[parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Karsten Merker]

Hallo everybody,

I experience a problem with the kernel cvs checkout from dec 16, 15:40
CET. I am trying to build a kernel with support for STICON, but for some
reason CONFIG_FBCON_STI does not get set when doing a "make menuconfig",
although it should get set.

I chose 

"Character Devices -> Support for console on virtual terminal"
"Console Drivers -> Frame-buffer support -> Generic STI frame buffer
                                            device support"
"Console Drivers -> STI console"

This results in a .config containing

CONFIG_VT=y
CONFIG_VT_CONSOLE=y
CONFIG_FB_STI=y
CONFIG_STI_CONSOLE=y

but CONFIG_FBCON_STI, which is needed to get drivers/video/fbcon-sti.c
compiled and linked, ist not declared, although this should be done by
drivers/video/Config.in which contains the following code:

      if [ "$CONFIG_FB_STI" = "y" ]; then
         define_tristate CONFIG_FBCON_STI y
      fi

$CONFIG_FB_STI is set to "y" as can be seen in .config, nonetheless
CONFIG_FBCON_STI is not declared. I have been looking for the reason for
this behaviour, but have not found a solution. Probably I am missing
something dead simple, does anybody have an idea?

Greetings,
Karsten
-- 
#include <standard_disclaimer>
Nach Paragraph 28 Abs. 3 Bundesdatenschutzgesetz widerspreche ich der
Nutzung oder Uebermittlung meiner Daten fuer Werbezwecke oder fuer die
Markt- oder Meinungsforschung.

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Helge Deller]

> $CONFIG_FB_STI is set to "y" as can be seen in .config, nonetheless
> CONFIG_FBCON_STI is not declared. 

Hi Karsten,

I can't reproduce your problems with a freshly checked-out tree. Maybe you 
have wrong permissions on .config or a similiar silly problem ?

If you check out, please remove your .config and make a new "make menuconfig".
Now remember to DISABLE (!!) "Support for console on serial port" under 
"Character Devices" and enable STI-console and STI-fb again.

Greetings,

	Helge

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Karsten Merker]

On Sun, Dec 17, 2000 at 09:20:51PM +0100, Helge Deller wrote:

> > $CONFIG_FB_STI is set to "y" as can be seen in .config, nonetheless
> > CONFIG_FBCON_STI is not declared. 
> 
> Hi Karsten,
> 
> I can't reproduce your problems with a freshly checked-out tree. Maybe you 
> have wrong permissions on .config or a similiar silly problem ?
> 
> If you check out, please remove your .config and make a new "make menuconfig".
> Now remember to DISABLE (!!) "Support for console on serial port" under 
> "Character Devices" and enable STI-console and STI-fb again.

I did that, but - as I have found out now - something more I should better
not have done :-).
I had activated "Console drivers -> Frame buffer support -> Advanced low
level driver options", and whenever this is selected, CONFIG_FBCON_STI is
not defined. If I unselect it, everything is ok. This is because the code
sequence

      if [ "$CONFIG_FB_STI" = "y" ]; then
         define_tristate CONFIG_FBCON_STI y
      fi

is not called when CONFIG_FBCON_ADVANCED ist set to "y":

   bool '  Advanced low level driver options' CONFIG_FBCON_ADVANCED
   if [ "$CONFIG_FBCON_ADVANCED" = "y" ]; then
      tristate '    Monochrome support' CONFIG_FBCON_MFB
      tristate '    2 bpp packed pixels support' CONFIG_FBCON_CFB2
      tristate '    4 bpp packed pixels support' CONFIG_FBCON_CFB4
      tristate '    8 bpp packed pixels support' CONFIG_FBCON_CFB8
      [... etc. ...]
   else
      [... etc. ...]
      if [ "$CONFIG_FB_STI" = "y" ]; then
         define_tristate CONFIG_FBCON_STI y
      fi
   fi

I had not seen this from the beginning as there are lots of nested
if-then-else blocks, in which I just had missed the point. I had not
thought that selecting CONFIG_FBCON_ADVANCED would be a problem, as the
additional options available when CONFIG_FBCON_ADVANCED is set do not seem
to contain anything regarding the definition of CONFIG_FBCON_STI and I
thought it would only show or hide the submenu. My mistake...

