Skip to contentSkip to conversation

related to swap?

OOPS Ian Kumlien 2 participants 4 messages Apr 5, 2006 – Apr 9, 2006

lore
Share & tools
lore
#0 lore raw
Quoted 5 lines
Quoted 2 lines · Ian Kumlien

Yes, i run a tainted kernel! either live with it or ignore this mail =)

Quoted 7 lines
Quoted 3 lines

starting swap lead to a deadlock within 15 mins

I have never had the energy to perform a full memtext86+

It would be useful if you could perform a memtest overnight one night, then run a non-patched and non-tained 2.6.16.1 kernel, and try to reproduce the problems.

As i said, i really doubt that the memory is at fault here, it has done several passes over the memory but not all tests. I can give it a go though, but i really doubt it'll find anything.

The kernel i run is a plain 2.6.16.1 from kernel.org (i have heard that you can actually compile gentoos own these days)

Since this is my *cough* desktop, running it without that ability is kinda a show stopper, thats why i included the thing above.

But the thing is, my laptop runs with the same compiler, "same" nvidia driver and the "same" kernel ("same" as in 32 bit not 64 bit). Eventhough "same" in this case usually means nothing, i doubt that one would have a serius bug and the other wouldn't, ie it's most likley a bug related to 64 bits or one or more of the drivers involved.

The only errors i get in dmesg atm is:
KERNEL: assertion (!sk->sk_forward_alloc) failed at net/core/stream.c
(283)
KERNEL: assertion (!sk->sk_forward_alloc) failed at net/ipv4/af_inet.c
(150)

Which is related to TSO, from what i gather, but i can't turn off tso on forcedeth... (i suspected this to cause corruption a while back....)

And:
eth0: too many iterations (6) in nv_nic_irq.

Which seems to be a warning of some sort (haven't checked the source yet).

PS. Reply-to-all is your friend... =)
DS.

Signature · 1 line
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net
#1 lore raw
Quoted 10 lines · Ian Kumlien
Quoted 10 lines
Quoted 6 lines · Ian Kumlien

Yes, i run a tainted kernel! either live with it or ignore this mail =)

starting swap lead to a deadlock within 15 mins

I have never had the energy to perform a full memtext86+

Quoted 9 lines
Quoted 3 lines

It would be useful if you could perform a memtest overnight one night, then run a non-patched and non-tained 2.6.16.1 kernel, and try to reproduce the problems.

As i said, i really doubt that the memory is at fault here, it has done several passes over the memory but not all tests. I can give it a go though, but i really doubt it'll find anything.

If it doesn't cost you much time (ie. do it overnight) it could save some developers a lot of time.

Quoted 3 lines

The kernel i run is a plain 2.6.16.1 from kernel.org (i have heard that you can actually compile gentoos own these days)

OK, good.

Quoted 3 lines

Since this is my *cough* desktop, running it without that ability is kinda a show stopper, thats why i included the thing above.

But if the problem can be reproduced in 15 minutes, it shouldn't be too hard to get a trace without nvidia loaded.

Quoted 6 lines

But the thing is, my laptop runs with the same compiler, "same" nvidia driver and the "same" kernel ("same" as in 32 bit not 64 bit). Eventhough "same" in this case usually means nothing, i doubt that one would have a serius bug and the other wouldn't, ie it's most likley a bug related to 64 bits or one or more of the drivers involved.

Quoted 9 lines

The only errors i get in dmesg atm is:
KERNEL: assertion (!sk->sk_forward_alloc) failed at net/core/stream.c
(283)
KERNEL: assertion (!sk->sk_forward_alloc) failed at net/ipv4/af_inet.c
(150)

Which is related to TSO, from what i gather, but i can't turn off tso on forcedeth... (i suspected this to cause corruption a while back....)

If your network hardware or driver is flakey, try compiling a kernel without that as well before reproducing this swap problem.

