tldr; DMA buffers aren't normal memory, expecting that you can use them like that (like calling get_user_pages works, or that they're accounting like any other normal memory) cannot be guaranteed.
Since some userspace only runs on integrated devices, where all buffers are actually all resident system memory, there's a huge temptation to assume that a struct page is always present and useable like for any more pagecache backed mmap. This has the potential to result in a uapi nightmare.
To stop this gap require that DMA buffer mmaps are VM_PFNMAP, which blocks get_user_pages and all the other struct page based infrastructure for everyone. In spirit this is the uapi counterpart to the kernel-internal CONFIG_DMABUF_DEBUG.
Motivated by a recent patch which wanted to swich the system dma-buf heap to vm_insert_page instead of vm_insert_pfn.
v2:
Jason brought up that we also want to guarantee that all ptes have the pte_special flag set, to catch fast get_user_pages (on architectures that support this). Allowing VM_MIXEDMAP (like VM_SPECIAL does) would still allow vm_insert_page, but limiting to VM_PFNMAP will catch that.
From auditing the various functions to insert pfn pte entires
(vm_insert_pfn_prot, remap_pfn_range and all it's callers like dma_mmap_wc) it looks like VM_PFNMAP is already required anyway, so this should be the correct flag to check for.
v3: Change to WARN_ON_ONCE (Thomas Zimmermann)
References: https://lore.kernel.org/lkml/CAKMK7uHi+mG0z0HUmNt13QCCvutuRVjpcR0NjRL12k-WbW... Acked-by: Christian König christian.koenig@amd.com Acked-by: Thomas Zimmermann tzimmermann@suse.de Cc: Thomas Zimmermann tzimmermann@suse.de Cc: Jason Gunthorpe jgg@ziepe.ca Cc: Suren Baghdasaryan surenb@google.com Cc: Matthew Wilcox willy@infradead.org Cc: John Stultz john.stultz@linaro.org Signed-off-by: Daniel Vetter daniel.vetter@intel.com Cc: Sumit Semwal sumit.semwal@linaro.org Cc: "Christian König" christian.koenig@amd.com Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org -- Resending this so I can test the next patches for vgem/shmem in intel-gfx-ci.
No immediate plans to merge this patch here since ttm isn't addressed yet (and there we have the hugepte issue, for which I don't think we have a clear consensus yet). -Daniel --- drivers/dma-buf/dma-buf.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index 63d32261b63f..d19b1cf6c34f 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -130,6 +130,7 @@ static struct file_system_type dma_buf_fs_type = { static int dma_buf_mmap_internal(struct file *file, struct vm_area_struct *vma) { struct dma_buf *dmabuf; + int ret;
if (!is_dma_buf_file(file)) return -EINVAL; @@ -145,7 +146,11 @@ static int dma_buf_mmap_internal(struct file *file, struct vm_area_struct *vma) dmabuf->size >> PAGE_SHIFT) return -EINVAL;
- return dmabuf->ops->mmap(dmabuf, vma); + ret = dmabuf->ops->mmap(dmabuf, vma); + + WARN_ON_ONCE(!(vma->vm_flags & VM_PFNMAP)); + + return ret; }
static loff_t dma_buf_llseek(struct file *file, loff_t offset, int whence) @@ -1260,6 +1265,8 @@ EXPORT_SYMBOL_GPL(dma_buf_end_cpu_access); int dma_buf_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma, unsigned long pgoff) { + int ret; + if (WARN_ON(!dmabuf || !vma)) return -EINVAL;
@@ -1280,7 +1287,11 @@ int dma_buf_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma, vma_set_file(vma, dmabuf->file); vma->vm_pgoff = pgoff;
- return dmabuf->ops->mmap(dmabuf, vma); + ret = dmabuf->ops->mmap(dmabuf, vma); + + WARN_ON_ONCE(!(vma->vm_flags & VM_PFNMAP)); + + return ret; } EXPORT_SYMBOL_GPL(dma_buf_mmap);
We want to stop gup, which isn't the case if we use vmf_insert_page and VM_MIXEDMAP, because that does not set pte_special.
