07.05.2019 9:09, Baolin Wang пишет: > Since we've introduced one device node parameter for __dma_request_channel(), > thus change to the correct function prototype. > > Signed-off-by: Baolin Wang <baolin.w...@linaro.org> > --- > drivers/soc/tegra/fuse/fuse-tegra20.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/soc/tegra/fuse/fuse-tegra20.c > b/drivers/soc/tegra/fuse/fuse-tegra20.c > index 49ff017..e2571b6 100644 > --- a/drivers/soc/tegra/fuse/fuse-tegra20.c > +++ b/drivers/soc/tegra/fuse/fuse-tegra20.c > @@ -110,7 +110,7 @@ static int tegra20_fuse_probe(struct tegra_fuse *fuse) > dma_cap_zero(mask); > dma_cap_set(DMA_SLAVE, mask); > > - fuse->apbdma.chan = __dma_request_channel(&mask, dma_filter, NULL); > + fuse->apbdma.chan = __dma_request_channel(&mask, dma_filter, NULL, > NULL); > if (!fuse->apbdma.chan) > return -EPROBE_DEFER; > >
1) Kernel should be kept bisect'able by not having intermediate patches that break compilation. Hence you should squash the changes into a single patch. 2) Better to replace __dma_request_channel() with dma_request_channel() since they are equal. -- Dmitry
