On 19.01.2022 09:14, Sergey Senozhatsky wrote: > Christoph suggests [1] that invalidating vmap range before > direct mapping range makes more sense. > > [1]: https://lore.kernel.org/all/20220111085958.GA22795@xxxxxx/ > Signed-off-by: Sergey Senozhatsky <senozhatsky@xxxxxxxxxxxx> Right. Acked-by: Marek Szyprowski <m.szyprowski@xxxxxxxxxxx> > --- > .../media/common/videobuf2/videobuf2-dma-contig.c | 12 ++++++------ > 1 file changed, 6 insertions(+), 6 deletions(-) > > diff --git a/drivers/media/common/videobuf2/videobuf2-dma-contig.c b/drivers/media/common/videobuf2/videobuf2-dma-contig.c > index 7c4096e62173..0e3f264122af 100644 > --- a/drivers/media/common/videobuf2/videobuf2-dma-contig.c > +++ b/drivers/media/common/videobuf2/videobuf2-dma-contig.c > @@ -132,12 +132,12 @@ static void vb2_dc_prepare(void *buf_priv) > if (!buf->non_coherent_mem) > return; > > - /* For both USERPTR and non-coherent MMAP */ > - dma_sync_sgtable_for_device(buf->dev, sgt, buf->dma_dir); > - > /* Non-coherent MMAP only */ > if (buf->vaddr) > flush_kernel_vmap_range(buf->vaddr, buf->size); > + > + /* For both USERPTR and non-coherent MMAP */ > + dma_sync_sgtable_for_device(buf->dev, sgt, buf->dma_dir); > } > > static void vb2_dc_finish(void *buf_priv) > @@ -152,12 +152,12 @@ static void vb2_dc_finish(void *buf_priv) > if (!buf->non_coherent_mem) > return; > > - /* For both USERPTR and non-coherent MMAP */ > - dma_sync_sgtable_for_cpu(buf->dev, sgt, buf->dma_dir); > - > /* Non-coherent MMAP only */ > if (buf->vaddr) > invalidate_kernel_vmap_range(buf->vaddr, buf->size); > + > + /* For both USERPTR and non-coherent MMAP */ > + dma_sync_sgtable_for_cpu(buf->dev, sgt, buf->dma_dir); > } > > /*********************************************/ Best regards -- Marek Szyprowski, PhD Samsung R&D Institute Poland
