The debugfs_create_*() functions never return NULL, so checking their
return value for NULL is pointless. Secondly, the debugfs subsystem is
designed to be as simple as possible, so if one of the debugfs_create_*()
method in a hierarchy fails, the following methods should silently return
the passed erroneous parental dentry. Finally, the code should work no
matter whether anything debugfs-related fails.
To make code simpler and debugfs-independent, stop checking the
debugfs_create_*() return values.
If the debugfs file system is unavailable, skip the debugfs node
initialization altogether to preserve some memory space.
Link: https://lore.kernel.org/r/20230113171409.30470-12-Sergey.Semin@baikalelectronics.ru
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The debugfs_entries structure declared in dw-edma-v0-debugfs.c contains the
debugfs node register address. The address is declared as dma_addr_t type,
but is cast to "void *".
Change the type to "void __iomem *" and drop the unnecessary casts.
Link: https://lore.kernel.org/r/20230113171409.30470-11-Sergey.Semin@baikalelectronics.ru
Fixes: 305aebeff8 ("dmaengine: Add Synopsys eDMA IP version 0 debugfs support")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The DMA engine core manages dma_device.chancnt itself, e.g., in
dma_async_device_register(). DMA device drivers should not initialize
chancnt because it causes the wrong number of channels printed in the
device summary.
Drop the dw-edma chancnt initialization.
Link: https://lore.kernel.org/r/20230113171409.30470-10-Sergey.Semin@baikalelectronics.ru
Fixes: e63d79d1ff ("dmaengine: Add Synopsys eDMA IP core driver")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The Synopsys PCIe Endpoint IP prototype kit can be attached via any PCI
host controller, including one where the PCI bus address space is different
from the CPU address space. Therefore, we need to make sure the source and
destination addresses of the DMA slave devices are converted to the PCI bus
address space; otherwise DMA transactions may cause memory corruption.
Add a new dw_edma_pcie_address() interface to perform this translation by
using pcibios_resource_to_bus().
Link: https://lore.kernel.org/r/20230113171409.30470-9-Sergey.Semin@baikalelectronics.ru
Fixes: 41aaff2a2a ("dmaengine: Add Synopsys eDMA IP PCIe glue-logic")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
Since 9575632052 ("dmaengine: make slave address physical"), the source
and destination addresses of the DMA slave device have been converted to
physical addresses in the CPU address space. It's the DMA device driver's
responsibility to convert them to the DMA bus address space. In case of the
DW eDMA device, the source or destination peripheral (slave) devices reside
in PCI bus space. Thus we need to perform the PCI Host/Endpoint windows-
based (i.e. DT "ranges" property) address translation; otherwise the eDMA
transactions won't work as expected (or can be even harmful) if the CPU and
PCI address spaces don't match.
Note 1: Even though the DMA interleaved template has both source and
destination addresses declared as dma_addr_t, only the CPU memory range
should be mapped to be seen by the DMA device since it's a subject of the
DMA getting towards the system side. The device part must not be mapped
since the slave device resides in the PCI bus space, which isn't affected
by IOMMUs or iATU translations. DW PCIe eDMA generates corresponding
MWr/MRd TLPs on its own.
Note 2: This functionality is mainly required for the remote eDMA setup
since the CPU address must be manually translated into the PCI bus space
before being written to LLI.{SAR,DAR}. If eDMA is embedded in the locally
accessible DW PCIe Root Port/Endpoint, software-based translation isn't
required since hardware will translate it via the Outbound iATU as long as
the DMA_BYPASS flag is cleared. If DMA_BYPASS is set or there is no
Outbound iATU entry that contains the SAR or DAR (for Read and Write
channel respectively), there won't be any translation performed but DMA
will proceed with the corresponding source/destination address as-is.
Link: https://lore.kernel.org/r/20230113171409.30470-8-Sergey.Semin@baikalelectronics.ru
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The interleaved DMA transfer support added by 85e7518f42 ("dmaengine:
dw-edma: Add device_prep_interleave_dma() support") seems contradictory to
what the DMA engine defines. The next conditional statements:
if (!xfer->xfer.il->numf)
return NULL;
if (xfer->xfer.il->numf > 0 && xfer->xfer.il->frame_size > 0)
return NULL;
mean that numf can't be zero and frame_size must always be zero, otherwise
the transfer won't be executed. Furthermore, the transfer execution method
takes the frame size from the dma_interleaved_template.sgl[] array for each
frame. That array in accordance with [1] is supposed to be of
dma_interleaved_template.frame_size size, which as we discovered before the
code expects to be zero. So judging by the dw_edma_device_transfer()
implementation, the method implies the dma_interleaved_template.sgl[] array
being of dma_interleaved_template.numf size, which is wrong. Since the
dw_edma_device_transfer() method doesn't permit
dma_interleaved_template.frame_size being non-zero, the multi-chunk
interleaved transfer turns to be unsupported even though the code implies
having it supported.
Add fully functioning support of interleaved DMA transfers.
First of all, dma_interleaved_template.frame_size is supposed to be greater
or equal to one thus having at least simple linear chunked frames.
Secondly, we can create a walk-through over all the chunks and frames by
initializing the number of the eDMA burst transactions as a multiple of
dma_interleaved_template.numf and dma_interleaved_template.frame_size and
getting the frame_size-modulo of the iteration step as an index of the
dma_interleaved_template.sgl[] array.
[1] include/linux/dmaengine.h: doc struct dma_interleaved_template
Link: https://lore.kernel.org/r/20230113171409.30470-7-Sergey.Semin@baikalelectronics.ru
Fixes: 85e7518f42 ("dmaengine: dw-edma: Add device_prep_interleave_dma() support")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The DW eDMA controller always increments both source and destination
addresses. Permitting DMA interleaved transfers with no src_inc/dst_inc
flags set may lead to unexpected behaviour for the device users.
Terminate interleaved transfers if at least one of the
dma_interleaved_template.{src_inc,dst_inc} flag is initialized to "false".
Note that in addition, we need to increase the source and destination
addresses after each iteration.
Link: https://lore.kernel.org/r/20230113171409.30470-6-Sergey.Semin@baikalelectronics.ru
Fixes: 85e7518f42 ("dmaengine: dw-edma: Add device_prep_interleave_dma() support")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
Interleaved DMA transfer support was added by 85e7518f42 ("dmaengine:
dw-edma: Add device_prep_interleave_dma() support"), but depending on the
selected channel, either source or destination address are left
uninitialized which was obviously wrong.
Initialize the destination address of the eDMA burst descriptors for
DEV_TO_MEM interleaved operations and the source address for MEM_TO_DEV
operations.
Link: https://lore.kernel.org/r/20230113171409.30470-5-Sergey.Semin@baikalelectronics.ru
Fixes: 85e7518f42 ("dmaengine: dw-edma: Add device_prep_interleave_dma() support")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The dw_edma_region.paddr field should be a memory base address visible by
the DW eDMA controller. If the DMA engine is embedded in the DW PCIe
Host/Endpoint controller, the address should belong to the Local CPU/
Application memory. If eDMA is remotely accessible across the PCI bus via
PCI memory IOs, the address should be part of the PCI bus memory space.
The latter case hasn't been well covered in the corresponding glue-driver.
Since pci_dev.resource[] contains resources defined in the CPU memory
space, they need to be converted to the PCI bus address space. Convert the
LL, DT and CSRs PCI memory ranges with pci_bus_address().
In addition, extend the dw_edma_region.paddr field size. The field normally
contains a memory range base address to be set in the DW eDMA Linked-List
pointer register or as a base address of the Linked-List data buffer. In
accordance with [1] the LL range is supposed to be created in the Local
CPU/Application memory, but depending on the DW eDMA utilization the memory
can be created as a part of the PCI bus address space (as in the case of
the DW PCIe Endpoint prototype kit).
