Commit graph

1,215,304 commits

Author SHA1 Message Date
Damien Le Moal
5444737e16 PCI: epf-test: Simplify pci_epf_test_raise_irq()
Change the interface of the function pci_epf_test_raise_irq() to directly
pass a pointer to the struct pci_epf_test_reg defining the test being
executed. This avoids the need for grabbing this pointer using the register
BAR address and simplifies the call sites as the IRQ type and IRQ numbers
do not have to be passed as arguments.

Link: https://lore.kernel.org/r/20230415023542.77601-8-dlemoal@kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
2023-06-23 15:00:50 -05:00
Damien Le Moal
62d48ec7ef PCI: epf-test: Simplify read/write/copy test functions
The function pci_epf_test_cmd_handler() uses the register BAR address as a
pointer to a struct pci_epf_test_reg to determine the command sent by the
host and to execute the test function accordingly. There is no need for
doing this assignment again in each of the read, write and copy test
functions. We can simply pass the reg pointer as an argument to the
functions pci_epf_test_write(), pci_epf_test_read() and
pci_epf_test_copy().

Link: https://lore.kernel.org/r/20230415023542.77601-7-dlemoal@kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
2023-06-23 15:00:40 -05:00
Damien Le Moal
349d5c840a PCI: epf-test: Use dmaengine_submit() to initiate DMA transfer
Instead of an open coded call to the tx_submit() operation of struct
dma_async_tx_descriptor, use the helper function dmaengine_submit().
No functional change is introduced with this.

Link: https://lore.kernel.org/r/20230415023542.77601-6-dlemoal@kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
2023-06-23 15:00:31 -05:00
Damien Le Moal
933f31a2fe PCI: epf-test: Fix DMA transfer completion detection
pci_epf_test_data_transfer() and pci_epf_test_dma_callback() are not
handling DMA transfer completion correctly, leading to completion
notifications to the RC side that are too early. This problem can be
detected when the RC side is running an IOMMU with messages such as:

  pci-endpoint-test 0000:0b:00.0: AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x001c address=0xfff00000 flags=0x0000]

When running the pcitest.sh tests: the address used for a previous
test transfer generates the above error while the next test transfer is
running.

Fix this by testing the DMA transfer status in pci_epf_test_dma_callback()
and notifying the completion only when the transfer status is DMA_COMPLETE
or DMA_ERROR. Furthermore, in pci_epf_test_data_transfer(), be paranoid and
check again the transfer status and always call dmaengine_terminate_sync()
before returning.

Link: https://lore.kernel.org/r/20230415023542.77601-5-dlemoal@kernel.org
Fixes: 8353813c88 ("PCI: endpoint: Enable DMA tests for endpoints with DMA capabilities")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Cc: stable@vger.kernel.org
2023-06-23 15:00:15 -05:00
Damien Le Moal
4aca56f8ea PCI: epf-test: Fix DMA transfer completion initialization
Reinitialize the transfer_complete DMA transfer completion before calling
tx_submit(), to avoid seeing the DMA transfer complete before the
completion is initialized, thus potentially losing the completion
notification.

Link: https://lore.kernel.org/r/20230415023542.77601-4-dlemoal@kernel.org
Fixes: 8353813c88 ("PCI: endpoint: Enable DMA tests for endpoints with DMA capabilities")
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Cc: stable@vger.kernel.org
2023-06-23 15:00:01 -05:00
Damien Le Moal
f6ec33979e PCI: endpoint: Move pci_epf_type_add_cfs() code
pci_epf_type_add_cfs() is called only from pci_ep_cfs_add_type_group() in
drivers/pci/endpoint/pci-ep-cfs.c, so there is no need to export this
function.  Move its code from pci-epf-core.c to pci-ep-cfs.c as a static
function.

