NONAKA Kimihiro | 4 Mar 08:03 2009
Picon

handling absolute X, Y position event in bsd_mouse.c

Hi,

Zaurus and hpc* port have a touch panel. As for them, the pointer
position is absolutely reported.
Currently It is never handled in
xorg-server/dist/hw/xfree86/os-support/bsd/bsd_mouse.c.
It came to be able to handle the pointer position of addressing absolutely
with the appended patch.
It referred to xorg-server/dist/hw/xfree86/os-support/solaris/sun_mouse.c
in making this patch.

I'll commit the patch if there is no objection.

Regards,
--

-- 
NONAKA Kimihiro
email: nonaka <at> netbsd.org
Attachment (xorg-bsd_mouse.c.diff): application/octet-stream, 3505 bytes
Brook Milligan | 4 Mar 15:47 2009

qgis triggers X11 errors

I am working on a package (qgis; http://www.qgis.org/) and am
encountering runtime errors related to X11 on the native X11 with
netbsd 5.0 but not on the modular Xorg with netbsd 4.0.  On the latter
platform, which has the following packages installed,

     modular-xorg-apps-1.5 Modular Xorg application meta-package
     modular-xorg-server-1.4.2nb3 Modular X11 server from modular X.org
     modular-xorg-fonts-1.0 Modular Xorg font meta-package
     modular-xorg-drivers-1.3 Modular Xorg driver meta-package
     modular-xorg-libs-1.0nb1 Modular Xorg library meta-package

qgis runs fine with no console errors.  On the former I immediately
get the following errors.  Unless I quit immediately, it will also
hang the system.

     Warning: X Error: BadDrawable (invalid Pixmap or Window parameter) 9
       Major opcode: 55 (X_CreateGC)
       Resource id:  0x0
     Warning: X Error: BadGC (invalid GC parameter) 13
       Major opcode: 56 (X_ChangeGC)
       Resource id:  0x1e001d3
     Warning: X Error: BadDrawable (invalid Pixmap or Window parameter) 9
       Major opcode: 70 (X_PolyFillRectangle)
       Resource id:  0x0
     Warning: X Error: BadGC (invalid GC parameter) 13
       Major opcode: 60 (X_FreeGC)
       Resource id:  0x1e001d3
     Warning: X Error: BadDrawable (invalid Pixmap or Window parameter) 9
       Major opcode: 55 (X_CreateGC)
       Resource id:  0x0
(Continue reading)

Iain Hibbert | 4 Mar 16:56 2009
Picon

Re: handling absolute X, Y position event in bsd_mouse.c

On Wed, 4 Mar 2009, NONAKA Kimihiro wrote:

> Zaurus and hpc* port have a touch panel. As for them, the pointer
> position is absolutely reported.
> Currently It is never handled in
> xorg-server/dist/hw/xfree86/os-support/bsd/bsd_mouse.c.
> It came to be able to handle the pointer position of addressing absolutely
> with the appended patch.
> It referred to xorg-server/dist/hw/xfree86/os-support/solaris/sun_mouse.c
> in making this patch.

did you check the bsd_mouse.c from our XFree86 also? I don't know if
anything there is significant but I had to add the W direction recently
and I noticed that the absolute pointer had not been carried through to
xorg. (there was something about ABSOLUTE_Z and ABSOLUTE_W there but I'm
not sure if they make sense?)

Its probably worth pushing the addition upstream though I made a ticket
about it (http://bugs.freedesktop.org/show_bug.cgi?id=8993) but it has not
been touched.  Do we have any insiders at X.org?

iain

NONAKA Kimihiro | 4 Mar 18:29 2009
Picon

Re: handling absolute X, Y position event in bsd_mouse.c

Hi,

2009/3/5 Iain Hibbert <plunky <at> rya-online.net>:

> did you check the bsd_mouse.c from our XFree86 also? I don't know if
> anything there is significant but I had to add the W direction recently
> and I noticed that the absolute pointer had not been carried through to
> xorg. (there was something about ABSOLUTE_Z and ABSOLUTE_W there but I'm
> not sure if they make sense?)

I see our XFree86 source's bsd_mouse.c. All necessary things for me
were written there.
However, I do not know what it is of ABSOLUTE_Z and ABSOLUTE_W.
I have only the device that reports on ABSOLUTE_X and ABSOLUTE_Y.

