On Sat, Nov 28, 2020 at 01:00:45PM +0530, Manivannan Sadhasivam wrote: > On Fri, Nov 20, 2020 at 01:55:59AM +0200, Cristian Ciocaltea wrote: > > The DMA controller present on the Actions Semi S500 SoC is compatible > > with the S900 variant, so add it to the list of devices supported by > > the Actions Semi Owl DMA driver. > > > > Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@xxxxxxxxx> > > I hope that you have verified both Memcpy and Slave transfers... I have been using 'dmatest' module as documented in: https://www.kernel.org/doc/html/latest/driver-api/dmaengine/dmatest.html I tested all the available channels and could not find any signs of possible issues. Bellow is an excerpt from the kernel ring buffer: [ 2661.884680] dmatest: dma0chan1-copy0: summary 300 tests, 0 failures 1653.48 iops 13249 KB/s (0) [ 2661.886567] dmatest: dma0chan2-copy0: summary 300 tests, 0 failures 1684.40 iops 12846 KB/s (0) [ 2661.888448] dmatest: dma0chan3-copy0: summary 300 tests, 0 failures 1730.62 iops 13648 KB/s (0) Should I perform some additional tests? Thanks, Cristi > Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@xxxxxxxxxx> > > Thanks, > Mani > > > --- > > drivers/dma/owl-dma.c | 1 + > > 1 file changed, 1 insertion(+) > > > > diff --git a/drivers/dma/owl-dma.c b/drivers/dma/owl-dma.c > > index 9fede32641e9..54e509de66e2 100644 > > --- a/drivers/dma/owl-dma.c > > +++ b/drivers/dma/owl-dma.c > > @@ -1082,6 +1082,7 @@ static struct dma_chan *owl_dma_of_xlate(struct of_phandle_args *dma_spec, > > static const struct of_device_id owl_dma_match[] = { > > { .compatible = "actions,s900-dma", .data = (void *)S900_DMA,}, > > { .compatible = "actions,s700-dma", .data = (void *)S700_DMA,}, > > + { .compatible = "actions,s500-dma", .data = (void *)S900_DMA,}, > > { /* sentinel */ }, > > }; > > MODULE_DEVICE_TABLE(of, owl_dma_match); > > -- > > 2.29.2 > >
