NetBSD Problem Report #49903

From www@NetBSD.org  Wed May 13 22:51:14 2015
Return-Path: <www@NetBSD.org>
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" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id B80A7A65CB
	for <gnats-bugs@gnats.NetBSD.org>; Wed, 13 May 2015 22:51:14 +0000 (UTC)
Message-Id: <20150513225113.B7AE4A65E1@mollari.NetBSD.org>
Date: Wed, 13 May 2015 22:51:13 +0000 (UTC)
From: sevan@NetBSD.org
Reply-To: sevan@NetBSD.org
To: gnats-bugs@NetBSD.org
Subject: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
X-Send-Pr-Version: www-1.0

>Number:         49903
>Category:       port-mips
>Synopsis:       Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    skrll
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed May 13 22:55:00 +0000 2015
>Closed-Date:    Sun Jul 10 13:39:53 +0000 2016
>Last-Modified:  Wed Nov 08 21:30:04 +0000 2017
>Originator:     venture37@geeklan.co.uk
>Release:        NetBSD 7.0_BETA
>Organization:
>Environment:
>Description:
During the installation as base.tgz is being uncompressed, the system panics with:
stopped in pid 43.1 (tar) at 801d5434

backtrace
0xc2c11ab0: 0+801d5434 (0,0,10,803f64b0) ra 0 sz 0
User-level: pid 43.1

ps
db_read_ptr: cannot find `allproc'


With previous snapshot, tar would generate malloc errors during untaring base.tgz.
malloc errors are also exhibited on NetBSD 6.1.5 too.
>How-To-Repeat:
Attempt installation of latest NetBSD/hpcmips snapshot
>Fix:

>Release-Note:

>Audit-Trail:

Responsible-Changed-From-To: install-manager->port-mips-maintainer
Responsible-Changed-By: martin@NetBSD.org
Responsible-Changed-When: Thu, 14 May 2015 05:14:09 +0000
Responsible-Changed-Why:
mips specific problem


From: Martin Husemann <martin@duskware.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
Date: Thu, 14 May 2015 07:14:16 +0200

 On Wed, May 13, 2015 at 10:55:00PM +0000, Sevan@NetBSD.org wrote:
 > During the installation as base.tgz is being uncompressed, the system panics with:
 > stopped in pid 43.1 (tar) at 801d5434

 What is the panic message itself?

 You should be able to get it from the db> prompt with the "dmesg" command.

 The earlier malloc errors are probably gone because sysinst now (if possible)
 enables swap early on low memory machines.

 Martin

From: Sevan Janiyan <sevan@NetBSD.org>
To: gnats-bugs@NetBSD.org, port-mips-maintainer@netbsd.org, 
 gnats-admin@netbsd.org, netbsd-bugs@netbsd.org, Janiyan@NetBSD.org
Cc: 
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips)
 whilst uncompressing base.tgz
Date: Thu, 14 May 2015 11:57:31 +0100

 On 14/05/2015 06:15, Martin Husemann wrote:
 > What is the panic message itself?
 > 
 > You should be able to get it from the db> prompt with the "dmesg" 
 > command.

 db> dmesg
 db_read_init: cannot find `msgbufenabled`

 > The earlier malloc errors are probably gone because sysinst now (if
 > possible) enables swap early on low memory machines.

 ah, I see.


 Sevan

From: Martin Husemann <martin@duskware.de>
To: Sevan Janiyan <sevan@NetBSD.org>
Cc: gnats-bugs@NetBSD.org
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
Date: Thu, 14 May 2015 13:13:51 +0200

 On Thu, May 14, 2015 at 11:57:31AM +0100, Sevan Janiyan wrote:
 > db> dmesg
 > db_read_init: cannot find `msgbufenabled`

 Hmm, not very helpfull - can you reproduce the issue by exiting sysinst
 and manually extracting base.tgz?