In the former case dw_edma_region.paddr should be a dma_addr_t, while in
the latter one it should be a pci_bus_addr_t. Since the corresponding CSRs
are always 64 bits wide, convert dw_edma_region.paddr to be u64, and let
the client make sure it has a valid address visible by the DW eDMA
controller. For instance, the DW eDMA PCIe glue-driver initializes the
field with addresses from the PCI bus memory space.
[1] DesignWare Cores PCI Express Controller Databook - DWC PCIe Root Port,
v.5.40a, March 2019, p.1103
Link: https://lore.kernel.org/r/20230113171409.30470-4-Sergey.Semin@baikalelectronics.ru
Fixes: 41aaff2a2a ("dmaengine: Add Synopsys eDMA IP PCIe glue-logic")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
If dw_edma_irq_request() fails to initialize an IRQ handler, any previously
requested IRQs will be left initialized.
Release the previously requested IRQs in the cleanup-on-error path of
dw_edma_irq_request().
Link: https://lore.kernel.org/r/20230113171409.30470-3-Sergey.Semin@baikalelectronics.ru
Fixes: e63d79d1ff ("dmaengine: Add Synopsys eDMA IP core driver")
Tested-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Serge Semin <Sergey.Semin@baikalelectronics.ru>
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
The interrupt handler (pt_core_irq_handler()) of the ptdma
driver can be called from interrupt context. The code flow
in this function can lead down to pt_core_execute_cmd() which
will attempt to grab a mutex, which is not appropriate in
interrupt context and ultimately leads to a kernel panic.
The fix here changes this mutex to a spinlock, which has
been verified to resolve the issue.
Fixes: fa5d823b16 ("dmaengine: ptdma: Initial driver for the AMD PTDMA")
Signed-off-by: Eric Pilmore <epilmore@gigaio.com>
Link: https://lore.kernel.org/r/20230119033907.35071-1-epilmore@gigaio.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Terminate vdesc when terminating an ongoing transfer.
This will ensure that the vdesc is present in the desc_terminated list
The descriptor will be freed later in desc_free_list().
This fixes the memory leaks which can happen when terminating an
ongoing transfer.
Fixes: ee17028009 ("dmaengine: tegra: Add tegra gpcdma driver")
Signed-off-by: Akhil R <akhilrajeev@nvidia.com>
Link: https://lore.kernel.org/r/20230118115801.15210-1-akhilrajeev@nvidia.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
platform_get_resource() and devm_ioremap_resource() are wrapped up in the
devm_platform_ioremap_resource() helper. Use the helper and get rid of the
local variable for struct resource *. We now have a function call less.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20221110152528.7821-1-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Since for_each_child_of_node() will increase the refcount of node, we need
to call of_node_put() manually when breaking out of the iteration.
Fixes: 9cd4360de6 ("dma: Add Xilinx AXI Video Direct Memory Access Engine driver support")
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Acked-by: Peter Korsgaard <peter@korsgaard.com>
Link: https://lore.kernel.org/r/20221122021612.1908866-1-liushixin2@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add runtime PM support which involves disabling/enabling controller's
clocks on runtime PM suspend/resume ops. The runtime suspend/resume is
done based on the work submitted to the controller: runtime resume is
happening on at_xdmac_start_xfer() and runtime suspend on
at_xdmac_tasklet().
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20221117131547.293044-2-claudiu.beznea@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
If the function sdma_load_context() fails, the sdma_desc will be
freed, but the allocated desc->bd is forgot to be freed.
We already met the sdma_load_context() failure case and the log as
below:
[ 450.699064] imx-sdma 30bd0000.dma-controller: Timeout waiting for CH0 ready
...
In this case, the desc->bd will not be freed without this change.
Signed-off-by: Hui Wang <hui.wang@canonical.com>
Reviewed-by: Sascha Hauer <s.hauer@pengutronix.de>
Link: https://lore.kernel.org/r/20221130090800.102035-1-hui.wang@canonical.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Follow the advice of the Documentation/filesystems/sysfs.rst and show()
should only use sysfs_emit() or sysfs_emit_at() when formatting the
value to be returned to user space.
Signed-off-by: ye xingchen <ye.xingchen@zte.com.cn>
Link: https://lore.kernel.org/r/202212061714501297954@zte.com.cn
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The sun6i DMA engine supports segment sizes up to 2^25-1 bytes. This is
explicitly stated in newer SoC documentation (H6, D1), and it is implied
in older documentation by the 25-bit width of the "bytes left in the
current segment" register field.
Exposing the real segment size limit (instead of the 64k default)
reduces the number of SG list segments needed for a transaction.
Reviewed-by: Jernej Skrabec <jernej.skrabec@gmail.com>
Signed-off-by: Samuel Holland <samuel@sholland.org>
Link: https://lore.kernel.org/r/20230101193605.50285-1-samuel@sholland.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The current global interrupt clear programming register offset
was not correct. Fix the programming with right offset
Fixes: ded1f3db4c ("dmaengine: tegra210-adma: prepare for supporting newer Tegra chips")
Cc: stable@vger.kernel.org
Signed-off-by: Mohan Kumar <mkumard@nvidia.com>
Link: https://lore.kernel.org/r/20230102064844.31306-1-mkumard@nvidia.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
ldma_cfg_init() will parse DT to retrieve certain configs.
However, that is called before ldma_dma_init_vXX(), which
will make some initialization to channel configs. It will
thus incorrectly overwrite certain configs that are declared
in DT.
To fix that, we move DT parsing after initialization.
Function name is renamed to better represent what it does.
Fixes: 32d31c79a1 ("dmaengine: Add Intel LGM SoC DMA support.")
Signed-off-by: Peter Harliman Liem <pliem@maxlinear.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/afef6fc1ed20098b684e0d53737d69faf63c125f.1672887183.git.pliem@maxlinear.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Drop the non-fatal probe deferral log for getting MSI domain.
This makes the kernel log clean and we do not get recurring logs
stating: "Failed to get MSI domain".
Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20230117051855.29644-1-j-choudhary@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The s3c24xx platform was removed and this driver is no longer
needed.
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The MMP_SRAM code is no longer used by the tdma driver because
the Kconfig symbol is not selected, so remove it along with its
former callsite.
Acked-By: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
The hdmi script already supported in sdma firmware. So add support hdmi
in sdma driver.
The design of hdmi script is different from common script such as sai.
There is no need to config buffer descriptor for HDMI. The cyclic
capability is achieved by the hdmi script. The slave config is so simple,
only config src_addr, dts_addr and direction DMA_TRANS_NONE.
Signed-off-by: Joy Zou <joy.zou@nxp.com>
Reviewed-by: Sascha Hauer <s.hauer@pengutronix.de>
Link: https://lore.kernel.org/r/20221115093823.2879128-3-joy.zou@nxp.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The HiSilicon DMA Engine is only present on HiSilicon SoCs. Hence add a
dependency on ARCH_HISI, to prevent asking the user about this driver
when configuring a kernel without HiSilicon SoC support.
Fixes: e9f08b6525 ("dmaengine: hisilicon: Add Kunpeng DMA engine support")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/363a1816d36cd3cf604d88ec90f97c75f604de64.1669044190.git.geert+renesas@glider.be
Signed-off-by: Vinod Koul <vkoul@kernel.org>
PSIL_EP_NATIVE endpoints may not have PEER registers for BCNT and thus
udma_decrement_byte_counters() should not try to decrement these counters.
This fixes the issue of crypto IPERF testing where the client side (EVM)
hangs without transfer of packets to the server side, seen since this
function was added.
