Linux on HP PA-RISC development discussion ()

Rolf Eike Beer | 6 Jun 2012 14:55

3.4.1 fails to boot on Apollo 705 in paging_init

Linux version 3.4.1 (root <at> voyager) (gcc version 4.5.3 (Gentoo 4.5.3-r2
p1.5, pie-0.4.7) ) #1 Wed Jun 6 14:15:43 CEST 2012
unwind_init: start = 0x1044e000, end = 0x1047faf0, entries = 12719
FP[0] enabled: Rev 3 Model 0
The 32-bit Kernel has started...
bootconsole [ttyB0] enabled
Initialized PDC Console for debugging.
Determining PDC firmware type: Snake.
model 00003020 00000481 00000000 00000000 77aac223 ffffffff 00000004
0000000d 00000000
vers  00000003
model 9000/705
Total Memory: 64 MB
      _______________________________
     < Your System ate a SPARC! Gah! >
      -------------------------------
             \   ^__^
                 (__)\       )\/\
                  U  ||----w |
                     ||     ||
swapper (pid 0): Protection id trap (code 7)

     YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI
PSW: 00000000000001000000000000001110 Not tainted
r00-03  0004000e 0004000e 1050104c 00000000
r04-07  00000001 104841cc 1048e000 104917b8
r08-11  104917d4 1048e024 00000000 00100000
r12-15  00004000 00000000 00537fff 104b58c8
r16-19  104ec800 00004000 1048e000 0000000e
r20-23  00000000 00000000 00000000 00000000

(Continue reading)

headers

Helge Deller | 13 Jun 2012 23:13

Re: 3.4.1 fails to boot on Apollo 705 in paging_init

Hi Rolf,

Did the machine worked with the old debian kernels?

According to http://hwdb.parisc-linux.org/view.php?type=machine&name=705 
it has a PA7000 (PCX-S)) CPU, which is PA1.1a generation.
Probably we still use some assembly which is not supported on PA1.1a yet.
My "oldest" testmachine was a 715/64 which is PA7100LC aka PA 1.1d 
generation...

I can try to look into this, but this will take some time since I'll 
leave for two-week vacation shortly...

Helge

On 06/06/2012 02:55 PM, Rolf Eike Beer wrote:
> Linux version 3.4.1 (root <at> voyager) (gcc version 4.5.3 (Gentoo 4.5.3-r2
> p1.5, pie-0.4.7) ) #1 Wed Jun 6 14:15:43 CEST 2012
> unwind_init: start = 0x1044e000, end = 0x1047faf0, entries = 12719
> FP[0] enabled: Rev 3 Model 0
> The 32-bit Kernel has started...
> bootconsole [ttyB0] enabled
> Initialized PDC Console for debugging.
> Determining PDC firmware type: Snake.
> model 00003020 00000481 00000000 00000000 77aac223 ffffffff 00000004
> 0000000d 00000000
> vers  00000003
> model 9000/705
> Total Memory: 64 MB
>        _______________________________

(Continue reading)

headers

Rolf Eike Beer | 13 Jun 2012 23:27

Re: 3.4.1 fails to boot on Apollo 705 in paging_init

Helge Deller wrote:

> Did the machine worked with the old debian kernels?

I have no idea, I'm running Gentoo. Current system is this:

Linux apollo 2.6.38.8 #2 Wed Feb 15 17:42:42 CET 2012 parisc PA7000
(PCX-S) 9000/705 GNU/Linux

> According to http://hwdb.parisc-linux.org/view.php?type=machine&name=705
> it has a PA7000 (PCX-S)) CPU, which is PA1.1a generation.
> Probably we still use some assembly which is not supported on PA1.1a yet.
> My "oldest" testmachine was a 715/64 which is PA7100LC aka PA 1.1d
> generation...
>
> I can try to look into this, but this will take some time since I'll
> leave for two-week vacation shortly...

I've reported the breakage in 2.6.39 in the cache code here a while ago
(that's the stuff that recently got some work), afterwards I found no
kernel that was working.

Eike
--
To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

headers

John David Anglin | 14 Jun 2012 03:58

Re: 3.4.1 fails to boot on Apollo 705 in paging_init

On 13-Jun-12, at 5:27 PM, Rolf Eike Beer wrote:

> I've reported the breakage in 2.6.39 in the cache code here a while  
> ago
> (that's the stuff that recently got some work), afterwards I found no
> kernel that was working.

Can you make sure that the latest set of cache fixes are present?  I'm
fairly certain they are all in 3.4.2.

If that doesn't resolve the crash, could you disassemble the code around
the failing instruction and submit to list?  Also, Helge might be able  
to test
your config.

I'm a bit skeptical that this is caused by unsupported assembly  
introduced
after 2.6.39.  If James' original tmpalias patch plus the recent fixes  
works
on top of 2.6.39, then a regression hunt may be in order.

Dave
--
John David Anglin	dave.anglin <at> bell.net

--
To unsubscribe from this list: send the line "unsubscribe linux-parisc" in
the body of a message to majordomo <at> vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

(Continue reading)