap > >> buffers cannot be exported by our driver, and the actual BO RSS size is the > >> one reported in its attached dmabuf structure. > >> > >> Signed-off-by: Adrián Larumbe <adrian.larumbe@collabora.com> > >> --- > >> drivers/gpu/drm/panfrost/panfrost_gem.c | 14 ++++++++++++++ > >> drivers/gpu/drm/panfrost/panfrost_gem.h | 5 +++++ > >> drivers/gpu/drm/panfrost/panfrost_mmu.c | 12 ++++++++---- > >> 3 files changed, 27 insertions(+), 4 deletions(-) > >> > >> diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.c b/drivers/gpu/drm/panfrost/panfrost_gem.c > >> index 7d8f83d20539..cb92c0ed7615 100644 > >> --- a/drivers/gpu/drm/panfrost/panfrost_gem.c > >> +++ b/drivers/gpu/drm/panfrost/panfrost_gem.c > >> @@ -208,6 +208,19 @@ static enum drm_gem_object_status panfrost_gem_status(struct drm_gem_object *obj > >> return res; > >> } > >> > >> +static size_t panfrost_gem_rss(struct drm_gem_object *obj) > >> +{ > >> + struct panfrost_gem_object *bo = to_panfrost_bo(obj); > >> + > >> + if (bo->is_heap) > >> + return bo->heap_rss_size; > >> + else if (bo->base.pages) { > >> + WARN_ON(bo->heap_rss_size); > >> + return bo->base.base.size; > >> + } else > >> + return 0; > > > >Nit: please add brackets on all conditional blocks, even if only the > >second one needs it. > > > >> +} > >> + > >> static const struct drm_gem_object_funcs panfrost_gem_funcs = { > >> .free = panfrost_gem_free_object, > >> .open = panfrost_gem_open, > >> @@ -220,6 +233,7 @@ static const struct drm_gem_object_funcs panfrost_gem_funcs = { > >> .vunmap = drm_gem_shmem_object_vunmap, > >> .mmap = drm_gem_shmem_object_mmap, > >> .status = panfrost_gem_status, > >> + .rss = panfrost_gem_rss, > >> .vm_ops = &drm_gem_shmem_vm_ops, > >> }; > >> > >> diff --git a/drivers/gpu/drm/panfrost/panfrost_gem.h b/drivers/gpu/drm/panfrost/panfrost_gem.h > >> index ad2877eeeccd..13c0a8149c3a 100644 > >> --- a/drivers/gpu/drm/panfrost/panfrost_gem.h > >> +++ b/drivers/gpu/drm/panfrost/panfrost_gem.h > >> @@ -36,6 +36,11 @@ struct panfrost_gem_object { > >> */ > >> atomic_t gpu_usecount; > >> > >> + /* > >> + * Object chunk size currently mapped onto physical memory > >> + */ > >> + size_t heap_rss_size; > >> + > >> bool noexec :1; > >> bool is_heap :1; > >> }; > >> diff --git a/drivers/gpu/drm/panfrost/panfrost_mmu.c b/drivers/gpu/drm/panfrost/panfrost_mmu.c > >> index d54d4e7b2195..67c206124781 100644 > >> --- a/drivers/gpu/drm/panfrost/panfrost_mmu.c > >> +++ b/drivers/gpu/drm/panfrost/panfrost_mmu.c > >> @@ -285,17 +285,19 @@ static void panfrost_mmu_flush_range(struct panfrost_device *pfdev, > >> pm_runtime_put_autosuspend(pfdev->dev); > >> } > >> > >> -static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, > >> +static size_t mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, > >> u64 iova, int prot, struct sg_table *sgt) > >> { > >> unsigned int count; > >> struct scatterlist *sgl; > >> struct io_pgtable_ops *ops = mmu->pgtbl_ops; > >> u64 start_iova = iova; > >> + size_t total = 0; > >> > >> for_each_sgtable_dma_sg(sgt, sgl, count) { > >> unsigned long paddr = sg_dma_address(sgl); > >> size_t len = sg_dma_len(sgl); > >> + total += len; > >> > >> dev_dbg(pfdev->dev, "map: as=%d, iova=%llx, paddr=%lx, len=%zx", mmu->as, iova, paddr, len); > >> > >> @@ -315,7 +317,7 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu, > >> > >> panfrost_mmu_flush_range(pfdev, mmu, start_iova, iova - start_iova); > >> > >> - return 0; > >> + return total; > >> } > >> > >> int panfrost_mmu_map(struct panfrost_gem_mapping *mapping) > >> @@ -447,6 +449,7 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, > >> pgoff_t page_offset; > >> struct sg_table *sgt; > >> struct page **pages; > >> + size_t mapped_size; > >> > >> bomapping = addr_to_mapping(pfdev, as, addr); > >> if (!bomapping) > >> @@ -518,10 +521,11 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as, > >> if (ret) > >> goto err_map; > >> > >> - mmu_map_sg(pfdev, bomapping->mmu, addr, > >> - IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); > >> + mapped_size = mmu_map_sg(pfdev, bomapping->mmu, addr, > >> + IOMMU_WRITE | IOMMU_READ | IOMMU_NOEXEC, sgt); > >> > >> bomapping->active = true; > >> + bo->heap_rss_size += mapped_size; > > > >The alloc-on-fault granularity is set static (2MB), so no need to > >make mmu_map_sg() return the mapped size, we can just do += SZ_2M if > >things worked. > > At the moment mmu_map_sg is treated as though it always succeeds in mapping the > page. Would it be alright if I changed it so that we take into account the > unlikely case that ops->map_pages might fail? Yep, that would probably be a good thing to gracefully handle allocation failures happening in ops->map_pages(), but I'd do that in a follow-up patch, because that's orthogonal to the fdinfo stuff. > Something like this: https://gitlab.collabora.com/-/snippets/323 Nit: I would change the mmu_unmap_range() prototype for something like: static void mmu_unmap_range(struct panfrost_mmu *mmu, u64 iova, size_t len); No need for this is_heap argument if you pass rss_size to mmu_unmap_range() for heap BOs. Note that ops->unmap_pages() can fail on mem allocation too, when an unmap triggers a 2M -> 4k page table split. But I don't think this can happen in panfrost, because, for regular BOs, we always map/unmap the whole BO, and for heaps, we map/unmap 2M at a time.