When the reference clock is 38.4MHz, using the current TBT PLL
fractional divider value results in a slightly off TBT link frequency.
This causes an endless loop of link training success followed by a bad
link signaling and retraining at least on a Dell WD19TB TBT dock. The
workaround provided by the HW team is to divide the fractional divider
value by two. This fixed the link training problem on the ThinkPad dock.
The same workaround is needed on some EHL platforms and for combo PHY
PLLs, these will be addressed in a follow-up.
Bspec: 49204
References: HSDES#22010772725
References: HSDES#14011861142
Reported-and-tested-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200629185848.20550-1-imre.deak@intel.com
The obj->lut_list is traversed when the object is closed as the file
table is destroyed during process termination. As this occurs before we
kill any outstanding context if, due to some bug or another, the closure
is blocked, then we fail to shootdown any inflight operations
potentially leaving the GPU spinning forever. As we only need to guard
the list against concurrent closures and insertions, the hold is short
and merits being treated as a simple spinlock.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200701084439.17025-1-chris@chris-wilson.co.uk
When graph isn't defined in a device-tree, the of_graph_get_remote_node()
prints a noisy error message, telling that port node is not found. This is
undesirable behaviour in our case because absence of a panel/bridge graph
is a valid case. Let's check the graph's presence in a device-tree before
proceeding with parsing of the graph.
Reviewed-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20200701074232.13632-3-digetx@gmail.com
I updated my system with Radeon VII from kernel 5.6 to kernel 5.7, and
following started to happen on each boot:
...
BUG: kernel NULL pointer dereference, address: 0000000000000128
...
CPU: 9 PID: 1940 Comm: modprobe Tainted: G E 5.7.2-200.im0.fc32.x86_64 #1
Hardware name: System manufacturer System Product Name/PRIME X570-P, BIOS 1407 04/02/2020
RIP: 0010:lock_bus+0x42/0x60 [amdgpu]
...
Call Trace:
i2c_smbus_xfer+0x3d/0xf0
i2c_default_probe+0xf3/0x130
i2c_detect.isra.0+0xfe/0x2b0
? kfree+0xa3/0x200
? kobject_uevent_env+0x11f/0x6a0
? i2c_detect.isra.0+0x2b0/0x2b0
__process_new_driver+0x1b/0x20
bus_for_each_dev+0x64/0x90
? 0xffffffffc0f34000
i2c_register_driver+0x73/0xc0
do_one_initcall+0x46/0x200
? _cond_resched+0x16/0x40
? kmem_cache_alloc_trace+0x167/0x220
? do_init_module+0x23/0x260
do_init_module+0x5c/0x260
__do_sys_init_module+0x14f/0x170
do_syscall_64+0x5b/0xf0
entry_SYSCALL_64_after_hwframe+0x44/0xa9
...
Error appears when some i2c device driver tries to probe for devices
using adapter registered by `smu_v11_0_i2c_eeprom_control_init()`.
Code supporting this adapter requires `adev->psp.ras.ras` to be not
NULL, which is true only when `amdgpu_ras_init()` detects HW support by
calling `amdgpu_ras_check_supported()`.
Before 9015d60c9e, adapter was registered by
-> amdgpu_device_ip_init()
-> amdgpu_ras_recovery_init()
-> amdgpu_ras_eeprom_init()
-> smu_v11_0_i2c_eeprom_control_init()
after verifying that `adev->psp.ras.ras` is not NULL in
`amdgpu_ras_recovery_init()`. Currently it is registered
unconditionally by
-> amdgpu_device_ip_init()
-> pp_sw_init()
-> hwmgr_sw_init()
-> vega20_smu_init()
-> smu_v11_0_i2c_eeprom_control_init()
Fix simply adds HW support check (ras == NULL => no support) before
calling `smu_v11_0_i2c_eeprom_control_{init,fini}()`.
Please note that there is a chance that similar fix is also required for
CHIP_ARCTURUS. I do not know whether any actual Arcturus hardware without
RAS exist, and whether calling `smu_i2c_eeprom_init()` makes any sense
when there is no HW support.
Cc: stable@vger.kernel.org
Fixes: 9015d60c9e ("drm/amdgpu: Move EEPROM I2C adapter to amdgpu_device")
Signed-off-by: Ivan Mironov <mironov.ivan@gmail.com>
Tested-by: Bjorn Nostvold <bjorn.nostvold@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Add a switch statement to simplify asic checks. Note
that BACO is not supported on APUs, so there is no
need to check them.
