Re: [gentoo-user] patching an ebuild file

[ralfconn]

Il 17/05/24 20:41, Neil Bothwick ha scritto:

On Fri, 17 May 2024 20:17:14 +0200, Alarig Le Lay wrote:


I can edit the ebuild and then rebuild the manifest but on every
update I have to repeat.

Is there a way to patch an ebuild in a similar way we can patch
sources?

You can make an overlay and mask the pacakges from ::gentoo

No need to mask anything, just set the priority of your overlay higher
than that for gentoo. Otherwise you could end up not getting updates if
you don't check the gentoo repo regularly.


Creating the local overlay did not work, portage kept on pulling in the 
original ebuilds (from a public overlay themselves). Then I found that 
the modifications I needed could be done more simply via 
/etc/portage/package.accept_keywords, but that did not work either. Then 
I finally found that for my crossdev environment you need to edit 
/usr/aarch64-unknown-linux-gnu/etc/portage/package.accept_keywords. 
Probably the same holds for the local overlay but I have not verified yet.


thanks,

raf

[gentoo-user] patching an ebuild file

[ralfconn]

For my raspberry cross-compilation project I need to do simple 
modifications locally to some ebuilds.


I can edit the ebuild and then rebuild the manifest but on every update 
I have to repeat.


Is there a way to patch an ebuild in a similar way we can patch sources?

thanks,

raffaele

Re: [gentoo-user] Graphics configuration for a Ryzen 7 7700X chip and water cooling.

[ralfconn]

Il 16/05/24 20:46, Dale ha scritto:

Question.  How are the compiles times between the old FX-8350 and the
newer Ryzen 9?  I currently have a FX-8350.  Plan to build to a new
Ryzen something, maybe 5 at first.  Just curious what difference in
speed you see.
I've not saved the merge times for the 8350 so I'll only give you the 
Ryzen 9 times, maybe you can compare with yours:


# qlop -mav net-libs/webkit-gtk
net-libs/webkit-gtk-2.44.1-r410: 41′22″ average for 1 merge
net-libs/webkit-gtk-2.42.5-r410: 19′45″ average for 2 merges
net-libs/webkit-gtk-2.42.4-r600: 47′39″ average for 1 merge
net-libs/webkit-gtk-2.42.4-r410: 48′55″ average for 1 merge
net-libs/webkit-gtk-2.42.3-r410: 21′09″ average for 1 merge

# qlop -mav firefox
www-client/firefox-126.0: 31′35″ average for 1 merge
www-client/firefox-125.0.3: 13′29″ average for 1 merge
www-client/firefox-125.0.2: 12′42″ average for 1 merge
www-client/firefox-125.0.1: 30′18″ average for 1 merge

The 2x or more difference in merge times I believe are due to the fact 
that sometimes I build the bigger packages on their own to avoid running 
out of memory, other times I don't so the load gets split amongst 
various compilations and time stretches. I think firefox with the 8350 
was in the hours range, so I had switched to the -bin since long time.


I have 64Gb of RAM to account for the 12cpus/24threads. Even so I can 
run out of memory if I try to build firefox+thunderbird+webkit-gtk at 
the same time, so I often use the --exclude emerge option with these 
behemoths.


I've also had a Ryzen 7 5700X/32Gb for a short time, then I passed it to 
my son and got me the 9. These are the merge times for the webkit-gtk, I 
switched to non-bin firefox only with the 9:


# qlop -mav net-libs/webkit-gtk
net-libs/webkit-gtk-2.40.5-r410: 26′52″ average for 1 merge
net-libs/webkit-gtk-2.40.4-r410: 25′16″ average for 1 merge
net-libs/webkit-gtk-2.40.3-r410: 59′46″ average for 1 merge
net-libs/webkit-gtk-2.40.2-r410: 32′47″ average for 2 merges

Not a huge difference compared to the 9, as foreseeable, after all it's 
the exact same architecture with some more pepper.


If you go for the Ryzen remember that its instruction set is not 
compatible with the Athlon's so if you built your 8350 system with e.g. 
-march=native (as I did) you need to recompile @world with a less 
restrictive -march before moving the disk to the Ryzen system otherwise 
it won't even boot.


raf

Re: [gentoo-user] Graphics configuration for a Ryzen 7 7700X chip and water cooling.

[ralfconn]

Il 15/05/24 16:23, Alan Mackenzie ha scritto:

As a somewhat tangential question, would it be worthwhile getting water
cooling in my new machine?  In particular, to reduce the noise it gives
off while building large packages such as clang and rust?  Or is water
cooling only sensible for really heavy users such as gamers?

For a Ryzen 9 5900X (105W TDP) here I use a Noctua CPU cooler NH-U12A 
PWM plus a Noctua NF-A12x25 PWM on the case, a pretty expensive solution 
and probably an overkill since even while building for Gentoo @24 
threads the noise is audible but a LOT less than the old FX-8350 (125W 
TDP) with the stock Wraith cooler. During normal work it's almost 
inaudibile. I don't play games.


raf

Re: [gentoo-user] cross-compiling environment for Rapberry

[ralfconn]

Il 01/05/24 19:54, ralfconn ha scritto:

Il 01/05/24 19:05, Michael ha scritto:

I've built the cross toolchain with 'crossdev --taget

aarch64-unknown-linux-gnu' per wiki [2]. I've selected a 23.0-split-usr
profile on the Pi since this is the one supported by the crossdev 
(or so

I understood from the error message when I tried with the non-split-usr
profile on the desktop).
Did you download the correct stage3?  There are split-usr and 
merged-usr tar
archives.  Since split-usr is or will be soon deprecated you better 
start with

merged-usr.
I wasn't clear: when I tried to select the merged-usr profile on the 
host and on the target I got an error invoking 
emerge-aarch64-unknown-linux-gnu on the _host_  telling me that the 
profile was 'merged-usr' but the directory layout wasn't. So the 
problem is in the directory layout created by crossdev on the host.The 
host has a merged-usr profile, I believe the problem is in crossdev 
itself. I tried to manually copy files in the crossdev environment to 
recreate the merged-usr layout but then the error on the host was even 
worse (I forgot what it was) so I went back to split.


Update: stage3 on the pi is merge-usr so the above is bound to fail 
sooner or later (as I discovered trying to emerge 
app-misc/resolve-march-native on the pi). But, the sys-apps/merge-usr 
script accepts a --root parameter, so it is possible to issue, on the host:


# merge-usr --root /usr/aarch64-unknown-linux-gnu

# PORTAGE_CONFIGROOT=/usr/aarch64-unknown-linux-gnu eselect profile set 
default/linux/arm64/23.0


and have on the host the same directory layout and profile of the ARM 
stage3 installed on the pi.


raffaele

Re: [gentoo-user] cross-compiling environment for Rapberry

[ralfconn]

Il 01/05/24 19:05, Michael ha scritto:

I've built the cross toolchain with 'crossdev --taget

aarch64-unknown-linux-gnu' per wiki [2]. I've selected a 23.0-split-usr
profile on the Pi since this is the one supported by the crossdev (or so
I understood from the error message when I tried with the non-split-usr
profile on the desktop).

