Handle has some limits and can't be used as unique buffer ID on systems
where display controller can scanout from CMA but GPU can work with both
CMA and VRAM.
Such systems have DRM/KMS and DRM/GPU drivers separated.
GBM frontend is always expecting handle for DRM/KMS driver.
In such system any attempt of importing the buffer with more
than 1 contiguous chunk into DRM/KMS driver will fail.
Using dma-buf inode as unique buffer ID is a common practice for
a last several years starting from [this kernel patch][1].
[1]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ed63bb1d1f8469586006a9ca63c42344401aa2ab
Change-Id: Ic3a69010d5da2f866a2252fc7e9eb29d67f8e1ed
Signed-off-by: Roman Stratiienko <r.stratiienko@gmail.com>
Some drivers may copy/convert the buffer during mapping and
in some cases stride of copied image can be different from
original. Android uses pixel_stride for CPU access and need
map_time stride instead of original stride in this cases.
Signed-off-by: Roman Stratiienko <r.stratiienko@gmail.com>
... to avoid compile-time error on C++:
error: ISO C++11 does not allow conversion from string
literal to 'char *' [-Werror,-Wwritable-strings]
Signed-off-by: Roman Stratiienko <r.stratiienko@gmail.com>
Change-Id: I176eb657f72e92d6b5c7c3b25c78c56f776c20ab
Crosvm currently makes the simplistic assumption that if driver==i915
then cached mappings to guest should be used for minigbm allocated
buffers. But this may not always be the correct choice. And other
drivers can do cached mappings to, in certain cases. But on ARM devices
in particular, we should be consistent in cachability when it comes to
CPU mappings. So add a new minigbm API which crosvm can use to
determine how to map to guest.
BUG=b:239718180, b:306548532
TEST=no artifact in Camera FOV Calibration of CtsVerifier on rex
Change-Id: I5c9b6346270f6a2eb83e6637a911f2153f6120ff
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3777567
Tested-by: Chia-I Wu <olv@google.com>
Auto-Submit: Rob Clark <robdclark@chromium.org>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Commit-Queue: Chia-I Wu <olv@google.com>
When MINIGBM_DEBUG=log_bos is set, log all bos created/imported.
BUG=none
TEST=add MINIGBM_DEBUG=log_bos to /etc/chrome_dev.conf and
see logs in /var/log/ui/ui.LATEST
Change-Id: Ib2b2d5c3181862b557dbc5499fef84f2c9ac8122
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/5014198
Commit-Queue: ChromeOS Auto Retry <chromeos-auto-retry@chromeos-bot.iam.gserviceaccount.com>
Reviewed-by: Dawn Han <dawnhan@google.com>
Tested-by: Chia-I Wu <olv@google.com>
The multi-planar formats are being allocated into a single plane,
and the minigbm exynos backend is dropped. So we can use `handle` instead of handles as of now.
Remove some of the helper functions that calculate the number of planes.
There is no need to check if there is more than one kernel buffer per buffer object.
Added a new function `drv_gem_close` in `drv_helpers` that can be used by `drv_gem_bo_destroy` and `drv_prime_bo_import`. It can be used to clean the unused gem_handle.
Bug=b:266776512
TEST=camera and screenshot work fine after deploying the change #strongbad
TEST=camera and screenshot work fine after deploying the change #kukui-arc-r
TEST=camera and screenshot work fine after deploying the change, tested a couple of apps in the Google Play Store, and checked file `dma_buf/bufinfo` to see there's no object leaking. #guybrush
Change-Id: Ib63559504e4cb8f1a32ae90170925146613d694f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/4297873
Commit-Queue: Dawn Han <dawnhan@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Tested-by: Dawn Han <dawnhan@google.com>
GFX10+ cannot support DRM_FORMAT_YVU420_ANDROID when the strides do not
meet hw requirement. When that happens, force the format to
DRM_FORMAT_YVU420 and log an error.
When the bo is used as a classic resource by virglrenderer, this is
hopefully fine because the guest only sees the guest storage (unless
there are bugs elsewhere). Otherwise, the logs can lead us to proper
fixes.
