Am Montag, den 17.08.2020, 15:06 +0200 schrieb Stefan Roese: > Octeon uses mapped addresses for virtual and physical memory. Its not > that easy to calculate the resulting addresses here. So lets remove > this BUG_ON() for Octeon in xhci_bulk_tx(). > > Signed-off-by: Stefan Roese <[email protected]> > Cc: Bin Meng <[email protected]> > Cc: Marek Vasut <[email protected]> > --- > > drivers/usb/host/xhci-ring.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c > index 092ed6eaf1..a762177c57 100644 > --- a/drivers/usb/host/xhci-ring.c > +++ b/drivers/usb/host/xhci-ring.c > @@ -726,8 +726,11 @@ int xhci_bulk_tx(struct usb_device *udev, unsigned long > pipe, > > BUG_ON(TRB_TO_SLOT_ID(field) != slot_id); > BUG_ON(TRB_TO_EP_INDEX(field) != ep_index); > - BUG_ON(*(void **)(uintptr_t)le64_to_cpu(event->trans_event.buffer) - > - buffer > (size_t)length); > + if (!IS_ENABLED(CONFIG_ARCH_OCTEON)) { > + BUG_ON(*(void **)(uintptr_t)le64_to_cpu( > + event->trans_event.buffer) - buffer > > + (size_t)length); > + }
why not remove the check? Or add at least a generic Kconfig option for special handling on archs with memory remapping. Arch or SoC specific config options in generic code are bad ;) > > record_transfer_result(udev, event, length); > xhci_acknowledge_event(ctrl); -- - Daniel
