Add support for the VIDIOC_QUERYCAP V4L2 iocts for the encoder video
device to report core driver capabilities.
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Add V4L2 format handling support for the encoder by adding
implementation of ENUM/S/G/TRY_FMT with necessary hooks.
This ensures that the encoder supports format negotiation consistent
with V4L2 expectation, enabling userspace applications to configure
resolution, pixel format and buffer layout properly.
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Add support for registering a V4L2 encoder video device to the iris
driver. The encoder device is registered with the name
"qcom-iris-encoder".
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Previously, the format validation relied on an array of supported
formats, which only listed formats for the OUTPUT plane. This caused
failures when validating formats for the CAPTURE plane.
Update the check to validate against the only supported format on the
CAPTURE plane, which is NV12.
Fixes: fde6161d91 ("media: iris: Add HEVC and VP9 formats for decoder")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Improve drain handling by ensuring the LAST flag is attached to final
capture buffer when drain response is received from the firmware.
Previously, the driver failed to attach the V4L2_BUF_FLAG_LAST flag when
a drain response was received from the firmware, relying on userspace to
mark the next queued buffer as LAST. This update fixes the issue by
checking the pending drain status, attaching the LAST flag to the
capture buffer received from the firmware (with EOS attached), and
returning it to the V4L2 layer correctly.
Fixes: d09100763b ("media: iris: add support for drain sequence")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Firmware can handle a dummy address for buffers with the EOS flag. To
ensure consistent behavior across all codecs, update the drain
command to always send a dummy buffer address.
This makes the drain handling uniform and avoids any codec specific
assumptions.
Fixes: 478c447861 ("media: iris: Add codec specific check for VP9 decoder drain handling")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
For HFI Gen1, the instances substate is changed to LOAD_RESOURCES only
when a START command is issues to the firmware. If STOP is called
without a prior START, the firmware may reject the command and throw
some erros.
Handle this by adding a substate check before issuing STOP command to
the firmware.
Fixes: 11712ce70f ("media: iris: implement vb2 streaming ops")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Remove earlier complex conditional checks in the non-streaming path that
attempted to verify if stop was called on a plane that was previously
started. These explicit checks are redundant, as vb2 already ensures
that stop is only called on ports that have been started, maintaining
correct buffer state management.
Fixes: 11712ce70f ("media: iris: implement vb2 streaming ops")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Update the vbuf flags appropriately in error cases before calling
v4l2_m2m_buf_done(). Previously, the flag update was skippied in error
scenario, which could result in incorrect state reporting for buffers.
Fixes: 17f2a485ca ("media: iris: implement vb2 ops for buf_queue and firmware response")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Currently, internal buffers are destroyed only if 'PENDING_RELEASE' flag
is set when a release response is received from the firmware, which is
incorrect. Internal buffers should always be destroyed when the firmware
explicitly releases it, regardless of whether the 'PENDING_RELEASE' flag
was set by the driver. This is specially important during force-stop
scenarios, where the firmware may release buffers without driver marking
them for release.
Fix this by removing the incorrect check and ensuring all buffers are
properly cleaned up.
Fixes: 73702f45db ("media: iris: allocate, initialize and queue internal buffers")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
A client (e.g., GST for encoder) can initiate streaming on the capture
port before the output port, causing the instance state to transition to
OUTPUT_STREAMING. When streaming is subsequently started on the output
port, the instance state advances to STREAMING, and the substate should
transition to LOAD_RESOURCES.
Previously, the code blocked the substate transition to LOAD_RESOURCES
if the instance state was OUTPUT_STREAMING. This update modifies the
logic to permit the substate transition to LOAD_RESOURCES when the
instance state is OUTPUT_STREAMING, thereby supporting this client
streaming sequence.
Fixes: 547f7b8c50 ("media: iris: add check to allow sub states transitions")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
The previous check to block capture port streaming before output port
was incorrect and caused some valid usecase to fail. While removing that
check allows capture port to enter streaming independently, it also
introduced firmware errors due to premature queuing of DPB buffers
before the firmware session was fully started which happens only when
streamon is called on output port.
Fix this by deferring DPB buffer queuing to the firmware until both
capture and output are streaming and state is 'STREAMING'.