Did you download the correct stage3?  There are split-usr and merged-usr tar
archives.  Since split-usr is or will be soon deprecated you better start with
merged-usr.
I wasn't clear: when I tried to select the merged-usr profile on the 
host and on the target I got an error invoking 
emerge-aarch64-unknown-linux-gnu on the _host_  telling me that the 
profile was 'merged-usr' but the directory layout wasn't. So the problem 
is in the directory layout created by crossdev on the host.The host has 
a merged-usr profile, I believe the problem is in crossdev itself. I 
tried to manually copy files in the crossdev environment to recreate the 
merged-usr layout but then the error on the host was even worse (I 
forgot what it was) so I went back to split.

The crossdev environment on the desktop knows nothing about the packages
installed on the Pi, so I copied /var/db/pkg from Pi to
/usr/aarch64-unknown-linux-gnu/var/db/pkg on the desktop. I ran emerge
--sync on the Pi and on the desktop approximatively at the same time.

 Now If I try a @system update on the Pi I get: 

pihole ~ # emerge -uDvN --ask @system

Why didn't you try '--usepkg y', or '--usepkgonly y' to emerge the precompiled
binaries on the pi?
That is the intention, after I'll manage to set up a workflow to 
crossbuild binary packages on the host :-)

 What am I missing to make the desktop use the same list of packages
to be upgraded as the Pi? 

thanks,

raffaele

[1] https://pi-hole.net/
[2] https://wiki.gentoo.org/wiki/Crossdev

Check the selected make.profile and the stage3 fs are the same on both
buildhost and guest.  Also diff the two /var/lib/portage/world and don't
forget to compare/grep for different USE flags between the two systems.
/etc/portage/make.conf on the pi and in 
/usr/aarch64-unknown-linux-gnu/etc/portage/make.conf on the host are 
almost identical except for host-only feature (e.g. buildpkg is only 
specified on the host side) and number of emerge jobs.

Finally, if nothing works as intended consider compiling in QEMU:

https://wiki.gentoo.org/wiki/Embedded_Handbook/General/
Compiling_with_QEMU_user_chroot


For the moment I am trying a different approach: create a list of 
packages to be updated on the pi and use it as input to 
the emerge-aarch64-unknown-linux-gnu command on the host.


Wow, even an emerge --sync takes forever on the pi, I was no longer used 
to such long delays!


raffaele

[gentoo-user] cross-compiling environment for Rapberry

[ralfconn]

Hello,

I recently got me a Raspberry Pi4b to use as a PiHole [1]. As a first 
step I put user-space Gentoo (i.e. aarch64 stage3) on it and now I am 
trying to set up my desktop to cross-compile binary packages for the PI, 
to keep the Pi up-to-date in reasonable computing time.


I've built the cross toolchain with 'crossdev --taget 
aarch64-unknown-linux-gnu' per wiki [2]. I've selected a 23.0-split-usr 
profile on the Pi since this is the one supported by the crossdev (or so 
I understood from the error message when I tried with the non-split-usr 
profile on the desktop).


The crossdev environment on the desktop knows nothing about the packages 
installed on the Pi, so I copied /var/db/pkg from Pi to 
/usr/aarch64-unknown-linux-gnu/var/db/pkg on the desktop. I ran emerge 
--sync on the Pi and on the desktop approximatively at the same time.


 Now If I try a @system update on the Pi I get: 

pihole ~ # emerge -uDvN --ask @system

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

Calculating dependencies... done!
Dependency resolution took 31.91 s (backtrack: 0/20).

[ebuild U  ] sys-libs/ncurses-6.4_p20240414:0/6::gentoo 
[6.4_p20230401:0/6::gentoo] USE="cxx (split-usr*) (tinfo) -ada -debug 
-doc -gpm -minimal -profile (-stack-realign) -static-libs -test -trace 
-verify-sig" 7647 KiB
[ebuild U  ] sys-apps/baselayout-2.15::gentoo [2.14-r2::gentoo] 
USE="(split-usr*) -build" 30 KiB
[ebuild U  ] dev-util/re2c-3.1::gentoo [2.2::gentoo] USE="-debug 
-test" 1415 KiB
[ebuild U  ] dev-util/pkgconf-2.2.0:0/5::gentoo [2.1.1:0/4::gentoo] 
USE="(native-symlinks) -test" 305 KiB
[ebuild U  ] app-portage/elt-patches-20240415::gentoo 
[20240315::gentoo] 46 KiB
[ebuild   R    ] app-alternatives/awk-4::gentoo  USE="gawk (split-usr*) 
-busybox -mawk -nawk" 0 KiB
[ebuild   R    ] app-alternatives/bzip2-1::gentoo USE="reference 
(split-usr*) -lbzip2 -pbzip2" 0 KiB
[ebuild   R    ] app-alternatives/tar-0::gentoo  USE="gnu (split-usr*) 
-libarchive" 0 KiB
[ebuild   R    ] app-alternatives/gzip-1::gentoo  USE="reference 
(split-usr*) -pigz" 0 KiB
[ebuild U  ] dev-libs/libxml2-2.12.6:2::gentoo [2.12.5:2::gentoo] 
USE="ftp python readline -debug -examples -icu -lzma -static-libs -test" 
PYTHON_TARGETS="python3_11 -python3_10 -python3_12" 2582 KiB
[ebuild U  ] sys-devel/gettext-0.22.5::gentoo [0.22.4::gentoo] 
USE="acl cxx ncurses nls openmp* (xattr) -doc -emacs -git -java 
-static-libs -verify-sig" 10031 KiB
[ebuild  N ] app-arch/libarchive-3.7.4:0/13::gentoo USE="acl bzip2 
e2fsprogs iconv lzma xattr zstd -blake2 -expat -lz4 -lzo -nettle 
-static-libs -test -verify-sig" 5291 KiB
[ebuild U  ] sys-apps/shadow-4.14.6:0/4::gentoo [4.14.2:0/4::gentoo] 
USE="acl nls pam (split-usr*) xattr -audit -cracklib (-selinux) -skey 
-su -systemd -verify-sig" 1764 KiB
[ebuild  N ] app-crypt/rhash-1.4.4-r1:0/1::gentoo  USE="nls ssl 
-debug -static-libs" 426 KiB
[ebuild U  ] dev-python/platformdirs-4.2.1::gentoo [4.2.0::gentoo] 
USE="-test" PYTHON_TARGETS="python3_11 -pypy3 -python3_10 -python3_12" 
20 KiB
[ebuild  N ] dev-python/backports-tarfile-1.1.1::gentoo USE="-test" 
PYTHON_TARGETS="python3_11 -pypy3 -python3_10" 83 KiB
[ebuild U  ] dev-python/trove-classifiers-2024.4.10::gentoo 
[2024.3.25::gentoo] USE="-test" PYTHON_TARGETS="python3_11 -pypy3 
-python3_10 -python3_12" 16 KiB
[ebuild U  ] dev-python/jaraco-context-5.3.0::gentoo [5.1.0::gentoo] 
USE="-test" PYTHON_TARGETS="python3_11 -pypy3 -python3_10 -python3_12" 
14 KiB
[ebuild U  ] dev-python/jaraco-functools-4.0.1::gentoo 
[4.0.0::gentoo] USE="-test" PYTHON_TARGETS="python3_11 -pypy3 
-python3_10 -python3_12" 18 KiB
[ebuild U  ] sys-apps/coreutils-9.5::gentoo [9.4-r1::gentoo] 
USE="acl nls openssl (split-usr*) xattr -caps -gmp -hostname -kill 
-multicall (-selinux) -static -test -vanilla -verify-sig" 5873 KiB
[ebuild  NS    ] dev-build/autoconf-2.72-r1:2.72::gentoo 
[2.71-r6:2.71::gentoo] USE="-emacs -verify-sig%" 1358 KiB
[ebuild U  ] app-eselect/eselect-pinentry-0.7.4::gentoo 
[0.7.3::gentoo] 0 KiB
[ebuild U  ] dev-perl/XS-Parse-Keyword-0.410.0::gentoo 
[0.380.0::gentoo] USE="-test" 87 KiB
[ebuild U  ] sys-apps/kmod-32-r1::gentoo [31::gentoo] USE="lzma 
(split-usr%*) (tools) zlib zstd -debug -doc -pkcs7 -static-libs 
(-python%)" PYTHON_TARGETS="(-python3_10%) (-python3_11%*)" 500 KiB
[ebuild U  ] app-crypt/pinentry-1.3.0-r2::gentoo [1.2.1-r4::gentoo] 
USE="ncurses -X -caps -efl -emacs -gtk -keyring -qt5 (-qt6) -verify-sig 
-wayland" 597 KiB
[ebuild U  ] dev-lang/tcl-8.6.14:0/8.6::gentoo 
[8.6.13-r1:0/8.6::gentoo] USE="threads -debug" 6916 KiB
[ebuild  N ] dev-libs/libuv-1.48.0:0/1::gentoo USE="-verify-sig" 
1285 KiB
[ebuild U  ] dev-perl/Syntax-Keyword-Try-0.290.0::gentoo 
[0.280.0::gentoo] USE="-test" 45 KiB
[ebuild U  ] app-text/po4a-0.71::gentoo [0.69::gentoo] USE="-test" 
4747 KiB
[ebuild U  ] app-portage/portage-utils-0.97::gentoo [0.96.1::gentoo] 
USE="openmp* 