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The release_firmware() function is NULL tolerant so we do not need
to check for NULL param before calling it.
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
There is an off-by-one bounds check on the index into arrays
table->mc_reg_address and table->mc_reg_table_entry[k].mc_data[j] that
can lead to reads and writes outside of arrays. Fix the bound checking
off-by-one error.
Addresses-Coverity: ("Out-of-bounds read/write")
Fixes: cc8dbbb4f6 ("drm/radeon: add dpm support for CI dGPUs (v2)")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In the cases where adev->jpeg.num_jpeg_inst is zero or the condition
adev->jpeg.harvest_config & (1 << i) is always non-zero the variable
ret is never set to an error condition and the function returns
an uninitialized value in ret. Since the only exit condition at
the end if the function is a success then explicitly return
0 rather than a potentially uninitialized value in ret.
Addresses-Coverity: ("Uninitialized scalar variable")
Fixes: 14f43e8f88 ("drm/amdgpu: move JPEG2.5 out from VCN2.5")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The function kobject_init_and_add alloc memory like:
kobject_init_and_add->kobject_add_varg->kobject_set_name_vargs
->kvasprintf_const->kstrdup_const->kstrdup->kmalloc_track_caller
->kmalloc_slab, in err branch this memory not free. If use
kmemleak, this path maybe catched.
These changes are to add kobject_put in kobject_init_and_add
failed branch, fix potential memleak.
Signed-off-by: Bernard Zhao <bernard@vivo.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
clk_s is checked twice in a row in ni_init_smc_spll_table().
fb_div should be checked instead.
Fixes: 69e0b57a91 ("drm/radeon/kms: add dpm support for cayman (v5)")
Cc: stable@vger.kernel.org
Signed-off-by: Denis Efremov <efremov@linux.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This fixes incorrect output when reading pp_dpm_dcefclk sysfs.
Output before this patch:
cat /sys/devices/pci0000:<snip>/pp_dpm_dcefclk
0: 506Mhz
1: 42949671Mhz *
2: 1266Mhz
v2: squash in arcturus fix
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If (smu)->ppt_funcs->intf is not initialized then
smu_send_smc_msg(smu, SMU_MSG_ReadSerialNumBottom32, &bottom32) and
smu_send_smc_msg(smu, SMU_MSG_ReadSerialNumTop32, &top32) will not
touch bottom32 and top32.
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Regression was introduced where setting max bpc property has no effect
on the atomic check and final commit. It has the same effect as max bpc
being stuck at 8.
[How]
Correctly propagate max bpc with the new connector state.
Signed-off-by: Stylon Wang <stylon.wang@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
DP link layer CTS specs updated to change the test parameters in test
4.2.1.1.
Before it requires source to delay 400us on aux no reply.
With the specs updates Errata5, it requires source to delay 3.2ms
(based on LTTPR aux timeout)
This causes our test to fail after updating with the latest test
equipment firmware.
[how]
the change is to allow LTTPR 3.2ms aux timeout delay by default.
And set back to 400us if LTTPR feature is not enabled.
We will set 3.2ms and always enable LTTPR non transparent mode
if LTTPR feature is enabled and LTTPR is present.
Signed-off-by: Wenjing Liu <wenjing.liu@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Krunoslav Kovac <Krunoslav.Kovac@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
Two issues:
1. Add read only operation support for query ddc data over aux.
2. Fix a bug where if read size is multiple of 16,
mot of the last read transaction will not be set to 0.
Signed-off-by: Wenjing Liu <wenjing.liu@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
If the front porch of the two timings differ, then there may not be
enough time while both streams are in vertical blank to perform a memory
clock change. This can hang the system.
[How]
Check the each streams timing.v_front_porch when determining if the two
streams are synchronizable.
Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
why:
seamless boots requires split of init_hw into hw and pipes to work. This
was implemented in dcn10_init_hw but did not apply yet to dcn30.
how:
Copy over dcn10_init_hw and adapt it to dcn30 using recent changes to
dcn3. Behavior will be different in init sequence.
Signed-off-by: Martin Leung <martin.leung@amd.com>
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
We try to to change new_clocks->dppclk_khz to 100000 when
new_clocks->dppclk_khz is 0
[How]
Don't change new_clocks->dppclk_khz value when new_clocks->dppclk_khz is
0
Signed-off-by: Peikang Zhang <peikang.zhang@amd.com>
Reviewed-by: Yongqiang Sun <yongqiang.sun@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
10K YCbCr420 does not need ODM 4:1, but it requires MPC 4 split
indicated on the flags.