Link: https://lore.kernel.org/r/20230415023542.77601-3-dlemoal@kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
2023-06-23 14:59:38 -05:00
Damien Le Moal
70b3740f2c PCI: endpoint: Automatically create a function specific attributes group
A PCI endpoint function driver can define function specific attributes
under its function configfs directory using the add_cfs() endpoint driver
operation. This is done by tying up the mkdir operation for the function
configfs directory to a call to the add_cfs() operation.  However, there
are no checks preventing the user from repeatedly creating function
specific attribute directories with different names, resulting in the same
endpoint specific attributes group being added multiple times, which also
result in an invalid reference counting for the attribute groups. E.g.,
using the pci-epf-ntb function driver as an example, the user creates the
function as follows:

  $ modprobe pci-epf-ntb
  $ cd /sys/kernel/config/pci_ep/functions/pci_epf_ntb
  $ mkdir func0
  $ tree func0
  func0/
  |-- baseclass_code
  |-- cache_line_size
  |-- ...
  `-- vendorid

  $ mkdir func0/attrs
  $ tree func0
  func0/
  |-- attrs
  |   |-- db_count
  |   |-- mw1
  |   |-- mw2
  |   |-- mw3
  |   |-- mw4
  |   |-- num_mws
  |   `-- spad_count
  |-- baseclass_code
  |-- cache_line_size
  |-- ...
  `-- vendorid

At this point, the function can be started by linking the EP controller.
However, if the user mistakenly creates again a directory:

  $ mkdir func0/attrs2
  $ tree func0
  func0/
  |-- attrs
  |   |-- db_count
  |   |-- mw1
  |   |-- mw2
  |   |-- mw3
  |   |-- mw4
  |   |-- num_mws
  |   `-- spad_count
  |-- attrs2
  |   |-- db_count
  |   |-- mw1
  |   |-- mw2
  |   |-- mw3
  |   |-- mw4
  |   |-- num_mws
  |   `-- spad_count
  |-- baseclass_code
  |-- cache_line_size
  |-- ...
  `-- vendorid

The endpoint function specific attributes are duplicated and cause a crash
when the endpoint function device is torn down:

  refcount_t: addition on 0; use-after-free.
  WARNING: CPU: 2 PID: 834 at lib/refcount.c:25 refcount_warn_saturate+0xc8/0x144
  CPU: 2 PID: 834 Comm: rmdir Not tainted 6.3.0-rc1 #1
  Hardware name: Pine64 RockPro64 v2.1 (DT)
  pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
  ...
  Call trace:
  refcount_warn_saturate+0xc8/0x144
  config_item_get+0x7c/0x80
  configfs_rmdir+0x17c/0x30c
  vfs_rmdir+0x8c/0x204
  do_rmdir+0x158/0x184
  __arm64_sys_unlinkat+0x64/0x80
  invoke_syscall+0x48/0x114
  ...

Fix this by modifying pci_epf_cfs_work() to execute the new function
pci_ep_cfs_add_type_group() which itself calls pci_epf_type_add_cfs() to
obtain the function specific attribute group and the group name (directory
name) from the endpoint function driver. If the function driver defines an
attribute group, pci_ep_cfs_add_type_group() then proceeds to register this
group using configfs_register_group(), thus automatically exposing the
function type specific configfs attributes to the user. E.g.:

  $ modprobe pci-epf-ntb
  $ cd /sys/kernel/config/pci_ep/functions/pci_epf_ntb
  $ mkdir func0
  $ tree func0
  func0/
  |-- baseclass_code
  |-- cache_line_size
  |-- ...
  |-- pci_epf_ntb.0
  |   |-- db_count
  |   |-- mw1
  |   |-- mw2
  |   |-- mw3
  |   |-- mw4
  |   |-- num_mws
  |   `-- spad_count
  |-- primary
  |-- ...
  `-- vendorid

With this change, there is no need for the user to create or delete
directories in the endpoint function attributes directory. The
pci_epf_type_group_ops group operations are thus removed.

Also update the documentation for the pci-epf-ntb and pci-epf-vntb function
drivers to reflect this change, removing the explanations showing the need
to manually create the sub-directory for the function specific attributes.