Fixes: 7c94dcfa8f ("dmaengine: ti: k3-udma: Reset UDMA_CHAN_RT byte counters to prevent overflow")
Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20221128085005.489964-1-j-choudhary@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
On driver unload any pending descriptors are flushed and pending
DMA descriptors are explicitly completed:
idxd_dmaengine_drv_remove() ->
drv_disable_wq() ->
idxd_wq_free_irq() ->
idxd_flush_pending_descs() ->
idxd_dma_complete_txd()
With this done during driver unload any remaining descriptor is
likely stuck and can be dropped. Even so, the descriptor may still
have a callback set that could no longer be accessible. An
example of such a problem is when the dmatest fails and the dmatest
module is unloaded. The failure of dmatest leaves descriptors with
dma_async_tx_descriptor::callback pointing to code that no longer
exist. This causes a page fault as below at the time the IDXD driver
is unloaded when it attempts to run the callback:
BUG: unable to handle page fault for address: ffffffffc0665190
#PF: supervisor instruction fetch in kernel mode
#PF: error_code(0x0010) - not-present page
Fix this by clearing the callback pointers on the transmit
descriptors only when workqueue is disabled.
Fixes: 403a2e2365 ("dmaengine: idxd: change MSIX allocation based on per wq activation")
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/37d06b772aa7f8863ca50f90930ea2fd80b38fc3.1670452419.git.reinette.chatre@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
On driver unload any pending descriptors are flushed at the
time the interrupt is freed:
idxd_dmaengine_drv_remove() ->
drv_disable_wq() ->
idxd_wq_free_irq() ->
idxd_flush_pending_descs().
If there are any descriptors present that need to be flushed this
flow triggers a "not present" page fault as below:
BUG: unable to handle page fault for address: ff391c97c70c9040
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
The address that triggers the fault is the address of the
descriptor that was freed moments earlier via:
drv_disable_wq()->idxd_wq_free_resources()
Fix the use after free by freeing the descriptors after any possible
usage. This is done after idxd_wq_reset() to ensure that the memory
remains accessible during possible completion writes by the device.
Fixes: 63c14ae6c1 ("dmaengine: idxd: refactor wq driver enable/disable operations")
Suggested-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/6c4657d9cff0a0a00501a7b928297ac966e9ec9d.1670452419.git.reinette.chatre@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The workqueue is enabled when the appropriate driver is loaded and
disabled when the driver is removed. When the driver is removed it
assumes that the workqueue was enabled successfully and proceeds to
free allocations made during workqueue enabling.
Failure during workqueue enabling does not prevent the driver from
being loaded. This is because the error path within drv_enable_wq()
returns success unless a second failure is encountered
during the error path. By returning success it is possible to load
the driver even if the workqueue cannot be enabled and
allocations that do not exist are attempted to be freed during
driver remove.
Some examples of problematic flows:
(a)
idxd_dmaengine_drv_probe() -> drv_enable_wq() -> idxd_wq_request_irq():
In above flow, if idxd_wq_request_irq() fails then
idxd_wq_unmap_portal() is called on error exit path, but
drv_enable_wq() returns 0 because idxd_wq_disable() succeeds. The
driver is thus loaded successfully.
idxd_dmaengine_drv_remove()->drv_disable_wq()->idxd_wq_unmap_portal()
Above flow on driver unload triggers the WARN in devm_iounmap() because
the device resource has already been removed during error path of
drv_enable_wq().
(b)
idxd_dmaengine_drv_probe() -> drv_enable_wq() -> idxd_wq_request_irq():
In above flow, if idxd_wq_request_irq() fails then
idxd_wq_init_percpu_ref() is never called to initialize the percpu
counter, yet the driver loads successfully because drv_enable_wq()
returns 0.
idxd_dmaengine_drv_remove()->__idxd_wq_quiesce()->percpu_ref_kill():
Above flow on driver unload triggers a BUG when attempting to drop the
initial ref of the uninitialized percpu ref:
BUG: kernel NULL pointer dereference, address: 0000000000000010
Fix the drv_enable_wq() error path by returning the original error that
indicates failure of workqueue enabling. This ensures that the probe
fails when an error is encountered and the driver remove paths are only
attempted when the workqueue was enabled successfully.
Fixes: 1f2bb40337 ("dmaengine: idxd: move wq_enable() to device.c")
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/e8d8116e5efa0fd14fadc5adae6ffd319f0e5ff1.1670452419.git.reinette.chatre@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
On DSA/IAX 1.0, TC-A and TC-B in GRPCFG are set as 1 to have best
performance and cannot be changed through sysfs knobs unless override
option is given.
The same values should be set on DSA 2.0 as well.
Fixes: ea7c8f598c ("dmaengine: idxd: restore traffic class defaults after wq reset")
Fixes: ade8a86b51 ("dmaengine: idxd: Set defaults for GRPCFG traffic class")
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20221209172141.562648-1-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The function set_completion_address is defined in the dma.c file, but not
called elsewhere, so remove this unused function.
drivers/dma/idxd/dma.c:66:20: warning: unused function 'set_completion_address'.
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=3416
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20221212033514.5831-1-jiapeng.chong@linux.alibaba.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Rx operation on SPI GSI DMA is currently not working.
As per GSI spec, link_rx bit is to be set on GO TRE on tx
channel whenever there is going to be a DMA TRE on rx
channel. This is currently set for duplex operation only.
Set the bit for rx operation as well.
This is part of changes required to bring up Rx.
Fixes: 94b8f0e58f ("dmaengine: qcom: gpi: set chain and link flag for duplex")
Signed-off-by: Vijaya Krishna Nivarthi <quic_vnivarth@quicinc.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/1671212293-14767-1-git-send-email-quic_vnivarth@quicinc.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
BCDMA CSI RX present on AM62Ax SoC is a dedicated DMA for servicing
Camera Serial Interface (CSI) IP. Add support for the same.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20221213164304.1126945-6-vigneshr@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
AM62A SoC has a BCDMA and PKTDMA as systems DMAs for service various
peripherals similar to AM64 SoC. Add support for the same.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20221213164304.1126945-5-vigneshr@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reusing loop iterator fails if BCHAN is not present as iterator is
uninitialized
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20221213164304.1126945-3-vigneshr@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
New support:
- Qualcomm SDM670, SM6115 and SM6375 GPI controller support
- Ingenic JZ4755 dmaengine support
- Removal of iop-adma driver
Updates:
- Tegra support for dma-channel-mask
- at_hdmac cleanup and virt-chan support for this driver
-----BEGIN PGP SIGNATURE-----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=pURR
-----END PGP SIGNATURE-----
Merge tag 'dmaengine-6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine
Pull dmaengine updates from Vinod Koul:
"New support:
- Qualcomm SDM670, SM6115 and SM6375 GPI controller support
- Ingenic JZ4755 dmaengine support
- Removal of iop-adma driver
Updates:
- Tegra support for dma-channel-mask
- at_hdmac cleanup and virt-chan support for this driver"
* tag 'dmaengine-6.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (46 commits)
dmaengine: Revert "dmaengine: remove s3c24xx driver"
dmaengine: tegra: Add support for dma-channel-mask
dt-bindings: dmaengine: Add dma-channel-mask to Tegra GPCDMA
dmaengine: idxd: Remove linux/msi.h include
dt-bindings: dmaengine: qcom: gpi: add compatible for SM6375
dmaengine: idxd: Fix crc_val field for completion record
dmaengine: at_hdmac: Convert driver to use virt-dma
dmaengine: at_hdmac: Remove unused member of at_dma_chan
dmaengine: at_hdmac: Rename "chan_common" to "dma_chan"
dmaengine: at_hdmac: Rename "dma_common" to "dma_device"
dmaengine: at_hdmac: Use bitfield access macros
dmaengine: at_hdmac: Keep register definitions and structures private to at_hdmac.c
dmaengine: at_hdmac: Set include entries in alphabetic order
dmaengine: at_hdmac: Use pm_ptr()
dmaengine: at_hdmac: Use devm_clk_get()
dmaengine: at_hdmac: Use devm_platform_ioremap_resource
dmaengine: at_hdmac: Use devm_kzalloc() and struct_size()
dmaengine: at_hdmac: Introduce atc_get_llis_residue()
dmaengine: at_hdmac: s/atc_get_bytes_left/atc_get_residue
dmaengine: at_hdmac: Pass residue by address to avoid unnecessary implicit casts
...
iommufd is the user API to control the IOMMU subsystem as it relates to
managing IO page tables that point at user space memory.
