Robert Foss
June 05, 2017
Reading time:
GPUs like those of Intel and Vivante support storing the contents of graphical buffers in different formats. Support for describing these formats using modifiers has now been added to Android and Mesa, enabling tiling artifact free running of Android on the iMX6 platform.
With modifier support added to Mesa and gbm_gralloc, it is now possible to boot Android on iMX6 platforms using no proprietary blobs at all. This makes iMX6 one of the very few embedded SOCs that needs no blobs at all to run.
Not only is that a great win for Open Source in general, but it also makes the iMX6 more attractive as a platform. A further positive point is that this lays the groundwork for the iMX8 platform, and supporting it will come much easier.
Modifiers are used to represent different properties of buffers. These properties can cover a range of different information about a buffer, for example compression and tiling.
For the case of the iMX6 and the Vivante GPU which it is equipped with, the modifiers are related to tiling. The reason being that buffers can be tiled in different ways (Tiled, Super Tiled, etc.) or not at all (Linear). Before sending buffers out to a display, they need to have the associated tiling information made available, so that the actual image that is being sent out is not tiled.
Support was added in two places; Mesa and gbm_gralloc. Mesa has had support added to many of the buffer allocation functions and to GBM (which is the API provided by Mesa, that gbm_gralloc uses).
gbm_gralloc in turn had support added for using a new GBM API call, GBM_BO_IMPORT_FD_MODIFIER, which imports a buffer object as well as accompanying information like modifier used by the buffer object in question.
Currently the modifiers work is in the process of being upstreamed, but in the meantime it can be found here. If you'd like to test this out yourself a How-To can be found here.
This work is built on efforts by a lot people:
This post has been a part of work undertaken by my employer Collabora, and has been funded by Zodiac Inflight Innovations.
19/12/2024
In the world of deep learning optimization, two powerful tools stand out: torch.compile, PyTorch’s just-in-time (JIT) compiler, and NVIDIA’s…
08/10/2024
Having multiple developers work on pre-merge testing distributes the process and ensures that every contribution is rigorously tested before…
15/08/2024
After rigorous debugging, a new unit testing framework was added to the backend compiler for NVK. This is a walkthrough of the steps taken…
01/08/2024
We're reflecting on the steps taken as we continually seek to improve Linux kernel integration. This will include more detail about the…
27/06/2024
With each board running a mainline-first Linux software stack and tested in a CI loop with the LAVA test framework, the Farm showcased Collabora's…
26/06/2024
WirePlumber 0.5 arrived recently with many new and essential features including the Smart Filter Policy, enabling audio filters to automatically…
Comments (0)
Add a Comment