Solved: Upgrading Hardware System won't boot or chroot

[cctsurf]

I've had this gentoo install going since 2006. I've upgraded hardware repeatedly over those years, but never run into this before.
When I previously upgraded hardware, I was always able to boot into the system or boot a livecd and chroot into it and rebuild the basics and then rebuild the system to be optimized for the new hardware. This time I've hit a brick wall.
I upgraded from a FX-8350 to a Ryzen 7 3700X and my system is truly borked.
When I attempt to chroot into the system I get

Code: Select all

Illegal instruction (core dumped)

I've been attempting to use https://wiki.gentoo.org/wiki/Fix_My_Gentoo, but I'm not certain which packages to replace... At this point, I'm pretty sure the code is incompatible with the new processor.

I have come up with a couple of options:
I'm looking for suggestions, if it were not such an old installation (or if it were, say Ubuntu), I'd just build a new system and drop my home directory on it. But I would like to save this old install.
I've already built the basic packages using the Fix_My_Gentoo methodology, I just don't know what they are... figure out what they are and drop them into my old install.
Any other ideas?
THANKS!

[eccerr0r]

Bah, someone said that AMD was better than Intel at backward compatibility. Now it's clear both are bad...

When you chroot() it will run the shell interpreter in the chroot. So this depends on bash (if that's your default shell interpreter) and its dependencies.

Top ones that should be redone to even get started with chroot:

-bash
-glibc

Then optionally
-readline
-internationalization

But that only gets you into chroot. Then you need to make sure python and its dependencies are working so you can emerge again ... Not sure if this is worth it over just salvaging your /var/lib/portage/world file and starting over with that?

(me likewise I have an install I've been running since 2004, luckily I explicitly did not use any CPU optimizations so it'd run on any hardware. The install still exists today and running with a public IP address. It now is in a VM...)

[Jaglover]

If you still have the old hardware you could do emerge -e @system with -march=x86-64.

[cctsurf]

If you still have the old hardware you could do emerge -e @system with -march=x86-64.

I attempted to do something like this, but I apparently did not go far enough. I do have the old hardware and I'm thinking about how I could do this. I changed the heart of my system, but didn't upgrade the drives (I've done that fairly recently). I'm not certain where to go from here, but somehow this fact should be added to the wiki somewhere... perhaps a "before you upgrade hardware" page... Seeing as

Bah, someone said that AMD was better than Intel at backward compatibility. Now it's clear both are bad..

I have to ask the same question as you, eccerr0r, the more I think about it, is it worth it? I have to say, I think there is an awful lot of old cruft in my old install. It might be good to have a fresh install. I notice that my installs are so old that the standard placement of everything in portage has changed...

I think I have a plan.
I'm going to copy this old install to a hard drive I have sitting around and assemble a working system on my old system and do the rebuild. I can use a different linux I have on my new computer for the time being and when that is rebuilt, I will have a way forward.
THANKS!

[eccerr0r]

Well, as said, I did not optimize everything to exactly the target CPU for machines I expect to upgrade in the future. Only target specific applications that could benefit from optimizations get special treatment, so for the most part I don't see too many issues when moving disks between machines.

Then there are machines like my Atom netbook whose disk(SSD) is pretty much useless for any other machine due to mechanical incompatibility. Now this machine gets optimizations because it already is way too slow and I suffer from this as I cannot even NFS chroot into this machine as I don't have any other machines that can run the Atom instruction set properly...

The thing that's killing me now?

SSE2

[figueroa]

Are you booting from a 64 bit live CD and trying to chroot into a 32 bit Gentoo system? That's what the error message sounds like.

[eccerr0r]

I thought if your kernel/cpu doesn't support the ELF binary type, it should give you an file format error instead of SIGILL ?

[OldTango]

I upgraded from a FX-8350 to a Ryzen 7 3700X and my system is truly borked.
When I attempt to chroot into the system I get

Code: Select all

Illegal instruction (core dumped)

I had this exact issue a couple of years back when I went from a FX-8350 to a Ryzen 1950X Threadripper. Not sure how I solved the problem exactly, but it came down to re-compiling @world on the FX-8350 system first with cflags both processors could understand and CPU_FLAGS_X86 either empty or with flags supported by both processors. After that was done I moved the drives to the Ryzen system and was able to chroot into the new system, set the proper CPU_FLAGS_X86 flags and re-compile @world using -march=native

I recently moved from the Ryzen 1950X Threadripper to a Ryzen 9 5950X without any problems.