Thanks a lot for your help.

Another thing I would like to ask (please forgive a PA-RISC-newbie if
this is a dumb question, at least this is not mentioned in the FAQ on
parisc-linux.org):

Has anybody tried to run Linux with serial console on a 715/100? When I
boot a kernel without support for STICON but with serial console enabled
(and console=ttyS0 as commandline parameter) the system crashes just after
displaying the "kernel command line" (system console is still in
"graphics" mode - does this make a difference?). 

When using STICON instead, the console output works in principle, but
there is pixel garbage throughout the screen (it looks like it is between
the character cells) after the second "mode switch" (when the first line
on the screen is "Console: switching to colour STI console 160x64"). I
have seen a similar effect on a 712/80 when using STICON. The keyboard
works on the 712/80 but is dead on the 715/100 (in both cases the keyboard
is a ps/2 model, connected directly on the 712/80 and via the small
external adaptor box on the 715/100).

Greetings,
Karsten
-- 
#include <standard_disclaimer>
Nach Paragraph 28 Abs. 3 Bundesdatenschutzgesetz widerspreche ich der Nutzung
oder Uebermittlung meiner Daten fuer Werbezwecke oder fuer die Markt- oder
Meinungsforschung.

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Helge Deller]

On Monday 18 December 2000 23:03, Karsten Merker wrote:
> On Sun, Dec 17, 2000 at 09:20:51PM +0100, Helge Deller wrote:
> > > $CONFIG_FB_STI is set to "y" as can be seen in .config, nonetheless
> > > CONFIG_FBCON_STI is not declared.
[snip]
> I had activated "Console drivers -> Frame buffer support -> Advanced low
> level driver options", and whenever this is selected, CONFIG_FBCON_STI is
> not defined. If I unselect it, everything is ok. 
[snip]

Ok, thanks a lot for your investigation. I hadn't realized that and since 
most of the developers use the default config, this problem has not shown up 
yet. 

> Has anybody tried to run Linux with serial console on a 715/100? When I
> boot a kernel without support for STICON but with serial console enabled
> (and console=ttyS0 as commandline parameter) the system crashes just after
> displaying the "kernel command line" (system console is still in
> "graphics" mode - does this make a difference?).

This doesn't happen on my 715/64, which is afaik nearly the same machine as 
yours. And, you don't need "console=ttyS0", but I don't think, that this is 
related to your problem.

> When using STICON instead, the console output works in principle, but
> there is pixel garbage throughout the screen (it looks like it is between
> the character cells) after the second "mode switch" (when the first line
> on the screen is "Console: switching to colour STI console 160x64"). I
> have seen a similar effect on a 712/80 when using STICON. 

Yes, this is a known problem - it happens here too -, but since some of the 
developers currently work on better STI-support this will get fixed in the 
future - and I think this problem was not there, before we merged with linus.

> The keyboard
> works on the 712/80 but is dead on the 715/100 (in both cases the keyboard
> is a ps/2 model, connected directly on the 712/80 and via the small
> external adaptor box on the 715/100).

Hmm, does the other end of this external adaptor on the /100 looks like a 
RJ45-connector? If yes, I assume, that you have a PS/2 keyboard on a 
HIL-adaptor and need the HIL keyboard driver, which should already be enabled 
in defconfig. And the HIL keyboard works for me...
If you could send me your bootlog I could take a closer look at it.

Greetings,

	Helge

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Karsten Merker]

On Wed, Dec 20, 2000 at 01:00:16AM +0100, Helge Deller wrote:

> > Has anybody tried to run Linux with serial console on a 715/100? When I
> > boot a kernel without support for STICON but with serial console enabled
> > (and console=ttyS0 as commandline parameter) the system crashes just after
> > displaying the "kernel command line" (system console is still in
> > "graphics" mode - does this make a difference?).
> 
> This doesn't happen on my 715/64, which is afaik nearly the same machine as 
> yours. And, you don't need "console=ttyS0", but I don't think, that this is 
> related to your problem.

