Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 921534 - ERROR: dev-lang/spidermonkey-102.15.0::gentoo failed (configure phase)
Summary: ERROR: dev-lang/spidermonkey-102.15.0::gentoo failed (configure phase)
Status: RESOLVED INVALID
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: All Linux
: Normal normal
Assignee: Mozilla Gentoo Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2024-01-07 00:54 UTC by Paweł Metelski
Modified: 2024-01-07 06:45 UTC (History)
0 users

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Paweł Metelski 2024-01-07 00:54:00 UTC
Configure script cannot find the suitable clang version.

I have sys-devel/clang-16.0.6 (stable) installed. Just to workaround this issue, I also unmasked and merged 17.0.6 and its dependencies. Doesn't help at all.

Reproducible: Always

Steps to Reproduce:
emerge -DuN world

All other packages build fine, this is the only world update blocker for me.
Actual Results:  
checking for clang for bindgen... /usr/lib/llvm/16/bin/clang++
checking for libclang for bindgen... /usr/lib/llvm/16/bin/../../../../lib/libclang.so
checking that libclang is new enough... 
ERROR: The libclang located at /usr/lib/llvm/16/bin/../../../../lib/libclang.so is too old (need at least 5.0).

Please make sure to update it or point to a newer libclang using
--with-libclang-path.


Expected Results:  
Just build.

I suppose you don't need my emerge --info if you can reproduce it. Otherwise, just let me know.
Comment 1 Paweł Metelski 2024-01-07 00:57:42 UTC
102.10.0 fails to build too, yet it was merged successfully on my system in June. Thus, something other than this ebuild was changed since.
Comment 2 Greg Kubaryk 2024-01-07 01:39:45 UTC
Please reopen when the required information (build.log as attachment, emerge --info as a comment) is provided.
Comment 3 Paweł Metelski 2024-01-07 02:49:30 UTC
I'm not a paid QA engineer. Either reproduce it on your side or mark "works for me". The community will decide if this is a proper resolution.
Comment 4 Paweł Metelski 2024-01-07 03:10:20 UTC
Okay, I added +duktape useflag for polkit and -introspection useflag for libsecret and don't need your legacy javascript shenanigans anymore.

Fixed by making my system free from toxic javascript community.
Comment 5 Greg Kubaryk 2024-01-07 03:53:59 UTC
(In reply to Paweł Metelski from comment #3)
> I'm not a paid QA engineer. Either reproduce it on your side or mark "works
> for me". The community will decide if this is a proper resolution.

Neither are any of us. We do appreciate bug reports but need actionable information, which in this case is described at the bottom of every failed build.
Comment 6 Joonas Niilola gentoo-dev 2024-01-07 06:45:19 UTC
Hey,

just managed to build spidermonkey-102.10.0 and 102.15.0 successfully with clang-16. I would need a full build log to have an idea what might be wrong, the text you posted isn't enough to guess. My initial impression is also that there's something weird going on in your system, so a full emerge --info would help a lot too in addition. These are commonly attached in any bug report to have as much information available as possible to try and reproduce the issue. 


> Okay, I added +duktape useflag for polkit and -introspection useflag for 
> libsecret and don't need your legacy javascript shenanigans anymore.
> 
> Fixed by making my system free from toxic javascript community.

"We" are not the "javascript community" here. We are just packaging upstream software. Also, we are here to help, but with the information you provided it's impossible to _guess_ what could be the issue. I think the trigger here was closing the bug as NEEDINFO, but that's also a common practice since there are people with very weird setups and they are reluctant in providing information about their systems, because they _know_ they've broken it on their own. These bugs would then stay open for a long time before maintainer closing them as NEEDINFO anyway.

Anyway I'm glad that 1) you found, 2) there is an alternative solution available that works for you!