NetBSD Problem Report #47935

From rjs@fdy2.co.uk  Sun Jun 16 11:30:51 2013
Return-Path: <rjs@fdy2.co.uk>
Received: from mail.netbsd.org (mail.netbsd.org [149.20.53.66])
	(using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "Postmaster NetBSD.org" (verified OK))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 6476E70B37
	for <gnats-bugs@gnats.NetBSD.org>; Sun, 16 Jun 2013 11:30:51 +0000 (UTC)
Message-Id: <20130616112714.AC6EC12306@ren.fdy2.co.uk>
Date: Sun, 16 Jun 2013 12:27:14 +0100 (BST)
From: rjs@fdy2.co.uk
Reply-To: rjs@fdy2.co.uk
To: gnats-bugs@NetBSD.org
Subject: x11/xf86-video-ati doesn't work
X-Send-Pr-Version: 3.95

>Number:         47935
>Category:       pkg
>Synopsis:       x11/xf86-video-ati doesn't work
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    pkg-manager
>State:          closed
>Class:          change-request
>Submitter-Id:   net
>Arrival-Date:   Sun Jun 16 11:35:00 +0000 2013
>Closed-Date:    Wed Jul 17 08:39:24 +0000 2013
>Last-Modified:  Tue Sep 03 05:55:00 +0000 2013
>Originator:     Robert Swindells
>Release:        Any
>Organization:
NetBSD
>Environment:
All except Linux
>Description:
The versions of xf86-video-ati above 7.0 will only work when the kernel
DRM code implements KMS, all non-KMS support has been removed.
>How-To-Repeat:
Run modular x11 with an AMD/ATI graphics card on anything other than
Linux.
>Fix:
Downgrade the x11/xf86-video-ati package to 6.14.6 or mark the existing
one as only for Linux and create a new package that builds 6.14.6.

>Release-Note:

>Audit-Trail:

State-Changed-From-To: open->feedback
State-Changed-By: wiz@NetBSD.org
State-Changed-When: Tue, 18 Jun 2013 08:53:51 +0000
State-Changed-Why:
Please try xf86-video-ati6 and report back if it works.


From: "Thomas Klausner" <wiz@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Tue, 18 Jun 2013 08:53:22 +0000

 Module Name:	pkgsrc
 Committed By:	wiz
 Date:		Tue Jun 18 08:53:22 UTC 2013

 Added Files:
 	pkgsrc/x11/xf86-video-ati6: DESCR Makefile PLIST distinfo

 Log Message:
 Import xf86-video-ati-6.14.5 as x11/xf86-video-ati6.

 This is the video driver for ATI Radeon cards for the modular Xorg
 server.

 This package contains the major version 6 of the driver, before
 DRI1/UMS/etc removal.

 This package is added in a try to address PR 47935.
 If it doesn't help, I'll remove it again.


 To generate a diff of this commit:
 cvs rdiff -u -r0 -r1.1 pkgsrc/x11/xf86-video-ati6/DESCR \
     pkgsrc/x11/xf86-video-ati6/Makefile pkgsrc/x11/xf86-video-ati6/PLIST \
     pkgsrc/x11/xf86-video-ati6/distinfo

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

From: "Thomas Klausner" <wiz@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Tue, 18 Jun 2013 17:13:11 +0000

 Module Name:	pkgsrc
 Committed By:	wiz
 Date:		Tue Jun 18 17:13:11 UTC 2013

 Modified Files:
 	pkgsrc/x11/xf86-video-ati6: Makefile distinfo
 Added Files:
 	pkgsrc/x11/xf86-video-ati6: options.mk

 Log Message:
 Update to 6.14.6 and make KMS a default-off option.
 Still for PR 47935.


 To generate a diff of this commit:
 cvs rdiff -u -r1.1 -r1.2 pkgsrc/x11/xf86-video-ati6/Makefile \
     pkgsrc/x11/xf86-video-ati6/distinfo
 cvs rdiff -u -r0 -r1.1 pkgsrc/x11/xf86-video-ati6/options.mk

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

From: Tobias Nygren <tnn@NetBSD.org>
To: gnats-bugs@NetBSD.org
Cc: Thomas Klausner <wiz@netbsd.org>
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Tue, 25 Jun 2013 21:15:32 +0200

 On Tue, 18 Jun 2013 08:55:01 +0000 (UTC)
 Thomas Klausner <wiz@netbsd.org> wrote:

 >  This package is added in a try to address PR 47935.
 >  If it doesn't help, I'll remove it again.

