I like to keep my current kernel sources and 2 older so that - just in case - I can boot an older kernel. Yes, this is from experience, though it has been many years. That is why I don't automate --depclean.
Instead, I run "emerge -pv --depclean", look at the output, and manually remove older ...

Also the output of gcc-config -l:

workstation ~ # gcc-config -l
[1] x86_64-pc-linux-gnu-11
[2] x86_64-pc-linux-gnu-12 *
[3] x86_64-pc-linux-gnu-13

And the contents of /var/tmp/portage/gui-libs/vte-common-0.74.2/work/vte-common-0.74.2-build/meson-logs/meson-log.txt:

,,, (I was getting post ...

Right you are. /var/tmp/portage/gui-libs/vte-common-0.74.2/temp/build.log:

* Package: gui-libs/vte-common-0.74.2:2.91
* Repository: gentoo
* Maintainer: gnome@gentoo.org
* USE: abi_x86_64 amd64 elibc_glibc kernel_linux
* FEATURES: network-sandbox preserve-libs sandbox userpriv usersandbox ...

I am at the last step of a very simple profile upgrade to 23 where I am running

emerge -av --emptytree @world

and I am failing with x11-libs/vte-common-0.74.2. It is failing:

...
* Call stack:
* ebuild.sh, line 136: Called src_configure
* environment, line 2850: Called meson_src_configure ...

I've run into this before but I think you used eselect to fix it back then...

Latest "emerge -uavDN world" has an existing preserved lib: dev-libs/libffi -3.4.2-r1. Wants me to run "emerge @preserved-rebuild" to fix it. So I run:

emerge -av @preserved-rebuild

and get back:

These are the ...

The instructions are just too broken, and this is my replacement for my work workstation (ie do not want to tinker with it as much as I would my home workstation).

Started all over (yay!), formatted /dev/sda2 == /boot == vfat, and basically let it force me into UEFI and VFAT.

And gems like when running grub-install (with EFI target) using /boot, and in the /etc/fstab saying that should be ext2, and..

I tried using just "pc"; apparently it is 32-bit only, and then grub2 *still* tells you it cannot find the efi directory.

My problem is getting grub2 to do the mbr install. I know that, after running into this problem before (the docs being *very* vague on what partition is what, what to mount ...

Do I have to use LILO to NOT use UEFI with the bootloader now?

First, I get told how to break it (mkfs.ext2 /dev/sda1 instead of using VFAT (???) for my boot partition).

Second, I get told I can not use EFI and use bios (no; it complains you are trying to do an EFI install)

My last install (not ...

The --with-bdeps y worked at home (32 bit). Am trying now at work (64 bit)

(edit) Posted too soon. At work, the --with-bdeps updated nothing. Trying @preserved-rebuild and it still is has some kind of dev-lang/php:7.0 dependency...

# emerge -av @preserved-rebuild

These are the packages that would be merged, in order:

Calculating dependencies... done!

emerge: there are no ebuilds to satisfy "dev-lang/php:7.0".
(dependency required by "@preserved-rebuild" [argument])

And running it because, after the --with-bdeps run (it ...

At home (32-bit) I must have unmerged PHP. And I removed tn5250 which let openssl update - now it can't find an ebuild for python 3.4 (eselect python list shows I have 2.7, 3.6 (fallback) and 3.5 (fallback) ) when doing @preserved-rebuild. For openssl-1.1.1d-r2 it wants to update:
/usr/lib/libssl.so ...

Nope; just 7.3.9

I should add that this happens at home (32-bit) and at work (64-bit)

This started with libffi.

!!! existing preserved libs:
>>> package: dev-libs/openssl-1.1.1d-r2
* - /usr/lib64/libssl.so.1.0.0
* used by /usr/bin/links (www-client/links-2.14-r1)
* used by /usr/lib64/libevent_openssl-2.1.so.6.0.2 (dev-libs/libevent-2.1.8)
* used by /usr/lib64/php5.6/bin/php ...

That did the trick. Thanks.

As the snippet shows, ld. --info output:


Portage 2.3.51 (python 2.7.15-final-0, default/linux/amd64/17.0/desktop, gcc-7.3.0, glibc-2.27-r6, 4.14.83-gentoo x86_64)
=================================================================
System uname: Linux-4.14.83-gentoo-x86_64-Intel-R-_Pentium-R-_CPU ...

While emerging firefox-60.5.1:


...
0:06.95 DEBUG: configure:2189:1: warning: return type defaults to 'int' [-Wimplicit-int]
0:06.95 DEBUG: /usr/lib/gcc/x86_64-pc-linux-gnu/7.3.0/../../../../x86_64-pc-linux-gnu/bin/ld: unrecognized option '--compress-debug-sections=zlib'
0:06.95 DEBUG: /usr/lib ...

Since I enabled the relocatable kernel a few versions ago, tn5250 program has always had a segmentation fault starting up. Anybody else? Especially if there is a workaround?

Thanks.

I have virtualbox, virtualbox-additions, virtualbox-extpack-oracle, virtualbox-modules emerged, but I have not been doing a @module-rebuild because I removed the 3 vbox modules from /etc/conf.d/modules because I don't currently run it (if I need another VM I'll do the @module-rebuild, manually ...

That solved it. I thought the problem was "you are not supposed to be using libopcodes any more in 2.26.1", when - as you surmised - it was "where the heck did your libopcodes-2.26.1.so go???"

The only "official" (known to portage) binutils I have is 2.26.1. The issue is, there is no libopcode in there (including in my backup from 2 weeks ago). I copied the 2.25.1 from my backup, but it does not show up in eselect (don't know how those get "registered").

I can restore my full system ...

No - make clean did nothing (and I did not expect it to; the assembler is trying to load a so that does not exist (libopcodes-2.26.1.so) (or, well, it is not in the directory where it is supposed to be)

[1] i686-pc-linux-gnu-2.26.1 *

I did manually copy 2.25.1 into /usr/lib/... but don't know how (or if I want it to) get it to show up in eselect.

Multilib 32-bit.

Did my usual morning "emerge -uavDN world" 2 days ago. It always recommends that I run "emerge --depclean" afterward, and I usually don't, because I like keeping 3 kernels in /boot. But I figured - what the heck; I never recompile any of the previous kernels, so go ahead - it is ...

Using gentoo.osuosl.org and lug.mtu.edu/gentoo and nothing after 2016-01-24 found (using websync). At home using rsync, I had a new update this morning...