soc: ti: k3-ringacc: add AM64 DMA rings support.

The DMAs in AM64 have built in rings compared to AM654/J721e/J7200 where a
separate and generic ringacc is used.

The ring SW interface is similar to ringacc with some major architectural
differences, like

They are part of the DMA (BCDMA or PKTDMA).

They are dual mode rings are modeled as pair of Rings objects which has
common configuration and memory buffer, but separate real-time control
register sets for each direction mem2dev (forward) and dev2mem (reverse).

The ringacc driver must be initialized for DMA rings use with
k3_ringacc_dmarings_init() as it is not an independent device as ringacc
is.

AM64 rings must be requested only using k3_ringacc_request_rings_pair(),
and forward ring must always be initialized/configured. After this any
other Ringacc APIs can be used without any callers changes.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-17-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: Add support for k3 event routers

In k3 architecture a DMA channel (in TR momde) can be triggered by global
events, origination from different modules.

The events for triggers can be sent from any module which is connected to
PSI-L fabric, but the event number to be sent is DMA channel specific, it
is only known after the channel itself is requested.

The router operation needs to be split up:
- route_allocate: configure the dma_spec for the DMA and store the
  configuration which is needed for the router's input
- set_event: callback used by the DMA driver to set the event number for
  the channel and enable the routing

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-16-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-psil: Add initial map for AM64

Add initial PSI-L map file for AM64.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-15-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-psil: Extend psil_endpoint_config for K3 PKTDMA

Additional fields needed for K3 PKTDMA to be able to handle the mapped
channels (channels are locked to handle specific threads) and flow ranges
for these mapped threads.
PKTDMA also introduces tflow for tx channels which can not be found in
K3 UDMA architecture.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-14-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dma: ti: Add document for K3 PKTDMA

New binding document for
Texas Instruments K3 Packet DMA (PKTDMA).

PKTDMA is introduced as part of AM64.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201208090440.31792-13-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dma: ti: Add document for K3 BCDMA

New binding document for
Texas Instruments K3 Block Copy DMA (BCDMA).

BCDMA is introduced as part of AM64.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201208090440.31792-12-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: dmatest: Use dmaengine_get_dma_device

By using the dmaengine_get_dma_device() to get the device for
dma_api use, the dmatest can support per channel coherency if it is
supported by the DMA controller.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-11-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: doc: client: Update for dmaengine_get_dma_device() usage

Client drivers should use the dmaengine_get_dma_device(chan) to get the
device pointer which should be used for DMA API for allocations and
mapping.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-10-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: Add support for per channel coherency handling

If the DMA device supports per channel coherency configuration (a channel
can be configured to have coherent or not coherent view) then a single
device (the DMA controller's device) can not be used for dma_api for all
channels as channels can have different coherency.

Introduce custom_dma_mapping flag for the dma_chan and a new helper to get
the device pointer to be used for dma_api for the given channel.

Client drivers should be updated to be able to support per channel
coherency by:

- dma_map_single(chan->device->dev, ptr, size, DMA_TO_DEVICE);
+ struct device *dma_dev = dmaengine_get_dma_device(chan);
+
+ dma_map_single(dma_dev, ptr, size, DMA_TO_DEVICE);

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-9-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: of-dma: Add support for optional router configuration callback

Additional configuration for the DMA event router might be needed for a
channel which can not be done during device_alloc_chan_resources callback
since the router information is not yet present for the drivers.

If there is a need for additional configuration for the channel if DMA
router is in use, then the driver can implement the device_router_config
callback.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-8-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma-glue: Configure the dma_dev for rings

Rings in RING mode should be using the DMA device for DMA API as in this
mode the ringacc will not access the ring memory in any ways, but the DMA
is.

Fix up the ring configuration and set the dma_dev unconditionally and let
the ringacc driver to select the correct device to use for DMA API.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-7-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma-glue: Get the ringacc from udma_dev

If of_xudma_dev_get() returns with the valid udma_dev then the driver
already got the ringacc, there is no need to execute
of_k3_ringacc_get_by_phandle() for each channel via the glue layer.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-6-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma-glue: Add function to get device pointer for DMA API

Glue layer users should use the device of the DMA for DMA mapping and
allocations as it is the DMA which accesses to descriptors and buffers,
not the clients

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-5-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma: Add support for second resource range from sysfw

Resource allocation via sysfw can use up to two ranges per resource subtype
to support more complex resource assignment, mainly for DMA channels.

Take the second range also into consideration when setting up the maps for
available resources.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-4-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma: Wait for peer teardown completion if supported

Set the TDTYPE if it is supported on the platform (j721e) which will cause
UDMAP to wait for the remote peer to finish the teardown before returning
the teardown completed message.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Tested-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-3-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma: Correct normal channel offset when uchan_cnt is not 0

According to different sections of the TRM, the hchan_cnt of CAP3 includes
the number of uchan in UDMA, thus the start offset of the normal channels
are hchan_cnt.

