This patch changes for a tx dma channel. There is a similar problem for a
rx one too. The code doesn't cause an error, but it is likely better to
change to just NULL checking.
2022年7月8日(金) 17:41 Dan Carpenter <dan.carpenter@xxxxxxxxxx>:
>
> The dma_request_channel() function doesn't return error pointers, it
> returns NULL.
>
> Fixes: fff86dfbbf82 ("PCI: endpoint: Enable DMA tests for endpoints with DMA capabilities")
> Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> ---
> It's surprising and unfortunate that dma_request_channel() returns NULL
> while dma_request_chan_by_mask() returns error pointers. These days
> static checkers will catch this so it's not as bad as it could be but
> still not ideal.
>
> drivers/pci/endpoint/functions/pci-epf-test.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/pci/endpoint/functions/pci-epf-test.c b/drivers/pci/endpoint/functions/pci-epf-test.c
> index 34aac220dd4c..eed6638ab71d 100644
> --- a/drivers/pci/endpoint/functions/pci-epf-test.c
> +++ b/drivers/pci/endpoint/functions/pci-epf-test.c
> @@ -221,7 +221,7 @@ static int pci_epf_test_init_dma_chan(struct pci_epf_test *epf_test)
> filter.dma_mask = BIT(DMA_MEM_TO_DEV);
> dma_chan = dma_request_channel(mask, epf_dma_filter_fn, &filter);
>
> - if (IS_ERR(dma_chan)) {
> + if (!dma_chan) {
> dev_info(dev, "Failed to get private DMA tx channel. Falling back to generic one\n");
> goto fail_back_rx;
> }
> --
> 2.35.1
>
Thanks,
Shunsuke.