It seems the "console=ttyS0" is the reason for the crash. To be sure, I
have taken a fresh kernel tree (checkout from today) and compiled the
kernel with the default configuration. It worked...
When I append a "console=ttyS0" to the kernel command line, the result is:

palo ipl karsten@excalibur.cologne.de Don Dez 21 22:08:22 CET 2000
0/vmlinux 2137117 bytes @ 0x6800
0/palo-cmdline '0/vmlinux HOME=/ TERM=linux root=/dev/nfs nfsroot=192.168.2.37 console=ttyS0'
Kernel: partition 0 file /vmlinux
ELF32 executable

Entry 00100158 first 00000000 n 5
Segment 0 load 00000000 size 304 mediaptr 0x1000
Segment 1 load 00100000 size 1434376 mediaptr 0x2000
Segment 2 load 00260000 size 170792 mediaptr 0x161000
Segment 3 load 0028c000 size 118432 mediaptr 0x18b000
Segment 4 load 002ac000 size 8192 mediaptr 0x1a8000
branching to kernel entry point 0x00100158
Set default PSW W bit to 0
PDC Console Initialized
The 32-bit Kernel has started...
Enabled FP coprocessor
Free memory starts at: 0xc02d7000
start_parisc(0x504d70,0x504d70,0x0,0x0)
PALO command line: 'HOME=/ TERM=linux root=/dev/nfs nfsroot=192.168.2.37 console=ttyS0'
PALO initrd 0-0
model   000060b0 00000481 00000000 00000000 779866f5 00000000 00000004 00000072 00000072
vers    0000000c
CPUID   vers 0 rev 0
model   9000/715
led_init: CPU_HVERSION 60b
Searching for devices in PDC firmware... processor hpa 0xfffbe000
 an older box...
Found devices:
1. Mirage GSC Builtin Graphics (10) at 0xf8000000, versions 0x13, 0x0, 0x85, 0x0, 0x0
2. Mirage Core BA (11) at 0xf0100000, versions 0x29, 0x0, 0x81, 0x0, 0x0
3. Mirage Core SCSI (10) at 0xf0106000, versions 0x29, 0x0, 0x82, 0x0, 0x0
4. Mirage Core LAN (802.3) (10) at 0xf0107000, versions 0x29, 0x0, 0x8a, 0x0, 0x0
5. Mirage Core RS-232 (10) at 0xf0105000, versions 0x29, 0x0, 0x8c, 0x0, 0x0
6. Mirage Core Centronics (10) at 0xf0102000, versions 0x29, 0x0, 0x74, 0x0, 0x0
7. Mirage Audio (10) at 0xf0104000, versions 0x29, 0x0, 0x7b, 0x0, 0x0
8. Mirage Core PC Floppy (10) at 0xf010a000, versions 0x29, 0x0, 0x83, 0x0, 0x0
9. Mirage Core PC Keyboard (10) at 0xf0108000, versions 0x29, 0x0, 0x84, 0x0, 0x0
10. Mirage Core PC Keyboard (10) at 0xf0108100, versions 0x29, 0x0, 0x84, 0x0, 0x0
11. Mirage Wax EISA BA (11) at 0xfc000000, versions 0x29, 0x0, 0x90, 0x0, 0x0
12. Mirage Wax BA (11) at 0xf0200000, versions 0x13, 0x0, 0x8e, 0x0, 0x0
13. Mirage 100 Wax HIL (10) at 0xf0201000, versions 0x13, 0x0, 0x73, 0x0, 0x0
14. Mirage Wax RS-232 (10) at 0xf0202000, versions 0x13, 0x0, 0x8c, 0x0, 0x0
15. Mirage 100 (0) at 0xfffbe000, versions 0x60b, 0x0, 0x4, 0x0, 0x81
16. Mirage 100 (1) at 0xfffbf000, versions 0x4b, 0x0, 0x9, 0x0, 0x0
That's a total of 16 devices.
CPU(s): 1 x PA7100LC (PCX-L) at 100.000000 MHz
Linux version 2.4.0-test10 (karsten@excalibur.cologne.de) (gcc version 2.96 20000925 (experimental)) #5 Don Dez 21 22:30:29 CET 2000
free_bootmem(0x2d7800, 0x3d28800)
initrd: 00000000-00000000
pagetable_init
On node 0 totalpages: 16384
zone(0): 8192 pages.
zone(1): 8192 pages.
zone(2): 0 pages.
Kernel command line: HOME=/ TERM=linux root=/dev/nfs nfsroot=192.168.2.37 console=ttyS0
trap_init