Fixes: daf4ad0499aa4 ("dmaengine: ti: k3-udma: Query throughput level information from hardware")
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-2-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

Merge tag 'tags/drivers_soc_for_5.11' into dmaengine/next

drivers: soc: TI SOC changes for 5.11

- ti_sci changes towards DMSS support
- Static warning fixes
- Kconfig update for Keystone ARM64 socs
- AM64X SOC family support

dmaengine: stm32-mdma: rework interrupt handler

To avoid multiple entries in MDMA interrupt handler for each flag&interrupt
enable, manage all flags set at once.

Signed-off-by: Amelie Delaunay <amelie.delaunay@st.com>
Link: https://lore.kernel.org/r/20201120143320.30367-5-amelie.delaunay@st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: stm32-dma: take address into account when computing max width

DMA_SxPAR or DMA_SxM0AR/M1AR registers have to be aligned on PSIZE or MSIZE
respectively. This means that bus width needs to be forced to 1 byte when
computed width is not aligned with address.

Signed-off-by: Amelie Delaunay <amelie.delaunay@st.com>
Link: https://lore.kernel.org/r/20201120143320.30367-4-amelie.delaunay@st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: stm32-dma: clean channel configuration when channel is freed

When dma_channel_release is called, it means that the channel won't be used
anymore with the configuration it had. To ensure a future client can safely
use the channel after it has been released, clean the configuration done
when channel was requested.

Signed-off-by: Amelie Delaunay <amelie.delaunay@st.com>
Link: https://lore.kernel.org/r/20201120143320.30367-3-amelie.delaunay@st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: stm32-dma: rework irq handler to manage error before xfer events

To better understand error that can be detected by the DMA controller,
manage the error flags before the transfer flags.
This way, it is possible to know if the FIFO error flag is set for an
over/underrun condition or a FIFO level error.
When a FIFO over/underrun condition occurs, the data is not lost because
peripheral request is not acknowledged by the stream until the over/
underrun condition is cleared. If this acknowledge takes too much time,
the peripheral itself may detect an over/underrun condition of its internal
buffer and data might be lost.
That's why in case the FIFO error flag is set, we check if the channel is
disabled or not, and if a Transfer Complete flag is set, which means that
the channel is disabled because of the end of transfer.
Because channel is disabled by hardware either by a FIFO level error, or by
an end of transfer.

Signed-off-by: Amelie Delaunay <amelie.delaunay@st.com>
Link: https://lore.kernel.org/r/20201120143320.30367-2-amelie.delaunay@st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: bam_dma: fix return of bam_dma_irq()

While performing suspend/resume, we were getting below kernel crash.

[   54.541672] [FTS][Info]gesture suspend...
[   54.605256] [FTS][Error][GESTURE]Enter into gesture(suspend) failed!
[   54.605256]
[   58.345850] irq event 10: bogus return value fffffff3
......

[   58.345966] [<ffff0000080830f0>] el1_irq+0xb0/0x124
[   58.345971] [<ffff000008085360>] arch_cpu_idle+0x10/0x18
[   58.345975] [<ffff0000081077f4>] do_idle+0x1ac/0x1e0
[   58.345979] [<ffff0000081079c8>] cpu_startup_entry+0x20/0x28
[   58.345983] [<ffff000008a80ed0>] rest_init+0xd0/0xdc
[   58.345988] [<ffff0000091c0b48>] start_kernel+0x390/0x3a4
[   58.345990] handlers:
[   58.345994] [<ffff0000085120d0>] bam_dma_irq

The reason for the crash we found is, bam_dma_irq() was returning
negative value when the device resumes in some conditions.

In addition, the irq handler should have one of the below return values.

IRQ_NONE            interrupt was not from this device or was not handled
IRQ_HANDLED         interrupt was handled by this device
IRQ_WAKE_THREAD     handler requests to wake the handler thread

Therefore, to resolve this crash, we have changed the return value to
IRQ_NONE.

Signed-off-by: Parth Y Shah <sparth1292@gmail.com>
Link: https://lore.kernel.org/r/1607322820-7450-1-git-send-email-sparth1292@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: add IAX configuration support in the IDXD driver

Add support to allow configuration of Intel Analytics Accelerator (IAX) in
addition to the Intel Data Streaming Accelerator (DSA). The IAX hardware
has the same configuration interface as DSA. The main difference
is the type of operations it performs. We can support the DSA and
IAX devices on the same driver with some tweaks.

IAX has a 64B completion record that needs to be 64B aligned, as opposed to
a 32B completion record that is 32B aligned for DSA. IAX also does not
support token management.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/160564555488.1834439.4261958859935360473.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dma: mtk-apdma: add bindings for MT8516 SOC

Add bindings to APDMA for MT8516 SoC. MT8516 is compatible with MT6577.

