0.9.3 on a B132L
Hi folks,
First of all, thanks for all the hard work that has gone into 0.9.3. Excellent!
I've got a couple of observations and/or questions. Apologies for the lack of dmesg output etc. - I'm at work and the workstation is at home. I have a B132L, with the Visualize EG, GRAPHICS(0), on board frame buffer (that's Gecko, if I'm not mistaken) and a 4071B in the GSC slot, GRAPHICS(2) (this thing is Coral in the dmesg output I believe).
Previously, when playing around with 0.9.2 I had set the display path to GRAPHICS(0), as the 4071B was unsupported. As it turned out, the PS/2 keyboard problems forced me to use a serial console in the end anyway, but at least the EG seemed to be supported.
So yesterday I booted up 0.9.3 in this configuration. The messages started to appear on the screen, then stopped after the 'if you don't see anything more' message. In a flash of inspiration, I moved the monitor cable to the 4071B, and lo and behold, all the output was there! The dmesg shows that its finding the 4071B before the EG, and using that as the console display even though the display path is set to GRAPHICS(0). Is this expected behaviour at this point in time, or is there something fishy going on.
Switching the display path to GRAPHICS(2) fixed things - 0.9.3 booted and displayed on the 4071B.
So far so good. Installation seemed to go mostly ok - I have two disks on the FW-Diff controller in this box. I guess I should be specifying the burst:3 parameter to avoid disk corruption.
The other problem I've experienced is keyboard lockup with the message 'LASI PS/2 keyboard timeout' scrolling up the screen. I was using a generic PS/2 keyboard at the time. Seems to kick in when the machine is installing from the cd and I'm messing about on another virtual console - although that is really just speculation. I can stick the B132L on the net tonight and log in to see what's going on next time the lockup occurs - anything in particular I should be looking for?
Regards,
-Andy
------------------------------------------------------------
Få din egen @start.no-adresse gratis på http://www.start.no/
[parisc-linux] Network bootproblem with 712/80
Hello all!
I am having a problem with my 712/80 when booting with bootp/tftp. With some effort I got it requesting kernel image (lifimage) over the network. The download starts but almost the same time it ends and it retries it over and over again. With Weird-solution's bootp and tftp server I can see that the download starts many times but all of them results error. When using the RedHat's bootp/tftp, the tcpdump shows (what I have understood from the output) that the HP is querying a file of name '.' (dot), this is quite strange...
Have I understood correctly that the lifimage (from the 0.9.3 installation cd) is the correct kernel image?
Tommi Aihkisalo
VTT Electronics @ Raahe
- ERROR 406: file corrupt: config.earth --- reboot universe? (Y/N)
Re: [parisc-linux] Network bootproblem with 712/80
Hello Tommi
Quoted 4 lines · Tommi Aihkisalo
Hello all!
I am having a problem with my 712/80 when booting with bootp/tftp. With some effort I got it requesting kernel image (lifimage) over the network. The
[SNIP]
I too have just tried a net boot of 0.9.3 from a remote location and also failed. I noticed MUCH too late that the lifimage contains the following files:
0/vmlinux32 3122078 bytes @ 0x7800
0/vmlinux64 3990942 bytes @ 0x302000
0/ramdisk 1946807 bytes @ 0x6d0800But PALO chooses a default kernel path of
0/linux
I assume that is simply that the PALO definition in the lifimage is incorrect. Once I get my remote button pusher to power cycle the system for me (she is out shopping right now) I will try an interactive boot & change the command line to boot the correct kernel.
If you get chance to try this before me then please let me know how it goes.
Cheers
PeteM
Re: [parisc-linux] Network bootproblem with 712/80
In reply to Tommi AihkisaloQuoted 11 lines · Tommi Aihkisalo
Hello all!
I am having a problem with my 712/80 when booting with bootp/tftp. With some effort I got it requesting kernel image (lifimage) over the network. The download starts but almost the same time it ends and it retries it over and over again. With Weird-solution's bootp and tftp server I can see that the download starts many times but all of them results error. When using the RedHat's bootp/tftp, the tcpdump shows (what I have understood from the output) that the HP is querying a file of name '.' (dot), this is quite strange...
Quoted 2 lines
Have I understood correctly that the lifimage (from the 0.9.3 installation cd) is the correct kernel image?
Sorry for the slow response. I just tried netbooting a 712 using the lifimage on the 0.9.3 ISO (dists/woody/main/disks-hppa/current/lifimage), and it booted fine. The palo command line does say 0/linux, while the lifimage contains linux32 and linux64 (or similar), but that is not a problem - palo figures out that it is on a 32 bit box and selects the 32 bit kernel correctly. I was using a Mandrake box as bootp/tftp server.
Richard
Re: [parisc-linux] Network bootproblem with 712/80
Quoted 11 lines · Richard Hirst
Quoted 10 lines · Tommi Aihkisalo
Hello all!
I am having a problem with my 712/80 when booting with bootp/tftp. With some effort I got it requesting kernel image (lifimage) over the network. The download starts but almost the same time it ends and it retries it over and over again. With Weird-solution's bootp and tftp server I can see that the download starts many times but all of them results error. When using the RedHat's bootp/tftp, the tcpdump shows (what I have understood from the output) that the HP is querying a file of name '.' (dot), this is quite strange...
Hi Tommi,
did you tried the newest versions of tftp-server on RH ? AFAIR the tftp-servers in 7.0 and 7.1 were broken and I fixed them during my time at RedHat.
Greetings,
Helge