BUG=b:265746435
TEST=subset of CTS and CTS Verifier on grunt, guybrush and skyrim
Change-Id: I0fda9e2f09bec8c26e87748b9d58c4f8e186e34d
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/4336266
Tested-by: Chia-I Wu <olv@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Ryan Neph <ryanneph@google.com>
Commit-Queue: Chia-I Wu <olv@google.com>
Remove the parameter to align with upstream gbm.
For some places that use the number of planes. The number is 1. The
multi-planar formats are being allocated into a single plane, and
minigbm exynos backend is dropped. So we can just set the plane
index to 0 if needed.
Fixed the format in i915.c.
Bug=b:266776512
TEST=camera works after deploying the change #strongbad
TEST=camera works after deploying the change #corsola
Change-Id: I0880917754c01b9d0f27d21f3c42d87f00a09f50
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/4249501
Commit-Queue: Dawn Han <dawnhan@google.com>
Tested-by: Dawn Han <dawnhan@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olv@google.com>
It can be used to preload a backend, currently only used to preload
amdgpu.
BUG=b:269664560
TEST="grep radeonsi /proc/$(pidof zygote)/maps" on grunt
Change-Id: I30d9f701f1da3bc243a9c81326a1ba0c5e0562de
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/4262034
Commit-Queue: Chia-I Wu <olv@google.com>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Tested-by: Chia-I Wu <olv@google.com>
Mesa's llvmpipe software renderer generates instructions that require
memory to be aligned properly, otherwise it will crash with general
protection exception. To support crosvm on headless VM with vkms we'd
like to always align memory properly on dumb driver.
BUG=b:239110721
TEST=Run betty-arc-r on headless VM with vkms enabled
Change-Id: I4b119f450cbb9b36d0d35d37be9ca97102924963
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/4134929
Tested-by: Yi Xie <yixie@chromium.org>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Commit-Queue: Yi Xie <yixie@chromium.org>
No need to introduce host storage and mapping to the guest for buffers
that only have CPU access. Fixes slow performance of sw encoders.
BUG=b:238158981, b:239333100
TEST=cts-tradefed run cts -m CtsVideoTestCases -t android.video.cts.VideoEncoderDecoderTest#testAvcGoog0Perf0320x0240
TEST=cts-tradefed run cts -m CtsCameraTestCases -t android.hardware.camera2.cts.RecordingTest#testSupportedVideoSizes[1]
Change-Id: Iaf88a750ae215d3a02b8415ab93be06b08a7f550
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3777563
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Auto-Submit: Rob Clark <robdclark@chromium.org>
Tested-by: Rob Clark <robdclark@chromium.org>
Commit-Queue: Yiwei Zhang <zzyiwei@chromium.org>
Commit-Queue: Rob Clark <robdclark@chromium.org>
The current bo_destroy callback only gets called if the bo has the
last references to the GEM BOs, and otherwise plain free gets
called.
However, this is an issue if bo->priv contains something per bo that
needs to be cleaned up. To solve this we introduce a new callback
to clean up things per bo instance.
BUG=b:185869479
TEST=none for this patch. See follow-on patch making use of this.
Change-Id: I9d48b3b5a70264adbc4de55a5c7b18e1a2209553
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3270683
Tested-by: Bas Nieuwenhuizen <basni@chromium.org>
Reviewed-by: Yiwei Zhang <zzyiwei@chromium.org>
Commit-Queue: Bas Nieuwenhuizen <basni@chromium.org>
It makes more sense to resovle the format and use_flags together to
avoid duplicates.
Add the helper to vc4 and dumb_driver backends since that's missed
from prior frontend refactoring.
Split the resolve helper in virtgpu_virgl for the 3d and 2d paths to
make code logic cleaner.
BUG=b:199524294
TEST=CQ
Change-Id: I89b633ea484f0fc5bb9b4e0548cad017b9970cc6
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3200143
Tested-by: Yiwei Zhang <zzyiwei@chromium.org>
Tested-by: Jason Macnak <natsu@google.com>
Reviewed-by: Chia-I Wu <olv@google.com>
Reviewed-by: Jason Macnak <natsu@google.com>
Reviewed-by: Dominik Behr <dbehr@chromium.org>
Commit-Queue: Yiwei Zhang <zzyiwei@chromium.org>
crosdomain specific metadata_cache should be protected by a backend
specific lock to reduce lock contention.
