On Thu, Jun 30, 2022 at 04:02:16PM +0200, Michael Grzeschik wrote: > For high-speed transfers the __dwc3_prepare_one_trb function is > calculating the multiplier setting for the trb based on the length of > the trb currently prepared. This assumption is wrong. For trbs with a > sglist the length of the actual request has to be taken instead. > > Signed-off-by: Michael Grzeschik <m.grzeschik@xxxxxxxxxxxxxx> > --- > drivers/usb/dwc3/gadget.c | 8 ++++---- > 1 file changed, 4 insertions(+), 4 deletions(-) > > diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c > index 8716bece107208..0fc3ecfaa48baf 100644 > --- a/drivers/usb/dwc3/gadget.c > +++ b/drivers/usb/dwc3/gadget.c > @@ -1186,7 +1186,7 @@ static void __dwc3_prepare_one_trb(struct dwc3_ep *dep, struct dwc3_trb *trb, > dma_addr_t dma, unsigned int length, unsigned int chain, > unsigned int node, unsigned int stream_id, > unsigned int short_not_ok, unsigned int no_interrupt, > - unsigned int is_last, bool must_interrupt) > + unsigned int is_last, bool must_interrupt, int req_length) Why not put this right next to 'length'? And wow, that's a crazy number of options for a function. Why is this even a separate function at all? Why can't it just be in dwc3_prepare_one_trb() as that's the only place this is called? thanks, greg k-h