Link: https://lore.kernel.org/r/20230415023542.77601-2-dlemoal@kernel.org
Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
2023-06-23 14:57:59 -05:00
Shunsuke Mie
37587673cd PCI: endpoint: Fix a Kconfig prompt of vNTB driver
vNTB driver and NTB driver have same Kconfig prompt. Changed to make it
distinguishable.

Link: https://lore.kernel.org/r/20230202103832.2038286-1-mie@igel.co.jp
Fixes: e35f56bb03 ("PCI: endpoint: Support NTB transfer between RC and EP")
Signed-off-by: Shunsuke Mie <mie@igel.co.jp>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
2023-06-23 14:57:47 -05:00
Evan Quan
1af3d0a8e8 drm/amd/pm: update the LC_L1_INACTIVITY setting to address possible noise issue
It is proved that insufficient LC_L1_INACTIVITY setting can cause audio
noise on some platform. With the LC_L1_INACTIVITY increased to 4ms, the
issue can be resolved.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:46:22 -04:00
Aric Cyr
111c1813a1 drm/amd/display: 3.2.240
This version brings along the following:
- DCN314 fixes
- DCN32x fixes
- New fast update sequence enablement
- DC mode clock switching enablement for DCN32x
- DP link loss fix
- New debugfs entry to set MST link settings

Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:46:12 -04:00
Alvin Lee
12a6e62bfd drm/amd/display: Enable dc mode clock switching for DCN32x
- DC mode clock switch interface was previously only executed
  for DCN303. Enable it for DCN32x so that the interface is called
  correctly
- Assign function pointers for DCN32x that are used in the dc mode
  interface
- Update the dc mode interface to work generically for each ASIC
- In update_clocks, make sure to consider softmax if we're in DC mode

Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:45:57 -04:00
Ilya Bakoulin
ed83fe2abc drm/amd/display: Fix 128b132b link loss handling
[Why]
We don't check 128b132b-specific bits in LANE_ALIGN_STATUS_UPDATED DPCD
registers when parsing link loss status, which can cause us to miss a
link loss notification from some sinks.

[How]
Add a 128b132b-specific status bit check.

Cc: stable@vger.kernel.org # 6.3+
Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Ilya Bakoulin <ilya.bakoulin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:45:48 -04:00
Dmytro Laktyushkin
effee878a8 drm/amd/display: fix odm k2 div calculation
Correct setting is div by 2 for odm. Seamless odm transitions
are enabled with enable_dp_dig_pixel_rate_div_policy debug flag.

Fixes: a2c7356f52 ("drm/amd/display: fix pixel rate update sequence")
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:44:48 -04:00
Fangzhi Zuo
0250a7145e drm/amd/display: Add MST Preferred Link Setting Entry
When using debugfs to change MST link settings, we need to wait until
the next stream update to apply the preferred link setting. So, trigger
a hotplug event right after the preferred link setting is applied.

Reviewed-by: Wayne Lin <wayne.lin@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:44:36 -04:00
Leo Chen
26518b3918 drm/amd/display: disable seamless boot if force_odm_combine is enabled
[Why & How]
Having seamless boot on while forcing debug option ODM combine 2 to 1
will cause some corruptions because of some missing programmings.

Cc: stable@vger.kernel.org # 6.1+
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Leo Chen <sancchen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:44:28 -04:00
Austin Zheng
2bf0ce3bec drm/amd/display: Disable DC Mode Capping On DCN321
Why:
Limiting clocks to DC mode max results in some
display modes to no longer be supported

How:
Disable the path that limits the clock values

Fixes: 3b718dcaf1 ("drm/amd/display: Filter out AC mode frequencies on DC mode systems")
Reviewed-by: Martin Leung <martin.leung@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Austin Zheng <austin.zheng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:43:32 -04:00
Alvin Lee
724617b94b drm/amd/display: enable the new fast update path for supported ASICs
The new fast update sequence is now supported on some ASICs. So, enable
it by default for all applicable ASICs.

Reviewed-by: Samson Tam <samson.tam@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:43:25 -04:00
Alvin Lee
873bbf2da2 drm/amd/display: Clear update flags at end of flip
Clear update flags so the next flip does not have any redundant
programming (if a subsequent flip does not have a stream or plane
update, the update flags are not cleared).

