NetBSD Problem Report #49364
From martin@duskware.de Wed Nov 5 13:30:18 2014
Return-Path: <martin@duskware.de>
Received: from mail.netbsd.org (mail.netbsd.org [149.20.53.66])
(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
(Client CN "mail.netbsd.org", Issuer "Postmaster NetBSD.org" (verified OK))
by mollari.NetBSD.org (Postfix) with ESMTPS id 80D14A6686
for <gnats-bugs@gnats.NetBSD.org>; Wed, 5 Nov 2014 13:30:18 +0000 (UTC)
From: martin@NetBSD.org
Reply-To: martin@NetBSD.org
To: gnats-bugs@NetBSD.org
Subject: arm kernel panics on exec
X-Send-Pr-Version: 3.95
>Number: 49364
>Category: port-arm
>Synopsis: arm kernel panics on exec
>Confidential: no
>Severity: serious
>Priority: high
>Responsible: port-arm-maintainer
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Wed Nov 05 13:35:00 +0000 2014
>Closed-Date: Sun Nov 09 11:10:59 +0000 2014
>Last-Modified: Mon Nov 10 20:00:00 +0000 2014
>Originator: Martin Husemann
>Release: NetBSD 7.99.1
>Organization:
The NetBSD Foundation, Inc.
>Environment:
System: NetBSD space-truckin.duskware.de 7.99.1 NetBSD 7.99.1 (CUBIETRUCK) #57: Wed Nov 5 10:29:20 CET 2014 martin@night-owl.duskware.de:/usr/src/sys/arch/evbarm/compile/CUBIETRUCK evbarm
Architecture: earmv7hfeb
Machine: evbarm
>Description:
Under VA pressure (or kernel memory, or whatever) execing new processes will
panic the kernel (pmap assertion: pg != NULL, details to be filled in).
This happens for me when extracting set files (especially the huge debug
ones) to wd0 and login in via ssh, or see below.
>How-To-Repeat:
Make sure you have plenty of space in /tmp (i.e. no tmpfs)
ssh into the box, then:
cd /usr/tests/fs/cd9660
atf-run | atf-report
Now wait like 10 minutes, then try to log in on the console
>Fix:
n/a
>Release-Note:
>Audit-Trail:
State-Changed-From-To: open->feedback
State-Changed-By: skrll@NetBSD.org
State-Changed-When: Sat, 08 Nov 2014 08:14:38 +0000
State-Changed-Why:
Possible fix commited
sys/arch/arm/arm32/pmap.c:1.312
From: "Matt Thomas" <matt@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/49364 CVS commit: src/sys/arch/arm/arm32
Date: Sat, 8 Nov 2014 08:01:34 +0000
Module Name: src
Committed By: matt
Date: Sat Nov 8 08:01:34 UTC 2014
Modified Files:
src/sys/arch/arm/arm32: pmap.c
Log Message:
When allocing a l1page, if a page isn't available, use uvm_wait to wait
for one to become available. Should fix PR/49364.
To generate a diff of this commit:
cvs rdiff -u -r1.311 -r1.312 src/sys/arch/arm/arm32/pmap.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: port-arm/49364 (arm kernel panics on exec)
Date: Sat, 8 Nov 2014 12:55:14 +0100
Doesn't seem to panic any more, but I can't login during the test run
either, neither via ssh nor on the console.