It takes over from drivers/vfio/vfio_iommu_type1.c (aka the VFIO
container) which is the VFIO specific interface for a similar idea.
We see a broad need for extended features, some being highly IOMMU device
specific:
- Binding iommu_domain's to PASID/SSID
- Userspace IO page tables, for ARM, x86 and S390
- Kernel bypassed invalidation of user page tables
- Re-use of the KVM page table in the IOMMU
- Dirty page tracking in the IOMMU
- Runtime Increase/Decrease of IOPTE size
- PRI support with faults resolved in userspace
Many of these HW features exist to support VM use cases - for instance the
combination of PASID, PRI and Userspace IO Page Tables allows an
implementation of DMA Shared Virtual Addressing (vSVA) within a
guest. Dirty tracking enables VM live migration with SRIOV devices and
PASID support allow creating "scalable IOV" devices, among other things.
As these features are fundamental to a VM platform they need to be
uniformly exposed to all the driver families that do DMA into VMs, which
is currently VFIO and VDPA.
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCY5ct7wAKCRCFwuHvBreF
YZZ5AQDciXfcgXLt0UBEmWupNb0f/asT6tk717pdsKm8kAZMNAEAsIyLiKT5HqGl
s7fAu+CQ1pr9+9NKGevD+frw8Solsw4=
=jJkd
-----END PGP SIGNATURE-----
Merge tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd
Pull iommufd implementation from Jason Gunthorpe:
"iommufd is the user API to control the IOMMU subsystem as it relates
to managing IO page tables that point at user space memory.
It takes over from drivers/vfio/vfio_iommu_type1.c (aka the VFIO
container) which is the VFIO specific interface for a similar idea.
We see a broad need for extended features, some being highly IOMMU
device specific:
- Binding iommu_domain's to PASID/SSID
- Userspace IO page tables, for ARM, x86 and S390
- Kernel bypassed invalidation of user page tables
- Re-use of the KVM page table in the IOMMU
- Dirty page tracking in the IOMMU
- Runtime Increase/Decrease of IOPTE size
- PRI support with faults resolved in userspace
Many of these HW features exist to support VM use cases - for instance
the combination of PASID, PRI and Userspace IO Page Tables allows an
implementation of DMA Shared Virtual Addressing (vSVA) within a guest.
Dirty tracking enables VM live migration with SRIOV devices and PASID
support allow creating "scalable IOV" devices, among other things.
As these features are fundamental to a VM platform they need to be
uniformly exposed to all the driver families that do DMA into VMs,
which is currently VFIO and VDPA"
For more background, see the extended explanations in Jason's pull request:
https://lore.kernel.org/lkml/Y5dzTU8dlmXTbzoJ@nvidia.com/
* tag 'for-linus-iommufd' of git://git.kernel.org/pub/scm/linux/kernel/git/jgg/iommufd: (62 commits)
iommufd: Change the order of MSI setup
iommufd: Improve a few unclear bits of code
iommufd: Fix comment typos
vfio: Move vfio group specific code into group.c
vfio: Refactor dma APIs for emulated devices
vfio: Wrap vfio group module init/clean code into helpers
vfio: Refactor vfio_device open and close
vfio: Make vfio_device_open() truly device specific
vfio: Swap order of vfio_device_container_register() and open_device()
vfio: Set device->group in helper function
vfio: Create wrappers for group register/unregister
vfio: Move the sanity check of the group to vfio_create_group()
vfio: Simplify vfio_create_group()
iommufd: Allow iommufd to supply /dev/vfio/vfio
vfio: Make vfio_container optionally compiled
vfio: Move container related MODULE_ALIAS statements into container.c
vfio-iommufd: Support iommufd for emulated VFIO devices
vfio-iommufd: Support iommufd for physical VFIO devices
vfio-iommufd: Allow iommufd to be used in place of a container fd
vfio: Use IOMMU_CAP_ENFORCE_CACHE_COHERENCY for vfio_file_enforced_coherent()
...
Resolve conflicts in drivers/vfio/vfio_main.c by using the iommfd version.
The rc fix was done a different way when iommufd patches reworked this
code.
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
This reverts cccc46ae36 ("dmaengine: remove s3c24xx driver") as it
causes regression due to missing header
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Adjust to reality and remove another layer of pointless Kconfig
indirection. CONFIG_GENERIC_MSI_IRQ is good enough to serve
all purposes.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20221111122014.524842979@linutronix.de
Add support for dma-channel-mask so that only the specified channels
are used. This helps to reserve some channels for the firmware.
This was initially achieved by limiting the channel number to 31 in
the driver and adjusting the register address to skip channel0 which
was reserved for a firmware. This is wrong and does not align with
the hardware.
Now, with this change, the driver can align more to the actual hardware
which has 32 channels. But this implies that there will be a break in the
ABI and the device tree need to be updated along with this change for the
driver to pickup the right interrupt corresponding to the channel
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/all/Y2EFoG1H9YpfxRjs@orome/
Signed-off-by: Akhil R <akhilrajeev@nvidia.com>
Link: https://lore.kernel.org/r/20221110171748.40304-4-akhilrajeev@nvidia.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Convert the driver to use the core virt-dma. The driver will be easier to
maintain as it uses the list handling and the tasklet from virt-dma.
With the conversion replace the election of a new transfer in the tasklet
with the election of the new transfer in the interrupt handler. With this
we have a shorter idle window as we remove the scheduling latency of the
tasklet. I chose to do this while doing the conversion to virt-dma,
because if I made a prerequisite patch with the new transfer election in
the irq handler, I would have to duplicate some virt-dma code in the
at_hdmac driver that would end up being removed at the virt-dma conversion
anyway. So do this in a single step.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-33-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
"dma_common" name was misleading and did not suggest that's actually
a struct dma_device underneath. Rename it so that readers can follow the
code easier. One may see some checks and a warning when running
checkpatch. Those have nothing to do with the rename and will be addressed
in a further patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-30-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Use the bitfield access macros in order to clean and to make the driver
easier to read. One will see some "line length exceeds 100 columns"
checkpatch warnings. I chose to not introduce new lines for regs
descriptions in order to preserve the style of the comments throughout
the definitions. Style can be fixed in a further patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-29-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Do not expose register definitions, structures and helpers via a .h file
because there are used only by at_hdmac.c. Since there are no other users,
remove the ambiguity and move all the .h contents to the .c file.
One may notice some checkpatch warnings and errors with this move. The move
was done "as it was", checkpatch complaints can be fixed in a further
patch.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-28-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Use the resource-managed kzalloc to simplify error logic. Memory allocated
with this function is automatically freed on driver detach. Use
struct_size() helper to calculate the size of the atdma structure with its
trailing flexible array. While here, move the mem allocation higher in the
probe method, as failing to allocate memory indicates a serious system
issue, and everything else does not matter anyway. All these help the code
look a bit cleaner.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-23-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
struct dma_tx_state defines residue as u32. atc_get_bytes_left() returned
an int which could be either an error or the value of the residue. This
could cause problems if the controller supported a u32 buffer transfer size
and the u32 value was past the max int can hold. Our controller does not
support u32 buffer transfer size, but even so, improve the code and pass
the residue by address to avoid unnecessary implicit casts and make
atc_get_bytes_left() return 0 on success or -errno on errors.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-20-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
atc_complete_all() had concurrency bugs, thus remove it:
1/ atc_complete_all() in its entirety was buggy, as when the atchan->queue
list (the one that contains descriptors that are not yet issued to the
hardware) contained descriptors, it fired just the first from the
atchan->queue, but moved all the desc from atchan->queue to
atchan->active_list and considered them all as fired. This could result in
calling the completion of a descriptor that was not yet issued to the
hardware.