Dumping Stack from c02ac000 to c02aca40:
c000 00000000 00000000 00000000 00000000 c027446c 00000000 00000000 ffffffff
c020 0000000a 00000000 00000000 00000000 00000000 00000000 ffffffff c02ac03c
c040 c02ac03c 00000000 c02ac000 c02ac000 c027960c 00000000 00000000 00000000
c060 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c080 00000000 c02ac000 c02ac000 00000000 00000000 00000000 c02ac098 c02ac098
c0a0 00000000 00000000 00000000 c02ac0ac c02ac0ac 00000000 00000000 00000000
c0c0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c0e0 00000000 c011c080 00000000 00000000 00000000 00000000 00000000 00000000
c100 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c120 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c140 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c160 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c180 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c1a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c1c0 00000000 fffffeff 00000000 ffffffff 00000000 c027501c ffffffff ffffffff
c1e0 ffffffff ffffffff ffffffff ffffffff 00800000 05000000 00000000 ffffffff
c200 ffffffff ffffffff 00000000 00000000 00000400 00000400 ffffffff ffffffff
c220 ffffffff ffffffff ffffffff ffffffff 00007377 61707065 72000000 00000000
c240 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c260 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c280 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c2a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c2c0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c2e0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c300 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c320 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c340 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c360 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c380 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c3a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c3c0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c3e0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c400 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c420 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c440 00000000 00000000 00000000 00000000 0027a000 00000000 c0279040 c0279064
c460 00000000 c0279204 00000000 00000000 00000000 c02ac470 00000000 00000000
c480 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c4a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c4c0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c4e0 00000000 00000000 00000000 00500098 00000000 00000000 00000000 00000000
c500 0000003f 00000000 7f7fffff 00000001 0027a000 00800327 00000000 00000000
c520 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c540 00000000 00000000 00000000 00000000 00000064 c022e4a0 00000001 c0226818
c560 00000000 00000000 00000000 c028da34 00000000 00000000 00000000 00000000
c580 00000000 00000000 c02c5a70 c02c5800 c02c5800 f0000704 00000001 00000001
c5a0 00000000 fc3ffe1f c028d844 00000008 00504d70 00000000 00000000 c02c5800
c5c0 c02c5800 c02c6728 c02c5800 c02af000 00000060 ffffffff c02c5a70 c0225a1c
c5e0 00000000 fc3ffe1f c028d844 c028cecc 00504d70 00000000 00000000 00000000
c600 c02c6728 c02c5800 c02c5800 c02c6728 00000030 0000000a 00000140 c0226e90
c620 c02c5800 c02c6728 c028d844 c0293170 00504d70 00000000 00000000 c02c5800
c640 00000000 00000002 f000b858 f0000704 00000001 00000001 00000000 fc3ffe1f
c660 c028d844 00000008 00504d70 c0294e14 00000000 c02c5800 c02c5800 c02d3da4
c680 c02c5800 c02c6728 c02ac624 00000000 c02c5800 0000003c 00000001 00000001
c6a0 00000000 fc3ffe1f c028d844 00000008 00504d70 00000000 00000000 c02c5800
c6c0 c02c5800 c02d3da4 c02744a0 0000000a 0000000a 0000003c 0000003e c0274000
c6e0 00000001 c02b1f55 00000004 c011ac7c 00000009 c02b1f55 00000000 00000002
c700 c02c6c24 00000000 c02805a0 c02c6c20 00000000 00000003 f0105000 00008000
c720 c028d844 00000008 00504d70 c028ed68 00000000 c02c5800 00000000 00000002
c740 002ae080 00000000 002ae000 000003a8 00000000 00000003 f0105000 00008000
c760 00000000 c02c5800 00100158 c0294d14 00504d70 00105000 00000000 c02c5800
c780 00100158 00260000 c02ac724 00105000 c02b1f55 c02ac648 c02c9000 00000000
c7a0 c02752bc c02751e0 04000000 c018d7a8 00004000 00000000 0000000a c027542c
c7c0 00000008 00000000 00004000 00002000 0000000a c0275374 c0233a1e fffffc00
c7e0 00000001 c02b1f5a 00000004 c010c4c4 0000000e c02b1f5a 00000000 fc3ffe1f
c800 0006000a c02b1980 c018d7a8 c02c2088 00000013 0000002f c02c2e6c c02805a0
c820 ffffffff c0274000 c028d844 fc3ffe1f 00000000 00000001 00000001 f0000704
c840 f000b858 00000002 00000000 00000000 c02c2e6c c02c6800 ffffffff 00000040
c860 00000093 00000002 00000000 c0260000 00002580 0000002f c02aca40 c0294cdc
c880 000d081f 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c8a0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c8c0 7fffffff 7fffffff 7f7fffff ffffffff 7f7fffff ffffffff 00000000 00000000
c8e0 00000000 00000000 41800000 00000000 00000010 00000010 00000000 00000000
c900 7fffffff 7fffffff 41000000 00000000 40800000 00000000 7fffffff 7fffffff
c920 41000000 00000000 7fffffff 7fffffff 40800000 00000000 41000000 00000000
c940 40300000 00000000 40200000 00000000 40200000 00000000 41800000 7fffffff
c960 40000000 00000000 40000000 00000000 40800000 00000000 41000000 00000000
c980 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
c9a0 00000000 00000000 c010c8fc c010c900 00000000 00000000 00000000 00000000
c9c0 00000000 00000000 00000000 00000000 0000001f 0e781096 00000000 0000000c
c9e0 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
ca00 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
ca20 00000000 00000000 00000000 c0106704 00000000 00000000 00000000 00000000

