View previous topic :: View next topic |
Author |
Message |
ThorstenHirsch Tux's lil' helper
Joined: 25 Sep 2003 Posts: 111
|
Posted: Wed Mar 30, 2005 5:29 pm Post subject: Fast Instruction Access MMU Miss |
|
|
Hi,
today I wanted to upgrade from 2.4.29-sparc-r1 to 2.6.11-gentoo-r4. Well, it worked more or less with the 1st kernel I made, more or less meaning that I forgot the sermouse module and console fb was not working anymore. So I built a new one. But since I am testing this new one I always get the following error while booting:
Quote: | SILO Version 1.4.8
boot:
Allocated 8 Megs of memory at 0x40000000 for kernel
Loaded Kernel version 2.6.11
Remapping the kernel... Fast Instruction Access MMU Miss
ok |
I already did a "reset-all" and a "test-all", but they couldn't help me, everything's working fine. The machine is an Ultra 10.
Google found some pages for this error message telling me that I should check the size of my kernel image. Meanwhile my kernel is not only < 3,5MB but even < 3500000 bytes, but I still get the same error message.
Any ideas?
Solaris 10 on the same machine is still working fine. |
|
Back to top |
|
|
ThorstenHirsch Tux's lil' helper
Joined: 25 Sep 2003 Posts: 111
|
Posted: Thu Mar 31, 2005 4:03 pm Post subject: |
|
|
Great. My new kernel (more things enabled) is telling me
Quote: |
SILO Version 1.4.9
boot:
Allocated 8 Megs of memory at 0x40000000 for kernel
Loaded Kernel version 2.6.11
Remapping the kernel... Data Access Exception |
...and as you can see, I've updated silo. But this error already came with silo 1.4.8 and the new kernel. |
|
Back to top |
|
|
ThorstenHirsch Tux's lil' helper
Joined: 25 Sep 2003 Posts: 111
|
Posted: Thu Mar 31, 2005 4:59 pm Post subject: |
|
|
And when I try to boot my old kernel (2.4.29-sparc-r1, the one I was working with over months!) I get a nice little kernel ooops just after this message:
Quote: | unable to handle kernel paging request at virtual address 0000000140000000 |
|
|
Back to top |
|
|
Drunkula Apprentice
Joined: 28 Jul 2003 Posts: 257 Location: Denton, TX - USA
|
Posted: Thu Mar 31, 2005 7:06 pm Post subject: |
|
|
Sounds like you hit the motherlode! Seriously though I get nervous whenever I update the kernel or SILO on my Ultra 2E. When it works I try to leave it as-is. Not sure why it seems so hit-or-miss on some boxes.
|
|
Back to top |
|
|
ThorstenHirsch Tux's lil' helper
Joined: 25 Sep 2003 Posts: 111
|
Posted: Fri Apr 01, 2005 8:54 am Post subject: |
|
|
Error message of the day:
Quote: |
SILO Version 1.4.9
boot:
Allocated 8 Megs of memory at 0x40000000 for kernel
Loaded Kernel version 2.6.11
Remapping the kernel... Illegal Instruction |
This message came when I took out memory banks 1+2 or memory banks 3+4 (so I took out 512MB and left in 512MB).
Then I gave up and re-put all the memory in. Now kernel 2.6 is booting - hooray! ...well, but it doesn't detect my hda, but well, this is probably my mistake in the kernel config. After all, my old kernel is working again. |
|
Back to top |
|
|
jbwillia Tux's lil' helper
Joined: 07 May 2004 Posts: 108 Location: TN, USA
|
Posted: Thu Jul 21, 2005 3:38 am Post subject: |
|
|
WOW, I just came across the most interesting solution to this problem that I could have imagined. I rebooted my machine (an U60 2x450MHz 512M 2x18.1G hd) many times and everytime I had gotten the fast instruction access MMU miss error. I even tried running reset and reset-all at the ok prompt and it wuld reset and then return the same result. At this point I was just tired of it, so I turned the machine off while I went to google away. I came across one page where the guy said he'd solved his problem with a hard reboot. I dismissed it thinking that by running the resets that I had that it couldn't do anything. However, I beg to to differ now; I asked my wife from across the room to turn the machine on as I was going to check boot the livecd and check the size of my kernel and whatnot. Soon after from across the room I saw the black screen ad kernel booting.
Sooo, either my wife has the magic touch or a hard reboot really was the solution. Either way, I figured the hard reboot was worth posting for some other poor soul who runs into this problem. _________________
Code: | #!/bin/bash
for everyexe in $(find /home/you -name "*.exe" | egrep -i ".*exe$")
do
kill $(pgrep you)
done |
|
|
Back to top |
|
|
HackingM2 Apprentice
Joined: 26 Jul 2004 Posts: 245 Location: Cambridge, England
|
Posted: Fri Oct 13, 2006 2:26 am Post subject: |
|
|
I've just encountered what looks to be the same problem using anything >= 2.6.14.
I have a 2.6.13-r5 kernel booting fine.
Anyone have any suggestions / solutions other that those already presented? |
|
Back to top |
|
|
HackingM2 Apprentice
Joined: 26 Jul 2004 Posts: 245 Location: Cambridge, England
|
Posted: Fri Oct 27, 2006 6:55 pm Post subject: |
|
|
I have since made some progress diagnosing this issue further (and edited the above post to reflect this) which I thought someone may find useful... |
|
Back to top |
|
|
Galahad Tux's lil' helper
Joined: 12 Feb 2003 Posts: 126
|
Posted: Mon Oct 30, 2006 2:13 pm Post subject: |
|
|
The firmware on some suns is a huge steaming pile
There are a lot of firmware commands on my netra that result in "Fast Data access mmu miss", which is probably some similar error. |
|
Back to top |
|
|
HackingM2 Apprentice
Joined: 26 Jul 2004 Posts: 245 Location: Cambridge, England
|
Posted: Mon Oct 30, 2006 2:57 pm Post subject: |
|
|
Galahad wrote: | The firmware on some suns is a huge steaming pile. |
Tell me about it.
This issue seems to occur using OBP versions 4.0.5, 4.3.2, 4.5.9 and 4.17.1 so I don't think it is firmware related.
Which version of OBP are you using? Can you boot a kernel >= 2.6.14 ? |
|
Back to top |
|
|
|