 Martin

From: David Brownlee <abs@absd.org>
To: Sevan Janiyan <sevan@netbsd.org>
Cc: "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>, port-mips-maintainer@netbsd.org, 
	"gnats-admin@netbsd.org" <gnats-admin@netbsd.org>, netbsd-bugs@netbsd.org, Janiyan@netbsd.org
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50
 (hpcmips) whilst uncompressing base.tgz
Date: Thu, 14 May 2015 13:06:07 +0100

 On 14 May 2015 at 11:57, Sevan Janiyan <sevan@netbsd.org> wrote:
 > On 14/05/2015 06:15, Martin Husemann wrote:
 >> What is the panic message itself?
 >>
 >> You should be able to get it from the db> prompt with the "dmesg"
 >> command.
 >
 > db> dmesg
 > db_read_init: cannot find `msgbufenabled`
 >
 >> The earlier malloc errors are probably gone because sysinst now (if
 >> possible) enables swap early on low memory machines.

 Though the swap is set to kick in at less than 32M of memory. I would
 think we could extend that to less than 64M if it helps this case...

From: Sevan Janiyan <sevan@NetBSD.org>
To: David Brownlee <abs@absd.org>
Cc: "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips)
 whilst uncompressing base.tgz
Date: Thu, 14 May 2015 13:54:50 +0100

 On 14/05/2015 13:06, David Brownlee wrote:
 > On 14 May 2015 at 11:57, Sevan Janiyan <sevan@netbsd.org> wrote:
 >> On 14/05/2015 06:15, Martin Husemann wrote:
 >>> What is the panic message itself?
 >>>
 >>> You should be able to get it from the db> prompt with the "dmesg"
 >>> command.
 >>
 >> db> dmesg
 >> db_read_init: cannot find `msgbufenabled`
 >>
 >>> The earlier malloc errors are probably gone because sysinst now (if
 >>> possible) enables swap early on low memory machines.
 > 
 > Though the swap is set to kick in at less than 32M of memory. I would
 > think we could extend that to less than 64M if it helps this case...
 > 

 That would definitely help as the IBM WorkPad Z50 only goes up to 48MB
 of RAM as standard
 (there are pads for what may be for additional RAM onboard and there's
 been talk on the hpcmips list of soldering some on there, to test it out
 but nothing yet :) )

From: Sevan Janiyan <sevan@NetBSD.org>
To: Martin Husemann <martin@duskware.de>
Cc: gnats-bugs@NetBSD.org
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips)
 whilst uncompressing base.tgz
Date: Fri, 15 May 2015 10:40:46 +0100

 On 14/05/2015 12:13, Martin Husemann wrote:
 > On Thu, May 14, 2015 at 11:57:31AM +0100, Sevan Janiyan wrote:
 >> db> dmesg
 >> db_read_init: cannot find `msgbufenabled`
 > 
 > Hmm, not very helpfull - can you reproduce the issue by exiting sysinst
 > and manually extracting base.tgz?
 > 
 > Martin

 Dropped out to /bin/sh from installer

 newfs /dev/wd0a

 mount_msdos /dev/wd0e /mnt
 I have all the installation files on a FAT partition on the same CF card

 mount /dev/wd0a /mnt2
 tar -zxvf mnt/sets/base.tgz -C /mnt2

 Panicked again

 pid 17(tar): trap: cpu0, TLB miss (load or instr. fetch) in kernel mode
 status=0xf003, cause=0x8, epc=0x801d5434, vaddr=0x8
 tf=0xc2c13998 ksp=0xc2c13a38 ra=0x801d5435 ppl=0xc0c01000
 kernel: TLB miss (load or instr. fetch) trap
 Stopped in pid 17.1 (tar) at 801d5434: lw v0,8(s0)

 db> bt
 0xc2c13a38: 0+801d5435 (0,0,10,803f64b0) ra 0 sz 0
 User-level: pid 17.1



 Sevan

From: Sevan Janiyan <sevan@NetBSD.org>
To: David Brownlee <abs@absd.org>
Cc: "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips)
 whilst uncompressing base.tgz
Date: Fri, 15 May 2015 10:58:24 +0100

 On 14/05/2015 13:06, David Brownlee wrote:
 > On 14 May 2015 at 11:57, Sevan Janiyan <sevan@netbsd.org> wrote:
 >> On 14/05/2015 06:15, Martin Husemann wrote:
 >>> What is the panic message itself?
 >>>
 >>> You should be able to get it from the db> prompt with the "dmesg"
 >>> command.
 >>
 >> db> dmesg
 >> db_read_init: cannot find `msgbufenabled`
 >>
 >>> The earlier malloc errors are probably gone because sysinst now (if
 >>> possible) enables swap early on low memory machines.
 > 
 > Though the swap is set to kick in at less than 32M of memory. I would
 > think we could extend that to less than 64M if it helps this case...