Fixes: 0baae62463 ("drm/amd/display: Refactor fast update to use new HWSS build sequence")
Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:42:21 -04:00
Alvin Lee
c8f2935418 drm/amd/display: Fix pipe check condition for manual trigger
Condition for programming manually trigger used the wrong pipe (always
used top pipe instead of the one we are iterating through).

Fixes: 0baae62463 ("drm/amd/display: Refactor fast update to use new HWSS build sequence")
Reviewed-by: Samson Tam <samson.tam@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:41:04 -04:00
Sridevi Arvindekar
4a87495a82 drm/amd/display: add missing ABM registers
[Why]
We are currently missing some ABM registers.

[How]
Add the missing registers to dce_abm.h.

Reviewed-by: Jun Lei <jun.lei@amd.com>
Reviewed-by: Chris Park <chris.park@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Sridevi Arvindekar <sridevi.arvindekar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:40:57 -04:00
Austin Zheng
c5f78ea8d7 drm/amd/display: Add Clock Table Entry With Max DC Values
Why:
Certain display configs resulted in underflow

How:
Add an entry containing all max DC clock timings

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Austin Zheng <austin.zheng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:40:50 -04:00
Sung-huai Wang
0f48a4b836 drm/amd/display: add a NULL pointer check
[Why & How]

We have to check if stream is properly initialized before calling
find_matching_pll(), otherwise we might end up trying to deferecence a
NULL pointer.

Cc: stable@vger.kernel.org # 6.1+
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Sung-huai Wang <danny.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:40:40 -04:00
Daniel Miess
a99a4ff6ef Partially revert "drm/amd/display: Fix possible underflow for displays with large vblank"
This partially reverts commit de231189e7 ("drm/amd/display: Fix
possible underflow for displays with large vblank").

[Why]
The increased value of VBlankNomDefaultUS causes underflow at the
desktop of an IP KVM setup

[How]
Change the value from 800 back to 668

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:39:15 -04:00
Daniel Miess
bf0097c5c9 Revert "drm/amd/display: Move DCN314 DOMAIN power control to DMCUB"
This reverts commit e383b12709.

Controling hubp power gating using the DMCUB isn't stable so we
are reverting this change to move control back into the driver.

Cc: stable@vger.kernel.org # 6.3+
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:38:31 -04:00
Daniel Miess
85e41f1ed5 drm/amd/display: disable RCO for DCN314
[Why]
RCO is causing error messages on some DCN314 systems

[How]
Force disable RCO for DCN314

Fixes: 17fbdbda9c ("drm/amd/display: Enable dcn314 DPP RCO")
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:37:48 -04:00
Daniel Miess
65dae8ff4c drm/amd/display: disable power gating for DCN314
[Why]
Power gating is causing error messages on some DCN314 systems

[How]
Force disable power gating for DCN314

Fixes: 4cc1cebe08 ("drm/amd/display: Re-enable DPP/HUBP Power Gating")
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Acked-by: Hamza Mahfooz <hamza.mahfooz@amd.com>
Signed-off-by: Daniel Miess <daniel.miess@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:36:56 -04:00
Zhigang Luo
acbe761046 drm/amdgpu: Skip TMR for MP0_HWIP 13.0.6
For SRIOV VF, no TMR needed.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Zhigang Luo <Zhigang.Luo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:36:48 -04:00
Nathan Chancellor
44762718b3 drm/amdgpu: Move clocks closer to its only usage in amdgpu_parse_cg_state()
After commit 8020f0f931 ("drm/amd/amdgpu: enable W=1 for amdgpu"),
there is an instance of -Wunused-const-variable when CONFIG_DEBUG_FS is
disabled:

  drivers/gpu/drm/amd/amdgpu/../pm/amdgpu_pm.c:38:34: error: unused variable 'clocks' [-Werror,-Wunused-const-variable]
     38 | static const struct cg_flag_name clocks[] = {
        |                                  ^
  1 error generated.

clocks is only used when CONFIG_DEBUG_FS is set, so move the definition
into the CONFIG_DEBUG_FS block right above its only usage to clear up
the warning.

Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:36:12 -04:00
Evan Quan
fd21987274 drm/amd/pm: revise the ASPM settings for thunderbolt attached scenario
Also, correct the comment for NAVI10_PCIE__LC_L1_INACTIVITY_TBT_DEFAULT
as 0x0000000E stands for 400ms instead of 4ms.

Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:36:03 -04:00
Samuel Pitoiset
ea2c3c0855 drm/amdgpu: fix clearing mappings for BOs that are always valid in VM
Per VM BOs must be marked as moved or otherwise their ranges are not
updated on use which might be necessary when the replace operation
splits mappings.

This fixes random GPU hangs when replacing sparse mappings from the
userspace, while OP_MAP/OP_UNMAP works fine because always valid BOs
are correctly handled there.

Cc: stable@vger.kernel.org
Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:35:51 -04:00
Alex Sierra
03d400e760 drm/amdkfd: set coherent host access capability flag
This flag determines whether the host possesses coherent access to
the memory of the device.

Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:35:34 -04:00
Lijo Lazar
184d838482 drm/amdgpu: Add vbios attribute only if supported
Not all devices carry VBIOS version information. Add the device
attribute only if supported.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:35:26 -04:00
Alex Deucher
c09b3bf736 drm/amdgpu/atomfirmware: fix LPDDR5 width reporting
LPDDR5 channels are 32 bit rather than 64, report the width properly
in the log.

v2: Only LPDDR5 are 32 bits per channel.  DDR5 is 64 bits per channel

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2468
Acked-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:35:16 -04:00
Nathan Chancellor
4e3f85d1c0 drm/amdgpu: Remove CONFIG_DEBUG_FS guard around body of amdgpu_rap_debugfs_init()
After commit 8020f0f931 ("drm/amd/amdgpu: enable W=1 for amdgpu"),
there is an instance of -Wunused-const-variable when CONFIG_DEBUG_FS is
disabled:

  drivers/gpu/drm/amd/amdgpu/amdgpu_rap.c:110:37: error: unused variable 'amdgpu_rap_debugfs_ops' [-Werror,-Wunused-const-variable]
    110 | static const struct file_operations amdgpu_rap_debugfs_ops = {
        |                                     ^
  1 error generated.

There is no reason for the body of this function to be guarded when
CONFIG_DEBUG_FS is disabled, as debugfs_create_file() is a stub that
just returns an error pointer in that situation. Remove the preprocessor
guards so that the variable never appears unused, while not changing
anything at run time.

Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:33:09 -04:00
Mingtong Bao
3ec61983aa drm/amd/pm: remove unneeded variable
fix the following coccicheck warning:

drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c:1657:14-18: Unneeded variable: "size".

Signed-off-by: Mingtong Bao <baomingtong001@208suo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:33:05 -04:00
Lijo Lazar
025654ae42 drm/amdgpu: Move calculation of xcp per memory node
Its value is required for finding the memory id of xcp.

Fixes: d26ea1b346 ("drm/amdgpu: Add xcp manager num_xcp_per_mem_partition")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:31:55 -04:00
Jiadong Zhu
ef3c36a6e0 drm/amdgpu: Skip mark offset for high priority rings
Only low priority rings are using chunks to save the offset.
Bypass the mark offset callings from high priority rings.

Signed-off-by: Jiadong Zhu <Jiadong.Zhu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:31:50 -04:00
Wang Ming
1ff310b97f amd/display/dc: remove repeating expression
Identify issues that arise by using the tests/doubletest.cocci
semantic patch. Need to remove duplicate expression in if statement.

