On Thursday, April 10, 2014 at 12:15:30 PM, Wolfgang Denk wrote:
Dear Marek,
In message 201404101012.42527.marex@denx.de you wrote:
On Sunday, April 06, 2014 at 01:17:46 PM, Wolfgang Denk wrote: [...]
@@ -131,7 +133,9 @@ void usb_kbd_generic_poll(void)
/* Submit a interrupt transfer request */ maxp = usb_maxpacket(usb_kbd_dev, pipe); usb_submit_int_msg(usb_kbd_dev, pipe, data->new,
maxp > 8 ? 8 : maxp, ep->bInterval);
maxp > USB_KBD_PDATA_SIZE ?USB_KBD_PDATA_SIZE : maxp,ep->bInterval);Can we use min(8, USB_KBD_PDATA_SIZE) here instead of this multi-line monster call please ?
Agreed.
Also, this USB_KBD_PDATA_SIZE should instead be called "USB_KBD_LS_REPORT_S> IZE"
according to USB HID spec v1.11 section 5.6 :
Fine with me.
What worries me a bit is that 64-byte high-speed report, but I never saw a device that would generate those. This section 5.6 is also the only place that mentions the high-speed HID device report size limit.
I'm not an USB expert; I cannot comment on this.
Other than that, I don't see a problem.
So how should we proceed? Shall I prepare an updated patch - or Adrian, will you do that?
Adrian, can you please take all the discussion here and roll out a new patch please ?
Best regards, Marek Vasut