 As I have a swap partition allocated on the CF card, I thought I'd try
 to active it before starting with the manual uncompression of base.tgz.
 I couldn't find a copy of swapon, is the swap activated by some other
 means or is it just not included in the hpcmips minirootfs?


 Sevan

From: David Brownlee <abs@absd.org>
To: Sevan Janiyan <sevan@netbsd.org>
Cc: "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50
 (hpcmips) whilst uncompressing base.tgz
Date: Fri, 15 May 2015 11:51:01 +0100

 On 15 May 2015 at 10:58, Sevan Janiyan <sevan@netbsd.org> wrote:
 > On 14/05/2015 13:06, David Brownlee wrote:
 >> Though the swap is set to kick in at less than 32M of memory. I would
 >> think we could extend that to less than 64M if it helps this case...
 >
 > As I have a swap partition allocated on the CF card, I thought I'd try
 > to active it before starting with the manual uncompression of base.tgz.
 > I couldn't find a copy of swapon, is the swap activated by some other
 > means or is it just not included in the hpcmips minirootfs?

 Look for swapctl (which is the more general purpose cousin for
 swapon). Installer space being limited you only get the one :)

 I strongly suspect there is an underlying kernel or toolchain bug
 here, rather than out of memory, but its definitely worth a quick
 check

 Thanks

From: Sevan Janiyan <sevan@NetBSD.org>
To: David Brownlee <abs@absd.org>
Cc: "gnats-bugs@netbsd.org" <gnats-bugs@netbsd.org>
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips)
 whilst uncompressing base.tgz
Date: Fri, 15 May 2015 12:38:44 +0100

 On 15/05/2015 11:51, David Brownlee wrote:
 > Look for swapctl (which is the more general purpose cousin for
 > swapon). Installer space being limited you only get the one :)

 swapctl: not found
 Checked /bin, /sbin, /usr/bin and /usr/sbin just to be sure.

 Going back to the memory size for which swap is activated, I should also
 mention that though this system has 48MB RAM, it is common to find these
 WorkPads with 16MB of RAM as the 32MB expansion is rare.


 Sevan

From: Sevan Janiyan <sevan@NetBSD.org>
To: gnats-bugs@NetBSD.org
Cc: David Brownlee <abs@absd.org>
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50
 (hpcmips) whilst uncompressing base.tgz
Date: Mon, 3 Aug 2015 16:32:34 +0100

 On 15/05/2015 11:55, David Brownlee wrote:
 >  On 15 May 2015 at 10:58, Sevan Janiyan <sevan@netbsd.org> wrote:
 >  > On 14/05/2015 13:06, David Brownlee wrote:
 >  >> Though the swap is set to kick in at less than 32M of memory. I would
 >  >> think we could extend that to less than 64M if it helps this case...
 >  >
 >  > As I have a swap partition allocated on the CF card, I thought I'd try
 >  > to active it before starting with the manual uncompression of base.tgz.
 >  > I couldn't find a copy of swapon, is the swap activated by some other
 >  > means or is it just not included in the hpcmips minirootfs?
 >  
 >  Look for swapctl (which is the more general purpose cousin for
 >  swapon). Installer space being limited you only get the one :)
 >  
 >  I strongly suspect there is an underlying kernel or toolchain bug
 >  here, rather than out of memory, but its definitely worth a quick
 >  check


 I bumped the value compared against from running get_ramsize() to 64 &
 rolled a new release. I took the generic kernel * miniroot.fs from this
 & attempted to install 7.0-RC1 again. Same thing, panic on extracting
 the base set.
 http://nxr.netbsd.org/xref/src/usr.sbin/sysinst/disks.c?r=1.9#1203


 Sevan

From: Nick Hudson <skrll@netbsd.org>
To: gnats-bugs@NetBSD.org, port-mips-maintainer@netbsd.org,
 gnats-admin@netbsd.org, netbsd-bugs@netbsd.org, Sevan@NetBSD.org,
 Janiyan@NetBSD.org
Cc: 
Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50
 (hpcmips) whilst uncompressing base.tgz
Date: Sat, 25 Jun 2016 16:46:56 +0100

