linux-stable.git/drivers/dma, branch v5.10.7

dmaengine: idxd: off by one in cleanup code

2021-01-12T19:18:26+00:00

commit ff58f7dd0c1352a01de3a40327895bd51e03de3a upstream.

The clean up is off by one so this will start at "i" and it should start
with "i - 1" and then it doesn't unregister the zeroeth elements in the
array.

Fixes: c52ca478233c ("dmaengine: idxd: add configuration component of driver")
Signed-off-by: Dan Carpenter 
Acked-by: Dave Jiang 
Link: https://lore.kernel.org/r/X9nFeojulsNqUSnG@mwanda
Signed-off-by: Vinod Koul 
Signed-off-by: Greg Kroah-Hartman

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

2020-12-30T10:53:51+00:00

[ Upstream commit e2de925bbfe321ba0588c99f577c59386ab1f428 ]

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

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

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

2020-12-30T10:53:21+00:00

[ Upstream commit c95e6515a8c065862361f7e0e452978ade7f94ec ]

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

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

Merge tag 'dmaengine-fix-5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine

2020-11-20T18:23:49+00:00

Pull dmaengine fixes from Vinod Koul:
 "A solitary core fix and a few driver fixes:

  Core:

   - channel_register error handling

  Driver fixes:

   - idxd: wq config registers programming and mapping of portal size

   - ioatdma: unused fn removal

   - pl330: fix burst size

   - ti: pm fix on busy and -Wenum-conversion warns

   - xilinx: SG capability check, usage of xilinx_aximcdma_tx_segment,
     readl_poll_timeout_atomic variant"

* tag 'dmaengine-fix-5.10-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine:
  dmaengine: fix error codes in channel_register()
  dmaengine: pl330: _prep_dma_memcpy: Fix wrong burst size
  dmaengine: ioatdma: remove unused function missed during dma_v2 removal
  dmaengine: idxd: fix mapping of portal size
  dmaengine: ti: omap-dma: Block PM if SDMA is busy to fix audio
  dmaengine: xilinx_dma: Fix SG capability check for MCDMA
  dmaengine: xilinx_dma: Fix usage of xilinx_aximcdma_tx_segment
  dmaengine: xilinx_dma: use readl_poll_timeout_atomic variant
  dmaengine: ti: k3-udma: fix -Wenum-conversion warning
  dmaengine: idxd: fix wq config registers offset programming

dmaengine: fix error codes in channel_register()

2020-11-18T12:21:51+00:00

The error codes were not set on some of these error paths.

Also the error handling was more confusing than it needed to be so I
cleaned it up and shuffled it around a bit.

Fixes: d2fb0a043838 ("dmaengine: break out channel registration")
Signed-off-by: Dan Carpenter 
Reviewed-by: Peter Ujfalusi 
Link: https://lore.kernel.org/r/20201113101631.GE168908@mwanda
Signed-off-by: Vinod Koul

dmaengine: pl330: _prep_dma_memcpy: Fix wrong burst size

2020-11-16T17:14:23+00:00

Actually, burst size is equal to '1 << desc->rqcfg.brst_size'.
we should use burst size, not desc->rqcfg.brst_size.

dma memcpy performance on Rockchip RV1126
@ 1512MHz A7, 1056MHz LPDDR3, 200MHz DMA:

dmatest:

/# echo dma0chan0 > /sys/module/dmatest/parameters/channel
/# echo 4194304 > /sys/module/dmatest/parameters/test_buf_size
/# echo 8 > /sys/module/dmatest/parameters/iterations
/# echo y > /sys/module/dmatest/parameters/norandom
/# echo y > /sys/module/dmatest/parameters/verbose
/# echo 1 > /sys/module/dmatest/parameters/run

dmatest: dma0chan0-copy0: result #1: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #2: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #3: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #4: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #5: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #6: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #7: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #8: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000

Before:

  dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 48 iops 200338 KB/s (0)

After this patch:

  dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 179 iops 734873 KB/s (0)

After this patch and increase dma clk to 400MHz:

  dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 259 iops 1062929 KB/s (0)

Signed-off-by: Sugar Zhang 
Link: https://lore.kernel.org/r/1605326106-55681-1-git-send-email-sugar.zhang@rock-chips.com
Signed-off-by: Vinod Koul

dmaengine: ioatdma: remove unused function missed during dma_v2 removal

2020-11-16T17:12:28+00:00

Commit 7f832645d0e5 ("dmaengine: ioatdma: remove ioatdma v2 registration")
missed to remove dca2_tag_map_valid() during its removal. Hence, since
then, dca2_tag_map_valid() is unused and make CC=clang W=1 warns:

  drivers/dma/ioat/dca.c:44:19:
    warning: unused function 'dca2_tag_map_valid' [-Wunused-function]

So, remove this unused function and get rid of a -Wused-function warning.

Signed-off-by: Lukas Bulwahn 
Reviewed-by: Nathan Chancellor 
Link: https://lore.kernel.org/r/20201113081248.26416-1-lukas.bulwahn@gmail.com
Signed-off-by: Vinod Koul

dmaengine: idxd: fix mapping of portal size

2020-11-16T17:08:20+00:00

Portal size is 4k. Current code is mapping all 4 portals in a single chunk.
Restrict the mapped portal size to a single portal to ensure that submission
only goes to the intended portal address.

Fixes: c52ca478233c ("dmaengine: idxd: add configuration component of driver")
Signed-off-by: Dave Jiang 
Link: https://lore.kernel.org/r/160513342642.510187.16450549281618747065.stgit@djiang5-desk3.ch.intel.com
Signed-off-by: Vinod Koul

dmaengine: ti: omap-dma: Block PM if SDMA is busy to fix audio

2020-11-10T12:44:09+00:00

We now use cpu_pm for saving and restoring device context for deeper SoC
idle states. But for omap3, we must also block idle if SDMA is busy.

If we don't block idle when SDMA is busy, we eventually end up saving and
restoring SDMA register state on PER domain idle while SDMA is active and
that causes at least audio playback to fail.

Fixes: 4c74ecf79227 ("dmaengine: ti: omap-dma: Add device tree match data and use it for cpu_pm")
Reported-by: Peter Ujfalusi 
Signed-off-by: Tony Lindgren 
Tested-by: Peter Ujfalusi 
Acked-by: Peter Ujfalusi 
Link: https://lore.kernel.org/r/20201109154013.11950-1-tony@atomide.com
Signed-off-by: Vinod Koul

dmaengine: xilinx_dma: Fix SG capability check for MCDMA

2020-11-09T12:24:22+00:00

The SG capability is inherently present with Multichannel DMA operation.
The register used to check for this capability with other DMA driver types
is not defined for MCDMA.

Fixes: 6ccd692bfb7f ("dmaengine: xilinx_dma: Add Xilinx AXI MCDMA Engine driver support")
Signed-off-by: Matthew Murrian 
Signed-off-by: Radhey Shyam Pandey 
Link: https://lore.kernel.org/r/1604473206-32573-4-git-send-email-radhey.shyam.pandey@xilinx.com
Signed-off-by: Vinod Koul