[gentoo-user] nullmailer error

[ralfconn]

Hello,

after a recent world update nullmailer stopped working. 
/var/log/nullmailer/nullmailer.log shows for every message send attempt:


"smtp: Failed: Error completing TLS handshake: The encryption algorithm 
is not supported."


Downgrading gnutls to 3.8.3 fixed the issue for me. I opened a bug 
(https://bugs.gentoo.org/930752).


Does anybody see the same?

raffaele

Re: [gentoo-user] How to synchronise between 2 locations

[ralfconn]

Il 28/03/24 07:30, J. Roeleveld ha scritto:

Unison creates a local index of all files it syncronised. So when you move a
file around on one end, Unison will notice that because the file at the new
location has the same hash as the file at the old location. As a result, it
does not transmit the file anew to the remote host, but instead copies it
locally on the remote host.

Since Unison uses ssh underneath, you can use ssh’s transparent compression
to speed up the transfer.

Unison sounds interesting. How does it handle conflicts (eg, file is changed on
both sides?)

I use Unison GUI on one of the two machines (on the other peer it's just 
a program invoked from the ssh). When the analysis is complete, the GUI 
shows what it would do to sync the machines, indicating the conflicts 
and giving you the chance to choose what to do.


I believe it can be used from the command line or maybe even in batch 
mode instead of GUI but I never did it that way.


raffaele

Re: [gentoo-user] How to synchronise between 2 locations

[ralfconn]

Il 27/03/24 19:58, J. Roeleveld ha scritto:

Hi all,

I am looking for a way to synchronise a filesystem between 2 servers. Changes
can occur on both sides which means I need to have it synchronise in both
directions.

Does anyone have any thoughts on this?

Also, both servers are connected using a slow VPN link, which is why I can't
simply access files on the remote server.

--
Joost

Unison. I use it just for that but can't say anything about the VPN bit, 
my servers are on local network.


raffaele

Re: [gentoo-user] New profile, gcc-13.2.1_p20240210 fails to build. ATTN: Peter Humphrey.

[ralfconn]

Il 25/03/24 08:04, Dale ha scritto:

Here is my update.  I wanted to skip the system update and change
profiles first.  Then do the emerge -e world which would also update
anything that was new as well.  I'd only have to compile once tho.
Well, that may have caused a problem.  It may work for some but it
didn't here.  I first had to do my usual emerge -auDN world and get a
clean run.  I had one build to fail, had to work on that.  Anyway, where
it says update first, it is best to do that.  It might work if you
don't, might not.  I'm now up to the part where I recompile everything.
Oh the joy.  At least it is a cool night so the extra heat will keep me
warm.  ROFL

As I noted in another thread, the profile switch involves a change in 
the LDFLAGS, with the addition of the 'pack-relative-relocs' option. 
https://rfc.archlinux.page/0023-pack-relative-relocs/ suggests that 
adding this flag might trigger issues with tools not up-to-date. I 
suppose that is why the gentoo devs instruct to rebuild binutils/gcc 
before 'emerge -e', and that might be the reason why your first attempt 
worked only partially.


BTW my update to split-usr 23-0 went pretty smoothly, next will be to go 
with the merged-usr.


raf

Re: [gentoo-user] New profiles 23.0

[ralfconn]

Il 23/03/24 20:37, ralfconn ha scritto:
In the meanwhile I tried to switch to my merged 23.0 profile, in step 
9 binutils updates fine while gcc builds but fails to install with no 
error message, so for now I'm back to 'merged' 17.1. Tomorrow I'll try 
to analyze the install log better.


Looks like the problem was self-induced. I run a KSPP kernel, one 
recommended setting blocks ptrace. Looking at yesterday's gcc failed 
merge log I found that the kernel had blocked several attempts to call 
ptrace. I disabled the option, gcc built fine and now I am running the 
last step (rebuild world).


raf

Re: [gentoo-user] New profiles 23.0

[ralfconn]

Il 23/03/24 20:18, Michael ha scritto:

On Saturday, 23 March 2024 19:10:28 GMT you wrote:

Il 23/03/24 19:43, Michael ha scritto:

On Saturday, 23 March 2024 18:29:58 GMT ralfconn wrote:

Il 23/03/24 18:42, Michael ha scritto:
   > I suggest it would be best to take heed of the devs hard work and

read the

   > instructions they have provided instead of winging it:
   >
   > https://wiki.gentoo.org/wiki/Project:Toolchain/23.0_update_instructio
   > ns

I'm currently running a local merged profile:

# cat /var/db/repos/local/profiles/no-multilib-hardened-desktop/parent
gentoo:default/linux/amd64/17.1/no-multilib/hardened
gentoo:targets/desktop

$ euse -a | grep usr
split-usr   [+  D F ]

I suppose that before step 3 of the wiki I'd need to create a new local
merged profile, e.g.:

#cat
/var/db/repos/local/profiles/23.0-split-usr-no-multilib-hardened-desktop/
par ent gentoo:default/linux/amd64/23.0/split-usr/no-multilib/hardened
gentoo:targets/split-usr/desktop

Does that make sense?

thanks

raf

Update portage and check the profiles offered by 'eselect profile list'.
For example, I can see:

[51]  default/linux/amd64/23.0/split-usr/no-multilib/hardened (stable)

which should provide what you're after.

It's not a 'desktop' profile

I'd think once you emerge/update the desktop environment or main packages you
want, most of the desktop related USE choices will be applied anyway.  In the
first instance I'd select the above profile ([51]), update your toolchain and
try an emerge --pretend of @world to see what USE flag differences remain.  If
this approach leaves you short you can always create your own merged profile
as you had done.


