TGL is suffering of timeouts and fifo underruns when disabling
transcoder in MST mode, this is fixed by set TRANS_DDI_MODE_SELECT to
0(HDMI mode) when clearing DDI select.
Although BSpec disable sequence don't require this step, it is a
harmless change and it is also done by Windows driver.
Anyhow HW team was notified about that but it can take some time to
documentation to be updated.
A case that always lead to those issues is:
- do a modeset enabling pipe A and pipe B in the same MST stream
leaving A as master
- disable pipe A, promote B as master doing a full modeset in A
- enable pipe A, changing the master transcoder back to A(doing a
full modeset in B)
- Pow: underruns and timeouts
The transcoders involved will only work again when complete disabled
and their power wells turned off causing a reset in their registers.
v2: Setting TRANS_DDI_MODE_SELECT to default when clearing DDI select
not anymore when disabling TRANS_DDI, both work but this one looks
more safe. (Ville comment)
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: José Roberto de Souza <jose.souza@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200203225549.152301-1-jose.souza@intel.com
Currently, gcc spews a warning as:
drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_hubbub.c: In function ‘hubbub1_verify_allow_pstate_change_high’:
./include/drm/drm_print.h:316:2: warning: ‘debug_data’ may be used uninitialized in this function [-Wmaybe-uninitialized]
This is because the code checks against a static value although it's
basically a constant and guaranteed to be set.
This patch changes the type prefix from static to const for addressing
the compile warning above and also for letting the compiler optimize
better.
Fixes: 62d591a8e0 ("drm/amd/display: create new files for hubbub functions")
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Provide compute queues information in sysfs under /sys/class/kfd/kfd/proc.
The format is /sys/class/kfd/kfd/proc/<pid>/queues/<queue id>/XX where
XX are size, type, and gpuid three files to represent queue size, queue
type, and the GPU this queue uses. <queue id> folder and files underneath
are generated when a queue is created. They are removed when the queue is
destroyed.
Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
not turning off the mst hub before detection on reboot
causes us to not be able to light up displays with mst hook
[how]
on hw init, see if any displays are lit up. if so, turn them off
Signed-off-by: Joseph Gravenor <joseph.gravenor@amd.com>
Reviewed-by: Eric Yang <eric.yang2@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Starting from 14nm, the PLL is built into the PHY and the PLL is mapped
to PHY on 1 to 1 basis. In the code, the DP port is mapped to a PLL that was not
initialized. This causes DP to HDMI dongle to not light up the display.
[How]
Initializations added for PLL2 when creating resources.
Signed-off-by: Isabel Zhang <isabel.zhang@amd.com>
Reviewed-by: Eric Yang <eric.yang2@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
these registers should have been double buffered. SW workaround we will have SW program the more aggressive (lower) values
whenever we are upating this register, so we will not have underflow at expense of less optimzal request pattern.
[How]
there is a driver bug where we don't check for 0, which is uninitialzed HW default. since 0 is smaller than any value we need to program,
driver end up with not programming these registers
Signed-off-by: Tony Cheng <tony.cheng@amd.com>
Reviewed-by: Yongqiang Sun <yongqiang.sun@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
The DMUB tracebuffer is useful for understanding DMCUB execution state.
[How]
Add a "show" attribute debugfs so we can loop through the buffer
and print the entries.
The structs for the entry format are defined in the debugfs since
the tracebuffer header no longer exists in the DMUB service.
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Hersen Wu <hersenxs.wu@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY]
enum dc_log_type has more than 32 entries. User cannot set larger entries,
like LOG_DSC. Logs from LOG_GAMMA_DEBUG where being printed even though
flag was not enabled, because dal_logger_should_log check erroneously
passed.
[HOW]
Change struct dal_logger.mask and struct dc_init_data.mask to uint64_t.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY & HOW]
Having watermark ranges saved inside clk_mgr to be
available for debug at all times would be useful.
Add it to the clk_mgr_internal struct for reference.
Only populated for Renoir, unused for other asics.
Signed-off-by: Sung Lee <sung.lee@amd.com>
Reviewed-by: Yongqiang Sun <yongqiang.sun@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
hyperV flag should be passed from dm to DC, and override the
nv12 flip workaround flag.
[How]
Add flag to phy address config struct and pass the value in dm.
Signed-off-by: Yongqiang Sun <yongqiang.sun@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Previous logic is only good for 15W parts. Other configuration
need a smarter logic to match clk levels with pp table in the fuse.
[How]
Cache all 8 DPM level's clock data, find lvl that match each pstate
in the pp table and build input into DML base on that
Signed-off-by: Eric Yang <Eric.Yang2@amd.com>
Signed-off-by: Sung Lee <sung.lee@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
hwss should not guess what type of pipe lock is needed.
The caller of the lock function should know
the right type of pipe lock.
Decouple the setup of global lock outside of pipe control lock
logic.
Signed-off-by: Wenjing Liu <Wenjing.Liu@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY & HOW]
Previously drain clk was unconstrained and fill clk was constrained on fclk.