Regards,
--

-- 
NONAKA Kimihiro
Index: external/mit/xorg-server/dist/hw/xfree86/os-support/bsd/bsd_mouse.c
===================================================================
RCS file: /cvsroot/xsrc/external/mit/xorg-server/dist/hw/xfree86/os-support/bsd/bsd_mouse.c,v
retrieving revision 1.5
diff -u -r1.5 bsd_mouse.c
--- external/mit/xorg-server/dist/hw/xfree86/os-support/bsd/bsd_mouse.c	9 Feb 2009 09:49:39 -0000	1.5
+++ external/mit/xorg-server/dist/hw/xfree86/os-support/bsd/bsd_mouse.c	4 Mar 2009 17:12:06 -0000
 <at>  <at>  -413,7 +413,7  <at>  <at> 
 {
     MouseDevPtr pMse;
(Continue reading)

NONAKA Kimihiro | 5 Mar 03:19 2009
Picon

Re: handling absolute X, Y position event in bsd_mouse.c

Hi,

2009/3/5 NONAKA Kimihiro <nonakap <at> gmail.com>:
> I see our XFree86 source's bsd_mouse.c. All necessary things for me
> were written there.
> However, I do not know what it is of ABSOLUTE_Z and ABSOLUTE_W.
> I have only the device that reports on ABSOLUTE_X and ABSOLUTE_Y.

When it is our XFree86 source's bsd_mouse.c method, warning goes out,
and it has become a build error.

#   compile  xorgos/bsd_mouse.o
/usr/local/netbsd-tools/bin/i386--netbsdelf-gcc -O2  -Werror
-fno-strict-aliasing   -DSHAPE -DXKB -DLBX -DXAPPGROUP -DXCSECURITY
-DTOGCUP -DXF86BIGFONT -DDPMSExtension -DPIXPRIV  -DPANORAMIX -DRENDER
-DRANDR -DGCCUSESGAS  -DAVOID_GLYPHBLT -DSINGLEDEPTH -DXvExtension
-DXFree86Server -DXvMCExtension -DSMART_SCHEDULE  -DBUILDDEBUG
-DXResExtension -DNDEBUG -DXINPUT -DXFreeXDGA -DXF86VIDMODE
-DXORG_VERSION_CURRENT="(((1) * 10000000) + ((4) * 100000) + ((2) *
1000) + 0)" -I/usr/xsrc/external/mit/xorg-server/dist/hw/xfree86/common
 -I/usr/xsrc/external/mit/xorg-server/dist/hw/xfree86/os-support
-I/usr/xsrc/external/mit/xorg-server/dist/hw/xfree86/os-support/bsd
-I/usr/xsrc/external/mit/xorg-server/dist/hw/xfree86/os-support/bus
-I/usr/xsrc/external/mit/xorg-server/dist/hw/xfree86/os-support/misc
-I/usr/xsrc/external/mit/xorg-server/dist/include
-I/home/snapshot/20090303/root/i386/usr/X11R7/include/X11
-I/home/snapshot/20090303/root/i386/usr/X11R7/include/X11/extensions
-I/home/snapshot/20090303/root/i386/usr/X11R7/include/pixman-1
-I/usr/xsrc/external/mit/xorg-server/dist/mi
-I/usr/xsrc/external/mit/xorg-server/dist/hw/xfree86/os-support/shared
(Continue reading)

David Brownlee | 6 Mar 01:26 2009

Screen corruption under Xen DOM0 in firefox

 	I'm trying Xen on my NetBSD 5.0 laptop and one thing I've
 	moticed is that the Firefox window seems to suffer corruption
 	when focus is gained/lost (this is under gnome on a Thinkpad
 	T60p using the radeonhd driver on a ATI MOBILITY FireGL V5250

 	An example is at:
 	http://mono.org/abs/xen-x11-corruption.png

 	Note that the 'Tools' on the menu bar has been garbled, as has
 	part of the webpage.

 	Has anyone else seem anything similar?

