kernel/linux.git/drivers/gpu/host1x, branch v5.8.18

gpu: host1x: Put gather's BO on pinning error

2020-09-03T09:29:20+00:00

[ Upstream commit fd323e9ef0a19112c0c85b85afc4848c0518174b ] This patch fixes gather's BO refcounting on a pinning error. Gather's BO won't be leaked now if something goes wrong. Signed-off-by: Dmitry Osipenko Signed-off-by: Thierry Reding Signed-off-by: Sasha Levin

gpu: host1x: debug: Fix multiple channels emitting messages simultaneously

2020-08-19T06:26:20+00:00

[ Upstream commit 35681862808472a0a4b9a8817ae2789c0b5b3edc ] Once channel's job is hung, it dumps the channel's state into KMSG before tearing down the offending job. If multiple channels hang at once, then they dump messages simultaneously, making the debug info unreadable, and thus, useless. This patch adds mutex which allows only one channel to emit debug messages at a time. Signed-off-by: Dmitry Osipenko Signed-off-by: Thierry Reding Signed-off-by: Sasha Levin

gpu: host1x: Detach driver on unregister

2020-06-16T16:59:46+00:00

Currently when a host1x device driver is unregistered, it is not detached from the host1x controller, which means that the device will stay around and when the driver is registered again, it may bind to the old, stale device rather than the new one that was created from scratch upon driver registration. This in turn can cause various weird crashes within the driver core because it is confronted with a device that was already deleted. Fix this by detaching the driver from the host1x controller when it is unregistered. This ensures that the deleted device also is no longer present in the device list that drivers will bind to. Reported-by: Sowjanya Komatineni Signed-off-by: Thierry Reding Tested-by: Sowjanya Komatineni Signed-off-by: Thierry Reding

gpu: host1x: Register child devices

2020-06-16T11:46:18+00:00

In order to remove the dependency on the simple-bus compatible string, which causes the OF driver core to register all child devices, make the host1x driver explicitly register its children. Signed-off-by: Thierry Reding

gpu: host1x: Clean up debugfs in error handling path

2020-06-16T11:46:17+00:00

host1x_debug_init() must be reverted in an error handling path. This is already fixed in the remove function since commit 44156eee91ba ("gpu: host1x: Clean up debugfs on removal") Signed-off-by: Christophe JAILLET Signed-off-by: Thierry Reding

gpu: host1x: Use SMMU on Tegra124 and Tegra210

2020-04-28T09:44:07+00:00

Tegra124 and Tegra210 support addressing more than 32 bits of physical memory. However, since their host1x does not support the wide GATHER opcode, they should use the SMMU if at all possible to ensure that all the system memory can be used for command buffers, irrespective of whether or not the host1x firewall is enabled. Tested-by: Jon Hunter Signed-off-by: Thierry Reding

drm/tegra: Fix SMMU support on Tegra124 and Tegra210

2020-04-28T09:44:07+00:00

When testing whether or not to enable the use of the SMMU, consult the supported DMA mask rather than the actually configured DMA mask, since the latter might already have been restricted. Fixes: 2d9384ff9177 ("drm/tegra: Relax IOMMU usage criteria on old Tegra") Tested-by: Jon Hunter Signed-off-by: Thierry Reding

Merge tag 'drm/tegra/for-5.6-rc1-fixes' of git://anongit.freedesktop.org/tegra/linux into drm-next

2020-02-07T02:22:30+00:00

drm/tegra: Fixes for v5.6-rc1 These are a couple of quick fixes for regressions that were found during the first two weeks of the merge window. Signed-off-by: Dave Airlie From: Thierry Reding Link: https://patchwork.freedesktop.org/patch/msgid/20200206172753.2185390-1-thierry.reding@gmail.com

gpu: host1x: Set DMA direction only for DMA-mapped buffer objects

2020-02-06T17:23:12+00:00

The DMA direction is only used by the DMA API, so there is no use in setting it when a buffer object isn't mapped with the DMA API. Signed-off-by: Thierry Reding Tested-by: Dmitry Osipenko Reviewed-by: Dmitry Osipenko

drm/tegra: Reuse IOVA mapping where possible

2020-02-06T17:21:55+00:00

This partially reverts the DMA API support that was recently merged because it was causing performance regressions on older Tegra devices. Unfortunately, the cache maintenance performed by dma_map_sg() and dma_unmap_sg() causes performance to drop by a factor of 10. The right solution for this would be to cache mappings for buffers per consumer device, but that's a bit involved. Instead, we simply revert to the old behaviour of sharing IOVA mappings when we know that devices can do so (i.e. they share the same IOMMU domain). Cc: # v5.5 Reported-by: Dmitry Osipenko Signed-off-by: Thierry Reding Tested-by: Dmitry Osipenko Reviewed-by: Dmitry Osipenko