What CFLAGS did you use for the FX-8350 system?
I will try to find my old forum post on how I fixed the problem

Best Tango.....

[cctsurf]

I suffer from this as I cannot even NFS chroot into this machine as I don't have any other machines that can run the Atom instruction set properly...

Ah yes, the Atom instruction set. I have a Aspire One with the N270 Atom chip that my kids use for typing tutor, the keyboard fits their small hands really well. I had a gentoo install that I built for it using distcc. I wanted to build packages on the particular system that I just upgraded. I attempted virtualization and chroots, but never got it working... They always seemed to want to pass the Atom instructions to the processor and that, of course, failed.

Are you booting from a 64 bit live CD and trying to chroot into a 32 bit Gentoo system? That's what the error message sounds like.

I have an installation of Manjaro on the same drive and it boots just fine. It is 64 bit. I am attempting to chroot from there. On the old hardware, I have done the chroot a few times... I think I over-optimized for the older processor and it seems that the instruction set is not compatible with the new chip. (These optimizations are one of the reasons I love runing gentoo... I have been running really old hardware and the optimizations made it run really nicely.)

My plan is to copy this old installation to a different drive (I tried that today and the drive I had available failed on me) and install it on my old system and rebuild @system with -march=x86-64. I should then be able to chroot into it or even boot it on my new hardware and optimize it for my new system...

[eccerr0r]

It looks like the special xop instructions on the FX8350 simply got removed for Ryzen. Sounds just like 3dNow! that got lost on newer AMD architectures.

The Atom problem is sort of an offshoot versus a direct lineage thing. In fact newer Intel Core CPUs (and probably AMD too) will run Atom instructions, so that's not a huge issue.

It almost seems the only thing AMD retained over the years is x86-64 itself, and all of its instructions after that got dropped? I mean MMX, SSE*, AVX, etc. have been adopted it seems by all x86* CPUs but nothing AMD did?

[Tony0945]

Working from your install media install a recent stage3 tarball on your system. This is all of @system, so you should be able to chroot in and rebuild -e world.

3D now and something else changed over the years (AVX ?) I have no trouble building packages for K10 on a Ryzen, but I don't run them there. Similarly I use a 32-bit K10 install to build for a K6-3. Eix was the only problem. Apparently some packages detect what hardware they are running on, ignoring CFLAGS settings.

However the official stage3 seems to be built for i686. Just be sure to get the right stage3, i.e. don't mix OpenRC and systemd.
If you wish, do it in two steps: emerge -e @system first. But even on a ryzen gcc will take hours so I wouldn't. BTW, I tried turning off FORTRAN since I haven't used it since the 1970's. Only saved a few minutes. I think C++ is the big time consumer.

[eccerr0r]

It's not hard for any machine to build instructions for another machine provided they're turing complete, but actually running the instructions without emulation is the problem. The actual use of 3dNow! opcodes apparently will trigger SIGILLs on newer AMD machines anyway, so whether the concepts are now part of AVX ... well, it still won't run them using the AVX hardware.

The silly thing I've found is that at least for gcc when using -msse2, I've found a lot of the instructions being used are basically just shuffling registers around, and not actually doing SSE2 math... That's what really pisses me off. I suspect gcc is doing the same for the extended AMD instructions for a 0.0001% speedup... This is not to say all of them are doing such.

Ultimately I tend not to use -m at all on my CFLAGS with the exception of my Atom, and only use CPU_FLAGS_X86 for package dependent optimizations. This should generate and keep a base system pretty much generically compatible with any machine of the same class and only the special packages get their specific benefits and their SIGILLs. This is not the "optimal" solution for any specific machine, but it sure helps moving binaries between machines.

[Tony0945]

FWIW, I updated an Atlhon II machine (fam10) using march=native to a 2700X using march=native with no problem. Just "emerge -e @world"
Maybe some individual programs might have failed, but not gcc, python etc.

I really question the utility of some of those CISC instructions. Just give me a fast clock speed, fast memory, and fast disk. OFC, that's the whole CISC vs RISC argument. CISC seems to have won.

[figueroa]

Along these lines, my current CPU is an Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz and my make.conf CFLAGS setting is:

Code: Select all

CFLAGS="-O2 -march=native -pipe"

but I'm not happy with that original choice for the reason of future CPU compatibility issues noted in this thread. I'm considering changing that to:

Code: Select all

CFLAGS="-O2 -march=x86-64 -mtune=sandybridge -pipe"

and then

Code: Select all

emerge -e @world

Will that give me the safe, near optimization, and compatibility/flexibility I'd rather have? I've had this on the back burner for at least a year.

