search for: pgprot_writecombin

...memory) >> 12; - struct page *pages[npages]; - int i; - - /* phys_to_page does not exist on all platforms... */ - pages[0] = pfn_to_page(dma_to_phys(dev, node->handle) >> PAGE_SHIFT); - for (i = 1; i < npages; i++) - pages[i] = pages[0] + i; - - return vmap(pages, npages, VM_MAP, pgprot_writecombine(PAGE_KERNEL)); -} - -static void __iomem * -gk20a_instobj_cpu_map_iommu(struct nvkm_memory *memory) -{ - struct gk20a_instobj_iommu *node = gk20a_instobj_iommu(memory); - int npages = nvkm_memory_size(memory) >> 12; - - return vmap(node->pages, npages, VM_MAP, - pgprot_writecombine(P...

...handle)), handle & ~PAGE_MASK, size, dir); } @@ -622,9 +622,9 @@ static void __free_from_contiguous(struct device *dev, struct page *page, static inline pgprot_t __get_dma_pgprot(unsigned long attrs, pgprot_t prot) { - prot = dma_get_attr(DMA_ATTR_WRITE_COMBINE, attrs) ? - pgprot_writecombine(prot) : - pgprot_dmacoherent(prot); + prot = (attrs & DMA_ATTR_WRITE_COMBINE) ? + pgprot_writecombine(prot) : + pgprot_dmacoherent(prot); return prot; } @@ -744,7 +744,7 @@ static void *__dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, .gfp = gfp, .prot = pr...

...s[npages]; > - int i; > - > - /* phys_to_page does not exist on all platforms... */ > - pages[0] = pfn_to_page(dma_to_phys(dev, node->handle) >> PAGE_SHIFT); > - for (i = 1; i < npages; i++) > - pages[i] = pages[0] + i; > - > - return vmap(pages, npages, VM_MAP, pgprot_writecombine(PAGE_KERNEL)); > -} > - > -static void __iomem * > -gk20a_instobj_cpu_map_iommu(struct nvkm_memory *memory) > -{ > - struct gk20a_instobj_iommu *node = gk20a_instobj_iommu(memory); > - int npages = nvkm_memory_size(memory) >> 12; > - > - return vmap(node->pages,...

..._caching to TTM_PL_FLAG_UNCACHED. It is not clear to me why this is needed. The BO being accessed through the BAR, they are correctly considered as IO memory and mapped using ttm_bo_ioremap(), so it really seems to be unhappy with the WC mapping itself. Note that if I go ahead and force the use of pgprot_writecombine() in ttm_io_prot() to get writecombined user-space mappings, pure DRM programs that map a buffer and try to read it fail similarly, while Mesa's glReadPixels() seems to be happy. I'm not sure what it does differently here. Cheers, Alex.

...ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -500,7 +500,7 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) pgprot_val(tmp) |= _PAGE_GUARDED; } #endif -#if defined(__ia64__) +#if defined(__ia64__) || defined(__arm__) if (caching_flags & TTM_PL_FLAG_WC) tmp = pgprot_writecombine(tmp); else -- 2.0.0

...NCACHED. It is not clear to me why this is needed. The BO > being accessed through the BAR, they are correctly considered as IO > memory and mapped using ttm_bo_ioremap(), so it really seems to be > unhappy with the WC mapping itself. > > Note that if I go ahead and force the use of pgprot_writecombine() in > ttm_io_prot() to get writecombined user-space mappings, pure DRM > programs that map a buffer and try to read it fail similarly, while > Mesa's glReadPixels() seems to be happy. I'm not sure what it does > differently here. > > Cheers, > Alex. -- Pengutronix...

...m_gem_shmem_mmap(struct file *filp, struct vm_area_struct *vma) return ret; } - /* VM_PFNMAP was set by drm_gem_mmap() */ - vma->vm_flags &= ~VM_PFNMAP; - vma->vm_flags |= VM_MIXEDMAP; + vma->vm_flags |= VM_IO | VM_MIXEDMAP | VM_DONTEXPAND | VM_DONTDUMP; + vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); + vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); + vma->vm_ops = &drm_gem_shmem_vm_ops; /* Remove the fake offset */ vma->vm_pgoff -= drm_vma_node_start(&shmem->base.vma_node); diff --git a/drivers/gpu/drm/panfrost/panfros...

...m_gem_shmem_mmap(struct file *filp, struct vm_area_struct *vma) return ret; } - /* VM_PFNMAP was set by drm_gem_mmap() */ - vma->vm_flags &= ~VM_PFNMAP; - vma->vm_flags |= VM_MIXEDMAP; + vma->vm_flags |= VM_IO | VM_MIXEDMAP | VM_DONTEXPAND | VM_DONTDUMP; + vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); + vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); + vma->vm_ops = &drm_gem_shmem_vm_ops; /* Remove the fake offset */ vma->vm_pgoff -= drm_vma_node_start(&shmem->base.vma_node); diff --git a/drivers/gpu/drm/panfrost/panfros...