BUG=b:201110412
TEST=CQ
Change-Id: Ic3cedd7ad7903e669f7fba2b5f45d88739fbdc97
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3188471
Commit-Queue: Yiwei Zhang <zzyiwei@chromium.org>
Tested-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Chia-I Wu <olv@google.com>
For guest running 3d virtgpu_virgl atop 2d virtgpu_virgl as host gbm
backend. Common and required scanout capable formats on the guest side
will fail virgl_supports_combination_natively check, resulting in
advertising no scanout support to the guest.
This change adds the same logic to resolve_use_flags before checking
support or asking for allocation to accommodate accordingly. Then
whether to strip scanout use_flag can align with native support on those
guest side scanout capable formats.
Advertising scanout for NV12 needs to go through virgl_add_combination
as well for the native support check. This is fixed by this CL.
The logic to strip scanout needs to be scanout use_flag specific to
avoid accidentally stripping scanout when the format fails texture check
but later passes the emulation check because scanout has been stripped.
This is also fixed by this CL.
BUG=b:200969382
TEST=CQ and camera interop works
Change-Id: I9126773a1ee49d4cdaf1f7186612d5f4f5c6200e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3180981
Tested-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Lepton Wu <lepton@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Yiwei Zhang <zzyiwei@chromium.org>
Only 2d virtgpu backend needs to fallback here because virtio primary
plane only allows DRM_FORMAT_XRGB8888.
Most our platforms cannot display YV12 (except msm), thus the fallback
is required for the converted DRM_FORMAT_YVU420_ANDROID. For virgl
backend, additionally append a BO_USE_LINEAR as a replacement for the
hack inside compute_virgl_bind_flags.
BUG=b:199524294
TEST=CQ
TEST=gralloctest alloc_combinations
TEST=android.media.cts.VideoDecoderRotationTest
Change-Id: Ic87838ea2aae2b0abf87ed898ad75a3d7e556471
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/3166775
Tested-by: Yiwei Zhang <zzyiwei@chromium.org>
Auto-Submit: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Yiwei Zhang <zzyiwei@chromium.org>
Vulkan on Android relis on VkImageDrmFormatModifierExplicitCreateInfoEXT
to import the gralloc allocated buffer, which requires buffer info of
the host storage.
BUG=b:186679287
TEST=venus wsi on arcvm and egl/gles
TEST=CtsNativeHardwareTestCases and related CtsGraphicsTestCases
Change-Id: I7ab0e80ab67a2daae9ea547a8d5abe9417af17d8
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2862161
Tested-by: Yiwei Zhang <zzyiwei@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Yiwei Zhang <zzyiwei@chromium.org>
When drivers start seeing this flag, we don't want them to reject the
allocation. Just have the option to exclude the flag from (for ex.)
bandwidth compressed formats.
BUG=b:168868719
TEST=None
Change-Id: Iee644d962129c9121925bc19395331a59e32d8a1
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2800148
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Rob Clark <robdclark@chromium.org>
Auto-Submit: Rob Clark <robdclark@chromium.org>
Commit-Queue: Rob Clark <robdclark@chromium.org>
This change provides a route to the host side "cross domain"
allocator, which is specialized for resource sharing across domain
boundaries. It uses the upcoming CONTEXT_INIT ioctl to initialize
the backend when available, and multiple timelines to perform
metadata queries.
While it would be awesome to use the revolutionary Address Space
Graphics (ASG) algorithm for metadata queries, it would be
non-trivial to pull off in minigbm.
Key aspects of the cross-domain allocator are:
* Intelligently falls back to OpenGL texture allocation host-side
when external memory is not available. The fallback path is
named "virgl", even though it encompasses gfxstream and 2D mode.
More refactorings will be added in the future for further
clarity.
* Uses host Vulkan or minigbm to perform metadata query, and uses
a cache to minimize vmexits.
* No shadow memory. Only zero-copy blobs will be supported if the
blob is mappable. Shareable blobs may be compressed or tiled if
not mappable.
* A commitment to sharing code across Google projects and
cross-platform GPU virtualization.
The main goal here is to enable faster interation/testing, so this
code is just a prototype. It should be fine to merge via minigbm,
since the project has a very chill philosophy and even proudly
accepts code that doesn't make any sense (crrev.com/c/2583188)
[so long as existing users are not broken].
