27.08.2020 18:54, Thierry Reding пишет: ... >> The Tegra DRM has a very special quirk for ARM32 that was added in this >> commit [2] and driver relies on checking of whether explicit or implicit >> IOMMU is used in order to activate the quirk. >> >> [2] >> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=273da5a046965ccf0ec79eb63f2d5173467e20fa >> >> Once the implicit IOMMU is used for the DRM driver, the quirk no longer >> works (if I'm not missing something). This problem needs to be resolved >> before implicit IOMMU could be used by the Tegra DRM on ARM32. ... > I do have a patch lying around somewhere that implements the mapping > cache that was referenced in the above commit. Let me know if I should > dig that up and send it out. Hello, Thierry! It certainly will be interesting to take a look at yours patch! I think that the caching shouldn't be strictly necessary for keeping the current workaround working and it should be possible to keep the code as-is by replacing the domain-type checking with the SoC-generation check in the Tegra DRM driver. In general, IMO it should be better to stash the complex changes until we'll get closer to adopting the new UAPI as it will certainly touch the aspect of the per-device mappings. But if yours patch is less than 100 LOC, then maybe we could consider applying it right now!