The motivation here is to stop get_user_pages from working on buffer object mmaps in general. Quoting some discussion with Thomas:
On Thu, Jul 22, 2021 at 08:22:43PM +0200, Thomas Zimmermann wrote:
Am 13.07.21 um 22:51 schrieb Daniel Vetter:
We want to stop gup, which isn't the case if we use vmf_insert_page
What is gup?
get_user_pages. It pins memory wherever it is, which badly wreaks at least ttm and could also cause trouble with cma allocations. In both cases becaue we can't move/reuse these pages anymore.
Now get_user_pages fails when the memory isn't considered "normal", like with VM_PFNMAP and using vm_insert_pfn. For consistency across all dma-buf I'm trying (together with Christian König) to roll this out everywhere, for fewer surprises.
E.g. for 5.14 iirc we merged a patch to do the same for ttm, where it closes an actual bug (ttm gets really badly confused when there's suddenly pinned pages where it thought it can move them).
cma allcoations already use VM_PFNMAP (because that's what dma_mmap is using underneath), as is anything that's using remap_pfn_range. Worst case we have to revert this patch for shmem helpers if it breaks something, but I hope that's not the case. On the ttm side we've also had some fallout that we needed to paper over with clever tricks. v2: With this shmem gem helpers now definitely need CONFIG_MMU (0day)
v3: add more depends on MMU. For usb drivers this is a bit awkward, but really it's correct: To be able to provide a contig mapping of buffers to userspace on !MMU platforms we'd need to use the cma helpers for these drivers on those platforms. As-is this wont work.
Also not exactly sure why vm_insert_page doesn't go boom, because that definitely wont fly in practice since the pages are non-contig to begin with.
v4: Explain the entire motivation a lot more (Thomas)
Acked-by: Thomas Zimmermann tzimmermann@suse.de Signed-off-by: Daniel Vetter daniel.vetter@intel.com Cc: Maarten Lankhorst maarten.lankhorst@linux.intel.com Cc: Maxime Ripard mripard@kernel.org Cc: Thomas Zimmermann tzimmermann@suse.de Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch --- drivers/gpu/drm/Kconfig | 2 +- drivers/gpu/drm/drm_gem_shmem_helper.c | 4 ++-- drivers/gpu/drm/gud/Kconfig | 2 +- drivers/gpu/drm/tiny/Kconfig | 4 ++-- drivers/gpu/drm/udl/Kconfig | 1 + 5 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 0d372354c2d0..314eefa39892 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -211,7 +211,7 @@ config DRM_KMS_CMA_HELPER
config DRM_GEM_SHMEM_HELPER bool - depends on DRM + depends on DRM && MMU help Choose this if you need the GEM shmem helper functions
diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index a61946374c82..cc96d1c3570e 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -542,7 +542,7 @@ static vm_fault_t drm_gem_shmem_fault(struct vm_fault *vmf) } else { page = shmem->pages[page_offset];
- ret = vmf_insert_page(vma, vmf->address, page); + ret = vmf_insert_pfn(vma, vmf->address, page_to_pfn(page)); }
mutex_unlock(&shmem->pages_lock); @@ -612,7 +612,7 @@ int drm_gem_shmem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma) return ret; }
- vma->vm_flags |= VM_MIXEDMAP | VM_DONTEXPAND; + vma->vm_flags |= VM_PFNMAP | VM_DONTEXPAND; vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); if (shmem->map_wc) vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); diff --git a/drivers/gpu/drm/gud/Kconfig b/drivers/gpu/drm/gud/Kconfig index 1c8601bf4d91..9c1e61f9eec3 100644 --- a/drivers/gpu/drm/gud/Kconfig +++ b/drivers/gpu/drm/gud/Kconfig @@ -2,7 +2,7 @@
config DRM_GUD tristate "GUD USB Display" - depends on DRM && USB + depends on DRM && USB && MMU select LZ4_COMPRESS select DRM_KMS_HELPER select DRM_GEM_SHMEM_HELPER diff --git a/drivers/gpu/drm/tiny/Kconfig b/drivers/gpu/drm/tiny/Kconfig index d31be274a2bd..1ceb93fbdc50 100644 --- a/drivers/gpu/drm/tiny/Kconfig +++ b/drivers/gpu/drm/tiny/Kconfig @@ -44,7 +44,7 @@ config DRM_CIRRUS_QEMU