BUG=b:173630595
TEST=launch virtual machine with 2D mode
TEST=launch virtual machine with 3D mode
Change-Id: Ie33a46f19e5cdd82a2ac03bcf2351f4a8f294970
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2691716
Reviewed-by: Lingfeng Yang <lfy@google.com>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
The early version of the modifier assumed per plane modifiers.
The current version is only one modifier per buffer object.
BUG=none
TEST=compile
Change-Id: Ic3899118d44cb172ee2e4eae346e98cdf8328cf9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2654590
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
Read MINIGBM_DEBUG env var when creating a gbm driver
Store value as a flag inside driver
Avoid allocating compressed buffers in the backends
when the flag disables compression
BUG=b:172215587
Change-Id: Idbd6f0aebc1782c1bf5921a6438310a87212d1f8
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2535658
Tested-by: Pilar Molina Lopez <pmolinalopez@google.com>
Commit-Queue: Pilar Molina Lopez <pmolinalopez@google.com>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
With real HW protection on the horizon, let's repurpose the
protection flag to mean that.
Currently, our protected buffer scheme on the Android side
allocates a dummy fd, which is sent to Chrome. Chrome
associates that dummy fd with an unmappable shared memory
buffer. In the entire process, minigbm doesn't really do
anything.
We prevent buffers allocated with the protected
flag from being mapped, but since it's a dummy fd it's not
really useful.
Chrome doesn't use the protected flag yet, but hopefully will
so in the future, but with real HW protection.
BUG=
TEST=
Change-Id: I57be26926539471f062ffeff33b523a3899c35f5
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2466958
Reviewed-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Jeffrey Kardatzke <jkardatzke@google.com>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
In ARC++, the wayland sevice and the video stack rely on
GRALLOC_DRM_GET_STRIDE and (*lock_ycbcr) with zero flags to return
the metadata associated with the buffer.
In the past, we've simply returned the metadata that was calculated
during allocation.
Since the current virtio-gpu API relies on shadow buffers, there's
actually two different sets of metadata:
1) The metadata of the shadow buffer --> useful for mapping
2) The metadata of the host resource --> useful for passing to Chrome
For the wayland_service and video stack, we want to return (2).
For the Android framework, we want to return (1).
BUG=b:132939420
TEST=compile
Change-Id: I1134d651396ba68e064eaf2e3cad3cb3225d7c5c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/1681383
Reviewed-by: David Stevens <stevensd@chromium.org>
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
If the driver does not implement the new function, or
if modifier is INVALID this should return exactly the
same as before.
For LINEAR, DRI should return exactly the same as
minigbm for all the YUV formats.
BUG=b:149819940
TEST=Use with followup patch and login on a Zork device + play a YT
video.
Change-Id: I6ea3b5827876844e510794b85212be51e5dfd68f
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2093211
Tested-by: Bas Nieuwenhuizen <basni@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Bas Nieuwenhuizen <basni@chromium.org>
This reverts commit 08d8dbf094.
The original change returned early from drv_bo_create_with_modifiers,
which broke reference counting. This must have hit some race condition
in the tests, as they no longer flake after fixing reference counting.
Original change's description:
> Revert "minigbm: introduce test allocation"
>
> This reverts commit f0e607c7d4.
>
> Reason for revert: caused flaky regressions across the board.
>
> BUG=b:150997559
> Exempt-From-Owner-Approval: revert.
>
> Original change's description:
> > minigbm: introduce test allocation
> >
> > This change introduces a GBM_TEST_ALLOC flag to minigbm, which allows
> > for the creation of fake buffers that can be used to determine buffer
> > metadata without actually allocating a full buffer. The new flag is
> > supported by the i915 backends. This flag also alleviates the need to
> > cache buffers when virtio_gpu queries metadata properties.
