Bug #2705

panic: assertion "pv->pv_m == p" in pmap_remove_pv_pte

Added by kaz 4 months ago. Updated 3 months ago.

Status:ClosedStart date:08/08/2014
Priority:NormalDue date:
Assignee:-% Done:

0%

Category:VM subsystem
Target version:3.9.x

Description

A recently-installed machine running a recent git (my hardware had problems with the current release) has been crashing frequently when X is running; it hasn't consistently provided a core dump (maybe I rebooted it too quickly?), but I have two of them (attached). The first (3) was with X running, the second (4) was when I tried to open firefox. Firefox does sometimes open successfully. The crashes are frequent enough I've been keeping it in console.

core.txt.4 (133 KB) kaz, 08/08/2014 11:57 AM

core.txt.3 (122 KB) kaz, 08/08/2014 11:57 AM

info.3 (567 Bytes) kaz, 08/08/2014 11:57 AM

info.4 (567 Bytes) kaz, 08/08/2014 11:57 AM

dmesg.txt Magnifier - dmesg (10.3 KB) kaz, 08/09/2014 09:19 AM

pciconf.txt Magnifier - pciconf -lv (4.07 KB) kaz, 08/09/2014 09:19 AM

dmesg.displaycorrupt - after corruption (10.3 KB) kaz, 08/11/2014 11:32 AM

Xorg.log.displaycorrupt - after corruption (27 KB) kaz, 08/11/2014 11:32 AM

messages.displaycorrupt - after corruption (4.39 KB) kaz, 08/11/2014 11:32 AM


Related issues

Related to Bug #2707: physical memory corruption with intel gpu 4500HD New 08/09/2014

History

#1 Updated by jorisgio 4 months ago

Could you please post a dmesg/pciconf ?

#2 Updated by kaz 4 months ago

dmesg / pciconf attached

#3 Updated by kaz 3 months ago

  • Status changed from New to Closed

This computer just had an episode of all kinds of display corruption (some weird colors on the screen, letters garbled in firefox). I think this is likely a RAM problem. I haven't been able to get all the way through a memtest pass because this old laptop overheats; I'll probably try sticking the RAM in another computer to test it properly. Anyway, I'll reopen this if further investigation suggests it isn't a hardware fault but otherwise this can just be ignored.

#4 Updated by kaz 3 months ago

Update: it doesn't seem to be my RAM after all -- I've tried running with either of my DIMMs removed, and the same display corruption occurred in either case. What I'm seeing is mainly letters being replaced with other letters, both in rxvt and firefox, starting soon (but not immediately) after I start up X. It usually locks up shortly after the corruption begins. It has started happening consistently with X running with xf86-video-intel. I'm trying vesa now; no problems so far.

I've attached some logs from after the display corruption begins, although I don't see anything unusual there.

#5 Updated by jorisgio 3 months ago

  • Related to Bug #2707: physical memory corruption with intel gpu 4500HD added

#6 Updated by jorisgio 3 months ago

  • Related to deleted (Bug #2707: physical memory corruption with intel gpu 4500HD)

#7 Updated by jorisgio 3 months ago

  • Related to Bug #2707: physical memory corruption with intel gpu 4500HD added

#8 Updated by jorisgio 3 months ago

It could be issue #2707, but i'm not 100% certain.I'm linking the too bugs for now.h

Also available in: Atom PDF