Fixes: 11712ce70f ("media: iris: implement vb2 streaming ops")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
One internal buffer which is allocated only once per session was not
being freed during session close because it was not being tracked as
part of internal buffer list which resulted in a memory leak.
Add the necessary logic to explicitly free the untracked internal buffer
during session close to ensure all allocated memory is released
properly.
Fixes: 73702f45db ("media: iris: allocate, initialize and queue internal buffers")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Add error reporting for unreleased PERSIST internal buffers in
iris_check_num_queued_internal_buffers(). This ensures all buffer types
are checked and logged if not freed during session close, helping to
detect memory leaks and improve driver robustness. No change to buffer
lifecycle or allocation logic.
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Initialize the count variable to zero before counting unreleased
internal buffers in iris_check_num_queued_internal_buffers().
This prevents stale values from previous iterations and ensures accurate
error reporting for each buffer type. Without this initialization, the
count could accumulate across types, leading to incorrect log messages.
Fixes: d2abb1ff5a ("media: iris: Verify internal buffer release on close")
Cc: stable@vger.kernel.org
Reviewed-by: Vikash Garodia <quic_vgarodia@quicinc.com>
Tested-by: Vikash Garodia <quic_vgarodia@quicinc.com> # X1E80100
Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-HDK
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8650-HDK
Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
Tested-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org> # x1e80100-crd
Signed-off-by: Bryan O'Donoghue <bod@kernel.org>
Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
The default width is 4 bytes for "syscon" devices, so "reg-io-width" is
redundant and can be dropped.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Andrew Jeffery <andrew@codeconstruct.com.au>
The variable num_cqs is of type u8 and does not require be16_to_cpu
conversion, so the redundant code is removed.
Signed-off-by: Qiang Liu <liuqiang@kylinos.cn>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Nitin Rawat <nitin.rawat@oss.qualcomm.com> says:
The patch series simplifies the UFS MCQ (Multi Circular Queue)
resource mapping in the Qualcomm UFS host controller driver by
replacing the complex multi-resource approach with a streamlined
single-resource implementation.
The current MCQ implementation uses multiple separate resource
mappings (RES_UFS, RES_MCQ, RES_MCQ_SQD, RES_MCQ_VS) with dynamic
resource allocation, which increases code complexity and potential for
resource mapping errors. This approach also doesn't align with the
device tree binding specification that defines a single 'mcq' memory
region.
Replace the multi-resource mapping with a single "mcq" resource that
encompasses the entire MCQ configuration space. The doorbell registers
(SQD, CQD, SQIS, CQIS) are accessed using predefined offsets relative
to the MCQ base address, providing clearer memory layout organization.
Tested on Qualcomm platforms SM8650 and SM8750 with UFS MCQ enabled.
Changes from v3:
1. Addressed Krzysztof comment to separate device tree and driver
patch independently in different patch series. This series caters
driver changes.
2. Addressed Manivannan's change to update commit text and remove
redundant null check in mcq code.
3. Addressed Manivannan's to Update few offsets as fixed definition
instead of enum.
Changes from v2:
1. Removed dt-bindings patch as existing binding supports required
reg-names format.
2. Added patch to refactor MCQ register dump logic for new resource
mapping.
3. Added patch to remove unused ufshcd_res_info structure from UFS core.
4. Changed reg-names from "ufs_mem" to "std" in device tree patches.
5. Reordered patches with driver changes first, then device tree changes.
6. Updated SM8750 MCQ region size from 0x2000 to 0x15000
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Refactor MCQ register dump to align with the new resource mapping. As
part of refactor, below changes are done:
- Update ufs_qcom_dump_regs() function signature to accept direct
base address instead of resource ID enum
- Modify ufs_qcom_dump_mcq_hci_regs() to use hba->mcq_base and
calculated addresses from MCQ operation info
- Replace enum ufshcd_res with direct memory-mapped I/O addresses
Additionally remove the ufshcd_res_info structure and associated enum
ufshcd_res definitions from the UFS host controller header. These were
previously used for MCQ resource mapping but are no longer needed
following recent refactoring to use direct base addresses instead of
multiple separate resource regions.