[eccerr0r]

I really question the utility of some of those CISC instructions. Just give me a fast clock speed, fast memory, and fast disk. OFC, that's the whole CISC vs RISC argument. CISC seems to have won.

I can't say that this is the best solution.
In encryption you have to swap bytes around a lot. A CISC instruction will do this in one clock cycle. Imagine a RISC machine given a whole bunch of bit shift and bitwise and/or operations to do the byte swap. Even if you can run each instruction 8 times as fast, if you need 16 instructions to do this, this is a huge net loss. Hence CISC wins in this case.

Of course a FPGA might do better, but even then, compare it to simply making an ASIC to do the same... but arent Intel/AMD CISC CPUs just another ASIC?

Granted over time a lot of old crufty instructions are still left in a CISC ISA... could that silicon be put to better use? Maybe, Maybe not... is it really that much silicon compared to the gobs of cache on the chips...

[cctsurf]

Well... I rebuilt @system on my old hardware with MARCH=x86-64 and now I can chroot into it on my new system. I can build packages, but many of them fail installation, including glibc. I assume that some package wasn't rebuilt with the @system that is used in the final installation.

This is the failure mode:

if test -r /var/tmp/portage/sys-libs/glibc-2.33-r5/image//usr/include/gnu/stubs-64.h && cmp -s /var/tmp/portage/sys-libs/glibc-2.33-r5/work/build-amd64-x86_64-pc-linux-gnu-nptl/stubs.h /var/tmp/portage/sys-libs/glibc-2.33-r5/image//usr/include/gnu/stubs-64.h; \
then echo 'stubs.h unchanged'; \
else /usr/lib/portage/python3.9/ebuild-helpers/xattr/install -c -m 644 /var/tmp/portage/sys-libs/glibc-2.33-r5/work/build-amd64-x86_64-pc-linux-gnu-nptl/stubs.h /var/tmp/portage/sys-libs/glibc-2.33-r5/image//usr/include/gnu/stubs-64.h; fi
rm -f /var/tmp/portage/sys-libs/glibc-2.33-r5/work/build-amd64-x86_64-pc-linux-gnu-nptl/stubs.h
/var/tmp/portage/sys-libs/glibc-2.33-r5/work/build-amd64-x86_64-pc-linux-gnu-nptl/elf/sln /var/tmp/portage/sys-libs/glibc-2.33-r5/work/build-amd64-x86_64-pc-linux-gnu-nptl/elf/symlink.list
rm -f /var/tmp/portage/sys-libs/glibc-2.33-r5/work/build-amd64-x86_64-pc-linux-gnu-nptl/elf/symlink.list
make[1]: Leaving directory '/var/tmp/portage/sys-libs/glibc-2.33-r5/work/glibc-2.33'
>>> Completed installing sys-libs/glibc-2.33-r5 into /var/tmp/portage/sys-libs/glibc-2.33-r5/image

* Final size of build directory: 610536 KiB (596.2 MiB)
* Final size of installed tree: 78296 KiB ( 76.4 MiB)

* ERROR: sys-libs/glibc-2.33-r5::gentoo failed:
* (no error message)
*
* Call stack:
* misc-functions.sh, line 601: Called install_qa_check
* misc-functions.sh, line 132: Called source 'install_symlink_html_docs'
* 60python-pyc, line 83: Called python_pyc_check
* 60python-pyc, line 15: Called python_get_sitedir 'install_hooks'
* environment, line 3160: Called _python_export '/usr/portage/metadata/install-qa-check.d/60python-pyc' 'python3_8'
* environment, line 683: Called die
* The specific snippet of code:
* PYTHON_SITEDIR=$("${PYTHON}" -c 'import sysconfig; print(sysconfig.get_path("purelib"))') || die;
*
* If you need support, post the output of `emerge --info '=sys-libs/glibc-2.33-r5::gentoo'`,
* the complete build log and the output of `emerge -pqv '=sys-libs/glibc-2.33-r5::gentoo'`.
* The complete build log is located at '/var/tmp/portage/sys-libs/glibc-2.33-r5/temp/build.log'.
* The ebuild environment file is located at '/var/tmp/portage/sys-libs/glibc-2.33-r5/temp/environment'.
* Working directory: '/var/tmp/portage/sys-libs/glibc-2.33-r5/image'
* S: '/var/tmp/portage/sys-libs/glibc-2.33-r5/work/glibc-2.33'
* Verifying compiled files in
!!! post install failed; exiting.