Data access rights fault in kernel: Code=26 regs=c02ac800 (Addr=0000000c)

     YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001100000000000001010
r0-3     00000000 c02b1980 c018d7a8 c02c2088
r4-7     00000013 0000002f c02c2e6c c02805a0
r8-11    ffffffff c0274000 c028d844 fc3ffe1f
r12-15   00000000 00000001 00000001 f0000704
r16-19   f000b858 00000002 00000000 00000000
r20-23   c02c2e6c c02c6800 ffffffff 00000040
r24-27   00000093 00000002 00000000 c0260000
r28-31   00002580 0000002f c02aca40 c0294cdc
sr0-3    00000000 00000000 00000000 00000000
sr4-7    00000000 00000000 00000000 00000000

IASQ: 00000000 00000000 IAOQ: c010c8fc c010c900
 IIR: 0e781096    ISR: 00000000  IOR: 0000000c
ORIG_R28: 00000000

The only difference to a working kernel is the additional kernel
parameter. What I find strange in this case is the fact that giving
"console=ttyS0" is mandatory on other systems - e.g. Linux/MIPS on a
DECstation does not give you anything on the serial port unless you
specify console=ttySx, even if serial console is the only option because
no other console driver is compiled into the kernel.

> > The keyboard
> > works on the 712/80 but is dead on the 715/100 (in both cases the keyboard
> > is a ps/2 model, connected directly on the 712/80 and via the small
> > external adaptor box on the 715/100).
> 
> Hmm, does the other end of this external adaptor on the /100 looks like a 
> RJ45-connector? If yes, I assume, that you have a PS/2 keyboard on a 
> HIL-adaptor and need the HIL keyboard driver, which should already be enabled 
> in defconfig. And the HIL keyboard works for me...
> If you could send me your bootlog I could take a closer look at it.

