On Tue, 28 Mar 2000, Lord Isildur wrote:

=> what speed is given to the line in the /etc/ttys?

I have the line set to "off" in /etc/ttys.  I use the modem for dialling
out only.

=> you could use the ioctl directly to change the speed, or use something
=> like cu or kermit to set the speed.. the 3000/300 has the same zilog uart
=> that the other 3000's (and the DS5000/240) have. 57600 is no sweat for
=> those. 

I have used kermit, and "set speed 57600", but the modem steadfastly
appears to stick to 9600 baud.  E.g., "ati4" yields:

>>>>>
U.S. Robotics Sportster 33600 Fax Settings...

   B0  E1  F1  M1  Q0  V1  X4  Y0
   BAUD=9600  PARITY=N  WORDLEN=8
   DIAL=TONE   ON HOOK

   &A3  &B1  &C1  &D2  &G0  &H1  &I0  &K1
   &M4  &N16  &P0  &R2  &S1  &T5  &U4  &Y1  

   S00=000  S01=000  S02=043  S03=013  S04=010  S05=008  S06=002
   S07=060  S08=002  S09=006  S10=014  S11=070  S12=050  S13=000
   S15=000  S16=000  S18=000  S19=000  S21=010  S22=017  S23=019
   S25=005  S27=000  S28=008  S29=020  S30=000  S31=128  S32=002
   S33=000  S34=000  S35=000  S36=014  S38=000  S39=000  S41=000
   S42=000
<<<<<

Although the DTE speed seems to be 9600, the DCE speed appears to be
higher.  For example, the stats for my last call are:

>>>>>
U.S. Robotics Sportster 33600 Fax Link Diagnostics...

Chars sent                98538      Chars Received           265706
Chars lost                    0
Octets sent              107208      Octets Received          268569
Blocks sent               10608      Blocks Received            6617
Blocks resent                10

Retrains Requested            0      Retrains Granted              0
Line Reversals                0      Blers                         0
Link Timeouts                 0      Link Naks                     3

Data Compression       V42BIS 2048/32
Equalization           Long
Fallback               Enabled
Protocol               LAPM
Speed                  28800/31200 
Last Call              00:45:11

Disconnect Reason is DTR dropped

OK
<<<<<

Like I said in my previous message, "ping" times are comparable to the
old V32 9600 modem I was using previously, and are SLOWER than the
M$-Windoze NT system belonging to someone else in the same house.

In addition, whenever I fire up a PPP connection, my console fills up
with syslog junk like this:

>>>>>
Mar 29 22:05:38 hawkwind /netbsd: ppp0: bad fcs a985
Mar 29 22:05:39 hawkwind /netbsd: z_decompress0: bad seq # 40, expected 39
Mar 29 22:05:39 hawkwind /netbsd: ppp0: decompress failed 1
Mar 29 22:05:47 hawkwind /netbsd: ppp0: bad fcs 1870
Mar 29 22:05:47 hawkwind /netbsd: z_decompress0: bad seq # 21, expected 20
Mar 29 22:05:47 hawkwind /netbsd: ppp0: decompress failed 1
Mar 29 22:05:52 hawkwind /netbsd: ppp_deflate0: exceeded mru (1508 > 1504)
Mar 29 22:05:57 hawkwind last message repeated 5 times
Mar 29 22:17:50 hawkwind /netbsd: ppp0: bad fcs 8dd3
Mar 29 22:17:50 hawkwind /netbsd: z_decompress0: bad seq # 998, expected 997
Mar 29 22:17:50 hawkwind /netbsd: ppp0: decompress failed 1
Mar 29 22:17:55 hawkwind /netbsd: ppp0: VJ uncompress failed on type comp
Mar 29 22:17:57 hawkwind /netbsd: ppp0: VJ uncompress failed on type comp
Mar 29 22:23:43 hawkwind /netbsd: ppp0: bad fcs b58b
Mar 29 22:23:44 hawkwind /netbsd: z_decompress0: bad seq # 395, expected 394
Mar 29 22:23:44 hawkwind /netbsd: ppp0: decompress failed 1
Mar 29 22:23:46 hawkwind /netbsd: ppp_deflate0: exceeded mru (1508 > 1504)
Mar 29 22:26:29 hawkwind /netbsd: ppp_deflate0: exceeded mru (1508 > 1504)
Mar 29 22:26:42 hawkwind last message repeated 17 times
Mar 29 22:26:54 hawkwind /netbsd: ppp0: bad fcs 4885
Mar 29 22:26:54 hawkwind /netbsd: z_decompress0: bad seq # 282, expected 281
[[etc.]]
<<<<<

This makes me think that bits are getting dropped somewhere, maybe?

I took a quick look at /sys/arch/alpha/tc/scc.c, and noticed there are a
lot of hardwired "9600"s in there.  Also, there's quite a few of this
kind of code:

>>>>>
#ifdef pmax
        /* XXX -- why on pmax, not on Alpha? */
        cterm.c_cflag  |= CLOCAL;
#endif
<<<<<

So why is it not on alpha?

Finally, I have my 3000/300 jumpered to select the serial console, and
the boot up sequence flags it as the console.  Would this hard-wire the
speed to 9600???

Cheers,

Paul.

e-mail: paul@gromit.dlib.vt.edu

"Without music to decorate it, time is just a bunch of boring production
 deadlines or dates by which bills must be paid."
        --- Frank Vincent Zappa