We want to change it to fill clk unconstrained and drain clock constrained
to dcfclk.
Signed-off-by: Sung Lee <sung.lee@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
The break in apply_ctx_interdependent_lock() may potentially
lead to early break from the loop leaving update plane unlocked
[How]
Remove break
Signed-off-by: Roman Li <roman.li@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
After locking refactor GSL is not acquired properly
resulting in immediate flip issues.
[How]
Do not copy old GSL state anymore since GSL is acquired
earlier now.
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
The DMCUB may be currently executing commands when the reset is
triggered.
Before issuing a reset we should first wait for the DMCUB to finish
its work.
[How]
Send the GPINT command for halting the firmware before reset.
Get the ack for the command then wait for the scratch register to
become the correct value.
We want this to take under ~40us or so at most before we force reset
to cover PHY delay sequence max time.
Each register read will be at least ~1-3us so don't bother using udelay.
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
The General Purpose Interrupt is used on the DMCUB to pass lightweight
commands via a register to the DMCUB.
This is limited to 32-bit command and 32-bit response.
This will be used for shutting down the firmware in a clean manner.
[How]
Add the command IDs and the data register to correctly format
the commands.
Add the interface functions to dmub_srv for sending and receiving the
commands.
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Eventually want to lock at a higher level in stack.
To do this, we need to be able to isolate the parts that need to be done
after pipe unlock.
[How]
Split out programming that is done post unlock.
Signed-off-by: Anthony Koo <Anthony.Koo@amd.com>
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY]
The optimized_require flag is needed to set watermarks and clocks lower
in certain conditions. This flag is set to true and then set to false
while programming front end in dcn20.
[HOW]
Do not set the flag to false while disabling plane.
Signed-off-by: Sung Lee <sung.lee@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Driver crash with psr feature enabled due to divide-by-zero error.
This is a regression after rework to calculate static screen frame
number entry time.
[How]
Correct order of operations to avoid divide-by-zero.
Signed-off-by: Roman Li <roman.li@amd.com>
Reviewed-by: Zhan Liu <Zhan.Liu@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
The call to dp_enable_link_phy are using default/invalid values for clock id
and link settings.
[How]
Move workaround code to after its parameter variables are determined.
Signed-off-by: George Shen <george.shen@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
For sriov, psp ip block has to be initialized before
ih block for the dynamic register programming interface
that needed for vf ih ring buffer. On the other hand,
current psp ip block hw_init function will initialize
xgmi session which actaully depends on interrupt to
return session context. This results an empty xgmi ta
session id and later failures on all the xgmi ta cmd
invoked from vf. xgmi ta session initialization has to
be done after ih ip block hw_init call.
to unify xgmi session init/fini for both bare-metal
sriov virtualization use scenario, move xgmi ta init
to xgmi_add_device call, and accordingly terminate xgmi
ta session in xgmi_remove_device call.
The existing suspend/resume sequence will not be changed.
v2: squash in return fix from Nirmoy
Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Frank Min <Frank.Min@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
we need to load SRM before we start HDCP. Because for S3 case the sysfs call will be
after we have already enabled HDCP, so we might not be using the latest SRM
[How]
Set srm before starting HDCP.
Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
PSP doesn't have the ability to store SRM in a non-volatile memory. And since
the kernel cannot write to the storage directly, we need usermode to facilitate
this
As per spec the SRM needs to be persistent so this interface is to be
called by the usermode anytime the system goes down/powers on
*boot/resume: load from storage
*shutdown/suspend: save to storage
[How]
Provide a sysfs interface so that the usermode can set/get srm at the right times
save to storage: call "cat /sys/class/drm/card0/device/hdcp_srm > file" after boot and resume
-driver calls psp_get_srm() to get the stored srm and outputs it
load from storage: call "cat file > /sys/class/drm/card0/device/hdcp_srm" before shutdown and suspend
-driver reads the file from sysfs and calls psp_set_srm() to send the SRM to PSP
v2:
-update commit description
-add comment about sysfs file handling in the code
v3:
- squash in use after free fix (Dan Carpenter)
Signed-off-by: Bhawanpreet Lakha <Bhawanpreet.Lakha@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
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 <treding@nvidia.com>
Tested-by: Dmitry Osipenko <digetx@gmail.com>
Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
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: <stable@vger.kernel.org> # v5.5
Reported-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
Tested-by: Dmitry Osipenko <digetx@gmail.com>
Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
Older Tegra devices only allow addressing 32 bits of memory, so whether
or not the host1x is attached to an IOMMU doesn't matter. host1x IOMMU
attachment is only needed on devices that can address memory beyond the
32-bit boundary and where the host1x doesn't support the wide GATHER
opcode that allows it to access buffers at higher addresses.
Cc: <stable@vger.kernel.org> # v5.5
Signed-off-by: Thierry Reding <treding@nvidia.com>
Tested-by: Dmitry Osipenko <digetx@gmail.com>
Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