 On 05/15/15 10:45, Sevan Janiyan wrote:
 > The following reply was made to PR port-mips/49903; it has been noted by GNATS.
 >
 > From: Sevan Janiyan <sevan@NetBSD.org>
 > To: Martin Husemann <martin@duskware.de>
 > Cc: gnats-bugs@NetBSD.org
 > Subject: Re: port-mips/49903: Panic during installation on WorkPad Z50 (hpcmips)
 >   whilst uncompressing base.tgz
 > Date: Fri, 15 May 2015 10:40:46 +0100
 >
 >   On 14/05/2015 12:13, Martin Husemann wrote:
 >   > On Thu, May 14, 2015 at 11:57:31AM +0100, Sevan Janiyan wrote:
 >   >> db> dmesg
 >   >> db_read_init: cannot find `msgbufenabled`
 >   >
 >   > Hmm, not very helpfull - can you reproduce the issue by exiting sysinst
 >   > and manually extracting base.tgz?
 >   >
 >   > Martin
 >   
 >   Dropped out to /bin/sh from installer
 >   
 >   newfs /dev/wd0a
 >   
 >   mount_msdos /dev/wd0e /mnt
 >   I have all the installation files on a FAT partition on the same CF card
 >   
 >   mount /dev/wd0a /mnt2
 >   tar -zxvf mnt/sets/base.tgz -C /mnt2
 >   
 >   Panicked again
 >   
 >   pid 17(tar): trap: cpu0, TLB miss (load or instr. fetch) in kernel mode
 >   status=0xf003, cause=0x8, epc=0x801d5434, vaddr=0x8
 >   tf=0xc2c13998 ksp=0xc2c13a38 ra=0x801d5435 ppl=0xc0c01000
 >   kernel: TLB miss (load or instr. fetch) trap
 >   Stopped in pid 17.1 (tar) at 801d5434: lw v0,8(s0)
 >   
 >   db> bt
 >   0xc2c13a38: 0+801d5435 (0,0,10,803f64b0) ra 0 sz 0
 >   User-level: pid 17.1

 OK, I reproduced this with gxemul.

 NIck

From: "Nick Hudson" <skrll@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49903 CVS commit: src/sys/arch/mips/mips
Date: Thu, 30 Jun 2016 12:57:35 +0000

 Module Name:	src
 Committed By:	skrll
 Date:		Thu Jun 30 12:57:35 UTC 2016

 Modified Files:
 	src/sys/arch/mips/mips: pmap.c

 Log Message:
 Fix MIPS3_NO_PV_UNCACHED alias handling by looping through the pv_list
 looking for bad aliases and removing the bad entries.  That is, revert
 to the code before the matt-mips64 merge.

 Additionally, fix the pmap_update call to not use the (recently
  removed/freed) pv for the pmap_t.

 Fixes the following two PRs

 PR/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
 PR/51226: Install bug for hpcmips NetBSD V7 using FTP Full installation


 To generate a diff of this commit:
 cvs rdiff -u -r1.222 -r1.223 src/sys/arch/mips/mips/pmap.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

Responsible-Changed-From-To: port-mips-maintainer->skrll
Responsible-Changed-By: skrll@NetBSD.org
Responsible-Changed-When: Thu, 30 Jun 2016 13:44:37 +0000
Responsible-Changed-Why:
I fixed it


State-Changed-From-To: open->pending-pullups
State-Changed-By: skrll@NetBSD.org
State-Changed-When: Thu, 30 Jun 2016 13:44:37 +0000
State-Changed-Why:
[pullup-7 #1200] mips pmap fixes


From: "Martin Husemann" <martin@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49903 CVS commit: [netbsd-7-0] src/sys/arch/mips/mips
Date: Sun, 10 Jul 2016 08:44:07 +0000

