linux-stable.git/drivers/dma, branch linux-6.3.y

dmaengine: at_hdmac: Extend the Flow Controller bitfield to three bits

2023-06-09T08:48:19+00:00

commit e14fd2af7a1d621c167dad761f729135a7a76ff4 upstream.

Some chips have two bits (e.g SAMA5D3), and some have three (e.g.
SAM9G45). A field width of three is compatible as long as valid
values are used for the different chips.

There is no current use of any value needing three bits, so the
fixed bug is relatively benign.

Fixes: d8840a7edcf0 ("dmaengine: at_hdmac: Use bitfield access macros")
Cc: stable@vger.kernel.org
Reviewed-by: Tudor Ambarus 
Signed-off-by: Peter Rosin 
Link: https://lore.kernel.org/r/e2c898ba-c3a3-5dd3-384b-0585661c79f2@axentia.se
Signed-off-by: Vinod Koul 
Signed-off-by: Greg Kroah-Hartman

dmaengine: at_hdmac: Repair bitfield macros for peripheral ID handling

2023-06-09T08:48:19+00:00

commit 2a6c7e8cc74e58ba94b8c897035a8ef7f7349f76 upstream.

The MSB part of the peripheral IDs need to go into the ATC_SRC_PER_MSB
and ATC_DST_PER_MSB fields. Not the LSB part.

This fixes a severe regression for TSE-850 devices (compatible
axentia,tse850v3) where output to the audio I2S codec (the main
purpose of the device) simply do not work.

Fixes: d8840a7edcf0 ("dmaengine: at_hdmac: Use bitfield access macros")
Cc: stable@vger.kernel.org
Signed-off-by: Peter Rosin 
Reviewed-by: Tudor Ambarus 
Link: https://lore.kernel.org/r/01e5dae1-d4b0-cf31-516b-423b11b077f1@axentia.se
Signed-off-by: Vinod Koul 
Signed-off-by: Greg Kroah-Hartman

dmaengine: pl330: rename _start to prevent build error

2023-06-09T08:47:48+00:00

[ Upstream commit a1a5f2c887252dec161c1e12e04303ca9ba56fa9 ]

"_start" is used in several arches and proably should be reserved
for ARCH usage. Using it in a driver for a private symbol can cause
a build error when it conflicts with ARCH usage of the same symbol.

Therefore rename pl330's "_start" to "pl330_start_thread" so that there
is no conflict and no build error.

drivers/dma/pl330.c:1053:13: error: '_start' redeclared as different kind of symbol
 1053 | static bool _start(struct pl330_thread *thrd)
      |             ^~~~~~
In file included from ../include/linux/interrupt.h:21,
                 from ../drivers/dma/pl330.c:18:
arch/riscv/include/asm/sections.h:11:13: note: previous declaration of '_start' with type 'char[]'
   11 | extern char _start[];
      |             ^~~~~~

Fixes: b7d861d93945 ("DMA: PL330: Merge PL330 driver into drivers/dma/")
Fixes: ae43b3289186 ("ARM: 8202/1: dmaengine: pl330: Add runtime Power Management support v12")
Signed-off-by: Randy Dunlap 
Cc: Jaswinder Singh 
Cc: Boojin Kim 
Cc: Krzysztof Kozlowski 
Cc: Russell King 
Cc: Vinod Koul 
Cc: dmaengine@vger.kernel.org
Cc: linux-riscv@lists.infradead.org
Link: https://lore.kernel.org/r/20230524045310.27923-1-rdunlap@infradead.org
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: at_xdmac: fix potential Oops in at_xdmac_prep_interleaved()

2023-06-09T08:47:47+00:00

[ Upstream commit 4d43acb145c363626d76f49febb4240c488cd1cf ]

There are two place if the at_xdmac_interleaved_queue_desc() fails which
could lead to a NULL dereference where "first" is NULL and we call
list_add_tail(&first->desc_node, ...).  In the first caller, the return
is not checked so add a check for that.  In the next caller, the return
is checked but if it fails on the first iteration through the loop then
it will lead to a NULL pointer dereference.

Fixes: 4e5385784e69 ("dmaengine: at_xdmac: handle numf > 1")
Fixes: 62b5cb757f1d ("dmaengine: at_xdmac: fix memory leak in interleaved mode")
Signed-off-by: Dan Carpenter 
Reviewed-by: Tudor Ambarus 
Link: https://lore.kernel.org/r/21282b66-9860-410a-83df-39c17fcf2f1b@kili.mountain
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: at_xdmac: do not enable all cyclic channels

2023-05-11T14:17:36+00:00

[ Upstream commit f8435befd81dd85b7b610598551fadf675849bc1 ]

Do not global enable all the cyclic channels in at_xdmac_resume(). Instead
save the global status in at_xdmac_suspend() and re-enable the cyclic
channel only if it was active before suspend.