>>> Failed to emerge sys-libs/glibc-2.33-r5, Log file:

>>> '/var/tmp/portage/sys-libs/glibc-2.33-r5/temp/build.log'

* Messages for package sys-libs/glibc-2.33-r5:

* ERROR: sys-libs/glibc-2.33-r5::gentoo failed:
* (no error message)
*
* Call stack:
* misc-functions.sh, line 601: Called install_qa_check
* misc-functions.sh, line 132: Called source 'install_symlink_html_docs'
* 60python-pyc, line 83: Called python_pyc_check
* 60python-pyc, line 15: Called python_get_sitedir 'install_hooks'
* environment, line 3160: Called _python_export '/usr/portage/metadata/install-qa-check.d/60python-pyc' 'python3_8'
* environment, line 683: Called die
* The specific snippet of code:
* PYTHON_SITEDIR=$("${PYTHON}" -c 'import sysconfig; print(sysconfig.get_path("purelib"))') || die;
*
* If you need support, post the output of `emerge --info '=sys-libs/glibc-2.33-r5::gentoo'`,
* the complete build log and the output of `emerge -pqv '=sys-libs/glibc-2.33-r5::gentoo'`.
* The complete build log is located at '/var/tmp/portage/sys-libs/glibc-2.33-r5/temp/build.log'.
* The ebuild environment file is located at '/var/tmp/portage/sys-libs/glibc-2.33-r5/temp/environment'.
* Working directory: '/var/tmp/portage/sys-libs/glibc-2.33-r5/image'
* S: '/var/tmp/portage/sys-libs/glibc-2.33-r5/work/glibc-2.33'

This is the emerge --info for it:

emerge --info =sys-libs/glibc-2.33-r5
Portage 3.0.20 (python 3.9.6-final-0, default/linux/amd64/17.1/desktop/plasma, gcc-11.2.0, glibc-2.33-r5, 5.13.8-1-MANJARO x86_64)
=================================================================
System Settings
=================================================================
System uname: Linux-5.13.8-1-MANJARO-x86_64-AMD_Ryzen_7_3700X_8-Core_Processor-with-glibc2.33
KiB Mem: 32792948 total, 20347688 free
KiB Swap: 16777212 total, 16754284 free
Timestamp of repository gentoo: Fri, 13 Aug 2021 04:00:01 +0000
Head commit of repository gentoo: c40c6f80af2032400cba5b068510cfb24a378c48
sh bash 5.1_p8
ld GNU ld (Gentoo 2.36.1 p5) 2.36.1
app-shells/bash: 5.1_p8::gentoo
dev-java/java-config: 2.3.1::gentoo
dev-lang/perl: 5.34.0-r2::gentoo
dev-lang/python: 2.7.18_p11::gentoo, 3.8.11::gentoo, 3.9.6_p1::gentoo, 3.10.0_rc1_p1::gentoo
dev-lang/rust: 1.54.0::gentoo
dev-util/cmake: 3.21.1::gentoo
sys-apps/baselayout: 2.7-r3::gentoo
sys-apps/openrc: 0.43.3::gentoo
sys-apps/sandbox: 2.24::gentoo
sys-devel/autoconf: 2.13-r1::gentoo, 2.71-r1::gentoo
sys-devel/automake: 1.13.4-r2::gentoo, 1.16.4::gentoo
sys-devel/binutils: 2.36.1-r2::gentoo
sys-devel/gcc: 11.2.0::gentoo
sys-devel/gcc-config: 2.4::gentoo
sys-devel/libtool: 2.4.6-r6::gentoo
sys-devel/make: 4.3::gentoo
sys-kernel/linux-headers: 5.13::gentoo (virtual/os-headers)
sys-libs/glibc: 2.33-r5::gentoo
Repositories:

gentoo
location: /usr/portage
sync-type: rsync
sync-uri: rsync://rsync.gentoo.org/gentoo-portage
priority: -1000
sync-rsync-verify-max-age: 24
sync-rsync-verify-jobs: 1
sync-rsync-verify-metamanifest: yes
sync-rsync-extra-opts:

