-
Notifications
You must be signed in to change notification settings - Fork 38
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[WA for ChimoHMI] Use 64K memory alignment instead of 2MB alignment #125
[WA for ChimoHMI] Use 64K memory alignment instead of 2MB alignment #125
Conversation
…le_android=0 for media cts" This reverts commit 0ebc3b5. Tracked-On: OAM-122074 Signed-off-by: Lin, Shenghua <[email protected]>
Vts checkVulkanDeviceMemoryReportSupport test requires that vulkan devices with API version >= 1.1 must support VK_EXT_device_memory_report. Tracked-On: OAM-122094 Signed-off-by: Huang Rui <[email protected]> Signed-off-by: shenghualin <[email protected]>
…f not busy" This reverts commit de912b6. This can avoid CTS incorrectly report "memory leak" if release BO immediately Tracked-On: OAM-122096 Signed-off-by: Lin, Shenghua <[email protected]>
Fix race condition when building ANV: In file included from src/intel/vulkan/grl/grl_metakernel_build_primref.c:40: ../src/intel/common/mi_builder.h:28:10: fatal error: 'genxml/genX_bits.h' file not found #include "genxml/genX_bits.h" This change is already included in commit 41b2ed6 (genxml: generate opencl packing headers) in the upstream. Hence when updating mesa version, this change can be safely removed. Test-done: build mesa multiple times Tracked-On: OAM-122542 Signed-off-by: Weifeng Liu <[email protected]>
Tracked-On: OAM-122380 Signed-off-by: Lin, Shenghua <[email protected]>
In case of legacy dataspace value is passed down by framework/av/media code. Tracked-On: OAM-122156 Signed-off-by: ZhuChenyanX <[email protected]>
Expected syntax:
|
0a93455
to
2874475
Compare
Improper Commit Message |
2874475
to
61055d2
Compare
Expected syntax:
|
61055d2
to
80b2321
Compare
Expected syntax:
|
80b2321
to
a678564
Compare
Improper Commit Message |
a678564
to
1141ad4
Compare
Improper Commit Message |
the caller (driCreateNewScreen3) will always call dri_destroy_screen() when these functions return failure, so releasing the screen is always wrong cc: mesa-stable Tracked-On: OAM-122605 Reviewed-by: Adam Jackson <[email protected]> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29021> Signed-off-by: Weifeng Liu <[email protected]> Signed-off-by: Mike Blumenkrantz <[email protected]>
this otherwise leads to a mismatch where some types of screen may have the mutex initialized while others don't, in which case dri_release_screen() will attempt to destroy an uninitialized mutex cc: mesa-stable Tracked-On: OAM-122605 Reviewed-by: Adam Jackson <[email protected]> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29021> Signed-off-by: Weifeng Liu <[email protected]> Signed-off-by: Mike Blumenkrantz <[email protected]>
Check the dGPU through whether has local memory. Calculate the i915 render node number for better compatibility. The GLES open one available render node, vulkan open all available render nodes. Enalbe the vulkan dGPU render. Tracked-On: OAM-122333 Signed-off-by: HeYue <[email protected]>
We have some code specific to Android only and it should be built only on Android so that this code base could possibly be built on GNU/Linux systems. Also replace \xC2\xA0 (non-breaking space) with normal space for consistence. Signed-off-by: Weifeng Liu <[email protected]>
When we have both dGPU and iGPU available, prefer dGPU for compositing because dGPU can cover more cases for us, e.g., allocating buffers in device memory. As a side work, also format the surrounding code for consistency. Tracked-On: OAM-124060 Signed-off-by: Weifeng Liu <[email protected]>
Ported from upstream: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30841?commit_id=1f9c40a8d1b2ec3702862371af3a934e9a0c9cda The default state doesn't seem well defined (or kernel driver bug maybe?). Let's just set it to disabled on platforms where we're not using it. Signed-off-by: Lionel Landwerlin <[email protected]> Cc: mesa-stable Found-by: Chuansheng Liu <[email protected]> Reviewed-by: Tapani Pälli <[email protected]> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30841> Tracked-On: OAM-124615 Signed-off-by: Lin, Shenghua <[email protected]>
Chimo app allocate descriptor pool so frequently, with 2MB alignment the 2Gb descriptor pool heap vm address space is used up quickly.
1141ad4
to
955c355
Compare
Improper Commit Message |
No description provided.