[How]
Make pixel encoding and resolution size specific workaround to enable
ODM combine on YCbCr420 high resolution modes.
Signed-off-by: Chris Park <Chris.Park@amd.com>
Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
If a typeC to HDMI dongle supports YCbCr420 pass through and VSC
colorimetry and pixel encoding formats in the Extended Receiver
Capability, we shall allow VSC SDP to be used.
[How]
The Extended Receiver Capability field shall check the
EXTENDED_RECEIVER_CAPABILITY_FIELD_PRESENT bit in the
TRAINING_AUX_RD_INTERVAL register. Removed DPCD rev checking for VSC
SDP.
Signed-off-by: Derek Lai <Derek.Lai@amd.com>
Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Acked-by: Tony Cheng <Tony.Cheng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
DMCUB firmware version is now available from firmware metadata block.
We should be passing this into dmub_srv so we can know when to apply
firmware version specific functionality like using CW4 only instead
of the REGION4.
[How]
We don't have the helpers for DM to actually extract out firmware
metadata block themselves.
We could add that and add helpers in DM to grab this, but not every
creation sequence has firmware instruction before dmub_srv_create.
Easiest way to handle this is to fill this in automatically per DM in
the place we do have it - when calculating the region parameters. But
only fill it in if DM already hasn't in case we need to override with
a specific version.
We aren't do anything firmware version specific in dmub_srv_create
today that does require fw_version, so while it's a little unituitive
to do it when calculating region parameters it works for now.
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Connector property output_bpc is available on DP/eDP only. New IGT tests
would benifit if this property works on HDMI.
[How]
Enable this read-only property on all types of connectors.
Signed-off-by: Stylon Wang <stylon.wang@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
All YCbCr420 resolutions 5K and above have tiling and discoloration
issues. The issue can be remedied by forcing ODM combine from 5K to 8K.
10K resolution requires ODM 4:1. The mechanism of what the real problem
is, that is inherent in ODM combine programming, doesn't seem to be
pointed at singular register programming (CLK, MPC, DCSURF, etc.), and
needs more in-depth programming sequence review for these new use case
scenarios. Until then, workaround to enable ODM combine is proposed.
While it is not our policy, HW spreadsheet also recommends turning on
ODM for these scenario to lower the voltage.
[How]
Make pixel encoding and resolution size specific workaround to enable
ODM combine on YCbCr420 high resolution modes.
Signed-off-by: Chris Park <Chris.Park@amd.com>
Reviewed-by: Charlene Liu <Charlene.Liu@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
There is using pixelclk AVFS for dppclk, that would cause issue.
[How]
To use dispclk AVFS for both dispclk and dppclk. There would choose
dppclk for request voltage when dispclk wouldn't be updated case. If
dispclk need to be updated, then it'll choose the bigger one from dppclk
and dispclk for request voltage.
Signed-off-by: Brandon Syu <Brandon.Syu@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
If optimized init is done in FW. DCN init be skipped in driver. This
need to be communicated between driver and fw and maintain backwards
compatibility.
[How]
Use DMUB scratch 0 bit 2 to indicate optimized init done in fw and
use DMUB scatch 4 bit 0 to indicate drive supports the optimized flow
so FW will perform it.
Signed-off-by: Eric Yang <Eric.Yang2@amd.com>
Reviewed-by: Tony Cheng <Tony.Cheng@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY]
Check max_tmds_clk_mhz firstly will restrict pixel clock under HDMI
1.4, thus HDMI2.0 port can't correctly support 4K 60Hz.
[HOW]
Fine tune the logic to check max_forum_tmds_clk_mhz firstly.
Signed-off-by: Dale Zhao <dale.zhao@amd.com>
Reviewed-by: Chris Park <Chris.Park@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
DC global validation can fail when userspace requests to draw large
plane without performing the clipping themselves.
This is observed in the IGT kms_plane panning tests for 4K displays
where they draw an 8K plane without any clipping while expecting only
the top 4K to be drawn.
[How]
DRM already has helpers to take care of the clipping necessary and to
mark whether a plane is visible or not, so make use of these helpers
in DM before passing the plane to DC.
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
Side-by-side and Top-and-bottom stereo configurations fail DML mode
validation due to Viewport exceeded.
This is because we consider the planes as being pipe split in pipe
population so we end up doubling the viewport width, eg. from 4k to 8k.
[How]
These pipes technically aren't hsplit, so add a check for determining
whether is_hsplit should be set.
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
