Tags

Linux 5.4

- Reverts a patch to avoid spinning forever when context's timeline
  is active but has no requests

drm fixes for 5.4

amdgpu:
- Remove experimental flag for navi14
- Fix confusing power message failures on older VI parts
- Hang fix for gfxoff when using the read register interface
- Two stability regression fixes for Raven

i915:
- Fix kernel oops on dumb_create ioctl on no crtc situation
- Fix bad ugly colored flash on VLV/CHV related to gamma LUT update
- Fix unity of the frequencies reported on PMU
- Fix kernel oops on set_page_dirty using better locks around it
- Protect the request pointer with RCU to prevent it being freed while we might need still
- Make pool objects read-only
- Restore physical addresses for fb_map to avoid corrupted page table

- Fix kernel oops on dumb_create ioctl on no crtc situation
- Fix bad ugly colored flash on VLV/CHV related to gamma LUT update
- Fix unity of the frequencies reported on PMU
- Fix kernel oops on set_page_dirty using better locks around it
- Protect the request pointer with RCU to prevent it being freed while we might need still
- Make pool objects read-only
- Restore physical addresses for fb_map to avoid corrupted page table

- Fix ttm bo refcnt when using the new gem obj mmap hook (Thomas)

Cc: Gerd Hoffmann <kraxel@redhat.com>

- Includes gvt-next-fixes-2019-11-12
- Fix Bugzilla #112051: Fix detection for a CMP-V PCH
- Fix Bugzilla #112256: Corrupted page table at address on plymouth splash
- Fix Bugzilla #111594: Avoid losing RC6 when HuC authentication is used
- Fix for OA/perf metric coherency, restore GT coarse power gating workaround
- Avoid atomic context on error capture
- Avoid MST bitmask overflowing to EDP/DPI input select
- Fixes to CI found dmesg splats

Linux 5.4-rc8

sound fixes for 5.4-rc8 or final

A few small last-minute fixes for USB-audio and HD-audio as well
as for PCM core:

- A race fix for PCM core between stopping and closing a stream
- USB-audio regressions in the recent descriptor validation code
  and relevant changes
- A read of uninitialized value in USB-audio spotted by fuzzer
- A fix for USB-audio race at stopping a stream
- Intel HD-audio platform fixes

drm fixes for 5.4-rc8

i915:
- MOCS table fixes for EHL and TGL
- Update Display's rawclock on resume
- GVT's dmabuf reference drop fix

amdgpu:
- Fix a potential crash in firmware parsing

sun4i:
- One fix to the dotclock dividers range for sun4i

- PMU "Frequency" is reported as accumulated cycles
- Avoid OOPS in dumb_create IOCTL when no CRTCs
- Mitigation for userptr put_pages deadlock with trylock_page
- Fix to avoid freeing heartbeat request too early
- Fix LRC coherency issue
- Fix Bugzilla #112212: Avoid screen corruption on MST
- Error path fix to unlock context on failed context VM SETPARAM
- Always consider holding preemption a privileged op in perf/OA
- Preload LUTs if the hw isn't currently using them to avoid color flash on VLV/CHV
- Protect context while grabbing its name for the request
- Don't resize aliasing ppGTT size
- Smaller fixes picked by tooling

- MOCS table fixes for EHL and TGL
- Update Display's rawclock on resume
- GVT's dmabuf reference drop fix

- Fix memory leak in gpu debugfs node's release (Johan)

Cc: Johan Hovold <johan@kernel.org>

- One fix to the dotclock dividers range for sun4i

Core Changes:
- Update DSI data type and command definitions
- Add helpers for sending compression mode and PPS packets

Driver Changes:
- Update tiny/st7586 to reflect a definition change

Linux 5.4-rc7

drm fixes for 5.4-rc7

core:
- add missing documentation for GEM shmem madvise helpers
- Fix for a state dereference in atomic self-refresh helpers

fbdev:
- One compilation fix for c2p fbdev helpers

amdgpu:
- Fix navi14 display issue root cause and revert workaround
- GPU reset scheduler interaction fix
- Fix fan boost on multi-GPU
- Gfx10 and sdma5 fixes for navi
- GFXOFF fix for renoir
- Add navi14 PCI ID
- GPUVM fix for arcturus

radeon:
- Port an SI power fix from amdgpu

i915:
- Fix HPD poll to avoid kworker consuming a lot of cpu cycles.
- Do not use TBT type for non Type-C ports.

One RCU fix and fix for suspend GEM_BUG_ON (with dependencies).

sound fixes for 5.4-rc7

It became a bit largish, but all small and good for 5.4.

- A regression fix of ALSA timer code bug that sneaked in by a
  recent cleanup; never trust innocent-looking guys...

- Fix for compress API max size check signedness

- Fixes in HD-audio: CA0132 work stall, Intel Tigerlake HDMI

- A few fixes for SOF: memory leak, sanity-check and build fixes

- A collection of device-specific fixes: firewire, rockchip, ASoC
  HDMI, rsnd, ASoC HDA, stm32, TI, kirkwood, msm, max98373

 - Some new documentation for GEM shmem madvise helpers
 - Fix for a state dereference in atomic self-refresh helpers
 - One compilation fix for c2p fbdev helpers

 - Some new documentation for GEM shmem madvise helpers
 - Fix for a state dereference in atomic self-refresh helpers
 - One compilation fix for c2p fbdev helpers