Signed-off-by: Wang Ming <machel@vivo.com>
Reviewed-by: Ammar Faizi <ammarfaizi2@gnuweeb.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-23 15:31:42 -04:00
Jiri Pirko
29e4c95fae net/mlx5: Remove pointless vport lookup from mlx5_esw_check_port_type()
As xa_get_mark() returns false in case the entry is not present,
no need to redundantly check if vport is present. Remove the lookup.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:35 -07:00
Jiri Pirko
899862b653 net/mlx5: Remove redundant check from mlx5_esw_query_vport_vhca_id()
Since mlx5_esw_query_vport_vhca_id() could be called either from
mlx5_esw_vport_enable() or mlx5_esw_vport_disable() where the
the check is done, this is always false here.
Remove the redundant check.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:34 -07:00
Jiri Pirko
0d0946d648 net/mlx5: Remove redundant is_mdev_switchdev_mode() check from is_ib_rep_supported()
is_mdev_switchdev_mode() check is done in is_eth_rep_supported().
Function is_ib_rep_supported() calls is_eth_rep_supported().
Remove the redundant check from it.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:34 -07:00
Jiri Pirko
61955da523 net/mlx5: Remove redundant MLX5_ESWITCH_MANAGER() check from is_ib_rep_supported()
MLX5_ESWITCH_MANAGER() check is done in is_eth_rep_supported().
Function is_ib_rep_supported() calls is_eth_rep_supported().
Remove the redundant check from it.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:34 -07:00
Roi Dayan
15ddd72ee3 net/mlx5e: E-Switch, Fix shared fdb error flow
On error flow resources being freed in esw_master_egress_destroy_resources()
but pointers not being set to null if error flow is from creating a
bounce rule. Then in esw_acl_egress_ofld_cleanup() we try to access already
freed pointers. Fix it by resetting the pointers to null.
Also if error is from creating a second or later bounce rule then the
flow group and table being used and cannot and should not be freed.
Add a check to destroy the flow group and table if there are no bounce
rules.

mlx5_core.sf mlx5_core.sf.2: mlx5_destroy_flow_group:2306:(pid 2235): Flow group 4 wasn't destroyed, refcount > 1
mlx5_core.sf mlx5_core.sf.2: mlx5_destroy_flow_table:2295:(pid 2235): Flow table 3 wasn't destroyed, refcount > 1

Fixes: 5e0202eb49 ("net/mlx5: E-switch, Handle multiple master egress rules")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:34 -07:00
Roi Dayan
ae4de89493 net/mlx5e: Remove redundant comment
The function comment says what it is and the comment
is redundant.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:34 -07:00
Roi Dayan
4575ab3b7d net/mlx5e: E-Switch, Pass other_vport flag if vport is not 0
When creating flow table for shared fdb resources, there is
only need to pass other_vport flag if vport is not 0 or
if the port is ECPF in BlueField.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:34 -07:00
Roi Dayan
70c3643839 net/mlx5e: E-Switch, Use xarray for devcom paired device index
To allow devcom events on E-Switch that is not a vport group manager,
use vhca id as an index instead of device index which might be shared
between several E-Switches. for example SF and its PF.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:33 -07:00
Roi Dayan
1552e9b518 net/mlx5e: E-Switch, Add peer fdb miss rules for vport manager or ecpf
Add peer fdb rules for E-Switch that are vport managers or ecpf device.
It is not needed for other devices.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:33 -07:00
Roi Dayan
1da9f36252 net/mlx5e: Use vhca_id for device index in vport rx rules
Device index is like PF index and limited to max physical ports.
For example, SFs created under PF the device index is the PF device index.
Use vhca_id which gets the FW index per vport, for vport rx rules
and vport pair events.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:33 -07:00
Roi Dayan
8ec91f5d07 net/mlx5: Lag, Remove duplicate code checking lag is supported
Remove duplicate function for checking if device has lag support.

Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:33 -07:00
Dan Carpenter
690ad62fc6 net/mlx5: Fix error code in mlx5_is_reset_now_capable()
The mlx5_is_reset_now_capable() function returns bool, not negative
error codes.  So if fast teardown is not supported it should return
false instead of -EOPNOTSUPP.

Fixes: 92501fa6e4 ("net/mlx5: Ack on sync_reset_request only if PF can do reset_now")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2023-06-23 12:27:33 -07:00