> >
> > BUG=b:145994510
> > TEST=play youtube with arcvm demo image plus this and virgl change
> >
> > Change-Id: I9c6819aa3b5b674e4bb33b0656f2a9f155b0884e
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/1980688
> > Tested-by: David Stevens <stevensd@chromium.org>
> > Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
> > Commit-Queue: David Stevens <stevensd@chromium.org>
>
> Bug: b:145994510
> Change-Id: I50079b7f0aabf38e1f373cac0f28c0e057eed760
> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2093923
> Commit-Queue: Ilja H. Friedel <ihf@chromium.org>
> Tested-by: Ilja H. Friedel <ihf@chromium.org>
> Reviewed-by: Ilja H. Friedel <ihf@chromium.org>
Bug: b:150997559, b:145994510
Change-Id: If0f02a4701bb6960b6413d6b0c00b481146914d9
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2094068
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: David Stevens <stevensd@chromium.org>
Tested-by: David Stevens <stevensd@chromium.org>
This reverts commit f0e607c7d4.
Reason for revert: caused flaky regressions across the board.
BUG=b:150997559
Exempt-From-Owner-Approval: revert.
Original change's description:
> minigbm: introduce test allocation
>
> This change introduces a GBM_TEST_ALLOC flag to minigbm, which allows
> for the creation of fake buffers that can be used to determine buffer
> metadata without actually allocating a full buffer. The new flag is
> supported by the i915 backends. This flag also alleviates the need to
> cache buffers when virtio_gpu queries metadata properties.
>
> BUG=b:145994510
> TEST=play youtube with arcvm demo image plus this and virgl change
>
> Change-Id: I9c6819aa3b5b674e4bb33b0656f2a9f155b0884e
> Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/1980688
> Tested-by: David Stevens <stevensd@chromium.org>
> Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
> Commit-Queue: David Stevens <stevensd@chromium.org>
Bug: b:145994510
Change-Id: I50079b7f0aabf38e1f373cac0f28c0e057eed760
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/2093923
Commit-Queue: Ilja H. Friedel <ihf@chromium.org>
Tested-by: Ilja H. Friedel <ihf@chromium.org>
Reviewed-by: Ilja H. Friedel <ihf@chromium.org>
This change introduces a GBM_TEST_ALLOC flag to minigbm, which allows
for the creation of fake buffers that can be used to determine buffer
metadata without actually allocating a full buffer. The new flag is
supported by the i915 backends. This flag also alleviates the need to
cache buffers when virtio_gpu queries metadata properties.
BUG=b:145994510
TEST=play youtube with arcvm demo image plus this and virgl change
Change-Id: I9c6819aa3b5b674e4bb33b0656f2a9f155b0884e
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/1980688
Tested-by: David Stevens <stevensd@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: David Stevens <stevensd@chromium.org>
There are two situations where guest access and host access to buffers
need to be synchronized to ensure consistency:
- If the guest CPU can write to the mapping and something in the host
besides the GPU might access the buffer, flush must be synchronous
to ensure the host sees any guest changes. Waiting is not necessary
if only the GPU can access the buffer because any subsequent
commands will be properly ordered.
- If the guest CPU can access the mapping and something in the host
can write to the buffer, then invalidate must be synchronous to
ensure the guest sees any host changes.
To perform this synchronization, have the virtio_gpu backend wait for
the transfer to/from the host to complete before returning from
flush/invalidate. In the future, support for fence-based synchronization
should also be added.
BUG=b:120456557 b:136733358
TEST=arc-codec-test, ArcVideoPlayer
Change-Id: If4ad293886a67d93d85b0d4a682b31c66597d4ef
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/1687736
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: David Stevens <stevensd@chromium.org>
Tested-by: David Stevens <stevensd@chromium.org>
Adds unneeded complication and extra kernel patches.
Hopefully with Chrome using modifiers we can remove this.
BUG=b:145747132
TEST=compile and run
Change-Id: I557bbd50828d7e39848c315a44a24587c9e2ce3a
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/platform/minigbm/+/1976277
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Commit-Queue: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Kristian H. Kristensen <hoegsberg@chromium.org>
Mesa drivers can't use it, in general.
I'm leaving in Rockchip since I haven't gotten any reports
about this test failing on kevin-arcnext.
