On Wed, Jan 23, 2002 at 09:49:41PM +0530, Francis S Nazareth wrote:
>
> From the first chunk of 64 bytes, the length is rightly calculated as 200
> - ACL_HDR_SIZE. (195); from the header information. But when the second
> packet comes, which doesn't contain an ACL header (since this is the
> continuation packet); the length variable is again calculated. ( since the
> second packet is also of size 64 bytes, which satisfies the condition
> bluetooth->bulk_packet_pos >= ACL_HDR_SIZE.) and this will result in a
> garbage value. The packet will never be submitted to the tty flip buffer,
> (unless in the exceptional case, where the new length (garbage) matches
> the original length.). The same case applies to any packet of length
> greater than bulk read buffer size. The point is, the packet_length should
> be calculated only from the first packet.
But packet_size is calculated from the start of the original buffer, not
the buffer that was just read:
packet_size = CHAR2INT16(bluetooth->bulk_buffer[4],bluetooth->bulk_buffer[3]);
So for your senario, packet_size will still be caculated properly,
right?
Or am I missing something?
Does your patch fix any problems that you have been seeing in the
driver?
thanks,
greg k-h
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
