[PATCH] ASoC: fsl_asrc: fix second front-end channel check

Commit b045248399d0 ("ASoC: fsl_asrc: Replace dma_request_slave_channel()
by dma_request_chan()") made ->get_dma_channel() propagate error
pointers, so the !NULL check added by commit b4136c0d69ea ("ASoC:
fsl_asrc: check the second front-end DMA channel") no longer catches
failures and the error pointer would be dereferenced at
tmp_chan->private.

Use IS_ERR() and propagate the error, keeping the release of the
previously acquired persistent Front-End channel.

Fixes: b045248399d0 ("ASoC: fsl_asrc: Replace dma_request_slave_channel() by dma_request_chan()")
Assisted-by: LLM
Signed-off-by: Slavin Liu <redacted>
---
 sound/soc/fsl/fsl_asrc_dma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/sound/soc/fsl/fsl_asrc_dma.c b/sound/soc/fsl/fsl_asrc_dma.c
index 0aa5db8973d6..097a0d5e8b31 100644
--- a/sound/soc/fsl/fsl_asrc_dma.c
+++ b/sound/soc/fsl/fsl_asrc_dma.c

@@ -251,10 +251,10 @@ static int fsl_asrc_dma_hw_params(struct snd_soc_component *component,
 
 		/* Get DMA request of Front-End */
 		tmp_chan = asrc->get_dma_channel(pair, dir);
-		if (!tmp_chan) {
+		if (IS_ERR(tmp_chan)) {
 			dma_release_channel(pair->dma_chan[!dir]);
 			pair->dma_chan[!dir] = NULL;
-			return -EINVAL;
+			return PTR_ERR(tmp_chan);
 		}
 		tmp_data = tmp_chan->private;
 		pair->dma_data.dma_request2 = tmp_data->dma_request;

-- 
2.34.1

On Sun, Sep 27, 2026 at 08:07:38AM +0800, Slavin Liu wrote:

Commit b045248399d0 ("ASoC: fsl_asrc: Replace dma_request_slave_channel()
by dma_request_chan()") made ->get_dma_channel() propagate error
pointers, so the !NULL check added by commit b4136c0d69ea ("ASoC:
fsl_asrc: check the second front-end DMA channel") no longer catches
failures and the error pointer would be dereferenced at
tmp_chan->private.

Use IS_ERR() and propagate the error, keeping the release of the
previously acquired persistent Front-End channel.

...

 			pair->dma_chan[!dir] = NULL;

-			return -EINVAL;
+			return PTR_ERR(tmp_chan);

I believe this has to be left untouched as it might leak to user space.

 		}

...

I also wondering why Sashiko hasn't pointed that out during the original
submission...

-- 
With Best Regards,
Andy Shevchenko