AuroraMiddleware/gtk
1
0
Fork 1
mirror of https://gitlab.gnome.org/GNOME/gtk.git synced 2024-12-28 06:21:14 +00:00
Code Issues Projects Releases Wiki Activity
78,115 Commits 1,241 Branches 809 Tags 591 MiB
cd7e1c1387
Commit Graph

11346 Commits

Author SHA1 Message Date
Benjamin Otte
cd7e1c1387 dmabuf: Modify gdk_gl_context_import_dmabuf() 
Move a bunch of code - namely the target selection - into the function.

I want to use the same code in the NGL renderer, so no need to duplicate
it.
 2023-11-07 05:12:57 +01:00
Benjamin Otte
61207f81c3 dmabuf: Set same YUV colorspace as elsewhere 
We use ITU.601 full range everywhere, so ensure it's used here, too.
 2023-11-07 05:12:57 +01:00
Matthias Clasen
1914adc344 dmabuf: Don't check for image storage 
I was wrong about this extension, and it isn't even a GL extension
to begin with, so revert all this.
 2023-10-29 14:45:20 -04:00
Matthias Clasen
6e236b9459 Revert "gl context: Check for more GL extensions" 
This reverts commit b8b5835fc6.
 2023-10-29 14:44:46 -04:00
Matthias Clasen
11715760f9 Revert "glcontext: Better debug spew" 
This reverts commit 30e14f73fa.
 2023-10-29 14:44:29 -04:00
Matthias Clasen
06dda4e12b Merge branch 'memory-format-swizzle-check' into 'main' 
gl: Check all swizzle values

See merge request GNOME/gtk!6524
 2023-10-26 02:38:35 +00:00
Matthias Clasen
4a749aee6e Fix swizzle values for some memory formats 
For opaque formats with 3 channels, we should use the default
GL_ALPHA, but for opaque formats with an ignored 4th channel,
we must use GL_ONE.
 2023-10-25 21:39:52 -04:00
Matthias Clasen
bb2b8bf2c3 dmabuf: Refine the handling of implicit modifiers 
If all formats with a given fourcc are external, treat the
fourcc with implicit modifiers as external too, following a
sugestion by Benjamin Otte.
 2023-10-24 22:33:45 -04:00
Matthias Clasen
eb048c91be dmabuf: Allow implicit modifiers 
Remove all the roadblocks we've put up to keep implicit modifiers
out. Our importing code already handles them as a signal that says
'No modifiers, please!'. Now we just hope for the best and pass
things along.

This is necessary since some drivers won't produce any explicit
modifiers.
 2023-10-24 22:33:45 -04:00
Matthias Clasen
8f90ddb906 dmabuf: Cache egl formats 
Only call eglQueryDmaBufModifiersEXT once, and reuse the results.
 2023-10-24 21:53:09 -04:00
Matthias Clasen
a90ffb117f docs: Some additions 
Add a link to the kernel docs about dmabufs.
 2023-10-24 10:50:46 -04:00
Matthias Clasen
29681717ed Fix annotations 
It is always org.gtk.Property, even if we are in gdk.
 2023-10-24 07:16:42 -04:00
Benjamin Otte
51a181af7a Merge branch 'wip/otte/for-main' into 'main' 
dmabuf: Seek back to the start

See merge request GNOME/gtk!6526
 2023-10-23 19:57:21 +00:00
Benjamin Otte
e2edf693f5 gdk: Make float32 report its true depth 
I have no idea how this is the only value that is wrong.
 2023-10-23 21:22:17 +02:00
Benjamin Otte
323300e92a dmabuf: Seek back to the start 
https://docs.kernel.org/driver-api/dma-buf.html recommends doing that,
so we do that.
 2023-10-23 17:57:57 +02:00
Benjamin Otte
e688be1cff gdk: Add some debug messages around dmabuf initialization 2023-10-23 17:57:57 +02:00
Matthias Clasen
625eb508e8 dmabuftexturebuilder: Fix doc formatting 2023-10-22 23:24:46 -04:00
Matthias Clasen
2718e1715d display: Add details to the docs 2023-10-22 23:24:27 -04:00
Matthias Clasen
6ddcad6b33 dmabuf: Update a comment 2023-10-22 23:14:07 -04:00
Matthias Clasen
30e14f73fa glcontext: Better debug spew 
Print out all the extensions we checked.
 2023-10-22 22:41:53 -04:00
Matthias Clasen
39e205aa78 dmabuf: Don't print unsupported formats 
We were confusingly printing "supported format" for dmabuf formats
that we end up not adding to our list of supported formats. Don't
do that, it is confusing. At the same time, we shuold print out
the linear formats we support via mmap.
 2023-10-22 22:29:44 -04:00
Matthias Clasen
79f98f3a13 Cosmetics 
Typo fix.
 2023-10-22 22:18:30 -04:00
Benjamin Otte
a579e3bc6d build: Fix up header includes 
These weren't fixed in the merge request in
  https://gitlab.gnome.org/GNOME/gtk/-/merge_requests/6489
to the change from
  https://gitlab.gnome.org/GNOME/gtk/-/merge_requests/6514
 2023-10-23 03:12:09 +02:00
Benjamin Otte
8f70f4b85e dmabuf: Add support for 3 plane YUV formats 2023-10-23 02:54:23 +02:00
Benjamin Otte
911198cddd dmabuf: Add NV16, NV61, NV24 and NV42 
and sanitize them and NV21.
 2023-10-23 02:54:23 +02:00
Matthias Clasen
a4d0a5bda5 dmabuf: We don't support external formats on GL 
Filter out external-only formats unless we are on GLES.
 2023-10-22 17:54:18 -04:00
Matthias Clasen
494154beb6 dmabuf: Add an EGL downloader 
Add an implementation of GdkDmabufDownloader that uses
gsk_renderer_render_texture + GL texture download.