config DRM_GM12U320 tristate "GM12U320 driver for USB projectors" - depends on DRM && USB + depends on DRM && USB && MMU select DRM_KMS_HELPER select DRM_GEM_SHMEM_HELPER help @@ -53,7 +53,7 @@ config DRM_GM12U320
config DRM_SIMPLEDRM tristate "Simple framebuffer driver" - depends on DRM + depends on DRM && MMU select DRM_GEM_SHMEM_HELPER select DRM_KMS_HELPER help diff --git a/drivers/gpu/drm/udl/Kconfig b/drivers/gpu/drm/udl/Kconfig index 1f497d8f1ae5..c744175c6992 100644 --- a/drivers/gpu/drm/udl/Kconfig +++ b/drivers/gpu/drm/udl/Kconfig @@ -4,6 +4,7 @@ config DRM_UDL depends on DRM depends on USB depends on USB_ARCH_HAS_HCD + depends on MMU select DRM_GEM_SHMEM_HELPER select DRM_KMS_HELPER help
intel-gfx-ci realized that something is not quite coherent anymore on some platforms for our i915+vgem tests, when I tried to switch vgem over to shmem helpers.
After lots of head-scratching I realized that I've removed calls to drm_clflush. And we need those. To make this a bit cleaner use the same page allocation tooling as ttm, which does internally clflush (and more, as neeeded on any platform instead of just the intel x86 cpus i915 can be combined with).
Unfortunately this doesn't exist on arm, or as a generic feature. For that I think only the dma-api can get at wc memory reliably, so maybe we'd need some kind of GFP_WC flag to do this properly.
v2: Add a TODO comment about what should be done to support this in other places (Thomas)
Acked-by: Thomas Zimmermann tzimmermann@suse.de Signed-off-by: Daniel Vetter daniel.vetter@intel.com Cc: Christian König christian.koenig@amd.com Cc: "Thomas Hellström" thomas.hellstrom@linux.intel.com Cc: Maarten Lankhorst maarten.lankhorst@linux.intel.com Cc: Maxime Ripard mripard@kernel.org Cc: Thomas Zimmermann tzimmermann@suse.de Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch --- drivers/gpu/drm/drm_gem_shmem_helper.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index cc96d1c3570e..0e0986dfbe0c 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -10,6 +10,10 @@ #include <linux/slab.h> #include <linux/vmalloc.h>
+#ifdef CONFIG_X86 +#include <asm/set_memory.h> +#endif + #include <drm/drm.h> #include <drm/drm_device.h> #include <drm/drm_drv.h> @@ -162,6 +166,16 @@ static int drm_gem_shmem_get_pages_locked(struct drm_gem_shmem_object *shmem) return PTR_ERR(pages); }
+ /* + * TODO: Allocating WC pages which are correctly flushed is only + * supported on x86. Ideal solution would be a GFP_WC flag, which also + * ttm_pool.c could use. + */ +#ifdef CONFIG_X86 + if (shmem->map_wc) + set_pages_array_wc(pages, obj->size >> PAGE_SHIFT); +#endif + shmem->pages = pages;
return 0; @@ -203,6 +217,11 @@ static void drm_gem_shmem_put_pages_locked(struct drm_gem_shmem_object *shmem) if (--shmem->pages_use_count > 0) return;
+#ifdef CONFIG_X86 + if (shmem->map_wc) + set_pages_array_wb(shmem->pages, obj->size >> PAGE_SHIFT); +#endif + drm_gem_put_pages(obj, shmem->pages, shmem->pages_mark_dirty_on_put, shmem->pages_mark_accessed_on_put);
Aside from deleting lots of code the real motivation here is to switch the mmap over to VM_PFNMAP, to be more consistent with what real gpu drivers do. They're all VM_PFNMP, which means get_user_pages doesn't work, and even if you try and there's a struct page behind that, touching it and mucking around with its refcount can upset drivers real bad.
v2: Review from Thomas: - sort #include - drop more dead code that I didn't spot somehow
v3: select DRM_GEM_SHMEM_HELPER to make it build (intel-gfx-ci)
v4: I got tricked by 0cf2ef46c6c0 ("drm/shmem-helper: Use cached mappings by default"), and we need WC in vgem because vgem doesn't have explicit begin/end cpu access ioctls.
