On Dec 15, 2008, at 7:58 AM, Boris Samorodov wrote: > Marcel Moolenaar <xcllnt_at_mac.com> writes: >> On Dec 13, 2008, at 5:36 AM, Boris Samorodov wrote: >> >>> That's for the line [*]: >>> ----- >>> --- pucdata.c (revision 185784) >>> +++ pucdata.c (working copy) >>> _at__at_ -1145,6 +1145,10 _at__at_ >>> case PUC_CFG_GET_TYPE: >>> *res = PUC_TYPE_SERIAL; >>> return (0); >>> + case PUC_CFG_INIT_PORT: >>> + bus_write_1((struct res *)res, 1 /* IER */, [*] >>> + (port >= 2) ? 0x10 : 0); >>> + return (0); >>> default: >>> break; >>> } >> >> Argh, (struct res *) must be (struct resource *)... > > No go. The same, ports are detected but show garbadge. One last thing that we could try. If the rclk for the upper 6 ports is wrong, then the baudrate will not be as programmed. For all ports we now have an rclk that's 8 times the default (i.e. RCLK*8). See if the baudrate is 8 times slower than what you specified. i.e. If you specified a baudrate of 115200, see if you need to program the other side for 14400. You may want to pick a baudrate that's yields a more standard rate when divided by 8. If this works, then the upper 6 port have a standard RCLK. More tricky, but worth to try is RCLK*10. Since we calculate based on RCLK*8, the baudrate will be 10/8 of what you expect. Specify a baudrate of 92160 to get 115200 (pick numbers that'll work). FYI, -- Marcel Moolenaar xcllnt_at_mac.comReceived on Mon Dec 15 2008 - 16:33:34 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:38 UTC