Fixes: e1f7c9eee707 ("dmaengine: at_xdmac: creation of the atmel eXtended DMA Controller driver")
Signed-off-by: Claudiu Beznea 
Link: https://lore.kernel.org/r/20230214151827.1050280-6-claudiu.beznea@microchip.com
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: at_xdmac: restore the content of grws register

2023-05-11T14:17:35+00:00

[ Upstream commit 7c5eb63d16b01c202aaa95f374ae15a807745a73 ]

In case the system suspends to a deep sleep state where power to DMA
controller is cut-off we need to restore the content of GRWS register.
This is a write only register and writing bit X tells the controller
to suspend read and write requests for channel X. Thus set GRWS before
restoring the content of GE (Global Enable) regiter.

Fixes: e1f7c9eee707 ("dmaengine: at_xdmac: creation of the atmel eXtended DMA Controller driver")
Signed-off-by: Claudiu Beznea 
Link: https://lore.kernel.org/r/20230214151827.1050280-5-claudiu.beznea@microchip.com
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: at_xdmac: do not resume channels paused by consumers

2023-05-11T14:17:35+00:00

[ Upstream commit 44fe8440bda545b5d167329df88c47609a645168 ]

In case there are DMA channels not paused by consumers in suspend
process (valid on AT91 SoCs for serial driver when no_console_suspend) the
driver pauses them (using at_xdmac_device_pause() which is also the same
function called by dmaengine_pause()) and then in the resume process the
driver resumes them calling at_xdmac_device_resume() which is the same
function called by dmaengine_resume()). This is good for DMA channels
not paused by consumers but for drivers that calls
dmaengine_pause()/dmaegine_resume() on suspend/resume path this may lead to
DMA channel being enabled before the IP is enabled. For IPs that needs
strict ordering with regards to DMA channel enablement this will lead to
wrong behavior. To fix this add a new set of functions
at_xdmac_device_pause_internal()/at_xdmac_device_resume_internal() to be
called only on suspend/resume.

Fixes: e1f7c9eee707 ("dmaengine: at_xdmac: creation of the atmel eXtended DMA Controller driver")
Signed-off-by: Claudiu Beznea 
Link: https://lore.kernel.org/r/20230214151827.1050280-4-claudiu.beznea@microchip.com
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: at_xdmac: fix imbalanced runtime PM reference counter

2023-05-11T14:17:35+00:00

[ Upstream commit e53957e1ec5196671e49a48f90a5c9555153189a ]

In case there are channels not paused during suspend (which on AT91 case
is valid for serial driver when no_console_suspend boot argument is used)
the at_xdmac_runtime_suspend_descriptors() was called more than
one time due to at_xdmac_off(). To fix this add a new argument to
at_xdmac_off() to specify if runtime PM reference counter needs to be
decremented for queued active descriptors. Along with it moved the
at_xdmac_runtime_suspend_descriptors() call under at_xdmac_chan_is_paused()
check on suspend path as for the rest of channels the suspend is delayed
by atmel_xdmac_prepare() in case channel is enabled. Same approach has
been applied on resume path.

Fixes: 650b0e990cbd ("dmaengine: at_xdmac: add runtime pm support")
Signed-off-by: Claudiu Beznea 
Link: https://lore.kernel.org/r/20230214151827.1050280-3-claudiu.beznea@microchip.com
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: at_xdmac: disable/enable clock directly on suspend/resume

2023-05-11T14:17:35+00:00

[ Upstream commit 2de5ddb5e68c94b781b3789bca1ce52000d7d0e0 ]

Runtime PM APIs for at_xdmac just plays with clk_enable()/clk_disable()
letting aside the clk_prepare()/clk_unprepare() that needs to be
executed as the clock is also prepared on probe. Thus instead of using
runtime PM force suspend/resume APIs use
clk_disable_unprepare() + pm_runtime_put_noidle() on suspend and
clk_prepare_enable() + pm_runtime_get_noresume() on resume. This
approach as been chosen instead of using runtime PM force suspend/resume
with clk_unprepare()/clk_prepare() as it looks simpler and the final
code is better.

While at it added the missing pm_runtime_mark_last_busy() on suspend before
decrementing the reference counter.

Fixes: 650b0e990cbd ("dmaengine: at_xdmac: add runtime pm support")
Signed-off-by: Claudiu Beznea 
Link: https://lore.kernel.org/r/20230214151827.1050280-2-claudiu.beznea@microchip.com
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin

dmaengine: dw-edma: Fix to enable to issue dma request on DMA processing

2023-05-11T14:17:35+00:00

[ Upstream commit 970b17dfe264a9085ba4e593730ecfd496b950ab ]

The issue_pending request is ignored while driver is processing a DMA
request. Fix to issue the pending requests on any dma channel status.

Fixes: e63d79d1ffcd ("dmaengine: Add Synopsys eDMA IP core driver")
Signed-off-by: Shunsuke Mie 
Link: https://lore.kernel.org/r/20230411101758.438472-2-mie@igel.co.jp
Signed-off-by: Vinod Koul 
Signed-off-by: Sasha Levin