 Hi,

 I'm using the -ati6 package on sparc64 with the updated X11 server and
 it works whereas the -ati package doesn't so please keep this legacy
 package around. The X server and Mesalib needed some separate patches
 to work, but surprisingly few changes were required. Will send them in
 a separate PR.

 All the best,
 -Tobias

State-Changed-From-To: feedback->closed
State-Changed-By: rjs@NetBSD.org
State-Changed-When: Wed, 17 Jul 2013 08:39:24 +0000
State-Changed-Why:
Problem is reported to be fixed.


From: "Kai-Uwe Eckhardt" <kuehro@gmx.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Sun, 18 Aug 2013 17:50:00 +0200 (CEST)

 Hi, 

 on 6.99.23 i386 with a 9250 RV280 card (AGP) the Xserver runs and
 I can start xterm from .xinitrc but the mouse moves only in
 horizontal direction (PS2). After reinstalling the system with native
 Xorg everything works. The same problem was reported in this mail:
 http://mail-index.netbsd.org/pkgsrc-users/2013/06/27/msg018240.html

 I didn't have time to analyse the problem and it might be unrelated
 to xf86-video-ati6, but for me modular xorg was not usable with
 video-ati6. This bug is closed and I would leave it that way
 but I wanted the problem to be documented here.

From: christos@zoulas.com (Christos Zoulas)
To: gnats-bugs@NetBSD.org, pkg-manager@netbsd.org, gnats-admin@netbsd.org, 
	pkgsrc-bugs@netbsd.org, rjs@fdy2.co.uk
Cc: 
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Sun, 18 Aug 2013 12:17:18 -0400

 On Aug 18,  4:00pm, kuehro@gmx.de ("Kai-Uwe Eckhardt") wrote:
 -- Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6

 | The following reply was made to PR pkg/47935; it has been noted by GNATS.
 | 
 | From: "Kai-Uwe Eckhardt" <kuehro@gmx.de>
 | To: gnats-bugs@NetBSD.org
 | Cc: 
 | Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
 | Date: Sun, 18 Aug 2013 17:50:00 +0200 (CEST)
 | 
 |  Hi, 
 |  
 |  on 6.99.23 i386 with a 9250 RV280 card (AGP) the Xserver runs and
 |  I can start xterm from .xinitrc but the mouse moves only in
 |  horizontal direction (PS2). After reinstalling the system with native
 |  Xorg everything works. The same problem was reported in this mail:
 |  http://mail-index.netbsd.org/pkgsrc-users/2013/06/27/msg018240.html
 |  
 |  I didn't have time to analyse the problem and it might be unrelated
 |  to xf86-video-ati6, but for me modular xorg was not usable with

 This happens when the mouse code does not send an ioctl to change
 to the new protocol (that uses a time_t that is 64 bits), so the xserver
 gets confused about the mouse struct format it receives.

 christos

From: Thomas Klausner <wiz@NetBSD.org>
To: Christos Zoulas <christos@zoulas.com>
Cc: NetBSD bugtracking <gnats-bugs@NetBSD.org>
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Sun, 18 Aug 2013 23:32:48 +0200

 On Sun, Aug 18, 2013 at 12:17:18PM -0400, Christos Zoulas wrote:
 > This happens when the mouse code does not send an ioctl to change
 > to the new protocol (that uses a time_t that is 64 bits), so the xserver
 > gets confused about the mouse struct format it receives.

 So we need an update for input-mouse? Do you remember which part of the code handles this?
  Thomas

From: "OBATA Akio" <obache@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Mon, 19 Aug 2013 10:26:21 +0900

 On Mon, 19 Aug 2013 06:35:01 +0900, Thomas Klausner <wiz@netbsd.org> wrote:

 >  So we need an update for input-mouse? Do you remember which part of the code handles this?

 http://mail-index.netbsd.org/tech-pkg/2013/06/13/msg011553.html

 Just patched to NetBSD in-tree xsrc, not sent to upstream,
 some patches had been pulled in pkgsrc formerly, but lost with last pkgsrc update,
 because upstream rewrote codes, and cannot be applied anymore.

 -- 
 OBATA Akio / obache@NetBSD.org

From: christos@zoulas.com (Christos Zoulas)
To: Thomas Klausner <wiz@NetBSD.org>
Cc: NetBSD bugtracking <gnats-bugs@NetBSD.org>
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Mon, 19 Aug 2013 01:38:25 -0400

 On Aug 18, 11:32pm, wiz@NetBSD.org (Thomas Klausner) wrote:
 -- Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6

 | On Sun, Aug 18, 2013 at 12:17:18PM -0400, Christos Zoulas wrote:
 | > This happens when the mouse code does not send an ioctl to change
 | > to the new protocol (that uses a time_t that is 64 bits), so the xserver
 | > gets confused about the mouse struct format it receives.
 | 
 | So we need an update for input-mouse? Do you remember which part of the code handles this?

