NetBSD Problem Report #55610

From gson@gson.org  Thu Aug 27 11:07:30 2020
Return-Path: <gson@gson.org>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 7B0AD1A923A
	for <gnats-bugs@gnats.NetBSD.org>; Thu, 27 Aug 2020 11:07:30 +0000 (UTC)
Message-Id: <20200827110725.8EE06253F75@guava.gson.org>
Date: Thu, 27 Aug 2020 14:07:25 +0300 (EEST)
From: gson@gson.org (Andreas Gustafsson)
Reply-To: gson@gson.org (Andreas Gustafsson)
To: gnats-bugs@NetBSD.org
Subject: pmax testbed panics with "address error"
X-Send-Pr-Version: 3.95

>Number:         55610
>Category:       port-pmax
>Synopsis:       pmax testbed panics with "address error"
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    skrll
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Aug 27 11:10:00 +0000 2020
>Closed-Date:    Sun Oct 11 15:07:59 +0000 2020
>Last-Modified:  Sat Dec 12 21:10:01 +0000 2020
>Originator:     Andreas Gustafsson
>Release:        NetBSD-current, source date >= 2020.07.23.15.24.37
>Organization:
>Environment:
System: NetBSD
Architecture: mips
Machine: pmax
>Description:

>How-To-Repeat:

The TNF NetBSD/pmax testbed consistently panics during the
closed_slave test case of the t_ttypty test:

  kernel/kqueue/read/t_ttypty (68/875): 3 test cases
      closed_slave: [ 2970.8125030] pid 22491(t_ttypty): trap: cpu0, address error (load or I-fetch) in kernel mode
  [ 2970.8125030] status=0x8ff34, cause=0x10, epc=0x8016b0f8, vaddr=0xfe6
  [ 2970.8125030] tf=0x808a99f0 ksp=0x808a9a90 ra=0x8016b0d0 ppl=0
  [ 2970.8125030] kernel: address error (load or I-fetch) trap
  Stopped in pid 22491.22491 (t_ttypty) at	netbsd:ptyfs_get_node+0x90:
  lw	v0,8(s0)
  db>

The problem started with this commit:

  2020.07.23.12.15.59 skrll src/sys/arch/mips/include/mips_param.h 1.42

Logs:

  http://releng.netbsd.org/b5reports/pmax/commits-2020.07.html#2020.07.23.12.15.59

>Fix:

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: port-pmax-maintainer->skrll
Responsible-Changed-By: skrll@NetBSD.org
Responsible-Changed-When: Sat, 29 Aug 2020 08:02:47 +0000
Responsible-Changed-Why:
I'll take a look


From: Andreas Gustafsson <gson@gson.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: port-pmax/55610: pmax testbed panics with "address error"
Date: Sun, 11 Oct 2020 17:57:40 +0300

 This problem appears to be fixed; the pmax testbed now gets a bit
 further in the ATF test suite and panics differently as reported in
 port-pmax/55715.  It's not clear to me what fixed it.
 -- 
 Andreas Gustafsson, gson@gson.org

State-Changed-From-To: open->closed
State-Changed-By: gson@NetBSD.org
State-Changed-When: Sun, 11 Oct 2020 15:07:59 +0000
State-Changed-Why:
This panic no longer happens, instead it panics later as reported in port-pmax/55715.


From: Nick Hudson <nick.hudson@gmx.co.uk>
To: gnats-bugs@netbsd.org, port-pmax-maintainer@netbsd.org,
 netbsd-bugs@netbsd.org, gnats-admin@netbsd.org,
 Andreas Gustafsson <gson@gson.org>
Cc: 
Subject: Re: port-pmax/55610 (pmax testbed panics with "address error")
Date: Sat, 12 Dec 2020 21:08:34 +0000

 Likely caused by corruption fixed by

 Module Name:=C2=A0=C2=A0=C2=A0 src
 Committed By:=C2=A0=C2=A0=C2=A0 uwe
 Date:=C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 Sat Dec 12 18:35:59 UTC 2020

 Modified Files:
  =C2=A0=C2=A0=C2=A0 src/sys/kern: vfs_cache.c

 Log Message:
 vfs_cache - do not write outside allocated memory.

 In cache_cpu_init allocate struct nchcpu, not struct nchstats_percpu.
 cache_update_stats expects the former (that contains two percpu
 structs).=C2=A0 Stat updates where smashing whatever was in the next pool
 allocation.

 Reported by martin@ as crashes in ptyfs on Krups.


 To generate a diff of this commit:
 cvs rdiff -u -r1.147 -r1.148 src/sys/kern/vfs_cache.c

>Unformatted:

Home	PR Database Search