From owner-freebsd-hackers Sat Dec 23 02:20:24 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id CAA02130 for hackers-outgoing; Sat, 23 Dec 1995 02:20:24 -0800 (PST) Received: from mailserv.uni-tuebingen.de (mailserv.uni-tuebingen.de [134.2.250.102]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id CAA02122 for ; Sat, 23 Dec 1995 02:20:20 -0800 (PST) Received: from pc-micha.zdv.uni-tuebingen.de by mailserv.uni-tuebingen.de with SMTP (PP); Sat, 23 Dec 1995 11:20:11 +0100 Received: from pc-micha.zdv.uni-tuebingen.de by pc-micha.zdv.uni-tuebingen.de (8.7.3/1.34) id LAA00256; Sat, 23 Dec 1995 11:03:20 +0100 (MET) Date: Sat, 23 Dec 1995 11:03:20 +0100 (MET) From: Micha Class Reply-To: Micha Class Subject: fbsd-current stability / sio-problem To: freebsd-hackers@freefall.freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Sender: owner-hackers@FreeBSD.ORG Precedence: bulk Hi Hackers, after all this discussion about stability of -current, I do have a real "Show-Stopper" here: Teh recent changes in sio.c that add support for some chips called 16650 make my ppp-access hanging. The problem is that my 16550-chips are detected as 16650, and the fifo-size ist set wrongly to 32 instead of 16 chars. The detection code looks like this: /* * Check for the Startech ST16C650 chip. * it has a shadow register under the com_iir, * which can only be accessed when cfcr == 0xff */ { u_char i, j; i = inb(iobase + com_iir); printf(" i: %0x,",i); outb(iobase + com_cfcr, 0xff); outb(iobase + com_iir, 0x0); outb(iobase + com_cfcr, CFCR_8BITS); j = inb(iobase + com_iir); printf(" j: %0x,",j); outb(iobase + com_iir, i); if (i != j) { printf(" 16550A"); } else { com->tx_fifo_size = 32; printf(" 16650"); ... The added printfs show 0xC1 for i and j on booting! I do have no infos about the 16650-chip, but this detection-code gets me into real trouble! (Btw. if I change the "com->tx_fifo_size = 32" to "com->tx_fifo_size = 16" everything works fine, that should be the prove that only the fifo-size is the Problem.) Micha ------------------------------------------------------------------------- michael class, hauptstr. 83, 72127 wankheim, frg E-Mail: michael.class@zdv.uni-tuebingen.de Phone: +49 7071 29-7539 (work) +49 7071 38365 (private) -------------------------------------------------------------------------