(Due to my mistake the last message was sent to me only instead of the list)

Unfortunately not, the non-desktop profile is very stripped-down 
compared to the desktop one. There's maybe 20 or more USE flags present 
in my merged profile that are missing from [51].


In the meanwhile I tried to switch to my merged 23.0 profile, in step 9 
binutils updates fine while gcc builds but fails to install with no 
error message, so for now I'm back to 'merged' 17.1. Tomorrow I'll try 
to analyze the install log better.


BTW, the only differences comparing emerge --info before and after the 
switch are:


LDFLAGS="-Wl,-O1 -Wl,--as-needed"

vs

LDFLAGS="-Wl,-O1 -Wl,--as-needed -Wl,-z,pack-relative-relocs"

and

USE = "cli fortran"

that after the switch disappear in favour of the new ones

USE = "lzma zstd"

(at least, in my 'crooked' local profile)

Re: [gentoo-user] New profiles 23.0

[ralfconn]

Il 23/03/24 18:42, Michael ha scritto:
> I suggest it would be best to take heed of the devs hard work and 
read the

> instructions they have provided instead of winging it:
>
> https://wiki.gentoo.org/wiki/Project:Toolchain/23.0_update_instructions
>
I'm currently running a local merged profile:

# cat /var/db/repos/local/profiles/no-multilib-hardened-desktop/parent
gentoo:default/linux/amd64/17.1/no-multilib/hardened
gentoo:targets/desktop

$ euse -a | grep usr
split-usr   [+  D F ]

I suppose that before step 3 of the wiki I'd need to create a new local 
merged profile, e.g.:


#cat 
/var/db/repos/local/profiles/23.0-split-usr-no-multilib-hardened-desktop/parent

gentoo:default/linux/amd64/23.0/split-usr/no-multilib/hardened
gentoo:targets/split-usr/desktop

Does that make sense?

thanks

raf

Re: [gentoo-user] gnucash will not generate a report

[ralfconn]

Il 11/03/24 23:44, Thelma ha scritto:

I tried to generate a report in GnuCash but I'm getting and empty page.

Can anybody confirm!

You don't specify which report, I tried 'Cash flow' and 'Assets over 
time' and they work regularly.


raf

$ eix -I gnucash
[I] app-office/gnucash
 Available versions:  4.13^t (~)5.1^t (~)5.3^t (~)5.4-r1^t (~)5.5^t 
{aqbanking debug doc examples gnome-keyring +gui keyring mysql nls ofx 
postgres python quotes register2 smartcard sqlite test 
PYTHON_SINGLE_TARGET="python3_10 python3_11 python3_12"}
 Installed versions:  5.5^t(20:17:46 01/21/24)(gui nls -aqbanking 
-debug -doc -examples -gnome-keyring -mysql -ofx -postgres -python 
-quotes -smartcard -sqlite -test PYTHON_SINGLE_TARGET="python3_11 
-python3_10 -python3_12")

Re: [gentoo-user] Emerge trouble with firefox and thunderbird ...

[ralfconn]

Il 10/03/24 23:44, Walter Dnes ha scritto:

On Sun, Mar 10, 2024 at 08:43:46PM +0100, ralfconn wrote


Given the warning message reported by Peter ("Enable USE=clang unless
you have a very good reason not to.")

   That message comes from sys-libs/compiler-rt which is a dedicated
runtime lib for clang.  It makes sense to use clang if you're building
the clang toolchain... just like using gcc if you're building the gcc
toolchain.



Correct, my mistake. My point was that it's worth filing a bug since 
it's possible that GCC support will be dropped upstream. But, the only 
'news' I found on this regard [1] is quite old and it looks like GCC is 
still being supported.


raf

[1] 
https://blog.mozilla.org/nfroyd/2018/05/29/when-implementation-monoculture-right-thing/

Re: [gentoo-user] Emerge trouble with firefox and thunderbird ...

[ralfconn]

Il 10/03/24 15:08, Peter Humphrey ha scritto:

On Sunday, 10 March 2024 07:17:27 GMT Walter Dnes wrote:


   So there are at least 2 people who've found out that Firefox can and
*MUST* be built with USE="-clang".

Ah. I'll change my USE flag straight away.

Thanks Walter.


This got me wondering, here I don't have clang in my make.config. The 
main packages using it are:


[I] www-client/firefox
 Installed versions:  123.0.1(rapid)(21:16:32 03/06/24)(X clang ...

[I] mail-client/thunderbird
 Installed versions:  115.8.1(21:14:19 03/07/24)(X clang ...

[I] app-office/libreoffice
 Installed versions:  7.6.5.2^t(21:42:53 03/06/24)(... -clang ...

libreoffice sees it unset because it is not present in the global uses. 
firefox and thunderbird instead set it in the ebuild:


$ grep clang /var/db/repos/gentoo/www-client/firefox/firefox-123.0.ebuild
IUSE="+clang cpu_flags_arm_neon dbus debug eme-free hardened hwaccel"

$ grep clang 
/var/db/repos/gentoo/mail-client/thunderbird/thunderbird-115.8.1.ebuild

IUSE="+clang cpu_flags_arm_neon dbus debug eme-free hardened hwaccel"

Both packages have no issues here with +clang. Given the warning message 
reported by Peter ("Enable USE=clang unless you have a very good reason 
not to.") and the fact that gentoo developers decided to switch it on 
specifically for these packages it would probably be a better idea to 
file a bugzilla rather than forcing the use of GCC, which might fix the 
issue now but lead to problems later. As usual, YMMV.


raf

Re: [gentoo-user] Emerge trouble with firefox and thunderbird ...

[ralfconn]

Il 03/03/24 10:47, Wols Lists ha scritto:

I'm getting this output from

emerge --update --newuse --deep --with-bdeps=y @world


Calculating dependencies... done!
 * Dependencies could not be completely resolved due to
 * the following required packages not being installed:
 *
 *   >=dev-libs/icu-73.1:0/73.1= pulled in by:
 * www-client/firefox-115.6.0
 *
 * Have you forgotten to do a complete update prior to depclean? The
 * most comprehensive command for this purpose is as follows:
 *
 *   emerge --update --newuse --deep --with-bdeps=y @world
 *
 * Note that the --with-bdeps=y option is not required in many
 * situations. Refer to the emerge manual page (run `man emerge`)
 * for more information about --with-bdeps.
 *
 * Also, note that it may be necessary to manually uninstall
 * packages that no longer exist in the repository, since it may not
 * be possible to satisfy their dependencies.
thewolery /usr/local/bin #

icu is at 74.2

firefox failed to update ...

*  www-client/firefox
  Latest version available: 115.8.0
  Latest version installed: 115.6.0
  Size of files: 496,244 KiB
  Homepage:  https://www.mozilla.com/firefox
  Description:   Firefox Web Browser
  License:   MPL-2.0 GPL-2 LGPL-2.1

as did thunderbird ...

*  mail-client/thunderbird
  Latest version available: 115.8.0
  Latest version installed: 115.6.0
  Size of files: 528,920 KiB
  Homepage:  https://www.thunderbird.net/
  Description:   Thunderbird Mail Client
  License:   MPL-2.0 GPL-2 LGPL-2.1

Andy ideas? Or is the mozilla emerge stuff slightly broken on my 
system? I've been having trouble with those two for the last few weeks 
...