2/ when in tasklet at atc_advance_work() time, atchan->active_list was
queried without holding the lock of the chan. This can result in
atchan->active_list concurrency problems between the tasklet and
issue_pending().
Fixes: dc78baa2b9 ("dmaengine: at_hdmac: new driver for the Atmel AHB DMA Controller")
Reported-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/lkml/13c6c9a2-6db5-c3bf-349b-4c127ad3496a@axentia.se/
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-8-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
tx_submit is supposed to push the current transaction descriptor to a
pending queue, waiting for issue_pending() to be called. issue_pending()
must start the transfer, not tx_submit(), thus remove atc_dostart() from
atc_tx_submit(). Clients of at_xdmac that assume that tx_submit() starts
the transfer must be updated and call dma_async_issue_pending() if they
miss to call it.
The vdbg print was moved to after the lock is released. It is desirable to
do the prints without the lock held if possible, and because the if
statement disappears there's no reason why to do the print while holding
the lock.
Fixes: dc78baa2b9 ("dmaengine: at_hdmac: new driver for the Atmel AHB DMA Controller")
Reported-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/lkml/13c6c9a2-6db5-c3bf-349b-4c127ad3496a@axentia.se/
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Link: https://lore.kernel.org/r/20221025090306.297886-1-tudor.ambarus@microchip.com
Link: https://lore.kernel.org/r/20221025090306.297886-3-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
When disabling dma channel, a TCF flag is set and as TCIE is enabled, an
interrupt is raised.
On a busy system, the interrupt may have latency and the user can ask for
dmaengine_resume while stm32-dma driver has not yet managed the complete
pause (backup of registers to restore state in resume).
To avoid such a case, instead of waiting the interrupt to backup the
registers, do it just after disabling the channel and discard Transfer
Complete interrupt in case the channel is paused.
Fixes: 099a9a94be ("dmaengine: stm32-dma: add device_pause/device_resume support")
Signed-off-by: Amelie Delaunay <amelie.delaunay@foss.st.com>
Link: https://lore.kernel.org/r/20221024083611.132588-1-amelie.delaunay@foss.st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
If device_register() fails, it should call put_device() to give
up reference, the name allocated in dev_set_name() can be freed
in callback function kobject_cleanup().
Fixes: 5b65781d06 ("dmaengine: ti: k3-udma-glue: Add support for K3 PKTDMA")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20221020062827.2914148-1-yangyingliang@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The of_xlate callback is supposed to return the channel after already
having 'grabbed' it for private use, so fill that in.
Fixes: b127315d9a ("dmaengine: apple-admac: Add Apple ADMAC driver")
Signed-off-by: Martin Povišer <povik+lin@cutebit.org>
Link: https://lore.kernel.org/r/20221019132324.8585-1-povik+lin@cutebit.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
When IDXD is not configurable, that means its WQ, engine, and group
configurations cannot be changed. But it can be disabled and its state
should be set as disabled regardless it's configurable or not.
Fix this by setting device state IDXD_DEV_DISABLED for read-only device
as well in idxd_device_clear_state().
Fixes: cf4ac3fef3 ("dmaengine: idxd: fix lockdep warning on device driver removal")
Signed-off-by: Fengqian Gao <fengqian.gao@intel.com>
Reviewed-by: Xiaochen Shen <xiaochen.shen@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220930032835.2290-1-fengqian.gao@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
>From Intel IAA spec [1], Intel IAA does not support batch processing.
Two batch related default values for IAA are incorrect in current code:
(1) The max batch size of device is set during device initialization,
that indicates batch is supported. It should be always 0 on IAA.
(2) The max batch size of work queue is set to WQ_DEFAULT_MAX_BATCH (32)
as the default value regardless of Intel DSA or IAA device during
work queue setup and cleanup. It should be always 0 on IAA.
Fix the issues by setting the max batch size of device and max batch
size of work queue to 0 on IAA device, that means batch is not
supported.
[1]: https://cdrdv2.intel.com/v1/dl/getContent/721858
Fixes: 23084545db ("dmaengine: idxd: set max_xfer and max_batch for RO device")
Fixes: 92452a72eb ("dmaengine: idxd: set defaults for wq configs")
Fixes: bfe1d56091 ("dmaengine: idxd: Init and probe for Intel data accelerators")
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220930201528.18621-2-xiaochen.shen@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The first IRQ is required, but IRQs 1 through (nb_phy_chans - 1) are
optional, because on some platforms (e.g. PXA168) there is a single IRQ
shared between all channels.
This change inhibits a flood of "IRQ index # not found" messages at
startup. Tested on a PXA168-based device.
Fixes: 7723f4c5ec ("driver core: platform: Add an error message to platform_get_irq*()")
Signed-off-by: Doug Brown <doug@schmorgal.com>
Link: https://lore.kernel.org/r/20220906000709.52705-1-doug@schmorgal.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
In current code, the following sysfs attributes are exposed to user to
show or update the values:
max_read_buffers (max_tokens)
read_buffer_limit (token_limit)
group/read_buffers_allowed (group/tokens_allowed)
group/read_buffers_reserved (group/tokens_reserved)
group/use_read_buffer_limit (group/use_token_limit)
>From Intel IAA spec [1], Intel IAA does not support Read Buffer
allocation control. So these sysfs attributes should not be supported on
IAA device.
Fix this issue by making these sysfs attributes invisible through
is_visible() filter when the device is IAA.
Add description in the ABI documentation to mention that these
attributes are not visible when the device does not support Read Buffer
allocation control.
[1]: https://cdrdv2.intel.com/v1/dl/getContent/721858
Fixes: fde212e44f ("dmaengine: idxd: deprecate token sysfs attributes for read buffers")
Fixes: c52ca47823 ("dmaengine: idxd: add configuration component of driver")
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20221022074949.11719-1-xiaochen.shen@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
shdma-arm.h was introduced with commit 1e69653d40 ("DMA: shdma: add
r8a73a4 DMAC data to the device ID table"), and its sole user was
removed with commit a19788612f ("dmaengine: sh: Remove R-Mobile APE6
support"). The latter mentions r8a73a4.dtsi but shdma support was
removed from that with commit cfda820377 ("ARM: dts: r8a73a4: Remove
non-functional DMA support"), so it seems this is safe to remove.
Signed-off-by: Stephen Kitt <steve@sk2.org>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20221028115336.1052782-1-steve@sk2.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The s3c24xx platform was removed and this driver is no longer
needed.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20221021203329.4143397-14-arnd@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The current kernel DMA with PASID support is based on the SVA with a flag
SVM_FLAG_SUPERVISOR_MODE. The IOMMU driver binds the kernel memory address
space to a PASID of the device. The device driver programs the device with
kernel virtual address (KVA) for DMA access. There have been security and
functional issues with this approach:
- The lack of IOTLB synchronization upon kernel page table updates.
(vmalloc, module/BPF loading, CONFIG_DEBUG_PAGEALLOC etc.)
- Other than slight more protection, using kernel virtual address (KVA)
has little advantage over physical address. There are also no use
cases yet where DMA engines need kernel virtual addresses for in-kernel
DMA.
This removes SVM_FLAG_SUPERVISOR_MODE support from the IOMMU interface.
The device drivers are suggested to handle kernel DMA with PASID through
the kernel DMA APIs.
The drvdata parameter in iommu_sva_bind_device() and all callbacks is not
needed anymore. Cleanup them as well.