...m_gem_shmem_mmap(struct file *filp, struct vm_area_struct *vma) return ret; } - /* VM_PFNMAP was set by drm_gem_mmap() */ - vma->vm_flags &= ~VM_PFNMAP; - vma->vm_flags |= VM_MIXEDMAP; + vma->vm_flags |= VM_IO | VM_MIXEDMAP | VM_DONTEXPAND | VM_DONTDUMP; + vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); + vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot); + vma->vm_ops = &drm_gem_shmem_vm_ops; /* Remove the fake offset */ vma->vm_pgoff -= drm_vma_node_start(&shmem->base.vma_node); diff --git a/drivers/gpu/drm/panfrost/panfros...

...oid *rockchip_gem_prime_vmap(struct drm_gem_object *obj) +int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) { struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); - if (rk_obj->pages) - return vmap(rk_obj->pages, rk_obj->num_pages, VM_MAP, - pgprot_writecombine(PAGE_KERNEL)); + if (rk_obj->pages) { + void *vaddr = vmap(rk_obj->pages, rk_obj->num_pages, VM_MAP, + pgprot_writecombine(PAGE_KERNEL)); + if (!vaddr) + return -ENOMEM; + dma_buf_map_set_vaddr(map, vaddr); + return 0; + } if (rk_obj->dma_attrs & DMA_ATTR_NO_KERNEL_M...

...oid *rockchip_gem_prime_vmap(struct drm_gem_object *obj) +int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) { struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); - if (rk_obj->pages) - return vmap(rk_obj->pages, rk_obj->num_pages, VM_MAP, - pgprot_writecombine(PAGE_KERNEL)); + if (rk_obj->pages) { + void *vaddr = vmap(rk_obj->pages, rk_obj->num_pages, VM_MAP, + pgprot_writecombine(PAGE_KERNEL)); + if (!vaddr) + return -ENOMEM; + dma_buf_map_set_vaddr(map, vaddr); + return 0; + } if (rk_obj->dma_attrs & DMA_ATTR_NO_KERNEL_M...

...m_gem_object *obj) > +int rockchip_gem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) > { > struct rockchip_gem_object *rk_obj = to_rockchip_obj(obj); > > - if (rk_obj->pages) > - return vmap(rk_obj->pages, rk_obj->num_pages, VM_MAP, > - pgprot_writecombine(PAGE_KERNEL)); > + if (rk_obj->pages) { > + void *vaddr = vmap(rk_obj->pages, rk_obj->num_pages, VM_MAP, > + pgprot_writecombine(PAGE_KERNEL)); > + if (!vaddr) > + return -ENOMEM; > + dma_buf_map_set_vaddr(map, vaddr); > + return 0; > + } > >...

For this v2 I have fixed the patches that are non-controversial (all Lucas' :)) and am resubmitting them in the hope that they will get merged. This will just leave the issue of Nouveau system-memory buffers mapping to be solved. This issue is quite complex, so let me summarize the situation and the data I have at hand. ARM caching is like a quantum world where Murphy's law constantly

DRM's fbdev console uses regular load and store operations to update framebuffer memory. The bochs driver on sparc64 requires the use of I/O-specific load and store operations. We have a workaround, but need a long-term solution to the problem. This patchset changes GEM's vmap/vunmap interfaces to forward pointers of type struct dma_buf_map and updates the generic fbdev emulation to use

DRM's fbdev console uses regular load and store operations to update framebuffer memory. The bochs driver on sparc64 requires the use of I/O-specific load and store operations. We have a workaround, but need a long-term solution to the problem. This patchset changes GEM's vmap/vunmap interfaces to forward pointers of type struct dma_buf_map and updates the generic fbdev emulation to use

This small series introduces TTM helper functions as well as Nouveau hooks that are needed to ensure buffer coherency on ARM. Most of this series is a forward-port of some patches Lucas Stach sent last year and that are also needed for Nouveau GK20A support: http://lists.freedesktop.org/archives/nouveau/2013-August/014026.html Another patch takes care of flushing the CPU write-buffer when

This is the first set of patches to make Nouveau work on Tegra. Those are only the obvious correctness fixes, a lot of optimization work remains to be done, but at least it's enough to get accel working and let the machine survive a piglit run. A new BO flag is introduced to allow userspace to hint the kernel about possible optimizations. Lucas Stach (6): drm/ttm: recognize ARM arch in

DRM's fbdev console uses regular load and store operations to update framebuffer memory. The bochs driver on sparc64 requires the use of I/O-specific load and store operations. We have a workaround, but need a long-term solution to the problem. This patchset changes GEM's vmap/vunmap interfaces to forward pointers of type struct dma_buf_map and updates the generic fbdev emulation to use

DRM's fbdev console uses regular load and store operations to update framebuffer memory. The bochs driver on sparc64 requires the use of I/O-specific load and store operations. We have a workaround, but need a long-term solution to the problem. This patchset changes GEM's vmap/vunmap interfaces to forward pointers of type struct dma_buf_map and updates the generic fbdev emulation to use

DRM's fbdev console uses regular load and store operations to update framebuffer memory. The bochs driver on sparc64 requires the use of I/O-specific load and store operations. We have a workaround, but need a long-term solution to the problem. This patchset changes GEM's vmap/vunmap interfaces to forward pointers of type struct dma_buf_map and updates the generic fbdev emulation to use