diff options
| author | Russell King <rmk+kernel@arm.linux.org.uk> | 2013-06-27 12:53:37 +0100 |
|---|---|---|
| committer | Russell King <rmk+kernel@arm.linux.org.uk> | 2013-10-31 14:48:47 +0000 |
| commit | c9bd5e690a439df044678d89e89e380cf9db7930 (patch) | |
| tree | e2f73839b2be962bd4b4f19279aafa38e7b72332 /sound/soc/fsl/fsl_dma.c | |
| parent | fa6a8d6d65b19ab44e5244ea499bcd553cc72343 (diff) | |
| download | linux-c9bd5e690a439df044678d89e89e380cf9db7930.tar.bz2 | |
DMA-API: sound: fix dma mask handling in a lot of drivers
This code sequence is unsafe in modules:
static u64 mask = DMA_BIT_MASK(something);
...
if (!dev->dma_mask)
dev->dma_mask = &mask;
as if a module is reloaded, the mask will be pointing at the original
module's mask address, and this can lead to oopses. Moreover, they
all follow this with:
if (!dev->coherent_dma_mask)
dev->coherent_dma_mask = mask;
where 'mask' is the same value as the statically defined mask, and this
bypasses the architecture's check on whether the DMA mask is possible.
Fix these issues by using the new dma_coerce_coherent_and_mask()
function.
Acked-by: Mark Brown <broonie@linaro.org>
Acked-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Diffstat (limited to 'sound/soc/fsl/fsl_dma.c')
| -rw-r--r-- | sound/soc/fsl/fsl_dma.c | 9 |
1 files changed, 3 insertions, 6 deletions
diff --git a/sound/soc/fsl/fsl_dma.c b/sound/soc/fsl/fsl_dma.c index 9cc5c1f82f09..f73c7eff8b23 100644 --- a/sound/soc/fsl/fsl_dma.c +++ b/sound/soc/fsl/fsl_dma.c @@ -298,14 +298,11 @@ static int fsl_dma_new(struct snd_soc_pcm_runtime *rtd) { struct snd_card *card = rtd->card->snd_card; struct snd_pcm *pcm = rtd->pcm; - static u64 fsl_dma_dmamask = DMA_BIT_MASK(36); int ret; - if (!card->dev->dma_mask) - card->dev->dma_mask = &fsl_dma_dmamask; - - if (!card->dev->coherent_dma_mask) - card->dev->coherent_dma_mask = fsl_dma_dmamask; + ret = dma_coerce_mask_and_coherent(card->dev, DMA_BIT_MASK(36)); + if (ret) + return ret; /* Some codecs have separate DAIs for playback and capture, so we * should allocate a DMA buffer only for the streams that are valid. |