Also add a comment why exactly vgem has to use wc.
v5: Don't set obj->base.funcs, it will default to drm_gem_shmem_funcs (Thomas)
v6: vgem also needs an MMU for remapping
v7: I absolutely butchered the rebases over the vgem mmap change and revert and broke the patch. Actually go back to v6 from before the vgem mmap changes.
Cc: Thomas Zimmermann tzimmermann@suse.de Acked-by: Thomas Zimmermann tzimmermann@suse.de Cc: John Stultz john.stultz@linaro.org Cc: Sumit Semwal sumit.semwal@linaro.org Cc: "Christian König" christian.koenig@amd.com Signed-off-by: Daniel Vetter daniel.vetter@intel.com Cc: Melissa Wen melissa.srw@gmail.com Cc: Chris Wilson chris@chris-wilson.co.uk --- drivers/gpu/drm/Kconfig | 5 +- drivers/gpu/drm/vgem/vgem_drv.c | 342 ++------------------------------ 2 files changed, 16 insertions(+), 331 deletions(-)
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 314eefa39892..28f7d2006e8b 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -272,7 +272,8 @@ source "drivers/gpu/drm/kmb/Kconfig"
config DRM_VGEM tristate "Virtual GEM provider" - depends on DRM + depends on DRM && MMU + select DRM_GEM_SHMEM_HELPER help Choose this option to get a virtual graphics memory manager, as used by Mesa's software renderer for enhanced performance. @@ -280,7 +281,7 @@ config DRM_VGEM
config DRM_VKMS tristate "Virtual KMS (EXPERIMENTAL)" - depends on DRM + depends on DRM && MMU select DRM_KMS_HELPER select DRM_GEM_SHMEM_HELPER select CRC32 diff --git a/drivers/gpu/drm/vgem/vgem_drv.c b/drivers/gpu/drm/vgem/vgem_drv.c index bf38a7e319d1..a87eafa89e9f 100644 --- a/drivers/gpu/drm/vgem/vgem_drv.c +++ b/drivers/gpu/drm/vgem/vgem_drv.c @@ -38,6 +38,7 @@
#include <drm/drm_drv.h> #include <drm/drm_file.h> +#include <drm/drm_gem_shmem_helper.h> #include <drm/drm_ioctl.h> #include <drm/drm_managed.h> #include <drm/drm_prime.h> @@ -50,87 +51,11 @@ #define DRIVER_MAJOR 1 #define DRIVER_MINOR 0
-static const struct drm_gem_object_funcs vgem_gem_object_funcs; - static struct vgem_device { struct drm_device drm; struct platform_device *platform; } *vgem_device;
-static void vgem_gem_free_object(struct drm_gem_object *obj) -{ - struct drm_vgem_gem_object *vgem_obj = to_vgem_bo(obj); - - kvfree(vgem_obj->pages); - mutex_destroy(&vgem_obj->pages_lock); - - if (obj->import_attach) - drm_prime_gem_destroy(obj, vgem_obj->table); - - drm_gem_object_release(obj); - kfree(vgem_obj); -} - -static vm_fault_t vgem_gem_fault(struct vm_fault *vmf) -{ - struct vm_area_struct *vma = vmf->vma; - struct drm_vgem_gem_object *obj = vma->vm_private_data; - /* We don't use vmf->pgoff since that has the fake offset */ - unsigned long vaddr = vmf->address; - vm_fault_t ret = VM_FAULT_SIGBUS; - loff_t num_pages; - pgoff_t page_offset; - page_offset = (vaddr - vma->vm_start) >> PAGE_SHIFT; - - num_pages = DIV_ROUND_UP(obj->base.size, PAGE_SIZE); - - if (page_offset >= num_pages) - return VM_FAULT_SIGBUS; - - mutex_lock(&obj->pages_lock); - if (obj->pages) { - get_page(obj->pages[page_offset]); - vmf->page = obj->pages[page_offset]; - ret = 0; - } - mutex_unlock(&obj->pages_lock); - if (ret) { - struct page *page; - - page = shmem_read_mapping_page( - file_inode(obj->base.filp)->i_mapping, - page_offset); - if (!IS_ERR(page)) { - vmf->page = page; - ret = 0; - } else switch (PTR_ERR(page)) { - case -ENOSPC: - case -ENOMEM: - ret = VM_FAULT_OOM; - break; - case -EBUSY: - ret = VM_FAULT_RETRY; - break; - case -EFAULT: - case -EINVAL: - ret = VM_FAULT_SIGBUS; - break; - default: - WARN_ON(PTR_ERR(page)); - ret = VM_FAULT_SIGBUS; - break; - } - - } - return ret; -} - -static const struct vm_operations_struct vgem_gem_vm_ops = { - .fault = vgem_gem_fault, - .open = drm_gem_vm_open, - .close = drm_gem_vm_close, -}; - static int vgem_open(struct drm_device *dev, struct drm_file *file) { struct vgem_file *vfile; @@ -159,266 +84,30 @@ static void vgem_postclose(struct drm_device *dev, struct drm_file *file) kfree(vfile); }
-static struct drm_vgem_gem_object *__vgem_gem_create(struct drm_device *dev, - unsigned long size) -{ - struct drm_vgem_gem_object *obj; - int ret; - - obj = kzalloc(sizeof(*obj), GFP_KERNEL); - if (!obj) - return ERR_PTR(-ENOMEM); - - obj->base.funcs = &vgem_gem_object_funcs; - - ret = drm_gem_object_init(dev, &obj->base, roundup(size, PAGE_SIZE)); - if (ret) { - kfree(obj); - return ERR_PTR(ret); - } - - mutex_init(&obj->pages_lock); - - return obj; -} - -static void __vgem_gem_destroy(struct drm_vgem_gem_object *obj) -{ - drm_gem_object_release(&obj->base); - kfree(obj); -} - -static struct drm_gem_object *vgem_gem_create(struct drm_device *dev, - struct drm_file *file, - unsigned int *handle, - unsigned long size) -{ - struct drm_vgem_gem_object *obj; - int ret; - - obj = __vgem_gem_create(dev, size); - if (IS_ERR(obj)) - return ERR_CAST(obj); - - ret = drm_gem_handle_create(file, &obj->base, handle); - if (ret) { - drm_gem_object_put(&obj->base); - return ERR_PTR(ret); - } - - return &obj->base; -} - -static int vgem_gem_dumb_create(struct drm_file *file, struct drm_device *dev, - struct drm_mode_create_dumb *args) -{ - struct drm_gem_object *gem_object; - u64 pitch, size; - - pitch = args->width * DIV_ROUND_UP(args->bpp, 8); - size = args->height * pitch; - if (size == 0) - return -EINVAL; - - gem_object = vgem_gem_create(dev, file, &args->handle, size); - if (IS_ERR(gem_object)) - return PTR_ERR(gem_object); - - args->size = gem_object->size; - args->pitch = pitch; - - drm_gem_object_put(gem_object); - - DRM_DEBUG("Created object of size %llu\n", args->size); - - return 0; -} - static struct drm_ioctl_desc vgem_ioctls[] = { DRM_IOCTL_DEF_DRV(VGEM_FENCE_ATTACH, vgem_fence_attach_ioctl, DRM_RENDER_ALLOW), DRM_IOCTL_DEF_DRV(VGEM_FENCE_SIGNAL, vgem_fence_signal_ioctl, DRM_RENDER_ALLOW), };
-static int vgem_mmap(struct file *filp, struct vm_area_struct *vma) -{ - unsigned long flags = vma->vm_flags; - int ret; - - ret = drm_gem_mmap(filp, vma); - if (ret) - return ret; - - /* Keep the WC mmaping set by drm_gem_mmap() but our pages - * are ordinary and not special. - */ - vma->vm_flags = flags | VM_DONTEXPAND | VM_DONTDUMP; - return 0; -} +DEFINE_DRM_GEM_FOPS(vgem_driver_fops);