Matthew Mondor | 6 Mar 01:49 2009
Picon

Re: Screen corruption under Xen DOM0 in firefox

On Fri, 6 Mar 2009 00:26:27 +0000 (GMT)
David Brownlee <abs <at> absd.org> wrote:

>  	I'm trying Xen on my NetBSD 5.0 laptop and one thing I've
>  	moticed is that the Firefox window seems to suffer corruption
>  	when focus is gained/lost (this is under gnome on a Thinkpad
>  	T60p using the radeonhd driver on a ATI MOBILITY FireGL V5250
> 
>  	An example is at:
>  	http://mono.org/abs/xen-x11-corruption.png
> 
>  	Note that the 'Tools' on the menu bar has been garbled, as has
>  	part of the webpage.
> 
>  	Has anyone else seem anything similar?
> 

Under a desktop with an ATI radeon card I had some corruption with native
firefox 3, but which only happened under some specific pages/tabs, so it
seemed like a gecko bug to me...  I've not had any such problem with
firefox 2 however.  No idea if this is related or not to your problem
though.  The type of corruption I observed appeared different too.
--

-- 
Matt

David Brownlee | 6 Mar 02:04 2009
Picon

Re: Screen corruption under Xen DOM0 in firefox

On Thu, 5 Mar 2009, Matthew Mondor wrote:

> David Brownlee <abs <at> absd.org> wrote:
>
>>  	I'm trying Xen on my NetBSD 5.0 laptop and one thing I've
>>  	moticed is that the Firefox window seems to suffer corruption
>>  	when focus is gained/lost (this is under gnome on a Thinkpad
>>  	T60p using the radeonhd driver on a ATI MOBILITY FireGL V5250
>>
>>  	An example is at:
>>  	http://mono.org/abs/xen-x11-corruption.png
>>
>>  	Note that the 'Tools' on the menu bar has been garbled, as has
>>  	part of the webpage.
>>
>>  	Has anyone else seem anything similar?
>
> Under a desktop with an ATI radeon card I had some corruption with native
> firefox 3, but which only happened under some specific pages/tabs, so it
> seemed like a gecko bug to me...  I've not had any such problem with
> firefox 2 however.  No idea if this is related or not to your problem
> though.  The type of corruption I observed appeared different too.

 	Interesting - I can reproduce it on firefox2 and thunderbird
 	on this system but not on openoffice (for example). Did you
 	see your issues on xen or straight i386? Were you using
 	a radeon or radeonhd driver? Thanks

--

-- 
 		David/absolute       -- www.NetBSD.org: No hype required --
(Continue reading)

Jared D. McNeill | 6 Mar 02:12 2009
Picon

Re: Screen corruption under Xen DOM0 in firefox

David Brownlee wrote:
>     I'm trying Xen on my NetBSD 5.0 laptop and one thing I've
>     moticed is that the Firefox window seems to suffer corruption
>     when focus is gained/lost (this is under gnome on a Thinkpad
>     T60p using the radeonhd driver on a ATI MOBILITY FireGL V5250
> 
>     An example is at:
>     http://mono.org/abs/xen-x11-corruption.png
> 
>     Note that the 'Tools' on the menu bar has been garbled, as has
>     part of the webpage.
> 
>     Has anyone else seem anything similar?

Does radeonhd use XAA? If so, and your X server isn't up-to-date (this 
has recently been made the default), add the following to the Device 
section of xorg.conf: Option "XAANoOffscreenPixmaps" "true".

David Brownlee | 6 Mar 02:18 2009
Picon

Re: Screen corruption under Xen DOM0 in firefox

On Thu, 5 Mar 2009, Jared D. McNeill wrote:

> David Brownlee wrote:
>>     I'm trying Xen on my NetBSD 5.0 laptop and one thing I've
>>     moticed is that the Firefox window seems to suffer corruption
>>     when focus is gained/lost (this is under gnome on a Thinkpad
>>     T60p using the radeonhd driver on a ATI MOBILITY FireGL V5250
>>
>>     An example is at:
>>     http://mono.org/abs/xen-x11-corruption.png
>>
>>     Note that the 'Tools' on the menu bar has been garbled, as has
>>     part of the webpage.
>>
>>     Has anyone else seem anything similar?
>
> Does radeonhd use XAA? If so, and your X server isn't up-to-date (this has 
> recently been made the default), add the following to the Device section of 
> xorg.conf: Option "XAANoOffscreenPixmaps" "true".

 	I've been using modular-x11 from pkgsrc (and have the latest
 	as of yesterday). I'll test out with that Option - thanks!

 	Out of curiosity, do people tend to run X from xsrc or
 	pkgsrc nowdays?

--

-- 
 		David/absolute       -- www.NetBSD.org: No hype required --

(Continue reading)


Gmane