db{0}> show uvm
Current UVM status:
pagesize=8192 (0x2000), pagemask=0x1fff, pageshift=13
, ncolors=1 258934 VM pages: 169872 active, 82960 inactive, 1092 wired, 1601 fr
ee
pages 2201 anon, 250395 file, 1328 exec
freemin=256, free-target=341, wired-max=86311
cpu0:
faults=20079, traps=26366, intrs=10126135, ctxswitch=22556408
softint=86732, syscalls=121857
cpu1:
faults=93273, traps=132938, intrs=9967258, ctxswitch=25485053
softint=6056, syscalls=215410
fault counts:
noram=0, noanon=0, pgwait=0, pgrele=0
ok relocks(total)=503(503), anget(retrys)=33645(0), amapcopy=16293
neighbor anon/obj pg=22944/180984, gets(lock/unlock)=47190/503
cases: anon=22491, anoncow=11159, obj=40227, prcopy=6962, przero=30155
daemon and swap counts:
woke=13091735, revs=246, scans=324910, obscans=281348, anscans=0
busy=0, freed=281348, reactivate=778, deactivate=395235
pageouts=0, pending=0, nswget=0
nswapdev=1, swpgavail=262266
swpages=262266, swpginuse=0, swpgonly=0, paging=0
db{0}> ps
PID LID S CPU FLAGS STRUCT LWP * NAME WAIT
2051 1 2 0 0 bfd40060 atf-run
1701 1 3 0 80 bfd405e0 atf-report pipe_rd
1373 1 3 0 80 bfc91680 atf-run select
39 1 3 0 80 bfd40b60 tcsh pause
1819 1 3 0 80 bfbb7120 tcsh pause
1772 1 3 1 80 bfd40320 rxvt select
1809 1 3 1 80 bfd408a0 sshd select
1786 1 3 1 80 bfd410e0 sshd select
1807 > 1 7 0 0 bff22880 getty
1386 > 1 7 1 0 bfbb6ba0 cron
1663 1 3 1 80 bfbb6e60 inetd kqueue
1133 1 3 1 80 bfbb73e0 sshd select
885 1 3 0 80 bfc90b80 ntpd pause
540 1 3 0 80 bfc90e40 syslogd kqueue
470 1 2 0 0 bfc91100 dhcpcd
1 1 3 0 80 bfd41920 init wait
0 49 3 1 200 bfc913c0 physiod physiod
0 48 3 0 200 bfd413a0 aiodoned aiodoned
0 47 3 0 200 bff22b40 ioflush syncer
0 46 3 1 200 bfd41660 pgdaemon pgdaemon
0 42 3 0 200 bff22e00 usb4 usbevt
0 41 3 1 200 bff230c0 usb3 usbevt
0 40 3 0 200 bff23380 usb2 usbevt
0 39 3 1 200 bff23640 usb1 usbevt
0 38 3 1 200 bff225c0 usb0 usbevt
0 37 3 1 200 bfd41be0 unpgc unpgc
0 35 3 0 200 bff22300 vmem_rehash vmem_rehash
0 34 3 0 200 bff22040 sdmmc0 mmctaskq
0 25 3 0 200 bff23900 iic0 iicintr
0 24 3 1 200 bff23bc0 atabus0 atath
0 23 3 0 200 bff66020 usbtask-dr usbtsk
0 22 3 0 200 bff662e0 usbtask-hc usbtsk
0 21 3 1 200 bff665a0 xcall/1 xcall
0 20 1 1 200 bff66860 softser/1
0 19 1 1 200 bff66b20 softclk/1
0 18 1 1 200 bff66de0 softbio/1
0 17 1 1 200 bff670a0 softnet/1
0 16 1 1 201 bff67360 idle/1
0 15 3 1 200 bff67620 pmfsuspend pmfsuspend
0 14 3 1 200 bff678e0 pmfevent pmfevent
0 13 3 0 200 bff67ba0 sopendfree sopendfr
0 12 3 1 200 bff94000 nfssilly nfssilly
0 11 3 0 200 bff942c0 cachegc cachegc
0 10 3 0 200 bff94580 vrele vrele
0 9 3 0 200 bff94840 vdrain vdrain
0 8 3 0 200 bff94b00 modunload mod_unld
0 7 3 0 200 bff94dc0 xcall/0 xcall
0 6 1 0 200 bff95080 softser/0
0 5 1 0 200 bff95340 softclk/0
0 4 1 0 200 bff95600 softbio/0
0 3 1 0 200 bff958c0 softnet/0
0 2 1 0 201 bff95b80 idle/0
0 1 3 0 200 8044acc0 swapper uvm
db{0}> bt
0xbfc2db4c: netbsd:comintr+0xc
0xbfc2db7c: netbsd:pic_dispatch+0xe8
0xbfc2dbd4: netbsd:armgic_irq_handler+0xd4
0xbfc2dc44: netbsd:irq_entry+0x60
0xbfc2dc94: netbsd:mi_switch+0x314
0xbfc2dcc4: netbsd:sleepq_block+0xb4
0xbfc2dd04: netbsd:mtsleep+0x138
0xbfc2dd24: netbsd:uvm_wait+0x7c
0xbfc2dd74: netbsd:uvm_km_kmem_alloc+0x270
0xbfc2dd8c: netbsd:pool_page_alloc+0x3c
0xbfc2ddcc: netbsd:pool_grow+0x38
0xbfc2de04: netbsd:pool_get+0x78
0xbfc2de4c: netbsd:pool_cache_get_slow+0x224
0xbfc2de84: netbsd:pool_cache_get_paddr+0x210
0xbfc2deac: netbsd:kmem_intr_alloc+0x68
