kernel/linux.git/drivers/gpu/drm/amd/pm, branch v5.14.8

drm/amdgpu: Disable PCIE_DPM on Intel RKL Platform

2021-09-26T12:10:19+00:00

[ Upstream commit b3dc549986eb7b38eba4a144e979dc93f386751f ] Due to high latency in PCIE clock switching on RKL platforms, switching the PCIE clock dynamically at runtime can lead to HDMI/DP audio problems. On newer asics this is handled in the SMU firmware. For SMU7-based asics, disable PCIE clock switching to avoid the issue. AMD provide a parameter to disable PICE_DPM. modprobe amdgpu ppfeaturemask=0xfff7bffb It's better to contorl PCIE_DPM in amd gpu driver, switch PCI_DPM by determining intel RKL platform for SMU7-based asics. Fixes: 1a31474cdb48 ("drm/amd/pm: workaround for audio noise issue") Ref: https://lists.freedesktop.org/archives/amd-gfx/2021-August/067413.html Signed-off-by: Koba Ko Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amd/pm: fix the issue of uploading powerplay table

2021-09-22T10:39:16+00:00

commit 5598d7c21a0bcab900f281dca4efbb1f80add0fe upstream. fix the issue of uploading powerplay table due to the dependancy of rlc. Signed-off-by: Kenneth Feng Reviewed-by: Jack Gui Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amd/pm: fix runpm hang when amdgpu loaded prior to sound driver

2021-09-22T10:39:15+00:00

commit 8b514e898ee7f861eb8863c647d258f71053af40 upstream. Current RUNPM mechanism relies on PMFW to master the timing for BACO in/exit. And that needs cooperation from sound driver for dstate change notification for function 1(audio). Otherwise(on sound driver missing), BACO cannot be kicked in correctly and hang will be observed on RUNPM exit. By switching back to legacy message way on sound driver missing, we are able to fix the runpm hang observed for the scenario below: amdgpu driver loaded -> runpm suspend kicked -> sound driver loaded Signed-off-by: Evan Quan Reported-and-tested-by: Pierre-Eric Pelloux-Prayer Reviewed-by: Lijo Lazar Reviewed-by: Guchun Chen Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amd/pm: Fix a bug in semaphore double-lock

2021-09-15T08:02:16+00:00

[ Upstream commit 544dcd74b7093ad4befac99b11d90331aa73348e ] Fix a bug in smu_cmn_send_msg_without_waiting() in that this function does not need to take the smu->message_lock mutex in order to send a message down to the SMU. The mutex is acquired by the caller of this function instead. Cc: Alex Deucher Cc: Changfeng Zhu Cc: Huang Rui Fixes: 5810323ba69289 ("drm/amd/pm: Fix a bug communicating with the SMU (v5)") Signed-off-by: Luben Tuikov Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amd/pm: Fix a bug communicating with the SMU (v5)

2021-09-15T08:02:14+00:00

[ Upstream commit 5810323ba692895b045e3f1b3e107605c3717dab ] This fixes a bug which if we probe a non-existing I2C device, and the SMU returns 0xFF, from then on we can never communicate with the SMU, because the code before this patch reads and interprets 0xFF as a terminal error, and thus we never write 0 into register 90 to clear the status (and subsequently send a new command to the SMU.) It is not an error that the SMU returns status 0xFF. This means that the SMU executed the last command successfully (execution status), but the command result is an error of some sort (execution result), depending on what the command was. When doing a status check of the SMU, before we send a new command, the only status which precludes us from sending a new command is 0--the SMU hasn't finished executing a previous command, and 0xFC--the SMU is busy. This bug was seen as the following line in the kernel log, amdgpu: Msg issuing pre-check failed(0xff) and SMU may be not in the right state! when subsequent SMU commands, not necessarily related to I2C, were sent to the SMU. This patch fixes this bug. v2: Add a comment to the description of __smu_cmn_poll_stat() to explain why we're NOT defining the SMU FW return codes as macros, but are instead hard-coding them. Such a change, can be followed up by a subsequent patch. v3: The changes are, a) Add comments to break labels in __smu_cmn_reg2errno(). b) When an unknown/unspecified/undefined result is returned back from the SMU, map that to -EREMOTEIO, to distinguish failure at the SMU FW. c) Add kernel-doc to smu_cmn_send_msg_without_waiting(), smu_cmn_wait_for_response(), smu_cmn_send_smc_msg_with_param(). d) In smu_cmn_send_smc_msg_with_param(), since we wait for completion of the command, if the result of the completion is undefined/unknown/unspecified, we print that to the kernel log. v4: a) Add macros as requested, though redundant, to be removed when SMU consolidates for all ASICs--see comment in code. b) Get out if the SMU code is unknown. v5: Rename the macro names. Cc: Alex Deucher Cc: Evan Quan Cc: Lijo Lazar Fixes: fcb1fe9c9e0031 ("drm/amd/powerplay: pre-check the SMU state before issuing message") Signed-off-by: Luben Tuikov Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amd/pm: change the workload type for some cards

2021-08-16T19:58:57+00:00

change the workload type for some cards as it is needed. Signed-off-by: Kenneth Feng Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher

Revert "drm/amd/pm: fix workload mismatch on vega10"

2021-08-16T19:58:44+00:00

This reverts commit 0979d43259e13846d86ba17e451e17fec185d240. Revert this because it does not apply to all the cards. Signed-off-by: Kenneth Feng Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher

drm/amd/pm: bug fix for the runtime pm BACO

2021-08-10T14:39:33+00:00

In some systems only MACO is supported. This is to fix the problem that runtime pm is enabled but BACO is not supported. MACO will be handled seperately. Signed-off-by: Kenneth Feng Reviewed-by: Jack Gui Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org

drm/amd/pm: Fix a memory leak in an error handling path in 'vangogh_tables_init()'

2021-08-06T15:36:55+00:00

'watermarks_table' must be freed instead 'clocks_table', because 'clocks_table' is known to be NULL at this point and 'watermarks_table' is never freed if the last kzalloc fails. Fixes: c98ee89736b8 ("drm/amd/pm: add the fine grain tuning function for vangogh") Signed-off-by: Christophe JAILLET Signed-off-by: Alex Deucher

drm/amd/pm: update smu v13.0.1 firmware header

2021-08-06T15:24:48+00:00

Update smu v13.0.1 firmware header for yellow carp. Signed-off-by: Xiaomeng Hou Reviewed-by: Aaron Liu Signed-off-by: Alex Deucher