Link: https://lore.kernel.org/linux-iommu/20210511194726.GP1002214@nvidia.com/
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Tested-by: Tony Zhu <tony.zhu@intel.com>
Link: https://lore.kernel.org/r/20221031005917.45690-4-baolu.lu@linux.intel.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
The iop32x platform was removed, so this driver is no longer
needed.
Cc: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Link: https://lore.kernel.org/r/20221019150410.3851944-10-arnd@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
There's a previously unknown part of the controller interface: We have
to assign SRAM carveouts to channels to store their in-flight samples
in. So, obtain the size of the SRAM from a read-only register and divide
it into 2K blocks for allocation to channels. The FIFO depths we
configure will always fit into 2K.
(This fixes audio artifacts during simultaneous playback/capture on
multiple channels -- which looking back is fully accounted for by having
had the caches in the DMA controller overlap in memory.)
Fixes: b127315d9a ("dmaengine: apple-admac: Add Apple ADMAC driver")
Signed-off-by: Martin Povišer <povik+lin@cutebit.org>
Link: https://lore.kernel.org/r/20221019132324.8585-2-povik+lin@cutebit.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
In current code, dev.max_batch_size and wq.max_batch_size attributes in
sysfs are exposed to user to show or update the values.
>From Intel IAA spec [1], Intel IAA does not support batch processing. So
these sysfs attributes should not be supported on IAA device.
Fix this issue by making the attributes of max_batch_size invisible in
sysfs through is_visible() filter when the device is IAA.
Add description in the ABI documentation to mention that the attributes
are not visible when the device does not support batch.
[1]: https://cdrdv2.intel.com/v1/dl/getContent/721858
Fixes: e7184b159d ("dmaengine: idxd: add support for configurable max wq batch size")
Fixes: c52ca47823 ("dmaengine: idxd: add configuration component of driver")
Signed-off-by: Xiaochen Shen <xiaochen.shen@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220930201528.18621-3-xiaochen.shen@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The function idma64_remove() returns zero unconditionally. Make it
return void.
This is a preparation for making platform remove callbacks return void.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20221014161250.468687-1-u.kleine-koenig@pengutronix.de
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The JZ4755 has 4 DMA channels per DMA unit, two idential DMA units.
The JZ4755 has the similar DMA engine to JZ4725b and it has the
same bug as JZ4725b, see commit a40c94be23.
At least the JZ_SOC_DATA_BREAK_LINKS flag make it work much better,
although not ideal.
Reviewed-by: Paul Cercueil <paul@crapouillou.net>
Tested-by: Siarhei Volkau <lis8215@gmail.com>
Signed-off-by: Siarhei Volkau <lis8215@gmail.com>
Link: https://lore.kernel.org/r/20221019063934.3278444-3-lis8215@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
When the idxd_user_drv driver is bound to a Work Queue (WQ) device
without IOMMU or with IOMMU Passthrough without Shared Virtual
Addressing (SVA), the application gains direct access to physical
memory via the device by programming physical address to a submitted
descriptor. This allows direct userspace read and write access to
arbitrary physical memory. This is inconsistent with the security
goals of a good kernel API.
Unlike vfio_pci driver, the IDXD char device driver does not provide any
ways to pin user pages and translate the address from user VA to IOVA or
PA without IOMMU SVA. Therefore the application has no way to instruct the
device to perform DMA function. This makes the char device not usable for
normal application usage.
Since user type WQ without SVA cannot be used for normal application usage
and presents the security issue, bind idxd_user_drv driver and enable user
type WQ only when SVA is enabled (i.e. user PASID is enabled).
Fixes: 448c3de8ac ("dmaengine: idxd: create user driver for wq 'device'")
Cc: stable@vger.kernel.org
Suggested-by: Arjan Van De Ven <arjan.van.de.ven@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Link: https://lore.kernel.org/r/20221014222541.3912195-1-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
New bool m2m_hw has been added at the end of stm32_mdma_chan_config struct
to support the STM32 DMA MDMA chaining.
m2m_hw is set true in stm32_mdma_slave_config() if peripheral_size is set,
but m2m_hw is never initialized false.
To ensure this case, and any further new update of the structure, memset it
to 0 before using it.
Fixes: 6968743227 ("dmaengine: stm32-mdma: add support to be triggered by STM32 DMA")
Signed-off-by: Amelie Delaunay <amelie.delaunay@foss.st.com>
Link: https://lore.kernel.org/r/20221017131413.202567-1-amelie.delaunay@foss.st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Combine all the SoC specific files into a single lib that can be
built-in or built as a module.
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>
Link: https://lore.kernel.org/r/20220929234820.940048-4-khilman@baylibre.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Currently k3-udma driver is built as separate platform drivers with a
shared probe and identical code path, just differnet platform data.
To enable to build as module, convert the separate platform driver
into a single module_platform_driver with the data selection done via
compatible string and of_match. The separate of_match tables are also
combined into a single table to avoid the multiple calls to
of_match_node()
Since all modern TI platforms using this are DT enabled, the removal
of separate platform_drivers should have no functional change.
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Signed-off-by: Kevin Hilman <khilman@baylibre.com>
Link: https://lore.kernel.org/r/20220929234820.940048-3-khilman@baylibre.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Devices with ee offset of 0x10000 should rather bind with SM6350
compatible, so the list will not unnecessarily grow for compatible
devices.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Richard Acayan <mailingradian@gmail.com>
Link: https://lore.kernel.org/r/20221018230352.1238479-3-krzysztof.kozlowski@linaro.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The drivers are transitioning from matching against lists of specific
compatible strings to matching against smaller lists of more generic
compatible strings. Add a message that the compatible strings with an
ee_offset of 0 are deprecated except for the SDM845 compatible string.
Signed-off-by: Richard Acayan <mailingradian@gmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20221018005740.23952-4-mailingradian@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The prandom_bytes() function has been a deprecated inline wrapper around
get_random_bytes() for several releases now, and compiles down to the
exact same code. Replace the deprecated wrapper with a direct call to
the real function. This was done as a basic find and replace.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Yury Norov <yury.norov@gmail.com>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu> # powerpc
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
New Support:
- AngeloGioacchino Del Regno added support for MT6795 SoC dma controller
- Christian Marangi updated qcom-adm controller binding to yaml
- Geert Uytterhoeven added yaml binding for Renesas r8a779g0 dma controller
- Luca Weiss added support for Qualcomm SM6350 GPI dma controller
Updates:
- Amelie Delaunay provided STM32 DMA-MDMA chaining support
- Andy Shevchenko updated hsu driver to use managed resources
- Dave Jiang & Jerry Snitselaar provided usual round of idxd driver updates
- Janne Grunau & Martin Povišer updated apple dma driver for iommu and pd
properties and removed use of devres for irqs
- Swati Agarwal added device_synchronize support for Xilinx zynqmp driver
-----BEGIN PGP SIGNATURE-----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=zRon
-----END PGP SIGNATURE-----
Merge tag 'dmaengine-6.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine
Pull dmaengine updates from Vinod Koul:
"New Support:
- MT6795 SoC dma controller (AngeloGioacchino Del Regno)
- qcom-adm controller yaml binding (Christian Marangi)
- Renesas r8a779g0 dma controller yaml binding (Geert Uytterhoeven)
- Qualcomm SM6350 GPI dma controller (Luca Weiss)
Updates:
- STM32 DMA-MDMA chaining support (Amelie Delaunay)
- make hsu driver use managed resources (Andy Shevchenko)
- the usual round of idxd driver updates (Dave Jiang & Jerry
Snitselaar)
- apple dma driver iommu and pd properties and remove use
of devres for irqs (Janne Grunau & Martin Povišer)
- device_synchronize support for Xilinx zynqmp driver (Swati
Agarwal)"
* tag 'dmaengine-6.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (60 commits)
dmaengine: ioat: remove unused declarations in dma.h
dmaengine: ti: k3-udma: Respond TX done if DMA_PREP_INTERRUPT is not requested
dmaengine: zynqmp_dma: Add device_synchronize support
dt-bindings: dma: add additional pbus reset to qcom,adm
dt-bindings: dma: rework qcom,adm Documentation to yaml schema
dt-bindings: dma: apple,admac: Add iommus and power-domains properties
dmaengine: dw-edma: Remove runtime PM support
dmaengine: idxd: add configuration for concurrent batch descriptor processing
dmaengine: idxd: add configuration for concurrent work descriptor processing
dmaengine: idxd: add WQ operation cap restriction support
dmanegine: idxd: reformat opcap output to match bitmap_parse() input
dmaengine: idxd: convert ats_dis to a wq flag
dmaengine: ioat: stop mod_timer from resurrecting deleted timer in __cleanup()
dmaengine: qcom-adm: fix wrong calling convention for prep_slave_sg
dmaengine: qcom-adm: fix wrong sizeof config in slave_config
dmaengine: ti: k3-psil: add additional TX threads for j721e
dmaengine: ti: k3-psil: add additional TX threads for j7200
dmaengine: apple-admac: Trigger shared reset
dmaengine: apple-admac: Do not use devres for IRQs
dmaengine: ti: edma: Remove some unused functions
...