Cheers,
Wol


Here I see:

[I] www-client/firefox
 Available versions:
 (esr)  115.7.0 115.8.0
 (rapid) (~)122.0.1 (~)123.0

You have 115.6.0 installed which apparently is out of tree. That may be 
confusing emerge. You could try to un-merge firefox, depclean and 
re-emerge it.


raf

Re: [gentoo-user] Re: gentoo-sources-6.7.7 build error

[ralfconn]

Il 02/03/24 18:17, Holger Hoffstätte ha scritto:
> On 2024-03-02 18:12, ralfconn wrote:
>> I have a build error with latest gentoo-sources (~amd64): # eix -I
>> gentoo-sources Installed versions: 6.7.6(6.7.6)^bs(10:37:31
>> 02/24/24)(-build -experimental -symlink) 6.7.7(6.7.7)^bs(18:43:20
>> 03/01/24)(-build -experimental -symlink) # eix -I gcc Installed
>> versions: 13.2.1_p20240210(13)^t(17:09:30 02/18/24)(cet cxx
>> default-stack-clash-protection default-znow fortran hardened nls
>> openmp pie sanitize ssp -ada -custom-cflags -d -debug -doc
>> -fixed-point -go -graphite -ieee-long-double -jit -libssp -lto
>> -modula2 -multilib -objc -objc++ -objc-gc -pch -pgo -systemtap -test
>> -valgrind -vanilla -vtv -zstd) fs/ntfs3/frecord.c: In function
>> ‘ni_read_frame’: fs/ntfs3/frecord.c:2460:16: error: unused variable
>> ‘i_size’ [-Werror=unused-variable]  2460 | loff_t i_size =
>> i_size_read(>vfs_inode);
>>
>> This is on a gentoo box with mixed profile
>> (local:no-multilib-hardened-desktop (dev)). The vanilla kernel
>> downloaded from kernel.org - with same config - has the same build
>> error. I thought it could be a hardened-related issue, but a
>> different gentoo box with an hardened-dektop profile builds fine. The
>> kernel config is different, though. By the way, looking at the
>> failing source it seems to me that the error is a false positive. I
>> suppose nobody else is seeing this, I see no reports on the net. Any
>> hints? Should I file a bug upstream? thanks, raffaele
>
> Already known:
> https://lore.kernel.org/stable/2024030214-scratch-compactly-638f@gregkh/
>
> So either set CONFIG_NTFS3_LZX_XPRESS=y or just wait.
>
>
Wow, faster than light, thanks!

In addition to the suggested fix, also disabling NTFS support completely 
works.


thanks again,

raffaele

[gentoo-user] gentoo-sources-6.7.7 build error

[ralfconn]

I have a build error with latest gentoo-sources (~amd64): # eix -I 
gentoo-sources Installed versions: 6.7.6(6.7.6)^bs(10:37:31 
02/24/24)(-build -experimental -symlink) 6.7.7(6.7.7)^bs(18:43:20 
03/01/24)(-build -experimental -symlink) # eix -I gcc Installed 
versions: 13.2.1_p20240210(13)^t(17:09:30 02/18/24)(cet cxx 
default-stack-clash-protection default-znow fortran hardened nls openmp 
pie sanitize ssp -ada -custom-cflags -d -debug -doc -fixed-point -go 
-graphite -ieee-long-double -jit -libssp -lto -modula2 -multilib -objc 
-objc++ -objc-gc -pch -pgo -systemtap -test -valgrind -vanilla -vtv 
-zstd) fs/ntfs3/frecord.c: In function ‘ni_read_frame’: 
fs/ntfs3/frecord.c:2460:16: error: unused variable ‘i_size’ 
[-Werror=unused-variable]  2460 | loff_t i_size = 
i_size_read(>vfs_inode);


This is on a gentoo box with mixed profile 
(local:no-multilib-hardened-desktop (dev)). The vanilla kernel 
downloaded from kernel.org - with same config - has the same build 
error. I thought it could be a hardened-related issue, but a different 
gentoo box with an hardened-dektop profile builds fine. The kernel 
config is different, though. By the way, looking at the failing source 
it seems to me that the error is a false positive. I suppose nobody else 
is seeing this, I see no reports on the net. Any hints? Should I file a 
bug upstream? thanks, raffaele

Re: [gentoo-user] dev-db/mysql-8.0.32-r2 and media-libs/opencv-4.8.1-r1 fails to emerge

[ralfconn]

Il 05/02/24 06:21, Dale ha scritto:

Howdy,

This is just a heads up.  I synced on Saturday, USA time, and was doing
my regular updates, weekly ones.  For some reason, mysql and opencv
would fail every time I attempted to upgrade.  I've had times where
something would fail on first attempt and just trying to update again
would allow it to finish.  I guess something was done out of order or
something.  Anyway, that didn't work this time.  I also searched BGO and
the forums with no one else having the problem.  After updating all I
could, I synced the tree again.  It's rare but sometimes one can sync
when the tree isn't quite complete.  I haven't seen that in a few years
but it is always on my list of things to try even tho it is so rare.  In
this case, syncing again was the solution.  I caught something in the
middle of a change or something and got something that wasn't quite up
to date.

TL;DR.  If the above fails, sync the tree again and retry.  It should
work then.  Time stamp for my original sync:  2024-02-03T20:37:56
Again, USA central time.

Hope this helps someone who runs into this and can't work around it.  :-D

Dale

:-)  :-)


It was a tracked opencv issue deriving from a dev-libs/protobuf update:

https://bugs.gentoo.org/909087

Probably you did not find it because by default the Bugzilla interface 
only searches for CONFIRMED, UNCONFIRMED, IN_PROGRESS. This one was 
fixed and closed at the speed of light, I found it yesterday only by 
chance because I did the search a short time before Comment 4...


I suppose that due to the urgency the developer did not change the 
ebuild version.


raf

Re: [gentoo-user] AMD microcode error?

[ralfconn]

Il 28/01/24 17:39, Peter Humphrey ha scritto:

Hello list,

For the first time ever, I received an mce error today:

[11473.528812] mce: [Hardware Error]: CPU 1: Machine Check: 0 Bank 14: 
9090909090909090
[11473.529657] mce: [Hardware Error]: TSC 0
[11473.530146] mce: [Hardware Error]: PROCESSOR 2:a20f10 TIME 1706457141 SOCKET 
0 APIC 2 microcode a201009

This is an AMD Ryzen M9 5900X.

Hits on the web suggest downgrading linux-firmware, which I've now done and
will await results. The latest upgrade was to version 20240115-r1, four days
ago.

Has anyone else experienced this?


No:

$ eix -I linux-firmware
[I] sys-kernel/linux-firmware
 Available versions:  (~)2023-r1^bstd 20231211^bstd 
20240115^bstd (~)20240115-r1^bstd ***l^bstd {compress-xz 
compress-zstd deduplicate initramfs +redistributable savedconfig 
unknown-license}
 Installed versions:  20240115-r1^bst(10:52:28 
01/27/24)(redistributable savedconfig -compress-xz -compress-zstd 
-deduplicate -initramfs -unknown-license)


$ grep -e "microcode\|model name" /proc/cpuinfo
model name    : AMD Ryzen 9 5900X 12-Core Processor
microcode    : 0xa20120e

raf

Re: [gentoo-user] downloading from cell phone to Gentoo

[ralfconn]

Il 18/01/24 04:28, Philip Webb ha scritto:

I want to be able to download photos from my new cellphone to Gentoo.
The phone is a Samsung A14 5G ; its pet name is Athene.
I use KDE to manage my desktops on my desktop machine ANB6.
...
Manjaro suggests soomething called 'MTP' is involved,
perhaps 'Media Transfer Protocol' (?).