Signed-off-by: Fabien Parent <fparent@baylibre.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201209114736.70625-1-fparent@baylibre.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: qcom: Fix ADM driver kerneldoc markup

Update the kerneldoc function headers to fix build warnings:

drivers/dma/qcom/qcom_adm.c:180: warning: Function parameter or member 'chan' not described in 'adm_free_chan'
drivers/dma/qcom/qcom_adm.c:190: warning: Function parameter or member 'burst' not described in 'adm_get_blksize'
drivers/dma/qcom/qcom_adm.c:466: warning: Function parameter or member 'chan' not described in 'adm_terminate_all'
drivers/dma/qcom/qcom_adm.c:466: warning: Excess function parameter 'achan' description in 'adm_terminate_all'
drivers/dma/qcom/qcom_adm.c:503: warning: Function parameter or member 'achan' not described in 'adm_start_dma'

Signed-off-by: Jonathan McDowell <noodles@earth.li>
Link: https://lore.kernel.org/r/20201126184602.GA1008@earth.li
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: add ATS disable knob for work queues

With the DSA spec 1.1 update, a knob to disable ATS for individually is
introduced. Add enabling code to allow a system admin to make the
configuration through sysfs.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/160530810593.1288392.2561048329116529566.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: drop of_match_ptr and mark of_device_id table as maybe unused

The driver can match only via the DT table so the main table should be
always used and the of_match_ptr does not have any sense (this also
allows ACPI matching via PRP0001, even though it is not relevant here).

The secondary match of_device_id tables (passed to of_match_node) should
be marked as maybe unused to fix compile testing (!CONFIG_OF on x86_64)
warnings:

    drivers/dma/ti/dma-crossbar.c:125:34: warning:
        ‘ti_am335x_master_match’ defined but not used [-Wunused-const-variable=]
    drivers/dma/ti/dma-crossbar.c:22:34: warning:
        ‘ti_dma_xbar_match’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201120162303.482126-6-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: stm32: mark of_device_id table as maybe unused

The driver uses a second of_device_id table in the probe()
function by passing it to of_match_node().  This code will be a no-op
for compile testing (!CONFIG_OF on x86_64):

    drivers/dma/stm32-dmamux.c:171:34: warning:
        ‘stm32_stm32dma_master_match’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201120162303.482126-5-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: sf: drop of_match_ptr from of_device_id table

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).  This fixes
compile warning (!CONFIG_OF on x86_64):

    drivers/dma/sf-pdma/sf-pdma.c:576:34: warning:
        ‘sf_pdma_dt_ids’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201120162303.482126-4-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: mv_xor: drop of_match_ptr from of_device_id table

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).  This fixes
compile warning (!CONFIG_OF on x86_64):

    drivers/dma/mv_xor.c:1281:34: warning:
        ‘mv_xor_dt_ids’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201120162303.482126-3-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: dw-axi-dmac: drop of_match_ptr from of_device_id table

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).  This fixes
compile warning (!CONFIG_OF on x86_64):

    drivers/dma/dw-axi-dmac/dw-axi-dmac-platform.c:984:34: warning:
        ‘dw_dma_of_id_table’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201120162303.482126-2-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: jz4780: drop of_match_ptr from of_device_id table

The driver can match only via the DT table so the table should be always
used and the of_match_ptr does not have any sense (this also allows ACPI
matching via PRP0001, even though it is not relevant here).  This fixes
compile warning (!CONFIG_OF on x86_64):

    drivers/dma/dma-jz4780.c:1031:34: warning:
        ‘jz4780_dma_dt_match’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201120162303.482126-1-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: mxs-dma: Remove the unused .id_table

The mxs-dma driver is only used by DT platforms and the .id_table
is unused.

Get rid of it to simplify the code.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201123193051.17285-1-festevam@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: mv_xor_v2: Fix error return code in mv_xor_v2_probe()

Return the corresponding error code when first_msi_entry() returns
NULL in mv_xor_v2_probe().

Fixes: 19a340b1a820430 ("dmaengine: mv_xor_v2: new driver")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Link: https://lore.kernel.org/r/20201124010813.1939095-1-chengzhihao1@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: imx-dma: Remove unused .id_table

Since 5.10-rc1 i.MX is a devicetree-only platform, so simplify the code
by removing the unused non-DT support.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201124143405.2764-1-festevam@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: qcom: Add GPI dma driver

This controller provides DMAengine capabilities for a variety of peripheral
buses such as I2C, UART, and SPI. By using GPI dmaengine driver, bus
drivers can use a standardize interface that is protocol independent to
transfer data between memory and peripheral.

Link: https://lore.kernel.org/r/20201109085450.24843-4-vkoul@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: add peripheral configuration

Some complex dmaengine controllers have capability to program the
peripheral device, so pass on the peripheral configuration as part of
dma_slave_config

Link: https://lore.kernel.org/r/20201109085450.24843-3-vkoul@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dmaengine: Document qcom,gpi dma binding

Add devicetree binding documentation for GPI DMA controller
implemented on Qualcomm SoCs

Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201109085450.24843-2-vkoul@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: qcom: Add ADM driver

Add the DMA engine driver for the QCOM Application Data Mover (ADM) DMA
controller found in the MSM8x60 and IPQ/APQ8064 platforms.

The ADM supports both memory to memory transactions and memory
to/from peripheral device transactions.  The controller also provides
flow control capabilities for transactions to/from peripheral devices.

The initial release of this driver supports slave transfers to/from
peripherals and also incorporates CRCI (client rate control interface)
flow control.

