Simplify and refactor the descriptor management by removing the redundant
Linked List Item (LLI) queue control logic from the AxiDMA driver.
The descriptor is split into virtual descriptor and hardware LLI so that
only hardware LLI memories are allocated from the DMA memory pool.
Up to 64 descriptors can be allocated within a PAGE_SIZE compare to 16
descriptors in previous version. This solves the problem where an
ALSA driver expects more than 16 DMA descriptors to run.
Signed-off-by: Sia Jee Heng <jee.heng.sia@intel.com>
Tested-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Eugeniy Paltsev <Eugeniy.Paltsev@synopsys.com>
Link: https://lore.kernel.org/r/20210125013255.25799-3-jee.heng.sia@intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The Intel Lightning Mountain (LGM) DMA controller is only present on
Intel Lightning Mountain SoCs. Hence add a dependency on X86, to
prevent asking the user about this driver when configuring a kernel
without Intel Lightning Mountain platform support.
While at it, fix a misspelling of "Intel".
Fixes: 32d31c79a1 ("dmaengine: Add Intel LGM SoC DMA support.")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20210129131702.2656060-1-geert+renesas@glider.be
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The DMACs (both SYS-DMAC and RT-DMAC) on R-Car V3U differ slightly from
the DMACs on R-Car Gen2 and other R-Car Gen3 SoCs:
1. The per-channel registers are located in a second register block.
Add support for mapping the second block, using the appropriate
offsets and stride.
2. The common Channel Clear Register (DMACHCLR) was replaced by a
per-channel register.
Update rcar_dmac_chan_clear{,_all}() to handle this.
As rcar_dmac_init() needs to clear the status before the individual
channels are probed, channel index and base address initialization
are moved forward.
Inspired by a patch in the BSP by Phong Hoang
<phong.hoang.wz@renesas.com>.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Tested-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Link: https://lore.kernel.org/r/20210128084455.2237256-5-geert+renesas@glider.be
Signed-off-by: Vinod Koul <vkoul@kernel.org>
When bam dma is "controlled remotely", thus far clocks were not controlled
from the Linux. In this scenario, Linux was disabling runtime pm in bam dma
driver and not doing any clock management in suspend/resume hooks.
With introduction of crypto engine bam dma, the clock is a rpmh resource
that can be controlled from both Linux and TZ/remote side. Now bam dma
clock is getting enabled during probe even though the bam dma can be
"controlled remotely". But due to clocks not being handled properly,
bam_suspend generates a unbalanced clk_unprepare warning during system
suspend.
To fix the above issue and to enable proper clock-management, this patch
enables runtim-pm and handles bam dma clocks in suspend/resume hooks if
the clock node is present irrespective of controlled_remotely property.
Signed-off-by: Thara Gopinath <thara.gopinath@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Tested-by: John Stultz <john.stultz@linaro.org>
Reviewed-by: Shawn Guo <shawn.guo@linaro.org>
Link: https://lore.kernel.org/r/20210126211859.790892-1-thara.gopinath@linaro.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Nikhil reported the misc interrupt handler can sometimes miss handling
the command interrupt when an error interrupt happens near the same time.
Have the irq handling thread continue to process the misc interrupts until
all interrupts are processed. This is a low usage interrupt and is not
expected to handle high volume traffic. Therefore there is no concern of
this thread running for a long time.
Fixes: 0d5c10b4c8 ("dmaengine: idxd: add work queue drain support")
Reported-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/161074755329.2183844.13295528344116907983.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
On Intel Tangier B0 and Anniedale the interrupt line, disregarding
to have different numbers, is shared between HSU DMA and UART IPs.
Thus on such SoCs we are expecting that IRQ handler is called in
UART driver only. hsu_pci_irq was handling the spurious interrupt
from HSU DMA by returning immediately. This wastes CPU time and
since HSU DMA and HSU UART interrupt occur simultaneously they race
to be handled causing delay to the HSU UART interrupt handling.
Fix this by disabling the interrupt entirely.