Home
location: /usr/local/portage
masters: gentoo
priority: 0

crossdev
location: /var/db/repos/portage-crossdev
masters: gentoo
priority: 10

brave-overlay
location: /var/lib/layman/brave-overlay
masters: gentoo
priority: 50

jorgicio
location: /var/lib/layman/jorgicio
masters: gentoo
priority: 50

ACCEPT_KEYWORDS="amd64 ~amd64"
ACCEPT_LICENSE="* PUEL"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-O2 -march=native -pipe"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/lib64/libreoffice/program/sofficerc /usr/share/config /usr/share/gnupg/qualified.txt"
CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/dconf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo"
CXXFLAGS="-O2 -march=native -pipe"
DISTDIR="/mnt/nfs_portage/distfiles"
EMERGE_DEFAULT_OPTS="--with-bdeps y --buildpkg-exclude "virtual/* sys-kernel/*-sources""
ENV_UNSET="CARGO_HOME DBUS_SESSION_BUS_ADDRESS DISPLAY GOBIN GOPATH PERL5LIB PERL5OPT PERLPREFIX PERL_CORE PERL_MB_OPT PERL_MM_OPT XAUTHORITY XDG_CACHE_HOME XDG_CONFIG_HOME XDG_DATA_HOME XDG_RUNTIME_DIR"
FCFLAGS="-O2 -pipe"
FEATURES="assume-digests binpkg-docompress binpkg-dostrip binpkg-logs binpkg-multi-instance buildpkg config-protect-if-modified distlocks ebuild-locks fixlafiles ipc-sandbox merge-sync multilib-strict network-sandbox news parallel-fetch pid-sandbox preserve-libs protect-owned qa-unresolved-soname-deps sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync xattr"
FFLAGS="-O2 -pipe"
GENTOO_MIRRORS="http://gentoo.osuosl.org/ http://gentoo.mirrors.easynews.com/linux/gentoo/ https://gentoo.osuosl.org/"
LANG="en_US.UTF-8"
LDFLAGS="-Wl,-O1 -Wl,--as-needed"
LINGUAS="en en_US"
MAKEOPTS="-j9"
PKGDIR="/mnt/nfs_portage/packages"
PORTAGE_CONFIGROOT="/"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --omit-dir-times --compress --force --whole-file --delete --stats --human-readable --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --exclude=/.git"
PORTAGE_TMPDIR="/var/tmp"
USE="X a52 aac aacs acl acpi activities alsa amd64 audiofile avahi avi bluetooth bluray branding bzip2 cairo cdda cddb cdio cdr cdrkit cg cli crypt css cups dbus declarative divx4linux dlloader dri dts dv dvb dvd dvdnav dvdr dvdread elogind emboss enca encode exif extras ffmpeg flac fontconfig fortran g3dvl gdbm gif gimp glut gpm gtk gui gzip iconv icu ieee1394 imagemagick java jpeg json kde kepler kerberos kipi kontact kwallet ladspa lame lcms libglvnd libmpv libnotify libtirpc linguas_en_US lingus_en live lm-sensors lv2 mad melt midi mjpeg mng mp2 mp3 mp4 mpeg mpm-worker mpv mtp musepack mythtv ncurses nls nptl nptlonly nsplugin ogg oggvorbis openal opengl openmp openvdb pam pango pcre pda pdf pdflib phonon plasma png policykit posix ppds pulseaudio qml qt5 readline redistributable scanner sdl seccomp semantic-desktop session smp snmp sockets speech spell split-usr srt ssl startup-notification subtitles svg swat szip tcpd theora threads tiff truetype udev udisks unicode upower ups_drivers_usbhid-ups usb uvm v4l vcd vim-syntax vorbis webcam webkit widgets wxwidgets x264 x265 xanim xattr xcb xinerama xml xmp xv xvfb xvid zlib" ABI_X86="64" ADA_TARGET="gnat_2018" ALSA_CARDS="hda_intel usb_audio" APACHE2_MODULES="authn_core authz_core socache_shmcb unixd actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache cgi cgid dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" CALLIGRA_FEATURES="karbon sheets words" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" CPU_FLAGS_X86="aes avx avx2 f16c fma3 mmx mmxext pclmul popcnt rdrand sha sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock greis isync itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf skytraq superstar2 timing tsip tripmate tnt ublox ubx" GRUB_PLATFORMS="efi-64 pc" INPUT_DEVICES="libinput" KERNEL="linux" L10N="en en_US la" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIBREOFFICE_EXTENSIONS="pdfimport presenter-console presenter-minimizer" LUA_SINGLE_TARGET="lua5-1" LUA_TARGETS="lua5-1" OFFICE_IMPLEMENTATION="libreoffice" PHP_TARGETS="php7-3 php7-4" POSTGRES_TARGETS="postgres12 postgres13" PYTHON_SINGLE_TARGET="python3_9" PYTHON_TARGETS="python3_9" RUBY_TARGETS="ruby27 ruby30" SANE_BACKENDS="hp" USERLAND="GNU" VIDEO_CARDS="amdgpu radeonsi" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq proto steal rawnat logmark ipmark dhcpmac delude chaos account"
Unset: CC, CPPFLAGS, CTARGET, CXX, INSTALL_MASK, LC_ALL, PORTAGE_BINHOST, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS, RUSTFLAGS