Signed-off-by: Nitin Rawat <quic_nitirawa@quicinc.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The current MCQ resource configuration involves multiple resource
mappings and dynamic resource allocation.
Simplify the resource mapping by directly mapping the single "mcq"
resource from device tree to hba->mcq_base instead of mapping multiple
separate resources (RES_UFS, RES_MCQ, RES_MCQ_SQD, RES_MCQ_VS).
It also uses predefined offsets for MCQ doorbell registers (SQD, CQD,
SQIS, CQIS) relative to the MCQ base,providing clearer memory layout
clarity.
Additionally update vendor-specific register offset UFS_MEM_CQIS_VS
offset from 0x8 to 0x4008 to align with the hardware programming guide.
The new approach assumes the device tree provides a single "mcq"
resource that encompasses the entire MCQ configuration space, making the
driver more maintainable and less prone to resource mapping errors.
The change aligns the driver implementation with the device tree binding
specification, which defines a single 'mcq' memory region rather than
multiple separate regions.
Co-developed-by: Ram Kumar Dwivedi <quic_rdwivedi@quicinc.com>
Signed-off-by: Ram Kumar Dwivedi <quic_rdwivedi@quicinc.com>
Signed-off-by: Nitin Rawat <quic_nitirawa@quicinc.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Peter Wang <peter.wang@mediatek.com> says:
These patches collectively enhance the UFS host driver's reliability,
power management efficiency, and error recovery mechanisms on MediaTek
platforms. They address critical issues and introduce optimizations
that improve system stability and performance.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Adjust the timing of device power control to ensure low power mode (LPM)
is entered only after VCC is turned off. Prevent VCCQ/VCCQ2 from
entering LPM prematurely, ensuring proper power management and device
stability.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Resolve the issue of unbalanced IRQ enablement by setting the
'is_mcq_intr_enabled' flag after the first successful IRQ enablement.
Ensure proper tracking of the IRQ state and prevent potential mismatches
in IRQ handling.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Address the issue where the host does not send adapt to the device after
PA_Init success. Ensure the adapt process is correctly initiated for
devices with IP version MT6899 and above, resolving communication issues
between the host and device.
Signed-off-by: Alice Chao <alice.chao@mediatek.com>
Reviewed-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Optimize the recovery flow by returning an error code immediately if a
wait idle timeout occurs, rather than waiting for the link to reach the
up state. Shorten the recovery process and improve error handling
efficiency when idle state transitions fail.
Signed-off-by: Sanjeev Y <sanjeev.y@mediatek.com>
Reviewed-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Disable auto-hibern8 during power mode transitions to prevent unintended
entry into auto-hibern8. Restore the original auto-hibern8 timer value
after completing the power mode change to maintain system stability and
prevent potential issues during power state transitions.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Add support for UFS PHY runtime power management by probing the PHY
device and enabling its runtime PM. Ensure the correct sequence of
operations during suspend and resume: PHY suspend -> UFS suspend -> UFS
resume -> PHY resume. Improve power management efficiency and system
stability with this enhancement.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Correct the system resume flow by turning MTCMOS on before setting LPM
to false. During system suspend, set LPM to true and turn MTCMOS
off. Ensure proper power management and system stability with the
updated resume sequence.
Signed-off-by: Alice Chao <alice.chao@mediatek.com>
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Reviewed-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Refine the system power management (PM) flow by skipping low power mode
(LPM) and MTCMOS settings if runtime PM is already applied. Prevent
redundant operations to ensure a more efficient PM process.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Improve the recovery process for failed resume operations. Log the
device's power status and return 0 if both resume and recovery fail to
prevent I/O hang.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Improve the recovery process for hibernation exit failures. Trigger the
error handler and break the suspend operation to ensure effective
recovery from hibernation errors. Activate the error handling mechanism
by ufshcd_force_error_recovery and scheduling the error handler work.
Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Comment out unused field 'residual_count' in a couple of structures, and
with this, fix the following -Wflex-array-member-not-at-end warnings:
drivers/scsi/pm8001/pm8001_hwi.h:342:33: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
drivers/scsi/pm8001/pm80xx_hwi.h:561:32: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]
Reviewed-by: Jack Wang <jinpu.wang@ionos.com>
Reviewed-by: John Garry <john.g.garry@oracle.com>
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Previous checks incorrectly tested the DMA addresses (dma_handle) for
NULL. Since dma_alloc_coherent() returns the CPU (virtual) address, the
NULL check should be performed on the *_base_addr pointer to correctly
detect allocation failures.
Update the checks to validate sqe_base_addr and cqe_base_addr instead of
sqe_dma_addr and cqe_dma_addr.
Fixes: 4682abfae2 ("scsi: ufs: core: mcq: Allocate memory for MCQ mode")
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com>
Reviewed-by: Alim Akhtar <alim.akhtar@samsung.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Reviewed-by: Peter Wang <peter.wang@mediatek.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Disabling the AES core in Shared ICE is not supported during power
collapse for UFS Host Controller v5.0, which may lead to data errors
after Hibern8 exit. To comply with hardware programming guidelines and
avoid this issue, issue a sync reset to ICE upon power collapse exit.
Hence follow below steps to reset the ICE upon exiting power collapse
and align with Hw programming guide.
a. Assert the ICE sync reset by setting both SYNC_RST_SEL and
SYNC_RST_SW bits in UFS_MEM_ICE_CFG
b. Deassert the reset by clearing SYNC_RST_SW in UFS_MEM_ICE_CFG
Signed-off-by: Palash Kambar <quic_pkambar@quicinc.com>
Reviewed-by: Manivannan Sadhasivam <mani@kernel.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The 88e1510 PHY has an erratum where the phy downshift counter is not
cleared after phy being suspended(BMCR_PDOWN set) and then later
resumed(BMCR_PDOWN cleared). This can cause the gigabit link to
intermittently downshift to a lower speed.
Disabling and re-enabling the downshift feature clears the counter,
allowing the PHY to retry gigabit link negotiation up to the programmed
retry count times before downshifting. This behavior has been observed
on copper links.
Signed-off-by: Rohan G Thomas <rohan.g.thomas@altera.com>
Reviewed-by: Matthew Gerlach <matthew.gerlach@altera.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20250906-marvell_fix-v2-1-f6efb286937f@altera.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Wei Fang says:
====================
ptp: add pulse signal loopback support for debugging
Some PTP devices support looping back the periodic pulse signal for
debugging. For example, the PTP device of QorIQ platform and the NETC v4
Timer has the ability to loop back the pulse signal and record the extts
events for the loopback signal. So we can make sure that the pulse
intervals and their phase alignment are correct from the perspective of
the emitting PHC's time base. In addition, we can use this loopback
feature as a built-in extts event generator when we have no external
equipment which does that. Therefore, add the generic debugfs interfaces
to the ptp_clock driver. The first two patch are separated from the
previous patch set [1]. The third patch is new added.
[1]: https://lore.kernel.org/imx/20250827063332.1217664-1-wei.fang@nxp.com/ #patch 3 and 9
====================
Link: https://patch.msgid.link/20250905030711.1509648-1-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Since the generic debugfs interfaces for setting the periodic pulse
signal loopback have been added to the ptp_clock driver, so convert
the vendor-defined debugfs interfaces to the generic interfaces.
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://patch.msgid.link/20250905030711.1509648-4-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The NETC Timer supports looping back the output pulse signal of Fiper-n
into Trigger-n input, so that users can leverage this feature to validate
some other features without external hardware support. For example, users
can use it to test external trigger stamp (EXTTS). And users can combine
EXTTS with loopback mode to check whether the generation time of PPS is
aligned with an integral second of PHC, or the periodic output signal
(PTP_CLK_REQ_PEROUT) whether is generated at the specified time.
Since ptp_clock_info::perout_loopback() has been added to the ptp_clock
driver as a generic interface to enable or disable the periodic output
signal loopback, therefore, netc_timer_perout_loopback() is added as a
callback of ptp_clock_info::perout_loopback().