Since gsk isn't threadsafe, we do the download in the main thread,
taking care to not disturb the current GL context of whatever is
going on there at the time.

And since gsk renderers are expensive to create, we cache it
in the display.

Note that gsk does not yet have any special support for
dmabuf textures, so for now, they will always get downloaded
and then reuploaded as GL textures.
 2023-10-22 16:42:17 -04:00
Matthias Clasen
8fb0ab2b43 gl context: Add private dmabuf API 
Add API to import a dmabuf into a texture,
and to export a texture to a dmabuf.

All this is implemented using the relevant
EGL and GL extensions.
 2023-10-22 16:39:24 -04:00
Matthias Clasen
b8b5835fc6 gl context: Check for more GL extensions 
These will be used in future commits.
 2023-10-22 16:39:24 -04:00
Matthias Clasen
463307655c display: Check for more EGL extensions 
We are going to rely on this in future commits.
 2023-10-22 16:39:24 -04:00
Benjamin Otte
e05764806a build: Change where we get drm_fourcc.h from 
The canonical source for userspace is libdrm, not kernel headers.

Fixes compilation on Arch Linux.
 2023-10-22 20:40:20 +02:00
Benjamin Otte
32ba1e389c build: Check for drm/drm_fourcc.h too 
It's part of the kernel headers package so I did not check for libdrm.

As a side effect I checked the #define to HAVE_DMABUF.
 2023-10-22 17:27:11 +02:00
Benjamin Otte
beb9e0c906 dmabuf: Add gdk_dmabuf_is_disjoint() 
No users yet, but Vulkan will use this later. So have it available
already.
 2023-10-22 17:03:15 +02:00
Benjamin Otte
98f02f6221 dmabuf: Reject the INVALID modifier 
Trying to use it is a programming error, applications should have code
that uses real modifiers.

Also add a check to the formatsbuilder so our code doesn't include the
invalid modifier by accident.

We don't really know how to deal with it, so better force applications
to figure out what to do.
 2023-10-22 17:03:15 +02:00
Benjamin Otte
13848969f4 dmabuf: Allow downloaders to fail initialization 
When adding the formats of a downloader, allow them to return FALSE to
mean "This method is not supported", which is a useful way to opt out
when checking GL or Vulkan extensions and finding out that the desired
one isn't supported.
 2023-10-22 17:03:13 +02:00
Matthias Clasen
9e3f537a15 dmabuf: Give downloaders a name 
Add name to the downloader struct, and add it
to the debug spew for newly created dmabuf
textures.
 2023-10-21 19:19:21 -04:00
Matthias Clasen
410240fe7b Merge branch 'matthiasc/for-main' into 'main' 
Cosmetics

See merge request GNOME/gtk!6509
 2023-10-21 20:34:36 +00:00
Matthias Clasen
73cf3a92f9 cosmetics 2023-10-21 16:10:09 -04:00
Matthias Clasen
dd15122ccb dmabuf: Add some debug spew 2023-10-21 14:44:28 -04:00
Benjamin Otte
cbfd8542c6 dmabuf: Require valid fds for all planes 
This seems to be what everyone does, so we should do it, too.

Previously it was assumed that an fd of -1 would mean reusing the
previous fd with a different offset, but that seems to be uncommon.
 2023-10-21 16:42:15 +02:00
Matthias Clasen
59ce7ab222 dmabuf: Improve the debug message 
Printing the format can only help.
 2023-10-20 20:03:27 -04:00
Matthias Clasen
0992837b08 dmabuf: Fix the mmap logic 
We want to unmap the buffers we mapped in
the first loop, so pass the right pointers
here.
 2023-10-20 20:03:11 -04:00
Matthias Clasen
aca02916ad dmabuf: Fix NV12 import 
We were not seeking in the right fd,
so our sizes were wrong, and crashes
happened.
 2023-10-20 19:53:33 -04:00
Matthias Clasen
8843c0d504 dmabuf: Fix format mapping for RGB formats 
The channel order was reversed.
 2023-10-20 19:22:00 -04:00
Matthias Clasen
a1c5c20e2c dmabuf: Add xrgb formats 2023-10-20 18:14:45 -04:00
Matthias Clasen
10250a0a6a Add XRGB memory formats 
These are not uncommon in dmabufs,
and supporting them is easy.

Fixes: #6172
 2023-10-20 17:58:06 -04:00
Matthias Clasen
bd974b08b3 Add GDK_MEMORY_A8B8G8R8_PREMULTIPLIED 
We did have 4 ordering variations of ARGB straight,
but only 3 premultiplied. Add the missing one.

Update all the places where we switch over memory formats.
 2023-10-20 16:42:40 -04:00
Benjamin Otte
008d9e5327 vulkan: Actually check for GDK_DEBUG=vulkan-disable 
The dmabuf code would happily init Vulkan even if it was told we don't
want it.
 2023-10-20 14:26:59 +02:00
Benjamin Otte
09723d79c6 dmabuf: Support YUYV and friends 
Same as the NV12 support, but necessary for the poor people stuck on
Intel laptops.
 2023-10-20 14:26:59 +02:00
Benjamin Otte
4586af5876 dmabuf: Add support for NV12 and NV21 
This is as-good-as-necessary and doesn't do any fancy colorspace magic
or chroma whatevering.

That's a task for the future.
 2023-10-20 14:26:59 +02:00