Windows 5.0.2 built with wxWidgets debug on?

Bug #1809622 reported by Seth Hillbrand on 2018-12-24
34
This bug affects 5 people
Affects Status Importance Assigned to Milestone
KiCad
High
Wayne Stambaugh

Bug Description

There's a report over at the forum (https://forum.kicad.info/t/win-v7-64-shutdown-and-kicad-v5-0-2-response/14306) that appears to indicate the we are shipping Windows builds with the debug information enabled.

The image from the forum is attached here.

tags: added: packaing windows
tags: added: packaging
removed: packaing
Wayne Stambaugh (stambaughw) wrote :

Would one of our package developers please confirm if this is the case. I haven't been able to trigger this but I don't use the packaged stable release of KiCad very often.

Changed in kicad:
status: New → Incomplete
Nick Østergaard (nickoe) wrote :

It is not build with debug on. But OP have not yet replied.

https://forum.kicad.info/t/win-v7-64-shutdown-and-kicad-v5-0-2-response/14306/5

Wayne Stambaugh (stambaughw) wrote :

AFAIR, there was a change in the CMakeList.txt file that used the default CMake build type setting which is RelWithDbgInfo. This build type would cause the assert dialog to be shown. I thought we had already addressed this because it was an issue with the 5.0.0 release. -DCMAKE_BUILD_TYPE=Release must be defined at configuration to ensure a true release build.

Seth Hillbrand (sethh) wrote :

Note that both the duplicate and this post[1] show wxAssert messages in the latest builds. We may not be building with debug on, but something is causing the asserts to show when they shouldn't. This may be a hard-coded define in some default header file?

[1] https://forum.kicad.info/t/3d-library-error/14615

Wayne Stambaugh (stambaughw) wrote :

Would someone who has access to the windows build scripts please check to see if -DCMAKE_BUILD_TYPE=Release is defined in the build configuration? If it's not, then this is the problem. If it is, then I have no idea why we would be seeing the assertions.

Nick Østergaard (nickoe) wrote :

I verified this before. See build log on
https://jenkins.simonrichter.eu/job/windows-kicad-msys2-stable/33/consoleText

Which did checkout this commit of kicad-winbuilder:
a8fb33b1c0cca9c7c64bdde7092ec326fa3cefc9

And it uses PKGBUILD-STABLE and indeed the Release value:
https://github.com/KiCad/kicad-winbuilder/blob/a8fb33b1c0cca9c7c64bdde7092ec326fa3cefc9/PKGBUILD-STABLE#L84

Wayne Stambaugh (stambaughw) wrote :

@Nick, I thought as much. I'm not sure what to do about this. The original bug reporter never responded to your request to test the package you provided and I cannot duplicate it. Since the version information was not provided, I have no idea if the reporter was using KiCad from one of our installers or something else.

Nick Østergaard (nickoe) wrote :

@Wayne, I think Kerusey tested the correct build on https://bugs.launchpad.net/kicad/+bug/1808655/comments/5

Kerusey Karyu (keruseykaryu) wrote :

Yes. I have installed corrected build form Nick's link provided in my report.
Both cases in My reports are related to the symbol (re)mapping. Although I saw other assert once in a completely different situation. At work, don't remember when exactly, it could have been during Gerber plot. Unfortunately, I did not grab the screen then.

Changed in kicad:
status: Incomplete → Triaged
importance: Undecided → High
assignee: nobody → Wayne Stambaugh (stambaughw)
milestone: none → 5.1.0
KiCad Janitor (kicad-janitor) wrote :

Fixed in revision 69d90a4e84c1c904c0d880d431a7aebfa991994d
https://git.launchpad.net/kicad/patch/?id=69d90a4e84c1c904c0d880d431a7aebfa991994d

Changed in kicad:
status: Triaged → Fix Committed
Nick Østergaard (nickoe) wrote :
Changed in kicad:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers