Bug 92640

Summary: No matching pixelformats found
Product: XQuartz Reporter: Zhongmin Hu <firmamentsess>
Component: New BugsAssignee: Jeremy Huddleston Sequoia <jeremyhu>
Status: RESOLVED MOVED QA Contact: Jeremy Huddleston Sequoia <jeremyhu>
Severity: major    
Priority: high CC: acvsaraiva, gnwiii
Version: 2.7.8 (xserver-1.16.4)Keywords: regression
Hardware: Other   
OS: Mac OS X (All)   
Whiteboard:
i915 platform: i915 features:
Attachments: glxinfo for 2.7.7
xdpyinfo for 2.7.7
glxinfo for 2.7.8
xdpyinfo for 2.7.8

Description Zhongmin Hu 2015-10-23 14:15:03 UTC
XQuartz works well by 2.7.7, but when I update to 2.7.8 I meet a problem as below when I start IDL:

No matching pixelformats found, perhaps try using LIBGL_ALLOW_SOFTWARE
Abort trap: 6
Comment 1 Jeremy Huddleston Sequoia 2015-10-23 17:05:41 UTC
Please attach the output of 'glxinfo' and 'xdpyinfo' when running XQuartz 2.7.7 and 2.7.8 for comparison purposes.
Comment 2 Jeremy Huddleston Sequoia 2015-10-23 17:29:45 UTC
*** Bug 92642 has been marked as a duplicate of this bug. ***
Comment 3 js54434 2015-11-02 00:32:03 UTC
Created attachment 119326 [details]
glxinfo for 2.7.7
Comment 4 js54434 2015-11-02 00:34:05 UTC
Created attachment 119327 [details]
xdpyinfo for 2.7.7
Comment 5 js54434 2015-11-02 00:34:33 UTC
Created attachment 119328 [details]
glxinfo for 2.7.8
Comment 6 js54434 2015-11-02 00:34:55 UTC
Created attachment 119329 [details]
xdpyinfo for 2.7.8
Comment 7 js54434 2015-11-02 00:35:38 UTC
I had the same bug. I've attached outputs for glxinfo and xdpyinfo.
Comment 9 Jeremy Huddleston Sequoia 2015-11-04 18:25:49 UTC
*** Bug 92816 has been marked as a duplicate of this bug. ***
Comment 10 Antonio Carlos Varela Saraiva 2015-11-04 18:53:21 UTC
I had the same problem and it seems that others are also facing this. I have tried all the betas like Jeremy said and only beta1 worked. Since beta2 I always got the same error message.

Hope this helps solving the bug...

Thanks
Comment 11 Jeremy Huddleston Sequoia 2015-11-04 19:59:05 UTC
The main change in beta2 over beta1 was the version of mesa.  It's unfortunate that this wasn't seen by anyone over the over 8 months since beta2 was released =(

As a workaround for now, you can backup libGL.1.dylib from the 2.7.7 install and use it to replace the one in 2.7.8

As for why this is happening, is there a way you could get me access to IDL to help triage this?
Comment 12 Jeremy Huddleston Sequoia 2015-11-04 19:59:45 UTC
Also, please add an update glxinfo output for the case of 2.7.8 with the libGL from 2.7.7, thanks.
Comment 13 js54434 2015-11-04 23:15:50 UTC
(In reply to Jeremy Huddleston Sequoia from comment #11)
> The main change in beta2 over beta1 was the version of mesa.  It's
> unfortunate that this wasn't seen by anyone over the over 8 months since
> beta2 was released =(
> 
> As a workaround for now, you can backup libGL.1.dylib from the 2.7.7 install
> and use it to replace the one in 2.7.8
> 
> As for why this is happening, is there a way you could get me access to IDL
> to help triage this?