=================================================================
Package Settings
=================================================================

sys-libs/glibc-2.33-r5::gentoo was built with the following:
USE="multiarch (multilib) ssp (static-libs) -audit -caps (-cet) -compile-locales (-crypt) -custom-cflags -doc -gd -headers-only -multilib-bootstrap -nscd -profile (-selinux) -static-pie -suid (-systemd) -systemtap -test (-vanilla)" ABI_X86="(64)"
CFLAGS="-march=x86-64 -mtune=bdver2 -pipe -O2"
CXXFLAGS="-march=x86-64 -mtune=bdver2 -pipe -O2"
FEATURES="usersync fixlafiles ebuild-locks protect-owned network-sandbox qa-unresolved-soname-deps sfperms unmerge-orphans userfetch strict preserve-libs multilib-strict unknown-features-warn news usersandbox merge-sync ipc-sandbox binpkg-multi-instance userpriv xattr config-protect-if-modified assume-digests binpkg-logs pid-sandbox distlocks binpkg-dostrip sandbox buildpkg binpkg-docompress unmerge-logs parallel-fetch"

[Phoenix591]

Try emerge -e system if you only did just emerge system before. -e will rebuild all of system's dependencies in addition to system.

Also, new with gcc 11, you can use -march=x86-64-v2 which is generic, but includes modern instructions like sse4.2 and some other modern ones

[cctsurf]

Try emerge -e system if you only did just emerge system before. -e will rebuild all of system's dependencies in addition to system.

Also, new with gcc 11, you can use -march=x86-64-v2 which is generic, but includes modern instructions like sse4.2 and some other modern ones

Thanks!
I did do an emerge -e system. I suppose I should have gone with "emerge -e @world"... I had the time away from the computer that it could have done it... Now I'm back to work, so I need the computer working... Good thing I have a backup installation of Manjaro...
I will keep x86-64-v2 in mind, much appreciated!

[cctsurf]

oof, interesting realization...
I thought I would identify the point of failure using "ebuild fetch, unpack, compile, install, package, qmerge"...
If I do it in the right order, I can install the errant packages with that method...hmmm.

[cctsurf]

Well... getting really close. Still need some help.
After rebuilding the whole -ev @world under chroot, I was finally able to get grub to install on the efi system (had difficulty with that after just the -ev @system rebuild), and boot into the gentoo installation! YAY!
As a side note, had to rebuild the genkernel to get it to boot correctly (genkernel all)... I'll compile a specific kernel for this system later.

I'm having difficulty building packages that rely on java. Libreoffice build failed on me, stating that

Code: Select all

configure: WARNING: You are building 64-bit binaries but the JDK /usr/lib64/openjdk-11/bin/java is 32-bit
configure: error: You should pass the --with-jdk-home option pointing to a 64-bit JDK

A quick "file /usr/lib64/openjdk-11/bin/java" showed it to be a 64-bit binary...
I thought a quick rebuild of openjdk might fix things...
that ended with

Code: Select all

configure: Found potential Boot JDK using configure arguments
configure: Potential Boot JDK found at /usr/lib64/openjdk-11 is incorrect JDK version (); ignoring
configure: (Your Boot JDK version must be one of: 10 11)
configure: error: The path given by --with-boot-jdk does not contain a valid Boot JDK
configure exiting with result code 1

I seem to need to rebuild my java system... but how do I do that?
(Edit) I installed openjdk-bin, uninstalled openjdk, and am now building openjdk. It seems to be building. Again, YAY! (/Edit)
Thanks!

[cctsurf]

With that, everything is rebuilt and I am up and running on my new hardware!
Thanks to everyone for their help and ideas!!