The hardware only supports a 32 bit physical address, so specifying
!PHYS_ADDR_T_64BIT gives maximum COMPILE_TEST coverage without having to
spend effort on kludging things in the code that will never actually be
needed on real hardware.

Signed-off-by: Andy Gross <agross@codeaurora.org>
Signed-off-by: Thomas Pedersen <twp@codeaurora.org>
Signed-off-by: Jonathan McDowell <noodles@earth.li>
Link: https://lore.kernel.org/r/20201114140233.GM32650@earth.li
Signed-off-by: Vinod Koul <vkoul@kernel.org>

drivers: soc: ti: knav_qmss_queue: Fix error return code in knav_queue_probe

Fix to return the error code from of_get_child_by_name() instaed of 0
in knav_queue_probe().

Fixes: 41f93af900a20d1a0a ("soc: ti: add Keystone Navigator QMSS driver")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: Fix reference imbalance in knav_dma_probe

The patch fix two reference leak.

  1) pm_runtime_get_sync will increment pm usage counter even it
     failed. Forgetting to call put operation will result in
     reference leak.

  2) The pm_runtime_enable will increase power disable depth. Thus
     a pairing decrement is needed on the error handling path to
     keep it balanced.

We fix it by: 1) adding call pm_runtime_put_noidle or
pm_runtime_put_sync in error handling. 2) adding pm_runtime_disable
in error handling, to keep usage counter and disable depth balanced.

Fixes: 88139ed030583 ("soc: ti: add Keystone Navigator DMA support")
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: pruss: Remove wrong check against *get_match_data return value

Since the of_device_get_match_data() doesn't return error code, remove
wrong IS_ERR test. Proper check against NULL pointer is already done
later before usage: if (data && data->...).

Additionally, proceeding with empty device data is valid (e.g. in case
of "ti,am3356-pruss").

Reported-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Grzegorz Jaszczyk <grzegorz.jaszczyk@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: Kconfig: Drop ARM64 SoC specific configs

With the integration of chip-id detection scheme in kernel[1], there
is no specific need to maintain multitudes of SoC specific config
options, discussed as per [2], we have deprecated the usage in other
places for v5.10-rc1. Drop the configuration for the follow on kernel.

[1] drivers/soc/ti/k3-socinfo.c commit 907a2b7e2fc7 ("soc: ti: add k3 platforms chipid module driver")

Signed-off-by: Nishanth Menon <nm@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: k3-ringacc: Provide documentation for 'k3_ring's 'state'

Fixes the following W=1 kernel build warning(s):

 drivers/soc/ti/k3-ringacc.c:163: warning: Function parameter or member 'state' not described in 'k3_ring'

Cc: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: wkup_m3_ipc: Document 'm3_ipc' parameter throughout

Fixes the following W=1 kernel build warning(s):

 drivers/soc/ti/wkup_m3_ipc.c:227: warning: Function parameter or member 'm3_ipc' not described in 'wkup_m3_set_mem_type'
 drivers/soc/ti/wkup_m3_ipc.c:236: warning: Function parameter or member 'm3_ipc' not described in 'wkup_m3_set_resume_address'
 drivers/soc/ti/wkup_m3_ipc.c:248: warning: Function parameter or member 'm3_ipc' not described in 'wkup_m3_request_pm_status'
 drivers/soc/ti/wkup_m3_ipc.c:268: warning: Function parameter or member 'm3_ipc' not described in 'wkup_m3_prepare_low_power'
 drivers/soc/ti/wkup_m3_ipc.c:322: warning: Function parameter or member 'm3_ipc' not described in 'wkup_m3_finish_low_power'
 drivers/soc/ti/wkup_m3_ipc.c:369: warning: Function parameter or member 'm3_ipc' not described in 'wkup_m3_set_rtc_only'
 drivers/soc/ti/wkup_m3_ipc.c:369: warning: Excess function parameter 'wkup_m3_wakeup' description in 'wkup_m3_set_rtc_only'

Cc: Santosh Shilimkar <ssantosh@kernel.org>
Cc: Dave Gerlach <d-gerlach@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: pm33xx: Remove set but unused variable 'ret'

Fixes the following W=1 kernel build warning(s):

 drivers/soc/ti/pm33xx.c: In function ‘am33xx_do_sram_idle’:
 drivers/soc/ti/pm33xx.c:138:6: warning: variable ‘ret’ set but not used [-Wunused-but-set-variable]

Cc: Santosh Shilimkar <ssantosh@kernel.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: knav_dma: Fix a kernel function doc formatting issue

Fixes the following W=1 kernel build warning(s):

 drivers/soc/ti/knav_dma.c:507: warning: Function parameter or member 'channel' not described in 'knav_dma_close_channel'

Cc: Santosh Shilimkar <ssantosh@kernel.org>
Cc: Sandeep Nair <sandeep_n@ti.com>
Cc: Cyril Chemparathy <cyril@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: knav_qmss_queue: Fix a whole host of function documentation issues