Test the generation time of PPS event:
$ echo 0 1 > /sys/kernel/debug/ptp0/perout_loopback
$ echo 1 > /sys/class/ptp/ptp0/pps_enable
$ testptp -d /dev/ptp0 -e 3
external time stamp request okay
event index 0 at 63.000000017
event index 0 at 64.000000017
event index 0 at 65.000000017
Test the generation time of the periodic output signal:
$ echo 0 1 > /sys/kernel/debug/ptp0/perout_loopback
$ echo 0 150 0 1 500000000 > /sys/class/ptp/ptp0/period
$ testptp -d /dev/ptp0 -e 3
external time stamp request okay
event index 0 at 150.000000014
event index 0 at 151.500000015
event index 0 at 153.000000014
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://patch.msgid.link/20250905030711.1509648-3-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
For some PTP devices, they have the capability to loop back the periodic
output signal for debugging, such as the ptp_qoriq device. So add the
generic interfaces to set the periodic output signal loopback, rather
than each vendor having a different implementation.
Show how many channels support the periodic output signal loopback:
$ cat /sys/kernel/debug/ptp<N>/n_perout_loopback
Enable the loopback of the periodic output signal of channel X:
$ echo <X> 1 > /sys/kernel/debug/ptp<N>/perout_loopback
Disable the loopback of the periodic output signal of channel X:
$ echo <X> 0 > /sys/kernel/debug/ptp<N>/perout_loopback
Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Wei Fang <wei.fang@nxp.com>
Link: https://patch.msgid.link/20250905030711.1509648-2-wei.fang@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Replace kzalloc() followed by copy_from_user() with memdup_user_nul() to
improve and simplify sdebug_error_write().
No functional changes intended.
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
This ethtool counter is meant to help with observing how many times the
congestion event was triggered but on query there was no state change.
This would help to indicate when a work item was scheduled to run too
late and in the meantime the congestion state changed back to previous
state.
While at it, do a driveby typo fix in documentation for
pci_bw_inbound_high.
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/1757237976-531416-3-git-send-email-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add devlink driverinit parameters for configuring the thresholds for
PCIe congestion events. These parameters are registered only when the
firmware supports this feature.
Update the mlx5 devlink docs as well on these new params.
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/1757237976-531416-2-git-send-email-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Saeed Mahameed says:
====================
devlink, mlx5: Add new parameters for link management and SRIOV/eSwitch configurations [part]
This patch series introduces several devlink parameters improving device
configuration capabilities, link management, and SRIOV/eSwitch, by adding
NV config boot time parameters.
Implement the following parameters:
a) total_vfs Parameter:
-----------------------
Adds support for managing the number of VFs (total_vfs) and enabling
SR-IOV (enable_sriov for mlx5) through devlink. These additions enhance
user control over virtualization features directly from standard kernel
interfaces without relying on additional external tools. total_vfs
functionality is critical for environments that require flexible num VF
configuration.
b) CQE Compression Type:
------------------------
Introduces a new devlink parameter, cqe_compress_type, to configure the
rate of CQE compression based on PCIe bus conditions. This setting
provides a balance between compression efficiency and overall NIC
performance under different traffic loads.
====================
Link: https://patch.msgid.link/20250907012953.301746-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Some devices support both symmetric (same value for all PFs) and
asymmetric, while others only support symmetric configuration. This
implementation prefers asymmetric, since it is closer to the devlink
model (per function settings), but falls back to symmetric when needed.
Example usage:
devlink dev param set pci/0000:01:00.0 name total_vfs value <u16> cmode permanent
devlink dev reload pci/0000:01:00.0 action fw_activate
echo 1 >/sys/bus/pci/devices/0000:01:00.0/remove
echo 1 >/sys/bus/pci/rescan
cat /sys/bus/pci/devices/0000:01:00.0/sriov_totalvfs
Signed-off-by: Vlad Dumitrescu <vdumitrescu@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Tested-by: Kamal Heib <kheib@redhat.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250907012953.301746-5-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Selects which algorithm should be used by the NIC in order to decide rate of
CQE compression dependeng on PCIe bus conditions.
Supported values:
1) balanced, merges fewer CQEs, resulting in a moderate compression ratio
but maintaining a balance between bandwidth savings and performance
2) aggressive, merges more CQEs into a single entry, achieving a higher
compression rate and maximizing performance, particularly under high
traffic loads.
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20250907012953.301746-3-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
