linux-spi.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] spi: cadence-quadspi: Fix pm_runtime unbalance on dma EPROBE_DEFER
@ 2025-10-08 13:38 Mattijs Korpershoek
  2025-10-08 15:33 ` Dan Carpenter
  0 siblings, 1 reply; 3+ messages in thread
From: Mattijs Korpershoek @ 2025-10-08 13:38 UTC (permalink / raw)
  To: Mark Brown, Khairul Anuar Romli, Dan Carpenter
  Cc: linux-spi, linux-kernel, Mattijs Korpershoek

In csqspi_probe(), when cqspi_request_mmap_dma() returns -EPROBE_DEFER,
we handle the error by jumping to probe_setup_failed.
In that label, we call pm_runtime_disable(), even if we never called
pm_runtime_enable() before.

Because of this, the driver cannot probe:

[    2.690018] cadence-qspi 47040000.spi: No Rx DMA available
[    2.699735] spi-nor spi0.0: resume failed with -13
[    2.699741] spi-nor: probe of spi0.0 failed with error -13

Only call pm_runtime_disable() if it was enabled by adding a new
label to handle cqspi_request_mmap_dma() failures.

Fixes: 04a8ff1bc351 ("spi: cadence-quadspi: fix cleanup of rx_chan on failure paths")
Signed-off-by: Mattijs Korpershoek <mkorpershoek@kernel.org>
---
This has been tested on a AM69 SK board.
---
 drivers/spi/spi-cadence-quadspi.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/spi-cadence-quadspi.c b/drivers/spi/spi-cadence-quadspi.c
index 8fb13df8ff8714d2ad5a019f0ae0ec3ee38833bb..81017402bc5661d08ff4e75017db954fda19ba2a 100644
--- a/drivers/spi/spi-cadence-quadspi.c
+++ b/drivers/spi/spi-cadence-quadspi.c
@@ -1995,7 +1995,7 @@ static int cqspi_probe(struct platform_device *pdev)
 	if (cqspi->use_direct_mode) {
 		ret = cqspi_request_mmap_dma(cqspi);
 		if (ret == -EPROBE_DEFER)
-			goto probe_setup_failed;
+			goto probe_dma_failed;
 	}
 
 	if (!(ddata && (ddata->quirks & CQSPI_DISABLE_RUNTIME_PM))) {
@@ -2019,9 +2019,10 @@ static int cqspi_probe(struct platform_device *pdev)
 
 	return 0;
 probe_setup_failed:
-	cqspi_controller_enable(cqspi, 0);
 	if (!(ddata && (ddata->quirks & CQSPI_DISABLE_RUNTIME_PM)))
 		pm_runtime_disable(dev);
+probe_dma_failed:
+	cqspi_controller_enable(cqspi, 0);
 probe_reset_failed:
 	if (cqspi->is_jh7110)
 		cqspi_jh7110_disable_clk(pdev, cqspi);

---
base-commit: 0d97f2067c166eb495771fede9f7b73999c67f66
change-id: 20251008-cadence-quadspi-fix-pm-runtime-bf8df9104577

Best regards,
-- 
Mattijs Korpershoek <mkorpershoek@kernel.org>


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] spi: cadence-quadspi: Fix pm_runtime unbalance on dma EPROBE_DEFER
  2025-10-08 13:38 [PATCH] spi: cadence-quadspi: Fix pm_runtime unbalance on dma EPROBE_DEFER Mattijs Korpershoek
@ 2025-10-08 15:33 ` Dan Carpenter
  2025-10-09  7:09   ` Mattijs Korpershoek
  0 siblings, 1 reply; 3+ messages in thread
From: Dan Carpenter @ 2025-10-08 15:33 UTC (permalink / raw)
  To: Mattijs Korpershoek
  Cc: Mark Brown, Khairul Anuar Romli, linux-spi, linux-kernel

On Wed, Oct 08, 2025 at 03:38:39PM +0200, Mattijs Korpershoek wrote:
> In csqspi_probe(), when cqspi_request_mmap_dma() returns -EPROBE_DEFER,
> we handle the error by jumping to probe_setup_failed.
> In that label, we call pm_runtime_disable(), even if we never called
> pm_runtime_enable() before.
> 
> Because of this, the driver cannot probe:
> 
> [    2.690018] cadence-qspi 47040000.spi: No Rx DMA available
> [    2.699735] spi-nor spi0.0: resume failed with -13
> [    2.699741] spi-nor: probe of spi0.0 failed with error -13
> 
> Only call pm_runtime_disable() if it was enabled by adding a new
> label to handle cqspi_request_mmap_dma() failures.
> 
> Fixes: 04a8ff1bc351 ("spi: cadence-quadspi: fix cleanup of rx_chan on failure paths")
> Signed-off-by: Mattijs Korpershoek <mkorpershoek@kernel.org>
> ---
> This has been tested on a AM69 SK board.

The patch seems correct, but the correct Fixes tag is:
Fixes: b07f349d1864 ("spi: spi-cadence-quadspi: Fix pm runtime unbalance")

regards,
dan carpenter


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] spi: cadence-quadspi: Fix pm_runtime unbalance on dma EPROBE_DEFER
  2025-10-08 15:33 ` Dan Carpenter
@ 2025-10-09  7:09   ` Mattijs Korpershoek
  0 siblings, 0 replies; 3+ messages in thread
From: Mattijs Korpershoek @ 2025-10-09  7:09 UTC (permalink / raw)
  To: Dan Carpenter, Mattijs Korpershoek
  Cc: Mark Brown, Khairul Anuar Romli, linux-spi, linux-kernel

Hi Dan,

Thank you for the review.

On Wed, Oct 08, 2025 at 18:33, Dan Carpenter <dan.carpenter@linaro.org> wrote:

> On Wed, Oct 08, 2025 at 03:38:39PM +0200, Mattijs Korpershoek wrote:
>> In csqspi_probe(), when cqspi_request_mmap_dma() returns -EPROBE_DEFER,
>> we handle the error by jumping to probe_setup_failed.
>> In that label, we call pm_runtime_disable(), even if we never called
>> pm_runtime_enable() before.
>> 
>> Because of this, the driver cannot probe:
>> 
>> [    2.690018] cadence-qspi 47040000.spi: No Rx DMA available
>> [    2.699735] spi-nor spi0.0: resume failed with -13
>> [    2.699741] spi-nor: probe of spi0.0 failed with error -13
>> 
>> Only call pm_runtime_disable() if it was enabled by adding a new
>> label to handle cqspi_request_mmap_dma() failures.
>> 
>> Fixes: 04a8ff1bc351 ("spi: cadence-quadspi: fix cleanup of rx_chan on failure paths")
>> Signed-off-by: Mattijs Korpershoek <mkorpershoek@kernel.org>
>> ---
>> This has been tested on a AM69 SK board.
>
> The patch seems correct, but the correct Fixes tag is:
> Fixes: b07f349d1864 ("spi: spi-cadence-quadspi: Fix pm runtime unbalance")

Right, I hesitated between the 04a8ff1bc351 and b07f349d1864. I will
send a v2 with an updated Fixes tag.

>
> regards,
> dan carpenter

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2025-10-09  7:09 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-10-08 13:38 [PATCH] spi: cadence-quadspi: Fix pm_runtime unbalance on dma EPROBE_DEFER Mattijs Korpershoek
2025-10-08 15:33 ` Dan Carpenter
2025-10-09  7:09   ` Mattijs Korpershoek

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).