0xbfc2df0c: netbsd:sys_posix_spawn+0x14c
0xbfc2df7c: netbsd:syscall+0x88
0xbfc2dfac: netbsd:swi_handler+0xa0
db{0}> mach cpu 1
kdb_trap: switching to cpu1
Stopped in pid 1386.1 (cron) at netbsd:pmap_pai_check.constprop.5+0x38: und e
1560003
db{1}> bt
0xbf773c34: netbsd:pmap_pai_check.constprop.5+0xc
0xbf773c84: netbsd:pmap_tlb_asid_acquire+0x6c
0xbf773ca4: netbsd:pmap_activate+0x7c
0xbf773cf4: netbsd:mi_switch+0x20c
0xbf773d24: netbsd:sleepq_block+0xb4
0xbf773d64: netbsd:mtsleep+0x138
0xbf773d84: netbsd:uvm_wait+0x7c
0xbf773db4: netbsd:pmap_create+0x94
0xbf773ddc: netbsd:uvmspace_init+0x6c
0xbf773e04: netbsd:uvmspace_alloc+0x44
0xbf773e34: netbsd:uvmspace_fork+0x2c
0xbf773e4c: netbsd:uvm_proc_fork+0x24
0xbf773ee4: netbsd:fork1+0x330
0xbf773f0c: netbsd:sys_fork+0x38
0xbf773f7c: netbsd:syscall+0x88
0xbf773fac: netbsd:swi_handler+0xa0
Martin
From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: port-arm/49364 (arm kernel panics on exec)
Date: Sun, 9 Nov 2014 12:05:37 +0100
This turns out to be PR kern/49175 again - the test complets (actually
fails, but that is a different issue) in 110 seconds (fast than I expected,
I really *like* this hardware) and then during cleanup wapbl strikes and
kills the machine for minutes.
Martin
State-Changed-From-To: feedback->closed
State-Changed-By: martin@NetBSD.org
State-Changed-When: Sun, 09 Nov 2014 11:10:59 +0000
State-Changed-Why:
Fixed
From: "Martin Husemann" <martin@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc:
Subject: PR/49364 CVS commit: [netbsd-7] src/sys/arch/arm
Date: Mon, 10 Nov 2014 19:57:27 +0000
Module Name: src
Committed By: martin
Date: Mon Nov 10 19:57:26 UTC 2014
Modified Files:
src/sys/arch/arm/arm [netbsd-7]: cpufunc_asm_armv7.S
src/sys/arch/arm/arm32 [netbsd-7]: bus_dma.c pmap.c
src/sys/arch/arm/pic [netbsd-7]: pic.c
Log Message:
Pull up following revision(s) (requested by skrll in ticket #209):
sys/arch/arm/pic/pic.c: revision 1.25
sys/arch/arm/arm/cpufunc_asm_armv7.S: revision 1.21
sys/arch/arm/arm32/pmap.c: revision 1.312
sys/arch/arm/arm32/bus_dma.c: revision 1.89
sys/arch/arm/arm32/pmap.c: revision 1.313
sys/arch/arm/arm32/pmap.c: revision 1.314
sys/arch/arm/arm32/pmap.c: revision 1.315
sys/arch/arm/arm32/pmap.c: revision 1.316
Include opt_multiprocessor.h
When allocing a l1page, if a page isn't available, use uvm_wait to wait
for one to become available. Should fix PR/49364.
Post a dmb before invalidating the cache in the post-{read,write}
operations to ensure that any/all cachelines brought in via speculation
are really flushed.
Ensure all memory operations are complete by before wfi. For example, the
cpu could have just been in uvm_pageidlezero.
In pmap_fault_fixup re-instate the TLB flush for the shared L1 case that
occurs for non-ARM_MMU_EXTENDED kernels.
This fixes rump/rumpkern/t_sp:stress_killer on rpi which is currently
non-ARM_MMU_EXTENDED
Remove an unnecessary flush that sneaked in as part of break-before-make
change.
Remove an XXXNH comment.
Update PTE_SYNC_CURRENT to add a dsb for armv7 - part of the
break-before-make fix.
To generate a diff of this commit:
cvs rdiff -u -r1.18.2.1 -r1.18.2.2 src/sys/arch/arm/arm/cpufunc_asm_armv7.S
cvs rdiff -u -r1.86 -r1.86.2.1 src/sys/arch/arm/arm32/bus_dma.c
cvs rdiff -u -r1.295.2.2 -r1.295.2.3 src/sys/arch/arm/arm32/pmap.c
cvs rdiff -u -r1.22.2.1 -r1.22.2.2 src/sys/arch/arm/pic/pic.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2014
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.
Home