Fixes the following W=1 kernel build warning(s):

 drivers/soc/ti/knav_qmss_queue.c:528: warning: Function parameter or member 'flags' not described in 'knav_queue_open'
 drivers/soc/ti/knav_qmss_queue.c:528: warning: Function parameter or member 'id' not described in 'knav_queue_open'
 drivers/soc/ti/knav_qmss_queue.c:528: warning: Function parameter or member 'name' not described in 'knav_queue_open'
 drivers/soc/ti/knav_qmss_queue.c:551: warning: Excess function parameter 'qh' description in 'knav_queue_close'
 drivers/soc/ti/knav_qmss_queue.c:551: warning: Function parameter or member 'qhandle' not described in 'knav_queue_close'
 drivers/soc/ti/knav_qmss_queue.c:583: warning: Excess function parameter 'qh' description in 'knav_queue_device_control'
 drivers/soc/ti/knav_qmss_queue.c:583: warning: Function parameter or member 'arg' not described in 'knav_queue_device_control'
 drivers/soc/ti/knav_qmss_queue.c:583: warning: Function parameter or member 'cmd' not described in 'knav_queue_device_control'
 drivers/soc/ti/knav_qmss_queue.c:583: warning: Function parameter or member 'qhandle' not described in 'knav_queue_device_control'
 drivers/soc/ti/knav_qmss_queue.c:635: warning: Excess function parameter 'data' description in 'knav_queue_push'
 drivers/soc/ti/knav_qmss_queue.c:635: warning: Excess function parameter 'qh' description in 'knav_queue_push'
 drivers/soc/ti/knav_qmss_queue.c:635: warning: Function parameter or member 'dma' not described in 'knav_queue_push'
 drivers/soc/ti/knav_qmss_queue.c:635: warning: Function parameter or member 'flags' not described in 'knav_queue_push'
 drivers/soc/ti/knav_qmss_queue.c:635: warning: Function parameter or member 'qhandle' not described in 'knav_queue_push'
 drivers/soc/ti/knav_qmss_queue.c:635: warning: Function parameter or member 'size' not described in 'knav_queue_push'
 drivers/soc/ti/knav_qmss_queue.c:655: warning: Excess function parameter 'qh' description in 'knav_queue_pop'
 drivers/soc/ti/knav_qmss_queue.c:655: warning: Function parameter or member 'qhandle' not described in 'knav_queue_pop'
 drivers/soc/ti/knav_qmss_queue.c:655: warning: Function parameter or member 'size' not described in 'knav_queue_pop'
 drivers/soc/ti/knav_qmss_queue.c:759: warning: Function parameter or member 'name' not described in 'knav_pool_create'
 drivers/soc/ti/knav_qmss_queue.c:759: warning: Function parameter or member 'num_desc' not described in 'knav_pool_create'
 drivers/soc/ti/knav_qmss_queue.c:759: warning: Function parameter or member 'region_id' not described in 'knav_pool_create'
 drivers/soc/ti/knav_qmss_queue.c:862: warning: Excess function parameter 'pool' description in 'knav_pool_destroy'
 drivers/soc/ti/knav_qmss_queue.c:862: warning: Function parameter or member 'ph' not described in 'knav_pool_destroy'
 drivers/soc/ti/knav_qmss_queue.c:892: warning: Excess function parameter 'pool' description in 'knav_pool_desc_get'
 drivers/soc/ti/knav_qmss_queue.c:892: warning: Function parameter or member 'ph' not described in 'knav_pool_desc_get'
 drivers/soc/ti/knav_qmss_queue.c:911: warning: Excess function parameter 'pool' description in 'knav_pool_desc_put'
 drivers/soc/ti/knav_qmss_queue.c:911: warning: Function parameter or member 'desc' not described in 'knav_pool_desc_put'
 drivers/soc/ti/knav_qmss_queue.c:911: warning: Function parameter or member 'ph' not described in 'knav_pool_desc_put'
 drivers/soc/ti/knav_qmss_queue.c:931: warning: Excess function parameter 'pool' description in 'knav_pool_desc_map'
 drivers/soc/ti/knav_qmss_queue.c:931: warning: Function parameter or member 'desc' not described in 'knav_pool_desc_map'
 drivers/soc/ti/knav_qmss_queue.c:931: warning: Function parameter or member 'dma' not described in 'knav_pool_desc_map'
 drivers/soc/ti/knav_qmss_queue.c:931: warning: Function parameter or member 'dma_sz' not described in 'knav_pool_desc_map'
 drivers/soc/ti/knav_qmss_queue.c:931: warning: Function parameter or member 'ph' not described in 'knav_pool_desc_map'
 drivers/soc/ti/knav_qmss_queue.c:931: warning: Function parameter or member 'size' not described in 'knav_pool_desc_map'
 drivers/soc/ti/knav_qmss_queue.c:956: warning: Excess function parameter 'pool' description in 'knav_pool_desc_unmap'
 drivers/soc/ti/knav_qmss_queue.c:956: warning: Function parameter or member 'dma' not described in 'knav_pool_desc_unmap'
 drivers/soc/ti/knav_qmss_queue.c:956: warning: Function parameter or member 'dma_sz' not described in 'knav_pool_desc_unmap'
 drivers/soc/ti/knav_qmss_queue.c:956: warning: Function parameter or member 'ph' not described in 'knav_pool_desc_unmap'
 drivers/soc/ti/knav_qmss_queue.c:975: warning: Excess function parameter 'pool' description in 'knav_pool_count'
 drivers/soc/ti/knav_qmss_queue.c:975: warning: Function parameter or member 'ph' not described in 'knav_pool_count'