Does anyone have any suggestions how to get this to work on Gentoo ?


No KDE nor GNOME here but this works in XFCE:

 $ eix -I gvfs
[I] gnome-base/gvfs
 Available versions:  1.50.6^t 1.52.1^t (~)1.52.2^t {afp archive 
bluray cdda elogind fuse gnome-online-accounts google gphoto2 +http ios 
keyring mtp nfs policykit samba systemd test +udev udisks zeroconf}
 Installed versions:  1.52.2^t(19:26:52 01/11/24)(cdda elogind fuse 
http mtp policykit udev udisks -afp -archive -bluray 
-gnome-online-accounts -google -gphoto2 -ios -keyring -nfs -samba 
-systemd -test -zeroconf)

 Homepage:    https://wiki.gnome.org/Projects/gvfs
 Description: Virtual filesystem implementation for GIO

Note the mtp USE which pulls in:

[I] media-libs/libmtp
 Available versions:  1.1.21(0/9) **(0/9)*l {+crypt doc 
examples static-libs}
 Installed versions:  1.1.21(0/9)(01:14:47 11/15/23)(crypt -doc 
-examples -static-libs)

 Homepage:    https://libmtp.sourceforge.net/
 Description: Implementation of Microsoft's Media Transfer 
Protocol (MTP)


The phone auto-mounts as soon as you switch Android to 'file transfer' 
and works well for me, less so for my daughter when she tries to 
transfer 10's of Gb of videos.


raf

Re: [gentoo-user] split-usr

[ralfconn]

Il 12/01/24 08:24, William Kenworthy ha scritto:
Some years back I did the usr-merge and my laptop has continued on 
more or less ok.


Now, I suddenly have a number of packages failing to build with 
internal collisions as they try and install (for example) a binary 
into /bin and /usr/bin and collide.  "emerge --info" is showing the 
split-usr flag enabled on my profile (5):


  [5]   default/linux/amd64/17.1/desktop (stable)
  [6]   default/linux/amd64/17.1/desktop/gnome (stable)
  [7]   default/linux/amd64/17.1/desktop/gnome/systemd (stable)
  [8] default/linux/amd64/17.1/desktop/gnome/systemd/merged-usr (stable)
  [9]   default/linux/amd64/17.1/desktop/plasma (stable)
  [10]  default/linux/amd64/17.1/desktop/plasma/systemd (stable)
  [11] default/linux/amd64/17.1/desktop/plasma/systemd/merged-usr 
(stable)

  [12]  default/linux/amd64/17.1/desktop/systemd (stable)
  [13]  default/linux/amd64/17.1/desktop/systemd/merged-usr (stable)

So which profile should I use for an XFCE4 desktop.  This laptop is 
openrc and predates the devils spawn (systemd) and I dont use gnome or 
plasma?


BillK

[5] is what I have been using here for years and I am also an 
XFCE/openrc addict. I recently switched to a merged 
desktop[5]/hardened[3] profile. split-usr is set but it's the first time 
I notice it and I don't know what the usr-merge is. No issues with 
merges. Not much help for your problem, I know.


raf

Re: [gentoo-user] emerge consistently 'hangs' on kernel/gentoo-sources

[ralfconn]

Il 17/12/23 15:17, Remco Rijnders ha scritto:

Hi all,

I've tried googling and didn't get very far so am afraid this might 
not be a

common thing...

Since a number of months, emerge hangs when doing a upgrade as soon as 
it gets

to the package gentoo-sources:

Installing (23 of 27) sys-kernel/gentoo-sources-6.1.67::gentoo

 * If you are upgrading from a previous kernel, you may be interested
 * in the following document:
 *   - General upgrade guide: https://wiki.gentoo.org/wiki/Kernel/Upgrade
 * For more info on this patchset, and how to report problems, see:
 * https://dev.gentoo.org/~mpagano/genpatches
WARNING: Will unset existing variable 'TMPDIR' to avoid clashing with 
genkernel config ...
WARNING: Will unset existing variable 'CHOST' to avoid clashing with 
genkernel config ...

* Gentoo Linux Genkernel; Version 4.3.6
* Using genkernel configuration from '/etc/genkernel.conf' ...
* Running with options: 
--kernel-config=/etc/kernels/kernel-config-6.1.19-gentoo-x86_64 all


* Working with Linux kernel 6.1.67-gentoo for x86_64
* Using kernel config file 
'/etc/kernels/kernel-config-6.1.19-gentoo-x86_64' ...

*
* Note: The version above is subject to change (depends on config and 
status of kernel sources).


* kernel: >> Initializing ...
* >> Running 'make mrproper' ...
* >> Running 'make oldconfig' ...
* >> Kernel version has changed (probably due to config 
change) since genkernel start:
*    We are now building Linux kernel 6.1.67-gentoo-x86_64 for 
x86_64 ...

* >> Compiling 6.1.67-gentoo-x86_64 bzImage ...
* >> Compiling 6.1.67-gentoo-x86_64 modules ...
* >> Installing 6.1.67-gentoo-x86_64 modules (and stripping) ...
* >> Generating module dependency data ...
* >> Compiling out-of-tree module(s) ...

It is consistent every time and at the same point. It has been stuck 
there for
24 hours now. Pressing Ctrl-C takes me back to the command line with 
no error
indication. Computer load is low, and this is the process tree for 
emerge:


root 19590  0.0  0.0   7916  4264 pts/5    S    09:15   0:00  
|   \_ bash
root 19936  0.6  1.9 322620 314152 pts/5   SN+  09:16   1:17  
|   \_ /usr/bin/python3.11 
/usr/lib/python-exec/python3.11/emerge --ask --update --newuse --de

ep @world
root  8996  0.1  1.8 324220 308860 pts/5   SN+  09:40   0:20  
|   \_ /usr/bin/python3.11 
/usr/lib/python-exec/python3.11/emerge --ask --update --newuse

--deep @world
root  9142  0.0  0.0  10892  7148 pts/5    SN+  09:40   0:00  
|   \_ bash /usr/lib/portage/python3.11/ebuild.sh 
postinst
root  9153  0.0  0.0  10892  5992 pts/5    SN+  09:40   0:00  
|   \_ bash 
/usr/lib/portage/python3.11/ebuild.sh postinst
root  9196  0.0  0.0  12724  9024 pts/5    SN+  09:40   0:00  
|   \_ /bin/bash /usr/bin/genkernel 
--kernel-config=/etc/kernels/kernel-config

-6.1.19-gentoo-x86_64 all
root 31236  0.0  0.4  71804 67416 pts/5    SN+  10:03   0:00  | \_ 
/usr/bin/python3.11 /usr/lib/python-exec/python3.11/emerge --ignore

-default-opts --buildpkg=n --usepkg=n --quiet-build=y @module-rebuild

Anyone has any ideas what might be wrong?