I think I have found a reason for my keyboard problems: it depends on the
systems console path. I have checked this by booting the same kernel
twice, one time with the system console path set to "graphics", where the
keyboard does not work, and one time with the system console path set to
"rs232": et voila - the keyboard works. Another interesting effect: when
booting from serial system console, there is no pixel garbage on STICON,
when booting from graphics, it is there. It looks like the firmware
console routines (is this the so called PDC-console?) somehow clash with
the STICON and the keyboard driver. This would explain why everything is
ok when system console input/output is directed to the serial port instead
to keyboard and video controller. Can anybody confirm my experiences?

Greetings,
Karsten
-- 
#include <standard_disclaimer>
Nach Paragraph 28 Abs. 3 Bundesdatenschutzgesetz widerspreche ich der Nutzung
oder Uebermittlung meiner Daten fuer Werbezwecke oder fuer die Markt- oder
Meinungsforschung.

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Matthew Wilcox]

On Thu, Dec 21, 2000 at 11:39:34PM +0100, Karsten Merker wrote:
> It seems the "console=ttyS0" is the reason for the crash. To be sure, I
> have taken a fresh kernel tree (checkout from today) and compiled the
> kernel with the default configuration. It worked...

specifying `console=ttyS0' before the checkin i recently did would
definitely have caused exactly that crash.  it was trying to do an outb
because the serial driver wasn't properly configured.  since no 715 has
PCI, outb crashed & burned.

> The only difference to a working kernel is the additional kernel
> parameter. What I find strange in this case is the fact that giving
> "console=ttyS0" is mandatory on other systems - e.g. Linux/MIPS on a
> DECstation does not give you anything on the serial port unless you
> specify console=ttySx, even if serial console is the only option because
> no other console driver is compiled into the kernel.

Yeah, this is because our console handling code was on crack.  It should
work now...

-- 
Revolutions do not require corporate support.

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Alex deVries]

Matthew Wilcox wrote:
> 
> On Thu, Dec 21, 2000 at 11:39:34PM +0100, Karsten Merker wrote:
> > It seems the "console=ttyS0" is the reason for the crash. To be sure, I
> > have taken a fresh kernel tree (checkout from today) and compiled the
> > kernel with the default configuration. It worked...
> 
> specifying `console=ttyS0' before the checkin i recently did would
> definitely have caused exactly that crash.  it was trying to do an outb
> because the serial driver wasn't properly configured.  since no 715 has
> PCI, outb crashed & burned.

Has that been fixed now?  Should serial on a 715/old work?  

(Also, you can have a 715 with a PCI bus if you have a GSC add on card)

- Alex

-- 
Alex deVries, Principal Solutions Architect, The Puffins at Linuxcare
613.562.2759 tel
alex@linuxcare.com, http://www.linuxcare.com/
Linuxcare, Support for the revolution.

Re: [parisc-linux] configuration problems with cvs from 20001216 15:40 CET

[Matthew Wilcox]

On Fri, Dec 22, 2000 at 11:01:08AM -0500, Alex deVries wrote:
> > specifying `console=ttyS0' before the checkin i recently did would
> > definitely have caused exactly that crash.  it was trying to do an outb
> > because the serial driver wasn't properly configured.  since no 715 has
> > PCI, outb crashed & burned.
> 
> Has that been fixed now?  Should serial on a 715/old work?  

Serial always worked, as did serial console.  Only if you specified
`console=ttyS0' would the Bad Things happen, because the serial console
didn't do a proper initialisation, it just hoped.  Now, the serial console
initialisation is done after the serial driver starts up, so all the
initialisation is done properly first.

> (Also, you can have a 715 with a PCI bus if you have a GSC add on card)

true... though i've never seen one.

-- 
Revolutions do not require corporate support.