Kernel fails to compile

[leyvi]

I've been trying to update my kernel for two weeks, but whenever I try, I get this generic error message:

[sunox]

Check out app-text/wgetpaste for sharing your config.

[CaptainBlood]

Do you have any 6.6.x working .config file?

Thks 4 ur attention, interest & support.
_________________
USE="-* ..." in /etc/portage/make.conf here, i.e. a countermeasure to portage implicit braces, belt & diaper paradigm
LT: "I've been doing a passable imitation of the Fontana di Trevi, except my medium is mucus. Sooo much mucus. "

[NeddySeagoon]

leyvi,

Try

[leyvi]

Ok, here's the link:
http://0x0.st/H6ey.txt

[leyvi]

Can someone take a look please?

[NeddySeagoon]

leyvi,

Here's what I've done to try to reproduce your problem here ..

[leyvi]

Thanks Ned!
I'll try using my config file from 6.6.4 (the last version to work correctly for me).

[NeddySeagoon]

leyvi,

I accepted the default answers to the questions that

[leyvi]

I’ve been using the variables “LLVM=1 LLVM_IAS=1”.
Does the fact that you were able to build it with just “LLVM=1” mean that I should stop using “LLVM_IAS”?
What does that flag IAS even do? (It’s not entirely clear to me).

[grknight]

[Hu]

Since the config works for Neddy, perhaps we should debug the actual error. leyvi: please pastebin the full build log, so we can see all error messages produced.

[NeddySeagoon]

leyvi,

[leyvi]

OK. But I don't know where the build logs are located...
When everything else failed, that was the first thing I tried to check.
I couldn't find it though.

[Hu]

You quoted the tail of the build output in your opening post. From where did you get that? Go there, and get the rest of it.

[leyvi]

Oh, I though there was a separate log file to the build output.
Here is a link to the pastebin:
http://0x0.st/H6Jf.log

[leyvi]

Now that I can actually see the top of the log (it was lost to the line limit in my terminal), I am able to see this:

[NeddySeagoon]

leyvi,

I don't have /etc/env.d/99xz and it worked for me.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

[leyvi]

Yeah... I tried disabling everything related to xz in the config, but it didn't change anything (tar was run with xz anyway).
But I did try this:

[Hu]

Do those custom XZ options work for building a simple tar file by hand? There are plenty of ways users can configure their tools to be completely broken. In my opinion, an unstated prerequisite for building anything is that your build tools are not broken, so specifically advising users not to configure their xz to be broken is only one of many things we would need to list.

[leyvi]

To clarify, I read about this in the xz article on the Gentoo Wiki.
I'm just saying that perhaps a warning should be added to that
article, so that people like me (who can be a bit clueless
sometimes) don't go changing this configuration, then wonder
why their kernel won't build. Here is the link:
https://wiki.gentoo.org/wiki/Xz-utils#Configuration
I don't think it would be a bad idea to add a disclaimer here.

[pietinger]

I would like to see the content of your previous XZ_DEFAULTS=
_________________
https://wiki.gentoo.org/wiki/User:Pietinger

[Hu]

Wikis are freely editable, so you could use the associated Talk: page to propose the disclaimer, and discuss with other Wiki contributors what the ideal phrasing would be.

What options had you put in your local environment for xz? The same ones shown on the Wiki (XZ_DEFAULTS="-9 -T0") or something more complex? The output from the failed kernel build looks to me like you put something very complicated in the environment, which is why I wanted to know if those complex options worked outside the kernel build.

[leyvi]

This was my xz config:

[Hu]

We have had a number of threads recently about systems failing to boot after a grub update. Perhaps one of those will have useful advice for you.