As the potential failure of the dma_set_mask(), we fix this
bug by checking its return value and performing proper error
handling.
Signed-off-by: harperchen <harperchen1110@gmail.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
* Convince the system that we can do DMA.
*/
pdev->dev.dma_mask = &viadev->pdev->dma_mask;
- dma_set_mask(&pdev->dev, 0xffffffff);
+ ret = dma_set_mask(&pdev->dev, 0xffffffff);
+ if (ret)
+ goto out_ctrl_hdl_free;
/*
* Fire up the capture port. The write to 0x78 looks purely
* OLPCish; any system will need to tweak 0x1e.