-static const struct file_operations vgem_driver_fops = { - .owner = THIS_MODULE, - .open = drm_open, - .mmap = vgem_mmap, - .poll = drm_poll, - .read = drm_read, - .unlocked_ioctl = drm_ioctl, - .compat_ioctl = drm_compat_ioctl, - .release = drm_release, -}; - -static struct page **vgem_pin_pages(struct drm_vgem_gem_object *bo) -{ - mutex_lock(&bo->pages_lock); - if (bo->pages_pin_count++ == 0) { - struct page **pages; - - pages = drm_gem_get_pages(&bo->base); - if (IS_ERR(pages)) { - bo->pages_pin_count--; - mutex_unlock(&bo->pages_lock); - return pages; - } - - bo->pages = pages; - } - mutex_unlock(&bo->pages_lock); - - return bo->pages; -} - -static void vgem_unpin_pages(struct drm_vgem_gem_object *bo) -{ - mutex_lock(&bo->pages_lock); - if (--bo->pages_pin_count == 0) { - drm_gem_put_pages(&bo->base, bo->pages, true, true); - bo->pages = NULL; - } - mutex_unlock(&bo->pages_lock); -} - -static int vgem_prime_pin(struct drm_gem_object *obj) +static struct drm_gem_object *vgem_gem_create_object(struct drm_device *dev, size_t size) { - struct drm_vgem_gem_object *bo = to_vgem_bo(obj); - long n_pages = obj->size >> PAGE_SHIFT; - struct page **pages; + struct drm_gem_shmem_object *obj;
- pages = vgem_pin_pages(bo); - if (IS_ERR(pages)) - return PTR_ERR(pages); + obj = kzalloc(sizeof(*obj), GFP_KERNEL); + if (!obj) + return NULL;
- /* Flush the object from the CPU cache so that importers can rely - * on coherent indirect access via the exported dma-address. + /* + * vgem doesn't have any begin/end cpu access ioctls, therefore must use + * coherent memory or dma-buf sharing just wont work. */ - drm_clflush_pages(pages, n_pages); - - return 0; -} - -static void vgem_prime_unpin(struct drm_gem_object *obj) -{ - struct drm_vgem_gem_object *bo = to_vgem_bo(obj); - - vgem_unpin_pages(bo); -} - -static struct sg_table *vgem_prime_get_sg_table(struct drm_gem_object *obj) -{ - struct drm_vgem_gem_object *bo = to_vgem_bo(obj); - - return drm_prime_pages_to_sg(obj->dev, bo->pages, bo->base.size >> PAGE_SHIFT); -} - -static struct drm_gem_object* vgem_prime_import(struct drm_device *dev, - struct dma_buf *dma_buf) -{ - struct vgem_device *vgem = container_of(dev, typeof(*vgem), drm); - - return drm_gem_prime_import_dev(dev, dma_buf, &vgem->platform->dev); -} - -static struct drm_gem_object *vgem_prime_import_sg_table(struct drm_device *dev, - struct dma_buf_attachment *attach, struct sg_table *sg) -{ - struct drm_vgem_gem_object *obj; - int npages; - - obj = __vgem_gem_create(dev, attach->dmabuf->size); - if (IS_ERR(obj)) - return ERR_CAST(obj); + obj->map_wc = true;
- npages = PAGE_ALIGN(attach->dmabuf->size) / PAGE_SIZE; - - obj->table = sg; - obj->pages = kvmalloc_array(npages, sizeof(struct page *), GFP_KERNEL); - if (!obj->pages) { - __vgem_gem_destroy(obj); - return ERR_PTR(-ENOMEM); - } - - obj->pages_pin_count++; /* perma-pinned */ - drm_prime_sg_to_page_array(obj->table, obj->pages, npages); return &obj->base; }