Thanks,
Nick

Signature · 2 lines
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 
#2 lore raw
Quoted 16 lines · Nick Piggin
Quoted 15 lines · Ian Kumlien
Quoted 13 lines
Quoted 6 lines · Ian Kumlien

Yes, i run a tainted kernel! either live with it or ignore this mail =)

starting swap lead to a deadlock within 15 mins

I have never had the energy to perform a full memtext86+

It would be useful if you could perform a memtest overnight one night, then run a non-patched and non-tained 2.6.16.1 kernel, and try to reproduce the problems.

Quoted 7 lines
Quoted 4 lines

As i said, i really doubt that the memory is at fault here, it has done several passes over the memory but not all tests. I can give it a go though, but i really doubt it'll find anything.

If it doesn't cost you much time (ie. do it overnight) it could save some developers a lot of time.

Yes, i'll try to do it this weekend... Since i need to keep this box running most of the time.

Quoted 4 lines
Quoted 2 lines

The kernel i run is a plain 2.6.16.1 from kernel.org (i have heard that you can actually compile gentoos own these days)

OK, good.

Kernel.org, it's the only flavor =)

Quoted 5 lines
Quoted 2 lines

Since this is my *cough* desktop, running it without that ability is kinda a show stopper, thats why i included the thing above.

But if the problem can be reproduced in 15 minutes, it shouldn't be too hard to get a trace without nvidia loaded.

The major bit is, this worked with single core with reiser... Someone else reported problems with XFS lately though, i'll have to read all about it...

Quoted 12 lines
Quoted 12 lines

But the thing is, my laptop runs with the same compiler, "same" nvidia driver and the "same" kernel ("same" as in 32 bit not 64 bit). Eventhough "same" in this case usually means nothing, i doubt that one would have a serius bug and the other wouldn't, ie it's most likley a bug related to 64 bits or one or more of the drivers involved.

The only errors i get in dmesg atm is:
KERNEL: assertion (!sk->sk_forward_alloc) failed at net/core/stream.c
(283)
KERNEL: assertion (!sk->sk_forward_alloc) failed at net/ipv4/af_inet.c
(150)

Quoted 5 lines
Quoted 2 lines

Which is related to TSO, from what i gather, but i can't turn off tso on forcedeth... (i suspected this to cause corruption a while back....)

If your network hardware or driver is flakey, try compiling a kernel without that as well before reproducing this swap problem.

Well, flakey and flakey, it's been running with pretty heavy network and io load for 4 days now... (with forcedeth, but there is no ethtool interface for disabling TSO, if that is the problem)

The sky2 driver tended to just die and lockup the box after a while, but i haven't tested Stepehns recent changes yet since i want something reliable first, to i have a base for comparison.

Oh, Btw:
free
total used free shared buffers
cached
Mem: 3056436 3039052 17384 0 4804
2001672
-/+ buffers/cache: 1032576 2023860
Swap: 0 0 0

If the memory was faulty, i'd have a corrupt filesystem by now imho... =)

Btw, sorry about being unclear, i spotted it after i mailed, but... Running without nvidia drivers on a nvdidia card is pretty horrible on a desktop machine, the kind of resolutions you can get would have been ok on a amiga, but with X, everything is sooo much bigger... =P

Also, sorry about the CC comment, but it took a day for your mail to hit my inbox, ie a day after it appeared on lkml, must be the greylists being overly efficient... =P

Signature · 1 line
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net
#3 lore raw
In reply to Nick Piggin
Quoted 7 lines · Nick Piggin
Quoted 3 lines · Ian Kumlien

As i said, i really doubt that the memory is at fault here, it has done several passes over the memory but not all tests. I can give it a go though, but i really doubt it'll find anything.

If it doesn't cost you much time (ie. do it overnight) it could save some developers a lot of time.

Well, the memory passes test 1-5 and 8. The 32 bit thingies all fail on 4 dimms, so either 4 dimms are broken or the test is broken for amd64.

Anyways, i noticed that the memory clocking got skewed when all the memory was in the machine, so i've removed 1gb of memory.

Quoted 4 lines
Quoted 2 lines

The kernel i run is a plain 2.6.16.1 from kernel.org (i have heard that you can actually compile gentoos own these days)

OK, good.

2.6.16.2 now... And a new nvidia driver...

Signature · 1 line
Ian Kumlien <pomac () vapor ! com> -- http://pomac.netswarm.net