On Sun, Jun 09, 2019 at 06:52:34PM +0300, Evgeny Zinoviev wrote:
So here's the patch for MacBooks, I hope it
doesn't break anything.
Unfortunately, the call to usb_poll_intr() in usb_check_key() will
memcpy maxPacketSize bytes. That would overflow 'struct usbkeyinfo'.
On 6/2/19 9:01 PM, Kevin O'Connor wrote:
> On Sun, Jun 02, 2019 at 05:39:11PM +0300, Evgeny Zinoviev wrote:
>> Hi folks.
>> I've recently ported coreboot on MBA 5,2 (13'' mid 2012 model) and
>> 10,1 (15'' mid 2012 retina model). The integrated keyboard on these
>> models is connected as a USB device, not PS/2. I have tested GRUB,
>> SeaBIOS and Tianocore payloads, and the keyboard works in GRUB and
>> Tianocore but not in SeaBIOS.
>> I was advised to send a log to this mailing list. It was created with
>> SEABIOS_DEBUG_LEVEL=9. I'm not yet familiar with SeaBIOS code, but with
>> your help perhaps I can debug and fix this.
> At a quick glance, it appears multiple USB keyboards/mice were found.
> SeaBIOS just chooses one randomly in that case. As a guess, it chose
> the wrong one. If you want to try and debug it, you could look at
> SeaBIOS mailing list -- seabios(a)seabios.org
> To unsubscribe send an email to seabios-leave(a)seabios.org
diff --git a/src/hw/usb-hid.c b/src/hw/usb-hid.c
index fa4d9a2..7d3561c 100644
@@ -59,7 +59,7 @@ usb_kbd_setup(struct usbdevice_s *usbdev
// XXX - this enables the first found keyboard (could be random)
- if (epdesc->wMaxPacketSize != 8)
+ if (epdesc->wMaxPacketSize != 8 && epdesc->wMaxPacketSize != 10)
// Enable "boot" protocol.
SeaBIOS mailing list -- seabios(a)seabios.org
To unsubscribe send an email to seabios-leave(a)seabios.org