Bug 621638 – x11-wm/icewm-1.3.8 "narrowing conversion error" under GCC 6.3.0

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Walter Dnes 2017-06-13 03:15:03 UTC]

Created attachment 476226 [details]
emerge --info

Blocks https://bugs.gentoo.org/show_bug.cgi?id=gcc-6

The following happens in both x86 and AMD64.

icesh.cc:249:1: error: narrowing conversion of '4294967295ul' from 'long unsigned int' to 'long int' inside { } [-Wnarrowing]

icesh.cc:261:1: error: narrowing conversion of '4294967295ul' from 'long unsigned int' to 'long int' inside { } [-Wnarrowing]

Build log and emerge --info attached

[Comment 1 Walter Dnes 2017-06-13 03:18:57 UTC]

Created attachment 476228 [details]
Failed build log

[Comment 2 Jonas Stein 2017-06-13 06:27:08 UTC]

Thank you.
Do you have the chance, to test x11-wm/icewm-1.3.12-r1 too?


https://github.com/bbidulock/icewm/blob/icewm-1-3-BRANCH/src/icesh.cc

[Comment 3 Walter Dnes 2017-06-13 18:10:30 UTC]

(In reply to Jonas Stein from comment #2)
> Thank you.
> Do you have the chance, to test x11-wm/icewm-1.3.12-r1 too?

Thank you for the suggestion. I keyworded x11-wm/icewm-1.3.12-r1 and built it a few minutes ago. It builds OK under GCC 6.3.0, and seems to run fine, although I've only tested it for a few minutes. Maybe the best solution for this bug is to stabilize x11-wm/icewm-1.3.12-r1.

[Comment 4 Peter Levine 2017-07-19 02:05:08 UTC]

Created attachment 485624 [details, diff]
icewm-1.3.8-gcc6.patch

I believe this error should only happen on platforms that define a long int to be 32 bits.  Unfortunately, on mine it's 64 bits so I'm not able to confirm that this patch works.

Could someone who notices the bug (especially on x86 arch) test this please?

[Comment 5 Andreas K. Hüttel 2017-10-02 11:14:25 UTC]

(In reply to Peter Levine from comment #4)
> Created attachment 485624 [details, diff] [details, diff]
> icewm-1.3.8-gcc6.patch
> 
> I believe this error should only happen on platforms that define a long int
> to be 32 bits.  Unfortunately, on mine it's 64 bits so I'm not able to
> confirm that this patch works.
> 
> Could someone who notices the bug (especially on x86 arch) test this please?

Not so many x86 around anymore, sorry. Stabilizing 12-r1 will be faster; thanks anyway for all of your patches.