Fixes: 4831e0d905 ("serial: 8250_mid: handle interrupt correctly in DMA case")
Signed-off-by: Ferry Toth <ftoth@exalondelft.nl>
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Link: https://lore.kernel.org/r/20210112223749.97036-1-ftoth@exalondelft.nl
Signed-off-by: Vinod Koul <vkoul@kernel.org>
The UDMA and BCDMA can provide higher throughput if the burst_size of the
channel is changed from it's default (which is 64 bytes) for Ultra-high
and high capacity channels.
This performance benefit is even more visible when the buffers are aligned
with the burst_size configuration.
The am654 does not have a way to change the burst size, but it is using
64 bytes burst, so increasing the copy_align from 8 bytes to 64 (and
clients taking that into account) can increase the throughput as well.
Numbers gathered on j721e:
echo 8000000 > /sys/module/dmatest/parameters/test_buf_size
echo 2000 > /sys/module/dmatest/parameters/timeout
echo 50 > /sys/module/dmatest/parameters/iterations
echo 1 > /sys/module/dmatest/parameters/max_channels
Prior this patch: ~1.3 GB/s
After this patch: ~1.8 GB/s
with 1 byte alignment: ~1.7 GB/s
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Tested-by: Kishon Vijay Abraham I <kishon@ti.com>
Link: https://lore.kernel.org/r/20210113114923.9231-3-peter.ujfalusi@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
BCDMA RX channels have one flow per channel, therefore set the rflow_cnt
to rchan_cnt.
Without this patch, request for BCDMA RX channel allocation fails as
rflow_cnt is 0 thus fails to reserve a rflow for the channel.
Fixes: 8844898028 ("dmaengine: ti: k3-udma: Add support for BCDMA channel TPL handling")
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20210112141403.30286-1-vigneshr@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
When building with CONFIG_UBSAN_UNSIGNED_OVERFLOW, clang decides not to
inline gpi_update_reg, which causes a linkage failure around __bad_mask:
ld.lld: error: undefined symbol: __bad_mask
>>> referenced by bitfield.h:119 (include/linux/bitfield.h:119)
>>> dma/qcom/gpi.o:(gpi_update_reg) in archive drivers/built-in.a
>>> referenced by bitfield.h:119 (include/linux/bitfield.h:119)
>>> dma/qcom/gpi.o:(gpi_update_reg) in archive drivers/built-in.a
If gpi_update_reg is not inlined, the mask value will not be known at
compile time so the check in field_multiplier stays in the final
object file, causing the above linkage failure. Always inline
gpi_update_reg so that this check can never fail.
Fixes: 5d0c3533a1 ("dmaengine: qcom: Add GPI dma driver")
Link: https://github.com/ClangBuiltLinux/linux/issues/1243
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Link: https://lore.kernel.org/r/20210112191214.1264793-1-natechancellor@gmail.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
A 'irq_dispose_mapping()' call is missing in the remove function.
Add it.
This is needed to undo the 'irq_of_parse_and_map() call from the probe
function and already part of the error handling path of the probe function.
It was added in the probe function only in commit d3f620b2c4 ("fsldma:
simplify IRQ probing and handling")
Fixes: 77cd62e808 ("fsldma: allow Freescale Elo DMA driver to be compiled as a module")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/r/20201212160516.92515-1-christophe.jaillet@wanadoo.fr
Signed-off-by: Vinod Koul <vkoul@kernel.org>
gcc points out an incorrect error handling loop:
drivers/dma/qcom/gpi.c: In function 'gpi_ch_init':
drivers/dma/qcom/gpi.c:1254:15: error: iteration 2 invokes undefined behavior [-Werror=aggressive-loop-optimizations]
1254 | struct gpii *gpii = gchan->gpii;
| ^~~~
drivers/dma/qcom/gpi.c:1951:2: note: within this loop
1951 | for (i = i - 1; i >= 0; i++) {
| ^~~
Change the loop to correctly walk backwards through the
initialized fields rather than off into the woods.
Fixes: 5d0c3533a1 ("dmaengine: qcom: Add GPI dma driver")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20210103135738.3741123-1-arnd@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>
If the dw_edma_alloc_burst() function fails then we free "chunk" but
it's still on the "desc->chunk->list" list so it will lead to a use
after free. Also the "->chunks_alloc" count is incremented when it
shouldn't be.
In current kernels small allocations are guaranteed to succeed and
dw_edma_alloc_burst() can't fail so this will not actually affect
runtime.
Fixes: e63d79d1ff ("dmaengine: Add Synopsys eDMA IP core driver")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Gustavo Pimentel <gustavo.pimentel@synopsys.com>
Link: https://lore.kernel.org/r/X9dTBFrUPEvvW7qc@mwanda
Signed-off-by: Vinod Koul <vkoul@kernel.org>
drivers/dma/qcom/gpi.c:1419:3: warning: format '%lu' expects argument of
type 'long unsigned int', but argument 8 has type 'size_t {aka unsigned
int}' [-Wformat=]
drivers/dma/qcom/gpi.c:1427:31: warning: format '%lu' expects argument of
type 'long unsigned int', but argument 3 has type 'size_t {aka unsigned
int}' [-Wformat=]
drivers/dma/qcom/gpi.c:1447:3: warning: format '%llx' expects argument of
type 'long long unsigned int', but argument 4 has type 'dma_addr_t {aka
unsigned int}' [-Wformat=]
drivers/dma/qcom/gpi.c:1447:3: warning: format '%llx' expects argument of
type 'long long unsigned int', but argument 5 has type 'phys_addr_t {aka
unsigned int}' [-Wformat=]
Signed-off-by: Xiaoming Ni <nixiaoming@huawei.com>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/20201218104137.59200-1-nixiaoming@huawei.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Pull dmaengine updates from Vinod Koul:
"The last dmaengine updates for this year :)
This contains couple of new drivers, new device support and updates to
bunch of drivers.
New drivers/devices:
- Qualcomm ADM driver
- Qualcomm GPI driver
- Allwinner A100 DMA support
- Microchip Sama7g5 support
- Mediatek MT8516 apdma
Updates:
- more updates to idxd driver and support for IAX config
- runtime PM support for dw driver
- TI drivers"
* tag 'dmaengine-5.11-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine: (75 commits)
soc: ti: k3-ringacc: Use correct error casting in k3_ringacc_dmarings_init
dmaengine: ti: k3-udma-glue: Add support for K3 PKTDMA
dmaengine: ti: k3-udma: Initial support for K3 PKTDMA
dmaengine: ti: k3-udma: Add support for BCDMA channel TPL handling
dmaengine: ti: k3-udma: Initial support for K3 BCDMA
soc: ti: k3-ringacc: add AM64 DMA rings support.
dmaengine: ti: Add support for k3 event routers
dmaengine: ti: k3-psil: Add initial map for AM64
dmaengine: ti: k3-psil: Extend psil_endpoint_config for K3 PKTDMA
dt-bindings: dma: ti: Add document for K3 PKTDMA
dt-bindings: dma: ti: Add document for K3 BCDMA
dmaengine: dmatest: Use dmaengine_get_dma_device
dmaengine: doc: client: Update for dmaengine_get_dma_device() usage
dmaengine: Add support for per channel coherency handling
dmaengine: of-dma: Add support for optional router configuration callback
dmaengine: ti: k3-udma-glue: Configure the dma_dev for rings
dmaengine: ti: k3-udma-glue: Get the ringacc from udma_dev
dmaengine: ti: k3-udma-glue: Add function to get device pointer for DMA API
dmaengine: ti: k3-udma: Add support for second resource range from sysfw
dmaengine: ti: k3-udma: Wait for peer teardown completion if supported
...
This commit adds support for PKTDMA in k3-udma glue driver. Use new
psil_endpoint_config struct to get static data for a given channel or a
flow during setup. Make sure that the RX flows being mapped to a RX
channel is within the range of flows that is been allocated to that RX
channel.
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201208090440.31792-21-peter.ujfalusi@ti.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
