summaryrefslogtreecommitdiff
path: root/Documentation/gpu
diff options
context:
space:
mode:
authorDaniel Vetter <daniel.vetter@ffwll.ch>2019-01-07 11:22:38 +0100
committerDaniel Vetter <daniel.vetter@ffwll.ch>2019-01-08 09:03:52 +0100
commitbe5cadc7e7b4cee83cdd19e58a9dc0eca9e23b69 (patch)
treed60ffab63408ca59e50fcb61fde192cfd1ab9085 /Documentation/gpu
parent1c95f662fceeb8ae2f34e3de9478e21fd31f09dd (diff)
drm/todo: Better defio support in the generic fbdev emulation
The current one essentially means you need CMA or a vmalloc backed object, which makes fbdev emulation a special case. Since implementing this will be quite a bit of work, capture the idea in a TODO. Cc: Noralf Trønnes <noralf@tronnes.org> Acked-by: Noralf Trønnes <noralf@tronnes.org> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190107102238.7789-1-daniel.vetter@ffwll.ch
Diffstat (limited to 'Documentation/gpu')
-rw-r--r--Documentation/gpu/todo.rst30
1 files changed, 30 insertions, 0 deletions
diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst
index 41da7b06195c..0a85dad876ae 100644
--- a/Documentation/gpu/todo.rst
+++ b/Documentation/gpu/todo.rst
@@ -209,6 +209,36 @@ Would be great to refactor this all into a set of small common helpers.
Contact: Daniel Vetter
+Generic fbdev defio support
+---------------------------
+
+The defio support code in the fbdev core has some very specific requirements,
+which means drivers need to have a special framebuffer for fbdev. Which prevents
+us from using the generic fbdev emulation code everywhere. The main issue is
+that it uses some fields in struct page itself, which breaks shmem gem objects
+(and other things).
+
+Possible solution would be to write our own defio mmap code in the drm fbdev
+emulation. It would need to fully wrap the existing mmap ops, forwarding
+everything after it has done the write-protect/mkwrite trickery:
+
+- In the drm_fbdev_fb_mmap helper, if we need defio, change the
+ default page prots to write-protected with something like this::
+
+ vma->vm_page_prot = pgprot_wrprotect(vma->vm_page_prot);
+
+- Set the mkwrite and fsync callbacks with similar implementions to the core
+ fbdev defio stuff. These should all work on plain ptes, they don't actually
+ require a struct page. uff. These should all work on plain ptes, they don't
+ actually require a struct page.
+
+- Track the dirty pages in a separate structure (bitfield with one bit per page
+ should work) to avoid clobbering struct page.
+
+Might be good to also have some igt testcases for this.
+
+Contact: Daniel Vetter, Noralf Tronnes
+
Put a reservation_object into drm_gem_object
--------------------------------------------