On Mon, Jun 24, 2013 at 06:42:28PM +0400, Denis V. Lunev wrote: > On 6/24/13 6:35 PM, Felipe Balbi wrote: > >Hi, > > > >On Mon, Jun 24, 2013 at 06:31:07PM +0400, Denis V. Lunev wrote: > >>with a length that isn't multiply by max pkt size for this endpoint. > >> > >>The commit 689d6eac (USB: UHCI: add native scatter-gather support(v1)) > >>introduced an implementation of scatter-gather list for UHCI. > >>This implementation has a bug when the size of a non-last sg > >>element was not multiply by TD's max-pkt-size. This bug was latent > >>till the commit 2851784f (usb/uhci: initialize sg_table properly) > >>which really initializes sg_table and enables SG lists in UHCI. > >> > >>The bug can be easily reproduced with Gadget Zero in full_speed mode > >>connected to a host with UHCI controller by the standard test from > >>tools/usb/ with command line > >> testusb -a -t 7 -c 2000 -s 4096 -v 41 > >>The test sends trash bytes to the gadget instead of only zero bytes > >>and so the test fails. > >> > >>This patch shortens TD's packet not only for the last TD in sg list, > >>but also for the last TD in sg element. > >> > >>Signed-off-by: Konstantin Filatov <kfila...@parallels.com> > >>Signed-off-by: Denis V. Lunev <d...@openvz.org> > >>Acked-by: Alan Stern <st...@rowland.harvard.edu> > >>CC: Ming Lei <tom.leim...@gmail.com> > >>CC: Sebastian Andrzej Siewior <bige...@linutronix.de> > >>CC: linux-usb@vger.kernel.org > >this is not the correct way to send a patch. Subject line is wrong, > >please fix that. > > > yep, sorry, fixed already.
saw as soon as I replied :-) -- balbi
signature.asc
Description: Digital signature