Fixes: abe44f ("minigbm: add support for BG24")
BUG=b:77876551, b:115564746
TEST=The following tests should pass on Eve/Grunt arc-next:
android.hardware.nativehardware.cts.AHardwareBufferNativeTests#SingleLayer_ColorTest_GpuColorOutputAndSampledImage_R8G8B8_UNORM
android.hardware.nativehardware.cts.AHardwareBufferNativeTests#SingleLayer_ColorTest_GpuColorOutputCpuRead_R8G8B8_UNORM
android.hardware.nativehardware.cts.AHardwareBufferNativeTests#SingleLayer_ColorTest_GpuColorOutputIsRenderable_R8G8B8_UNORM
android.hardware.nativehardware.cts.AHardwareBufferNativeTests#SingleLayer_ColorTest_GpuSampledImageCanBeSampled_R8G8B8_UNORM
Change-Id: Ic7aec07c89fdc21e0c8392238e833f7980062049
Reviewed-on: https://chromium-review.googlesource.com/1229439
Commit-Ready: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
When building against an older libdrm, the DRM_FORMAT_MOD_LINEAR enum
will be missing. It's OK to use DRM_FORMAT_MOD_NONE instead, because
this older libdrm has limited DRM_FORMAT_MOD_xx support and not all of
the backends actually need it.
Change-Id: I0c289c50de4ad1fe5a9b0848961f5bcb75bd4bf8
Reviewed-on: https://chromium-review.googlesource.com/971358
Commit-Ready: ChromeOS CL Exonerator Bot <chromiumos-cl-exonerator@appspot.gserviceaccount.com>
Tested-by: Alistair Strachan <astrachan@google.com>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
CL:737409 replaced DRM_FORMAT_MOD_NONE with DRM_FORMAT_MOD_INVALID in
struct kms_item. We use drv_query_kms() to add the scanout flag to most
buffer combinations. That means anything with DRM_FORMAT_MOD_LINEAR
will not get the scanout flag, since drv_query_kms returns combinations
with DRM_FORMAT_MOD_INVALID. This leads to artifacts in the zero latency
app, since we end up picking a X-tiled buffer since we don't advertise
a linear buffer with scanout.
Since all hardware can scanout linear buffers, replace DRM_FORMAT_MOD_INVALID
with DRM_FORMAT_MOD_LINEAR.
BUG=chromium:762324
TEST=no overlay artifacts running Zero latency app
Change-Id: Ib8a33b9e8dde3767337b9cd32c5b2849f9c8aa04
Reviewed-on: https://chromium-review.googlesource.com/823258
Commit-Ready: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
Everything else should use GTT (USWC preferably). We should also switch
scanouts to GTT after we enable display VM.
BUG=b:69941535,b:69647975
TEST=run WebGL Aquarium on Kahlee
Change-Id: I939b159551305f74102ba47eb551e2bed2e32e06
Signed-off-by: Dominik Behr <dbehr@chromium.org>
Reviewed-on: https://chromium-review.googlesource.com/801797
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Reviewed-by: Gurchetan Singh <gurchetansingh@chromium.org>
This sets better expectations for what we expect from the
backends.
BUG=chromium:764871
TEST=mmap_test
Change-Id: I7fb815b58fae8e9fbd73bf7c0263c7db44488844
Reviewed-on: https://chromium-review.googlesource.com/770519
Commit-Ready: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Joe Kniss <djmk@google.com>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Let's start allowing multiple mappings of the same buffer when
different map flags are passed in.
BUG=chromium:764871
TEST=mmap_test -g on Kevin, gbmtest
Change-Id: I4eb0b6f4c3572a92001696c2720d5a5f7d9d73a4
Reviewed-on: https://chromium-review.googlesource.com/758146
Commit-Ready: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Joe Kniss <djmk@google.com>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
Since some drivers (AMDGPU, Tegra) may have to do expensive tiling
and detiling operations, we should try to take advantage of the
access regions passed in by gralloc and gbm. Let's refactor struct
map_data so we can separate the actual mapping and access region.
Here is the Coccinelle rule used in this change:
@@ struct map_info *M; @@
- (M)
+ M->vma
In addition, struct map_data was also renamed to struct mapping.
BUG=chromium:764871
TEST= mmap_test -g on Kevin
Change-Id: Idb094aa3b5f81e45ce3a2f4fb2d9bf8fba32bf29
Reviewed-on: https://chromium-review.googlesource.com/758144
Commit-Ready: Gurchetan Singh <gurchetansingh@chromium.org>
Tested-by: Gurchetan Singh <gurchetansingh@chromium.org>
Reviewed-by: Joe Kniss <djmk@google.com>
Reviewed-by: Stéphane Marchesin <marcheu@chromium.org>