ioat_ring_alloc_order and ioat_ring_max_alloc_order have
been removed since commit cd60cd9613 ("dmaengine: IOATDMA:
Removing descriptor ring reshape"), so remove them.
Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20220911091817.3214271-1-cuigaosheng1@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
If the DMA consumer driver does not expect the callback for TX done, then
we need not perform the channel RT byte counter calculations and estimate
the completion but return complete on first attempt itself.This assumes
that the consumer who did not request DMA_PREP_INTERRUPT has its own
mechanism for understanding TX completion, example: MCSPI EOW interrupt
can be used as TX completion signal for a SPI transaction.
Signed-off-by: Vaishnav Achath <vaishnav.a@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20220914110049.5842-1-vaishnav.a@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
dmaengine_synchronize implementation is required to synchronize proper
termination of current transfers so that any memory resources are not freed
while still in use.
Implement this callback in the driver so that framework can use the same
(in dmaengine_terminate_sync/ dmaengine_synchronize).
Signed-off-by: Swati Agarwal <swati.agarwal@amd.com>
Link: https://lore.kernel.org/r/20220915090516.5812-1-swati.agarwal@amd.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Currently, the dw-edma driver enables the runtime_pm for parent device
(chip->dev) and increments/decrements the refcount during alloc/free
chan resources callbacks.
This leads to a problem when the eDMA driver has been probed, but the
channels were not used. This scenario can happen when the DW PCIe driver
probes eDMA driver successfully, but the PCI EPF driver decides not to
use eDMA channels and use iATU instead for PCI transfers.
In this case, the underlying device would be runtime suspended due to
pm_runtime_enable() in dw_edma_probe() and the PCI EPF driver would have
no knowledge of it.
Ideally, the eDMA driver should not be the one doing the runtime PM of
the parent device. The responsibility should instead belong to the client
drivers like PCI EPF.
So let's remove the runtime PM support from eDMA driver.
Cc: Serge Semin <fancer.lancer@gmail.com>
Cc: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Serge Semin <fancer.lancer@gmail.com>
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Link: https://lore.kernel.org/r/20220910054700.12205-1-manivannan.sadhasivam@linaro.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add sysfs knob to allow control of the number of batch descriptors that can
be concurrently processed by an engine in the group as a fraction of the
Maximum Work Descriptors in Progress value specfied in ENGCAP register.
This control knob is part of toggle for QoS control.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220917161222.2835172-6-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add sysfs knob to allow control of the number of work descriptors that can
be concurrently processed by an engine in the group as a fraction of the
Maximum Work Descriptors in Progress value specified in ENGCAP register.
This control knob is part of toggle for QoS control.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220917161222.2835172-5-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
DSA 2.0 add the capability of configuring DMA ops on a per workqueue basis.
This means that certain ops can be disabled by the system administrator for
certain wq. By default, all ops are available. A bitmap is used to store
the ops due to total op size of 256 bits and it is more convenient to use a
range list to specify which bits are enabled.
One of the usage to support this is for VM migration between different
iteration of devices. The newer ops are disabled in order to allow guest to
migrate to a host that only support older ops. Another usage is to
restrict the WQ to certain operations for QoS of performance.
A sysfs of ops_config attribute is added per wq. It is only usable when the
ops_config bit is set under WQ_CAP register. This means that this attribute
will return -EOPNOTSUPP on DSA 1.x devices. The expected input is a range
list for the bits per operation the WQ supports.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220917161222.2835172-4-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
To make input and output consistent and prepping for the per WQ operation
configuration support, change the output of opcap display to match the
input that is expected by bitmap_parse() helper function. The output will
be a bitmap with field width as the number of bits using the %*pb format
specifier for printk() family.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Co-developed-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Link: https://lore.kernel.org/r/20220917161222.2835172-3-fenghua.yu@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
User reports observing timer event report channel halted but no error
observed in CHANERR register. The driver finished self-test and released
channel resources. Debug shows that __cleanup() can call
mod_timer() after the timer has been deleted and thus resurrect the
timer. While harmless, it causes suprious error message to be emitted.
Use mod_timer_pending() call to prevent deleted timer from being
resurrected.
Fixes: 3372de5813 ("dmaengine: ioatdma: removal of dma_v3.c and relevant ioat3 references")
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/166360672197.3851724.17040290563764838369.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The calling convention for pre_slave_sg is to return NULL on error and
provide an error log to the system. Qcom-adm instead provide error
pointer when an error occur. This indirectly cause kernel panic for
example for the nandc driver that checks only if the pointer returned by
device_prep_slave_sg is not NULL. Returning an error pointer makes nandc
think the device_prep_slave_sg function correctly completed and makes
the kernel panics later in the code.
While nandc is the one that makes the kernel crash, it was pointed out
that the real problem is qcom-adm not following calling convention for
that function.
To fix this, drop returning error pointer and return NULL with an error
log.
Fixes: 03de6b2738 ("dmaengine: qcom-adm: stop abusing slave_id config")
Fixes: 5c9f8c2dbd ("dmaengine: qcom: Add ADM driver")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Cc: stable@vger.kernel.org # v5.11+
Link: https://lore.kernel.org/r/20220916041256.7104-1-ansuelsmth@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Fix broken slave_config function that uncorrectly compare the
peripheral_size with the size of the config pointer instead of the size
of the config struct. This cause the crci value to be ignored and cause
a kernel panic on any slave that use adm driver.
To fix this, compare to the size of the struct and NOT the size of the
pointer.
Fixes: 03de6b2738 ("dmaengine: qcom-adm: stop abusing slave_id config")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Cc: stable@vger.kernel.org # v5.17+
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20220915204844.3838-1-ansuelsmth@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
If a reset domain is attached to the device, obtain a shared reference
to it and trigger it. Typically on a chip the ADMAC controller will
share a reset domain with the MCA peripheral.
Signed-off-by: Martin Povišer <povik+lin@cutebit.org>
Link: https://lore.kernel.org/r/20220918095845.68860-5-povik+lin@cutebit.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
This is in advance of adding support for triggering the reset signal to
the peripheral, since registering the IRQ handler will have to be
sequenced with it.