I don't use genkernel so just a guess. It looks like the kernel got 
build and installed but genkernel is stuck 'building out of tree 
module'. Maybe it is trying to fetch some code and not managing to? Try 
checking /var/log/emerge-fetch.log


raf

[gentoo-user] harfbuzz fails with KSPP

[ralfconn]

Hello,

media-libs/harfbuzz-8.3.0 failed to build on this KSPP [1] box. The 
issue is known and due to the use of PTRACE during the build [2].


I fixed as suggested in the post, but I still wonder why would a /build/ 
process want to call a PTRACE ?


thanks,

raffaele

[1] 
https://kernsec.org/wiki/index.php/Kernel_Self_Protection_Project/Recommended_Settings


[2]https://forums.gentoo.org/viewtopic-t-1120327-start-0.html

[gentoo-user] hardened vs -bin packages

[ralfconn]

Hello,

I recently switched to an hardened 'profile'. I have several packages 
installed as -bin to reduce compile time:


thunderbird-bin
firefox-bin
libreoffice-bin
rust-bin

I suppose I'd better use the non-bin version of at least the thunderbird 
and firefox ones, to take advantage of the hardened toolchain features 
for these internet-connected applications. I'm not so sure of  
libreoffice (which I use seldom and only for local documents) and rust. 
Opinions?


thanks,

raffaele

Re: [gentoo-user] hardened vs desktop

[ralfconn]

Il 13/11/23 14:22, Peter Böhm ha scritto:

Am Montag, 13. November 2023, 11:19:26 CET schrieb ralfconn:

Hello,

I've been running the desktop profile for years. Now I'm thinking to
switch to the hardened. Since there is no 'hardened desktop' profile,
the hint I found online is to note the current desktop USEs, switch to
hardened and add the USEs not found there, but I wonder if it is really
the best option. Comparing the two profiles, hardened seems a sub-set of
desktop with the addition of:

cet
hardened
pie
ssp
xtpax

It seems to me easier to add these to the desktop rather the other way
round. Any gotcha's I am missing?

Yes, you are missing that the best solution is: Make a new profile which
contains both profiles. See more here:

https://forums.gentoo.org/viewtopic-p-8694188.html#8694188

(And you have to start with a hardened stage3)
Looks like a good alternative, thanks. Following the post I created the 
local profile 'hardened-desktop' and confirmed the USEs are the 
combination of the two profiles. I suppose the added benefit of this new 
profile is that it will inherit the changes eventually done to the 
parent profiles by the gentoo developers, correct?

P.S.: Maybe read also the first note from this article:

https://wiki.gentoo.org/wiki/User:Pietinger/Tutorials/Kernel_Hardening_with_KSPP


Thanks, this requires a bit more of study on my side which I'll 
certainly do as a second step. BTW, hardened-sources is no longer 
available so KSPP might be the only option.


raffaele

[gentoo-user] hardened vs desktop

[ralfconn]

Hello,

I've been running the desktop profile for years. Now I'm thinking to 
switch to the hardened. Since there is no 'hardened desktop' profile, 
the hint I found online is to note the current desktop USEs, switch to 
hardened and add the USEs not found there, but I wonder if it is really 
the best option. Comparing the two profiles, hardened seems a sub-set of 
desktop with the addition of:


cet
hardened
pie
ssp
xtpax

It seems to me easier to add these to the desktop rather the other way 
round. Any gotcha's I am missing?


thanks

raffaele

Re: [gentoo-user] Network throughput from main Gentoo rig to NAS box.

[ralfconn]

On 9/23/23 14:04, Dale wrote:

Howdy,

As most everyone knows, I redone my NAS box.  Before I had Truenas on it
but switched to Ubuntu server thingy called Jimmy.  Kinda like the
name.  lol  Anyway, Ubuntu has the same odd transfer pattern as the
Truenas box had.  I'm not sure if the problem is on the Gentoo end or
the Ubuntu end or something else.  I'm attaching a picture of Gkrellm so
you can see what I'm talking about.  It transfers a bit, then seems to
stop for some reason, then start up again and this repeats over and
over.  I'm expecting more of a consistent throughput instead of all the
idle time.  The final throughput is only around 29.32MB/s according to
info from rsync.  If it was not stopping all the time and passing data
through all the time, I think that would improve.  Might even double.

...
Has anyone ever seen something like this and know why it is idle for so
much of the time?  Anyone know if this can be fixed so that it is more
consistent, and hopefully faster?

I found a similar pattern when I checked some time ago, while 
transferring big (several Gb) files from one desktop to the other. I 
concluded the cause of the gaps was the destination PC's SATA spinning 
disk that needed to empty its cache before accepting more data. In 
theory the network is 1Gb/s (measured with iperf, it is really close to 
that) and the SATA is 6Gb/s so it should not be the limit, but I have 
strong doubts as how this speed is measured by the manufacturer. As 
indirect confirmation, when I switched the destination disk to a flash 
one the overall network transfer speed improved a lot (a lot lot!).


raffaele

Re: [gentoo-user] Problem loading r8169 kernel module

[ralfconn]

On 8/5/23 15:40, Peter Humphrey wrote:

Hello list,

I decided to find out why the kernel had trouble loading the r8169 module on my
Intel NUC server. I found 
 which seemed to 
tell me all I needed to know.

I got this result after installing sys-apps/fwts and running it:

-
Test 1 of 2: PCIe ASPM ACPI test.
PCIe ASPM is not controlled by Linux kernel.

ADVICE: BIOS reports that Linux kernel should not modify ASPM settings that
BIOS configured. It can be intentional because hardware vendors identified some
capability bugs between the motherboard and the add-on cards.
-

The BIOS offered two choices:

PCIe ASPM support   enable/disable
Native ACPI OS PCIe Support enable/disable

I've tried all four combinations of those settings and got almost the same
result: the boot-up console output complained that the r8169 module couldn't
be loaded because it was already in the kernel. (I think that's just a coder's
assumption from the inability to load the module.)

The only tiny difference was in the fwts log: that PCIe ASPM was or was not
controlled by the kernel.

Is there a way forward from here, or should I just ignore it and get on with
life?

I have the same module here but built into the kernel (no loadable 
module support) with no problems.


lspci | grep -i eth
05:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 
2.5GbE Controller (rev 05)


Regarding ASPM the syslog reports:

acpi PNP0A08:00: _OSC: OS supports [ExtendedConfig ASPM ClockPM Segments 
MSI HPX-Type3]
pci :00:03.1: ASPM: current common clock configuration is 
inconsistent, reconfiguring


And for r8169:

Loading firmware: rtl_nic/rtl8125b-2.fw
RTL8226B_RTL8221B 2.5Gbps PHY r8169-0-500:00: attached PHY driver 
(mii_bus:phy_addr=r8169-0-500:00, irq=MAC)


What is the error you get loading the r8169 module?

raffaele

Re: [gentoo-user] new machine : root partition mounted RO

[ralfconn]

On 5/25/23 08:02, Philip Webb wrote:

The new machine I've built is working well in other ways,
I've gone thro' the 'change-root' steps,
compiled the kernel -- 6.1.27-gentoo-r1 -- ,
configured Lilo -- which has never let me down in  20 years  -- ,
created Fstab & rebooted.  No problem booting into the new system,
except that it insists on mounting the root filesystem 'read-only'.