Cc: Santosh Shilimkar <ssantosh@kernel.org>
Cc: Sandeep Nair <sandeep_n@ti.com>
Cc: Cyril Chemparathy <cyril@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: knav_qmss_queue: Remove set but unchecked variable 'ret'

Fixes the following W=1 kernel build warning(s):

 drivers/soc/ti/knav_qmss_queue.c: In function ‘knav_setup_queue_pools’:
 drivers/soc/ti/knav_qmss_queue.c:1310:6: warning: variable ‘ret’ set but not used [-Wunused-but-set-variable]

Cc: Santosh Shilimkar <ssantosh@kernel.org>
Cc: Sandeep Nair <sandeep_n@ti.com>
Cc: Cyril Chemparathy <cyril@ti.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: omap-prm: Do not check rstst bit on deassert if already deasserted

If a rstctrl reset bit is already deasserted, we can just bail out early
not wait for rstst to clear. Otherwise we can have deassert fail for
already deasserted resets.

Fixes: c5117a78dd88 ("soc: ti: omap-prm: poll for reset complete during de-assert")
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: knav_qmss: fix reference leak in knav_queue_probe

pm_runtime_get_sync will increment pm usage counter even it
failed. Forgetting to pm_runtime_put_noidle will result in
reference leak in knav_queue_probe, so we should fix it.

Fixes: 41f93af900a20 ("soc: ti: add Keystone Navigator QMSS driver")
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

dmaengine: imx-sdma: Remove unused .id_table support

Since 5.10-rc1 i.MX is a devicetree-only platform and the existing
.id_table support in this driver was only useful for old non-devicetree
platforms.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201116202403.29749-1-festevam@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: sun6i: Add support for A100 DMA

The dma of a100 is similar to h6, with some minor changes to
support greater addressing capabilities.

Add support for it.

Signed-off-by: Yangtao Li <frank@allwinnertech.com>
Link: https://lore.kernel.org/r/719852c6a9a597bd2e82d01a268ca02b9dee826c.1604988979.git.frank@allwinnertech.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dma: allwinner,sun50i-a64-dma: Add A100 compatible

Add a binding for A100's dma controller.

Signed-off-by: Yangtao Li <frank@allwinnertech.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/f15a18e9b8868e8853db1b5a3d1e411b0ac1c63a.1604988979.git.frank@allwinnertech.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dmaengine: Add X2000 bindings.

Add the dmaengine bindings for the X2000 SoC from Ingenic.

Signed-off-by: 周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201107122016.89859-3-zhouyanjie@wanyeetech.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dmaengine: Add JZ4775 bindings.

Add the dmaengine bindings for the JZ4775 SoC from Ingenic.

Signed-off-by: 周琰杰 (Zhou Yanjie) <zhouyanjie@wanyeetech.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201107122016.89859-2-zhouyanjie@wanyeetech.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: pxa_dma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Cc: Daniel Mack <daniel@zonque.org>
Cc: Haojian Zhuang <haojian.zhuang@gmail.com>
Link: https://lore.kernel.org/r/20201027215252.25820-11-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ste_dma40: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Cc: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20201027215252.25820-10-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: moxart-dma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-9-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: hisi_dma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Zhou Wang <wangzhou1@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-8-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: k3dma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-7-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: milbeaut-xdmac: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-6-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: tegra210-adma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Cc: Laxman Dewangan <ldewangan@nvidia.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Link: https://lore.kernel.org/r/20201027215252.25820-5-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: sf-pdma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Cc: Green Wan <green.wan@sifive.com>
Link: https://lore.kernel.org/r/20201027215252.25820-4-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma: remove redundant irqsave and irqrestore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201027215252.25820-3-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ipu_idmac: remove redundant irqsave and restore in hardIRQ

Running in hardIRQ, disabling IRQ is redundant since hardIRQ has disabled
IRQ. This patch removes the irqsave and irqstore to save some instruction
cycles.

Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Link: https://lore.kernel.org/r/20201027215252.25820-2-song.bao.hua@hisilicon.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ti: k3-udma-glue: move psi-l pairing in channel en/dis functions

The NAVSS UDMA will stuck if target IP module is disabled by PM while PSI-L
threads are paired UDMA<->IP and no further transfers is possible. This
could be the case for IPs J721E Main CPSW (cpsw9g).

Hence, to avoid such situation do PSI-L threads pairing only when UDMA
channel is going to be enabled as at this time DMA consumer module expected
to be active already.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201030203000.4281-1-grygorii.strashko@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idma64: Switch to use __maybe_unused instead of ifdeffery

ifdeffery is prone to errors and makes code harder to read.
Switch to use __maybe_unused instead of ifdeffery.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20201104103131.89907-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: dw: Enable runtime PM

When consumer requests channel power on the DMA controller device
and otherwise on the freeing channel resources.