 Module Name:	src
 Committed By:	martin
 Date:		Sun Jul 10 08:44:07 UTC 2016

 Modified Files:
 	src/sys/arch/mips/mips [netbsd-7-0]: pmap.c

 Log Message:
 Pull up following revision(s) (requested by skrll in ticket #1200):
 	sys/arch/mips/mips/pmap.c: revision 1.221
 	sys/arch/mips/mips/pmap.c: revision 1.222
 	sys/arch/mips/mips/pmap.c: revision 1.223
 Fix a bug introduced by me in 1.214 where unmanaged mappings would be
 affected by calls to pmap_page_protect which is wrong.  Now PV_KENTER
 mappings are left intact.
 Thanks to chuq for spotting my mistake and reviewing this diff.
 Thanks to everyone who tested it as well.
 Fix PR/51288 reproducable panic on evbmips64-eb (erlite)
 pmap_page_remove from the previous change neglected to terminate the pv
 list correctly when it started with an initial unmanaged mapping and
 subsequent managed mappings.  Fix this.
 Fix MIPS3_NO_PV_UNCACHED alias handling by looping through the pv_list
 looking for bad aliases and removing the bad entries.  That is, revert
 to the code before the matt-mips64 merge.
 Additionally, fix the pmap_update call to not use the (recently
   removed/freed) pv for the pmap_t.
 Fixes the following two PRs
 PR/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
 PR/51226: Install bug for hpcmips NetBSD V7 using FTP Full installation


 To generate a diff of this commit:
 cvs rdiff -u -r1.214 -r1.214.6.1 src/sys/arch/mips/mips/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@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49903 CVS commit: [netbsd-7] src/sys/arch/mips/mips
Date: Sun, 10 Jul 2016 10:33:42 +0000

 Module Name:	src
 Committed By:	martin
 Date:		Sun Jul 10 10:33:42 UTC 2016

 Modified Files:
 	src/sys/arch/mips/mips [netbsd-7]: pmap.c

 Log Message:
 Pull up following revision(s) (requested by skrll in ticket #1200):
 	sys/arch/mips/mips/pmap.c: revision 1.221
 	sys/arch/mips/mips/pmap.c: revision 1.222
 	sys/arch/mips/mips/pmap.c: revision 1.223
 Fix a bug introduced by me in 1.214 where unmanaged mappings would be
 affected by calls to pmap_page_protect which is wrong.  Now PV_KENTER
 mappings are left intact.
 Thanks to chuq for spotting my mistake and reviewing this diff.
 Thanks to everyone who tested it as well.
 Fix PR/51288 reproducable panic on evbmips64-eb (erlite)
 pmap_page_remove from the previous change neglected to terminate the pv
 list correctly when it started with an initial unmanaged mapping and
 subsequent managed mappings.  Fix this.
 Fix MIPS3_NO_PV_UNCACHED alias handling by looping through the pv_list
 looking for bad aliases and removing the bad entries.  That is, revert
 to the code before the matt-mips64 merge.
 Additionally, fix the pmap_update call to not use the (recently
   removed/freed) pv for the pmap_t.
 Fixes the following two PRs
 PR/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
 PR/51226: Install bug for hpcmips NetBSD V7 using FTP Full installation


 To generate a diff of this commit:
 cvs rdiff -u -r1.214 -r1.214.2.1 src/sys/arch/mips/mips/pmap.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

State-Changed-From-To: pending-pullups->closed
State-Changed-By: skrll@NetBSD.org
State-Changed-When: Sun, 10 Jul 2016 13:39:53 +0000
State-Changed-Why:
pullups done


From: "Manuel Bouyer" <bouyer@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49903 CVS commit: [netbsd-6] src/sys/arch/mips/mips
Date: Sat, 27 Aug 2016 14:34:55 +0000

 Module Name:	src
 Committed By:	bouyer
 Date:		Sat Aug 27 14:34:55 UTC 2016

 Modified Files:
 	src/sys/arch/mips/mips [netbsd-6]: pmap.c

 Log Message:
 Pull up following revision(s) (requested by skrll in ticket #1390):
 	sys/arch/mips/mips/pmap.c: revision 1.221
 	sys/arch/mips/mips/pmap.c: revision 1.222
 	sys/arch/mips/mips/pmap.c: revision 1.223
 Fix a bug introduced by me in 1.214 where unmanaged mappings would be
 affected by calls to pmap_page_protect which is wrong.  Now PV_KENTER
 mappings are left intact.
 Thanks to chuq for spotting my mistake and reviewing this diff.
 Thanks to everyone who tested it as well.
 Fix PR/51288 reproducable panic on evbmips64-eb (erlite)
 pmap_page_remove from the previous change neglected to terminate the pv
 list correctly when it started with an initial unmanaged mapping and
 subsequent managed mappings.  Fix this.
 Fix MIPS3_NO_PV_UNCACHED alias handling by looping through the pv_list
 looking for bad aliases and removing the bad entries.  That is, revert
 to the code before the matt-mips64 merge.
 Additionally, fix the pmap_update call to not use the (recently
   removed/freed) pv for the pmap_t.
 Fixes the following two PRs
 PR/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
 PR/51226: Install bug for hpcmips NetBSD V7 using FTP Full installation


 To generate a diff of this commit:
 cvs rdiff -u -r1.207.2.3 -r1.207.2.4 src/sys/arch/mips/mips/pmap.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

From: "Soren Jacobsen" <snj@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49903 CVS commit: [netbsd-6-0] src/sys/arch/mips/mips
Date: Wed, 8 Nov 2017 21:28:18 +0000

 Module Name:	src
 Committed By:	snj
 Date:		Wed Nov  8 21:28:18 UTC 2017

 Modified Files:
 	src/sys/arch/mips/mips [netbsd-6-0]: pmap.c

 Log Message:
 Pull up following revision(s) (requested by skrll in ticket #1390):
 	sys/arch/mips/mips/pmap.c: 1.221-1.223
 Fix a bug introduced by me in 1.214 where unmanaged mappings would be
 affected by calls to pmap_page_protect which is wrong.  Now PV_KENTER
 mappings are left intact.
 Thanks to chuq for spotting my mistake and reviewing this diff.
 Thanks to everyone who tested it as well.
 Fix PR/51288 reproducable panic on evbmips64-eb (erlite)
 pmap_page_remove from the previous change neglected to terminate the pv
 list correctly when it started with an initial unmanaged mapping and
 subsequent managed mappings.  Fix this.
 Fix MIPS3_NO_PV_UNCACHED alias handling by looping through the pv_list
 looking for bad aliases and removing the bad entries.  That is, revert
 to the code before the matt-mips64 merge.
 Additionally, fix the pmap_update call to not use the (recently
   removed/freed) pv for the pmap_t.
 Fixes the following two PRs
 PR/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
 PR/51226: Install bug for hpcmips NetBSD V7 using FTP Full installation


 To generate a diff of this commit:
 cvs rdiff -u -r1.207.2.1.4.2 -r1.207.2.1.4.3 src/sys/arch/mips/mips/pmap.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

From: "Soren Jacobsen" <snj@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49903 CVS commit: [netbsd-6-1] src/sys/arch/mips/mips
Date: Wed, 8 Nov 2017 21:28:24 +0000

 Module Name:	src
 Committed By:	snj
 Date:		Wed Nov  8 21:28:24 UTC 2017

 Modified Files:
 	src/sys/arch/mips/mips [netbsd-6-1]: pmap.c

 Log Message:
 Pull up following revision(s) (requested by skrll in ticket #1390):
 	sys/arch/mips/mips/pmap.c: 1.221-1.223
 Fix a bug introduced by me in 1.214 where unmanaged mappings would be
 affected by calls to pmap_page_protect which is wrong.  Now PV_KENTER
 mappings are left intact.
 Thanks to chuq for spotting my mistake and reviewing this diff.
 Thanks to everyone who tested it as well.
 Fix PR/51288 reproducable panic on evbmips64-eb (erlite)
 pmap_page_remove from the previous change neglected to terminate the pv
 list correctly when it started with an initial unmanaged mapping and
 subsequent managed mappings.  Fix this.
 Fix MIPS3_NO_PV_UNCACHED alias handling by looping through the pv_list
 looking for bad aliases and removing the bad entries.  That is, revert
 to the code before the matt-mips64 merge.
 Additionally, fix the pmap_update call to not use the (recently
   removed/freed) pv for the pmap_t.
 Fixes the following two PRs
 PR/49903: Panic during installation on WorkPad Z50 (hpcmips) whilst uncompressing base.tgz
 PR/51226: Install bug for hpcmips NetBSD V7 using FTP Full installation


 To generate a diff of this commit:
 cvs rdiff -u -r1.207.2.1.6.2 -r1.207.2.1.6.3 src/sys/arch/mips/mips/pmap.c

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

>Unformatted:

Home	PR Database Search