DAVE (at https://www.ncnr.nist.gov/dave/download.html) is an open source application made with IDL, and I encountered the bug when I tried opening it. You might be able to see if the bug is fixed by running this program rather than having to install IDL.
Comment 14 Antonio Carlos Varela Saraiva 2015-11-05 19:11:34 UTC
I did the suggested test and it worked flawlessly. I used libGL.1.dylib from XQuartz 2.7.8 Beta1 and replaced on the installation of XQuartz 2.7.8. It was clear that libGL is the problem, so the question is, is there any way to solve this or those of us that use this version of IDL will have to always use the Workaround? In that case, how long "forever" might be? I mean, how long do you think we will be able to use the outdated version of libGL?

An for the glxinfo and xdpyinfo, I have no idea how I get those... If you tell me how, I'm pleased to get it for you.
Comment 15 George N. White III 2016-05-04 16:41:18 UTC
I came to this bug because I need to run NASA SeaDAS 6.4 on MacOS.  This program provides a 32-bit IDL runtime and 32-bit bina
Comment 16 George N. White III 2016-05-04 16:53:09 UTC
I came to this bug because I need to run NASA SeaDAS 6.4 on MacOS.  This program provides a 32-bit IDL 7.0 runtime and 32-bit binaries including
an IDL interface, IDLshare.so.  I also have IDL 8.5, but it does not install
32-bit libraries and the release notes indicate that 32-bit operation is "unsupported".  I see this bug using the 7.0 (i386) runtime, but (after building a 64-bit IDLshare.so) not with IDL 8.5 (x86_64).  This causes me to wonder if this bug occurs only for 32-bit IDL sessions or has been fixed in current IDL versions: 

It could be useful to see the IDL banner:

32-bit session:

$ seadas -em
IDL Version 7.0, Mac OS X (darwin i386 m32). (c) 2007, ITT Visual Information Solutions

% Embedded IDL: NASA GSFC SeaDAS Development, SeaDAS.
% Embedded IDL: NASA GSFC SeaDAS Development, SeaDAS.
No matching pixelformats found, perhaps try using LIBGL_ALLOW_SOFTWARE
/Users/gwhite/seadas-6.4.2/run/scripts/seadas: line 341: 35543 Abort trap: 6           $IDL_EXE $SDSBATCH

64-bit session:

$ seadas
IDL Version 8.5, Mac OS X (darwin x86_64 m64).
(c) 2015, Exelis Visual Information Solutions, Inc., a subsidiary of Harris Corporation.
Installation number: 95009.
Licensed for use by: Bedford Institute of Oceanography

SeaDAS Version 6.4.2 (pid = 35556)
Comment 17 Jeremy Huddleston Sequoia 2016-05-29 21:25:46 UTC
@js54434: I downloaded DAVE (dave_dev_macos_2016-05-24.zip) and launched it.

It opened a window for the application as well as an xterm window with this console output:


IDL Version 8.5.1, Mac OS X (darwin x86_64 m64).
(c) 2015, Exelis Visual Information Solutions, Inc., a subsidiary of Harris Corporation.

% Embedded IDL: NIST, 99368.
% Embedded IDL: NIST, 99368.
Display mode    : TrueColor
Color table size: 256
Number of colors: 16777216

DAVE System variables have been defined
% Program caused arithmetic error: Floating illegal operand
% Loaded DLM: XML.
Nutty font "-Misc-Fixed-Medium-R-SemiCondensed--13-120-75-75-C-60-ISO8859-1" with string " ".
Nutty font "-Misc-Fixed-Medium-R-SemiCondensed--13-120-75-75-C-60-ISO8859-1" with string " ".
DAVEvisContour::Init called...
DAVEvisSurface::Init called...
DAVEvisContour::Init called...
% Program caused arithmetic error: Floating underflow
% Program caused arithmetic error: Floating overflow
DAVEvisSurface::Init called...


Can you please explain to me how I use DAVE to trigger your error?
Comment 18 GitLab Migration User 2019-05-23 18:30:26 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/xserver/issues/756.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.