-static int vgem_prime_vmap(struct drm_gem_object *obj, struct dma_buf_map *map) -{ - struct drm_vgem_gem_object *bo = to_vgem_bo(obj); - long n_pages = obj->size >> PAGE_SHIFT; - struct page **pages; - void *vaddr; - - pages = vgem_pin_pages(bo); - if (IS_ERR(pages)) - return PTR_ERR(pages); - - vaddr = vmap(pages, n_pages, 0, pgprot_writecombine(PAGE_KERNEL)); - if (!vaddr) - return -ENOMEM; - dma_buf_map_set_vaddr(map, vaddr); - - return 0; -} - -static void vgem_prime_vunmap(struct drm_gem_object *obj, struct dma_buf_map *map) -{ - struct drm_vgem_gem_object *bo = to_vgem_bo(obj); - - vunmap(map->vaddr); - vgem_unpin_pages(bo); -} - -static int vgem_prime_mmap(struct drm_gem_object *obj, - struct vm_area_struct *vma) -{ - int ret; - - if (obj->size < vma->vm_end - vma->vm_start) - return -EINVAL; - - if (!obj->filp) - return -ENODEV; - - ret = call_mmap(obj->filp, vma); - if (ret) - return ret; - - vma_set_file(vma, obj->filp); - vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP; - vma->vm_page_prot = pgprot_writecombine(vm_get_page_prot(vma->vm_flags)); - - return 0; -} - -static const struct drm_gem_object_funcs vgem_gem_object_funcs = { - .free = vgem_gem_free_object, - .pin = vgem_prime_pin, - .unpin = vgem_prime_unpin, - .get_sg_table = vgem_prime_get_sg_table, - .vmap = vgem_prime_vmap, - .vunmap = vgem_prime_vunmap, - .vm_ops = &vgem_gem_vm_ops, -}; - static const struct drm_driver vgem_driver = { .driver_features = DRIVER_GEM | DRIVER_RENDER, .open = vgem_open, @@ -427,13 +116,8 @@ static const struct drm_driver vgem_driver = { .num_ioctls = ARRAY_SIZE(vgem_ioctls), .fops = &vgem_driver_fops,
- .dumb_create = vgem_gem_dumb_create, - - .prime_handle_to_fd = drm_gem_prime_handle_to_fd, - .prime_fd_to_handle = drm_gem_prime_fd_to_handle, - .gem_prime_import = vgem_prime_import, - .gem_prime_import_sg_table = vgem_prime_import_sg_table, - .gem_prime_mmap = vgem_prime_mmap, + DRM_GEM_SHMEM_DRIVER_OPS, + .gem_create_object = vgem_gem_create_object,
.name = DRIVER_NAME, .desc = DRIVER_DESC,
Hi Daniel,
On Thu, Aug 12, 2021 at 03:14:12PM +0200, Daniel Vetter wrote:
Aside from deleting lots of code the real motivation here is to switch the mmap over to VM_PFNMAP, to be more consistent with what real gpu drivers do. They're all VM_PFNMP, which means get_user_pages doesn't work, and even if you try and there's a struct page behind that, touching it and mucking around with its refcount can upset drivers real bad.
The only thing I understood of all this complicated stuff was "deleting lots of code" which is a good thing. You may want to s/VM_PFNMP/VM_PFNMAP/ before you push this.
Sam
On Thu, Aug 12, 2021 at 07:01:44PM +0200, Sam Ravnborg wrote:
Hi Daniel,
On Thu, Aug 12, 2021 at 03:14:12PM +0200, Daniel Vetter wrote:
Aside from deleting lots of code the real motivation here is to switch the mmap over to VM_PFNMAP, to be more consistent with what real gpu drivers do. They're all VM_PFNMP, which means get_user_pages doesn't work, and even if you try and there's a struct page behind that, touching it and mucking around with its refcount can upset drivers real bad.
The only thing I understood of all this complicated stuff was "deleting lots of code" which is a good thing. You may want to s/VM_PFNMP/VM_PFNMAP/ before you push this.
Fixed and patches 2-4 from this series pushed. -Daniel
dri-devel@lists.freedesktop.org
-
Daniel Vetter -
Daniel Vetter -
Sam Ravnborg