Signed-off-by: Martin Povišer <povik+lin@cutebit.org>
Link: https://lore.kernel.org/r/20220918095845.68860-4-povik+lin@cutebit.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
These functions are defined in the edma.c file, but not called elsewhere,
so delete these unused functions.
drivers/dma/ti/edma.c:746:31: warning: unused function 'to_edma_cc'.
drivers/dma/ti/edma.c:420:20: warning: unused function 'edma_param_or'.
drivers/dma/ti/edma.c:414:20: warning: unused function 'edma_param_and'.
drivers/dma/ti/edma.c:402:20: warning: unused function 'edma_param_write'.
drivers/dma/ti/edma.c:373:28: warning: unused function 'edma_shadow0_read'.
drivers/dma/ti/edma.c:396:28: warning: unused function 'edma_param_read'.
drivers/dma/ti/edma.c:355:20: warning: unused function 'edma_or_array'.
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2152
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20220914101943.83929-1-jiapeng.chong@linux.alibaba.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Since fault processing code has been removed, struct idxd_fault is not used any
more and can be removed as well.
Signed-off-by: Yuan Can <yuancan@huawei.com>
Link: https://lore.kernel.org/r/20220928014747.106808-1-yuancan@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Now that idxd_wq_disable_cleanup() sets the workqueue state to
IDXD_WQ_DISABLED, use a bitmap to track which workqueues have been
enabled. This will then be used to determine which workqueues
should be re-enabled when attempting a software reset to recover
from a device halt state.
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20220928154856.623545-3-jsnitsel@redhat.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
If we are calling idxd_wq_disable_cleanup(), the workqueue should be
in a disabled state. So set the workqueue state to IDXD_WQ_DISABLED so
that the state reflects that. Currently if there is a device failure,
and a software reset is attempted the workqueues will not be
re-enabled due to idxd_wq_enable() seeing that state as already being
IDXD_WQ_ENABLED.
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/20220928154856.623545-2-jsnitsel@redhat.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Driver registration fails on SOC imx8mn as its supplier, the clock
control module, is probed later than subsys initcall level. This driver
uses platform_driver_probe which is not compatible with deferred probing
and won't be probed again later if probe function fails due to clock not
being available at that time.
This patch replaces the use of platform_driver_probe with
platform_driver_register which will allow probing the driver later again
when the clock control module will be available.
The __init annotation has been dropped because it is not compatible with
deferred probing. The code is not executed once and its memory cannot be
freed.
Fixes: a580b8c542 ("dmaengine: mxs-dma: add dma support for i.MX23/28")
Co-developed-by: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20220921170556.1055962-1-dario.binacchi@amarulasolutions.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Since commit 0166dc11be ("of: make CONFIG_OF user selectable"), it
is possible to test-build any driver which depends on OF on any
architecture by explicitly selecting OF. Therefore depending on
COMPILE_TEST as an alternative is no longer needed.
It is actually better to always build such drivers with OF enabled,
so that the test builds are closer to how each driver will actually be
built on its intended target. Building them without OF may not test
much as the compiler will optimize out potentially large parts of the
code. In the worst case, this could even pop false positive warnings.
Dropping COMPILE_TEST here improves the quality of our testing and
avoids wasting time on non-existent issues.
Signed-off-by: Jean Delvare <jdelvare@suse.de>
Cc: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Cc: Vinod Koul <vkoul@kernel.org>
Link: https://lore.kernel.org/r/20220803223448.6f08095b@endymion.delvare
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add a cosmetic change and replace two if statements with a single if
statement with two conditions. In case the optional txstate parameter is
NULL, we return the dma_cookie_status, which is fine, no functional change
required.
Signed-off-by: Tudor Ambarus <tudor.ambarus@microchip.com>
Link: https://lore.kernel.org/r/20220802140630.243550-1-tudor.ambarus@microchip.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
We should call of_node_put() for the reference returned by
of_parse_phandle() in fail path or when it is not used anymore.
Here we only need to move the of_node_put() before the check.
Fixes: d702419134 ("dmaengine: ti: k3-udma: Add glue layer for non DMAengine users")
Signed-off-by: Liang He <windhl@126.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20220720073234.1255474-1-windhl@126.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
STM32_DMAMUX_MAX_DMA_REQUESTS is small (i.e. 32) and when the 'dma_inuse'
bitmap is allocated, there is already a check that 'dma_req' is <= this
limit.
So, there is no good reason to dynamically allocate this bitmap. This
just waste some memory and some cycles.
Use DECLARE_BITMAP with the maximum bitmap size instead.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/r/2d8c24359b2daa32ce0597a2949b7b2bebaf23de.1659211633.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Vinod Koul <vkoul@kernel.org>
>From the coccinelle check:
./drivers/dma/sf-pdma/sf-pdma.c
Error:line 409 is redundant because platform_get_irq() already prints an
error
./drivers/dma/sf-pdma/sf-pdma.c
Error:line 424 is redundant because platform_get_irq() already prints an
error
So,remove the unnecessary print function dev_err()
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: ye xingchen <ye.xingchen@zte.com.cn>
Link: https://lore.kernel.org/r/20220810062532.13425-1-ye.xingchen@zte.com.cn
Signed-off-by: Vinod Koul <vkoul@kernel.org>
UDMA_CHAN_RT_*BCNT_REG stores the real-time channel bytecount statistics.
These registers are 32-bit hardware counters and the driver uses these
counters to monitor the operational progress status for a channel, when
transferring more than 4GB of data it was observed that these counters
overflow and completion calculation of a operation gets affected and the
transfer hangs indefinitely.
This commit adds changes to decrease the byte count for every complete
transaction so that these registers never overflow and the proper byte
count statistics is maintained for ongoing transaction by the RT counters.
Earlier uc->bcnt used to maintain a count of the completed bytes at driver
side, since the RT counters maintain the statistics of current transaction
now, the maintenance of uc->bcnt is not necessary.
Signed-off-by: Vaishnav Achath <vaishnav.a@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20220802054835.19482-1-vaishnav.a@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The driver does not handle the failure case while calling
dma_set_mask_and_coherent API.
In case of failure, capture the return value of API and then report an
error.
Addresses-coverity: Unchecked return value (CHECKED_RETURN)
Signed-off-by: Swati Agarwal <swati.agarwal@xilinx.com>
Reviewed-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Link: https://lore.kernel.org/r/20220817061125.4720-4-swati.agarwal@xilinx.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Add missing cleanup in devm_platform_ioremap_resource().
When probe fails remove dma channel resources and disable clocks in
accordance with the order of resources allocated .
Signed-off-by: Swati Agarwal <swati.agarwal@xilinx.com>
Link: https://lore.kernel.org/r/20220817061125.4720-2-swati.agarwal@xilinx.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
STM32 MDMA can be triggered by STM32 DMA channels transfer complete.
In case of non-null struct dma_slave_config .peripheral_size, it means the
DMA client wants the DMA to trigger the MDMA.
stm32-mdma driver gets the request id, the mask_addr, and the mask_data in
struct stm32_mdma_dma_config passed by DMA with struct dma_slave_config
.peripheral_config/.peripheral_size.
Then, as DMA is configured in Double-Buffer mode, and MDMA channel will
transfer data from/to SRAM to/from DDR, then bursts are optimized.
Signed-off-by: Amelie Delaunay <amelie.delaunay@foss.st.com>
Link: https://lore.kernel.org/r/20220829154646.29867-7-amelie.delaunay@foss.st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
STM32 MDMA can be triggered by STM32 DMA channels transfer complete.
The "request line number" triggering STM32 MDMA is the STM32 DMAMUX channel
id set by stm32-dmamux driver in dma_spec->args[3].
stm32-dma driver fills the struct stm32_dma_mdma_config used to configure
the MDMA with struct dma_slave_config .peripheral_config/.peripheral_size.
Signed-off-by: Amelie Delaunay <amelie.delaunay@foss.st.com>
Link: https://lore.kernel.org/r/20220829154646.29867-6-amelie.delaunay@foss.st.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Serge Semin