 You are looking for:

 ./mouse.c:#if defined(__NetBSD__) && defined(WSCONS_SUPPORT) && defined(WSMOUSEIO_SETVERSION)
 ./mouse.c:           if (ioctl(pInfo->fd, WSMOUSEIO_SETVERSION, &version) == -1)
 ...

 christos

From: Thomas Klausner <wiz@NetBSD.org>
To: NetBSD bugtracking <gnats-bugs@NetBSD.org>
Cc: Christos Zoulas <christos@zoulas.com>
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Mon, 19 Aug 2013 11:20:45 +0200

 On Mon, Aug 19, 2013 at 05:40:01AM +0000, Christos Zoulas wrote:
 >  | So we need an update for input-mouse? Do you remember which part of the code handles this?
 >  
 >  You are looking for:
 >  
 >  ./mouse.c:#if defined(__NetBSD__) && defined(WSCONS_SUPPORT) && defined(WSMOUSEIO_SETVERSION)
 >  ./mouse.c:           if (ioctl(pInfo->fd, WSMOUSEIO_SETVERSION, &version) == -1)
 >  ...

 Thanks. I always wondered what these were for.

 I've added all NetBSD xsrc patches to xf86-input-mouse, please try again.
  Thomas

From: Thomas Klausner <wiz@NetBSD.org>
To: NetBSD bugtracking <gnats-bugs@NetBSD.org>
Cc: OBATA Akio <obache@NetBSD.org>
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Mon, 19 Aug 2013 11:22:03 +0200

 On Mon, Aug 19, 2013 at 01:30:01AM +0000, OBATA Akio wrote:
 >  Just patched to NetBSD in-tree xsrc, not sent to upstream,
 >  some patches had been pulled in pkgsrc formerly, but lost with last pkgsrc update,
 >  because upstream rewrote codes, and cannot be applied anymore.

 I've been sending our NetBSD-local xsrc changes upstream for some
 months now, and much has been accepted already.

 I've just sent the mouse diffs upstream.
  Thomas

From: "Kai-Uwe Eckhardt" <kuehro@gmx.de>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Fri, 23 Aug 2013 09:38:37 +0200 (CEST)

 The mouse works now. I reinstalled the system without native X11 and have built
 meta-pkgs/modular-xorg and xf86-video-ati6. The mouse works fine and
 the system is stable. There is still a small problem left. After exiting
 from fluxbox I get a coredump of the Xserver with this message:

 Server termninated successfully (0).

 Segmentation fault at address 0xb2817f31.




From: Jan Danielsson <jan.m.danielsson@gmail.com>
To: gnats-bugs@NetBSD.org
Cc: Kai-Uwe Eckhardt <kuehro@gmx.de>, pkg-manager@netbsd.org, 
 gnats-admin@netbsd.org, pkgsrc-bugs@netbsd.org, rjs@fdy2.co.uk
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Fri, 23 Aug 2013 10:20:31 +0200

 On 8/23/13 9:40 AM, Kai-Uwe Eckhardt wrote:
 [---]
 >  The mouse works now. I reinstalled the system without native X11 and have built
 >  meta-pkgs/modular-xorg and xf86-video-ati6. The mouse works fine and
 >  the system is stable. There is still a small problem left. After exiting
 >  from fluxbox I get a coredump of the Xserver with this message:
 >  
 >  Server termninated successfully (0).
 >  
 >  Segmentation fault at address 0xb2817f31.

    I am seeing this as well. (Well, another address; I'm using amd64,
 but the segfault happens there as well).

    /Jan

From: David Holland <dholland-pbugs@netbsd.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: PR/47935 CVS commit: pkgsrc/x11/xf86-video-ati6
Date: Tue, 3 Sep 2013 05:52:54 +0000

 On Fri, Aug 23, 2013 at 08:25:00AM +0000, Jan Danielsson wrote:
  >  >  There is still a small problem left. After exiting
  >  >  from fluxbox I get a coredump of the Xserver with this message:
  >  >  
  >  >  Server termninated successfully (0).
  >  >  
  >  >  Segmentation fault at address 0xb2817f31.
  >  
  >     I am seeing this as well. (Well, another address; I'm using amd64,
  >  but the segfault happens there as well).

 Can one or both of you file a separate PR on this? (if you haven't
 already, I'm too far behind on gnats to know)

 -- 
 David A. Holland
 dholland@netbsd.org

>Unformatted:

Home	PR Database Search