Note, in some cases consumer acquires channel at the ->probe() stage and
releases it at the ->remove() stage. It will mean that DMA controller device
will be powered during all this time if there is no assist from hardware
to idle it. The above mentioned cases should be investigated separately
and individually.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Link: https://lore.kernel.org/r/20201103183938.64752-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: define table offset multiplier

Convert table offset multiplier magic number to a define.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/160407311690.839435.6941865731867828234.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: Update calculation of group offset to be more readable

Create helper macros to make group offset calculation more readable.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/160407294683.839093.10740868559754142070.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

soc: ti: k3-socinfo: Add entry for AM64X SoC family

It's JTAG PARTNO is 0xBB38.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: k3-ringacc: Use correct device for allocation in RING mode

In RING mode the ringacc does not access the ring memory. In this access
mode the ringacc coherency does not have meaning.

If the ring is configured in RING mode, then the ringacc itself will not
access to the ring memory. Only the requester (user) of the ring is going
to read/write to the memory.
Extend the ring configuration parameters with a device pointer to be used
for DMA API when the ring is configured in RING mode.

Extending the ring configuration struct will allow per ring selection of
device to be used for allocation, thus allowing per ring coherency.

To avoid regression, fall back to use the ringacc dev in case the alloc_dev
is not provided.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: rm: Remove unused config() from ti_sci_rm_ringacc_ops

The ringacc driver has been converted to use the new set_cfg function to
configure the ring, the old config ops can be removed.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: k3-ringacc: Use the ti_sci set_cfg callback for ring configuration

Switch to the new set_cfg to configure the ring.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: rm: Add new ops for ring configuration

The sysfw ring configuration message has been extended to include virtid
and asel value for the ring.
Add the ASEL_VALID to TI_SCI_MSG_VALUE_RM_ALL_NO_ORDER as it is required
for DMA rings.

Instead of extending the current .config() ops - which would need same
patch change in the ringacc driver - add ti_sci_msg_rm_ring_cfg struct and
a new ops using it to configure the ring.

This will allow easy update path in case new members are added for the ring
configuration.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: rm: Remove ring_get_config support

The ring_get_cfg (0x1111 message) is not used and it is not supported by
sysfw for a long time.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: rm: Add support for extended_ch_type for tx channel

Sysfw added 'extended_ch_type' to the tx_ch_cfg_req message which should be
used when BCDMA block copy channels are configured:
extended_ch_type = 0 : the channel is split tx channel (tchan)
extended_ch_type = 1 : the channel is block copy channel (bchan)

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

soc: ti: ti_sci_inta_msi: Add support for second range in resource ranges

Allocate MSI entries for both first and second range if they are valid

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: rm: Add support for second resource range

Sysfw added support for a second range in the resource range API to be able
to describe complex allocations mainly for DMA channels.

Update the ti_sci part to consider the second range as well.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: Use struct ti_sci_resource_desc in get_range ops

Use the ti_sci_resource_desc directly and update it's start and num members
directly instead of requiring individual parameters for them.

This will allow easy extension of the RM parameters without changing API.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

firmware: ti_sci: rm: Add support for tx_tdtype parameter for tx channel

The system controller's resource manager have support for configuring the
TDTYPE of TCHAN_CFG register on j721e.
With this parameter the teardown completion can be controlled:
TDTYPE == 0: Return without waiting for peer to complete the teardown
TDTYPE == 1: Wait for peer to complete the teardown

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Reviewed-by: Tero Kristo <t-kristo@ti.com>
Tested-by: Keerthy <j-keerthy@ti.com>
Reviewed-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>

dmaengine: idxd: Add ABI documentation for shared wq

Add the sysfs attribute bits in ABI/stable for shared wq support.

Signed-off-by: Jing Lin <jing.lin@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/160382008649.3911367.10851752182908509837.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: Clean up descriptors with fault error

Add code to "complete" a descriptor when the descriptor or its completion
address hit a fault error when SVA mode is being used. This error can be
triggered due to bad programming by the user. A lock is introduced in order
to protect the descriptor completion lists since the fault handler will run
from the system work queue after being scheduled in the interrupt handler.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/160382008092.3911367.12766483427643278985.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: Add shared workqueue support

Add shared workqueue support that includes the support of Shared Virtual
memory (SVM) or in similar terms On Demand Paging (ODP). The shared
workqueue uses the enqcmds command in kernel and will respond with retry if
the workqueue is full. Shared workqueue only works when there is PASID
support from the IOMMU.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/160382007499.3911367.26043087963708134.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ppc4xx: remove xor_hw_desc assignment without reading

The xor_hw_desc local variable is assigned but never read:

    drivers/dma/ppc4xx/adma.c: In function ‘ppc440spe_desc_set_src_mult’:
    drivers/dma/ppc4xx/adma.c:562:17: warning: variable ‘xor_hw_desc’ set but not used [-Wunused-but-set-variable]

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201019155756.21445-2-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: ppc4xx: make ppc440spe_adma_chan_list static

The ppc440spe_adma_chan_list file-scope variable is not used outside of
the unit so it can be made static.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20201019155756.21445-1-krzk@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: at_xdmac: add AXI priority support and recommended settings

The sama7g5 version of the XDMAC supports priority configuration and
outstanding capabilities.
Add defines for the specific registers for this configuration, together
with recommended settings.
However the settings are very different if the XDMAC is a mem2mem or a
per2mem controller.
Thus, we need to differentiate according to device tree property.

Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Link: https://lore.kernel.org/r/20201016093918.290137-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: at_xdmac: add support for sama7g5 based at_xdmac

SAMA7G5 SoC uses a slightly different variant of the AT_XDMAC.
Added support by a new compatible and a layout struct that copes
to the specific version considering the compatible string.
Only the differences in register map are present in the layout struct.
I reworked the register access for this part that has the differences.
Also the Source/Destination Interface bits are no longer valid for this
variant of the XDMAC. Thus, the layout also has a bool for specifying
whether these bits are required or not.

Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Link: https://lore.kernel.org/r/20201016093850.290053-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: at_xdmac: adapt perid for mem2mem operations

The PERID in the CC register for mem2mem operations must match an unused
PERID.
The PERID field is 7 bits, but the selected value is 0x3f.
On later products we can have more reserved PERIDs for actual peripherals,
thus this needs to be increased to maximum size.
Changing the value to 0x7f, which is the maximum for 7 bits field.

Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Reviewed-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Link: https://lore.kernel.org/r/20201016093725.289880-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dt-bindings: dmaengine: at_xdmac: add compatible with microchip,sama7g5

Add compatible to sama7g5 SoC.

Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201016081754.288488-1-eugen.hristev@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: pl330: Remove unreachable code

_setup_req(..) never returns negative value.
Hence the condition ret < 0 is never met

Signed-off-by: Surendran K <surendran.k@samsung.com>
Link: https://lore.kernel.org/r/20201016103347.63084-1-surendran.k@samsung.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: stm32-mdma: Use struct_size() in kzalloc()

Make use of the new struct_size() helper instead of the offsetof() idiom.

Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20201008141828.GA20325@embeddedor
Signed-off-by: Vinod Koul <vkoul@kernel.org>

dmaengine: idxd: fix wq config registers offset programming

DSA spec v1.1 [1] updated to include a stride size register for WQ
configuration that will specify how much space is reserved for the WQ
configuration register set. This change is expected to be in the final
gen1 DSA hardware. Fix the driver to use WQCFG_OFFSET() for all WQ
offset calculation and fixup WQCFG_OFFSET() to use the new calculated
wq size.

[1]: https://software.intel.com/content/www/us/en/develop/download/intel-data-streaming-accelerator-preliminary-architecture-specification.html

Fixes: bfe1d56091c1 ("dmaengine: idxd: Init and probe for Intel data accelerators")
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/160383444959.48058.14249265538404901781.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>

Linux 5.10-rc1

treewide: Convert macro and uses of __section(foo) to __section("foo")

Use a more generic form for __section that requires quotes to avoid
complications with clang and gcc differences.

Remove the quote operator # from compiler_attributes.h __section macro.

Convert all unquoted __section(foo) uses to quoted __section("foo").
Also convert __attribute__((section("foo"))) uses to __section("foo")
even if the __attribute__ has multiple list entry forms.

Conversion done using the script at:

    https://lore.kernel.org/lkml/75393e5ddc272dc7403de74d645e6c6e0f4e70eb.camel@perches.com/2-convert_section.pl

Signed-off-by: Joe Perches <joe@perches.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@gooogle.com>
Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

kernel/sys.c: fix prototype of prctl_get_tid_address()

tid_addr is not a "pointer to (pointer to int in userspace)"; it is in
fact a "pointer to (pointer to int in userspace) in userspace".  So
sparse rightfully complains about passing a kernel pointer to
put_user().

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: remove kzfree() compatibility definition

Commit 453431a54934 ("mm, treewide: rename kzfree() to
kfree_sensitive()") renamed kzfree() to kfree_sensitive(),
but it left a compatibility definition of kzfree() to avoid
being too disruptive.

Since then a few more instances of kzfree() have slipped in.

Just get rid of them and remove the compatibility definition
once and for all.

Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

checkpatch: enable GIT_DIR environment use to set git repository location

If set, use the environment variable GIT_DIR to change the default .git
location of the kernel git tree.

If GIT_DIR is unset, keep using the current ".git" default.

Link: https://lkml.kernel.org/r/c5e23b45562373d632fccb8bc04e563abba4dd1d.camel@perches.com
Signed-off-by: Joe Perches <joe@perches.com>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Merge tag 'timers-urgent-2020-10-25' of git://git./linux/kernel/git/tip/tip

Pull timer fixes from Thomas Gleixner:
 "A time namespace fix and a matching selftest. The futex absolute
  timeouts which are based on CLOCK_MONOTONIC require time namespace
  corrected. This was missed in the original time namesapce support"

* tag 'timers-urgent-2020-10-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  selftests/timens: Add a test for futex()
  futex: Adjust absolute futex timeouts with per time namespace offset