It's only '/' : the other fs's listed in Fstab are mounted RW
& I can edit a test file & save it there ; if I try to do that in /etc ,
which is on the root partition, it refuses to save it.

Also, I can't create a user, as it can't alter  /etc .

Dmesg shows signs of something wrong ; the gist is

   nvme :03:00.0: platform quirk: setting simple suspend
   nvme nvme0: pci function :03:00.0

   nvme nvme0: Shutdown timeout set to 10 seconds
   nvme nvme0: 12/0/0 default/read/poll queues
   nvme0n1: p1 p2 p3 p4 p5 p6 p7 p8 p9 p10

   EXT4-fs (nvme0n1p2): orphan cleanup on readonly fs
   EXT4-fs (nvme0n1p2): mounted filesystem with ordered data mode.
Quota mode: none.
   VFS: Mounted root (ext4 filesystem) readonly on device 259:2.

   Adding 32767996k swap on /dev/nvme0n1p3.
Priority:-2 extents:1 across:32767996k SS
   EXT4-fs (nvme0n1p7): mounted filesystem with ordered data mode.
Quota mode: none.
   EXT4-fs (nvme0n1p8): mounted filesystem with ordered data mode.
Quota mode: none.
   EXT4-fs (nvme0n1p9): mounted filesystem with ordered data mode.
Quota mode: none.
   EXT4-fs (nvme0n1p10): mounted filesystem with ordered data mode.
Quota mode: none.

Those are the only mentions of a 'nvme' device (an M2 SSD).

As above, I've tested nvme0n1p10 & it's mounted RW.

I've tried 'fsck /dev/nvme0n1p2' & Gparted "check" (via System Rescue) ;
both report no problem very quickly :

   fsck : "clean, nn/nn files, nn/nn blocks".
   Gparted : "completed : calibrate ; check for errors + fix them ;
grow fs to fill system".

  lilo.conf  explicitly says 'read-write' twice, incl globally.

Could it be some kernel option ?  'grep' shows no 'read_only' or similar.

Can anyone offer advice ? -- please read my desciption above carefully.


root fs is mounted ro by the kernel and then remounted rw by the 
/etc/init.d/root openrc init script; I think the script uses fstab to 
find the rootfs. In my /var/log/rc.log I have:


root  | * Remounting root filesystem read/write ... 
[ ok ]


What does your rc.log say?

raf

Re: [gentoo-user] make[4]: *** No rule to make target

[ralfconn]

On 5/7/23 03:29, the...@sys-concept.com wrote:
Trying to compile "cifs" support in the kernel and getting an error 
during "make"


  CALL    scripts/checksyscalls.sh
  DESCEND objtool
make[4]: *** No rule to make target 
'/usr/lib/gcc/x86_64-pc-linux-gnu/11.3.0/include/stddef.h', needed by 
'/usr/src/linux-6.1.19-gentoo/tools/objtool/fixdep.o'.  Stop.
make[3]: *** [Makefile:47: 
/usr/src/linux-6.1.19-gentoo/tools/objtool/fixdep-in.o] Error 2
make[2]: *** 
[/usr/src/linux-6.1.19-gentoo/tools/build/Makefile.include:5: fixdep] 
Error 2

make[1]: *** [Makefile:73: objtool] Error 2
make: *** [Makefile:1442: tools/objtool] Error 2

Did you update GCC recently? What do 'gcc-config -l' and ' gcc 
--version' return in the terminal where you build the kernel?

[gentoo-user] source profile.env from non-interactive shell

[ralfconn]

Hello,

I have a bash script executed from cronie's anacron daemon which 
performs a local rsync backup and then emails the status via nullmailer 
. The latter part never worked until I understood that some 
nullmailer-inject options that I set via environment variables are not 
visible from the bash script when spawned by anacron (all was fine from 
an interactive shell). Adding a 'source /etc/profile.env' to the bash 
script solved the issue.


Are there any contraindication to sourcing the whole profile.env from a 
non-interactive shell? Potential security issues?


thanks,

raffaele

Re: [gentoo-user] NAS and replacing with larger drives

[ralfconn]

On 12/18/22 23:08, Dale wrote:

Mark Knecht wrote:



On Sun, Dec 18, 2022 at 12:20 PM Dale  wrote:

> root@truenas[~]# iperf -s
> 
> Server listening on TCP port 5001
> TCP window size: 64.0 KByte (default)
> 
>
>
> And nothing.  Several minutes later, still nothing.  And it 
continues to sit there.  I don't think it is working.  :/

>
> Still, odds are, whatever it is, I'm not likely going to be able to 
change it.  That poor old CPU just may not have the needed 
instruction set to be really fast for this.  Maybe a different 
encryption would be better.  I dunno.  It's temporary anyway.

>
> And still nothing.  It's been sitting there since I read the last 
message.  Still, nothing.

>
Sadly, you didn't read all of my instructions or apparently read the 
man page or help file


All you've done is start the server which listens for a connection.

Now go to your Gentoo Land machine that you want to backup and execute

iperf -c IP.ADDR.OF.SERVER

Wait 10 seconds and hit ctrl-C



Oh, I read that but didn't get that one worked with the other. Ooops.  
Thing is, I don't have the second command on my Gentoo install.  :/  
It sort of grumbles about that.  May look into that later.  Got other 
things in the air right now.


Dale

$ eix iperf
[I] net-misc/iperf
 Available versions:
 (2)    2.0.14a **2.*l
 (3)    3.12
   {debug ipv6 sctp threads}
 Installed versions:  3.12(3)(10:20:30 AM 10/08/2022)(-sctp)
 Homepage:    https://github.com/esnet/iperf
 Description: A TCP, UDP, and SCTP network bandwidth 
measurement tool

Re: [gentoo-user] Major problems with libpcre / UTF8

[ralfconn]

On 12/15/22 21:52, Walter Dnes wrote:
>    I just finished solving my babl problems, but more stuff shows up in
> libpcre.  First, here are my USE flags.  I don't see "utf8" anywhere.
>
> Calculating dependencies... done!
> [ebuild   R    ] dev-libs/libpcre-8.45-r1:3::gentoo USE="bzip2 cxx 
jit readline (split-usr) (unicode) zlib -libedit -pcre16 -pcre32 
-static-libs" 0 KiB

>
>    Can someone give me their output from "emerge -pv1 dev-libs/libpcre"
>
> mc (Midnight Commander) spews out a lot of...
>
> (mc:5796): GLib-CRITICAL **: 15:19:15.617: PCRE library is compiled 
without UTF8 support

>
>    Application windows have every second character missing in the window
> title; e.g. "xterm" shows as "xem".  I have a spreasheet that tracks my
> Presto card usage.  The title has gone from "presto.gnumeric-Gnumeric" to
> "pet.nmrc-Guei'.  Various PCRE-based searches fail in vaious apps.  HELP!


# emerge -pv1 dev-libs/libpcre

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

Calculating dependencies... done!
[ebuild   R    ] dev-libs/libpcre-8.45-r1:3::gentoo  USE="bzip2 cxx jit 
readline (split-usr) (unicode) zlib -libedit -pcre16 -pcre32 
-static-libs" ABI_X86="(64) -32 (-x32)" 1,542 KiB


# cat /etc/locale.gen
en_US ISO-8859-1
en_US.UTF-8 UTF-8
it_IT ISO-8859-1
it_IT.UTF-8 UTF-8

Re: [gentoo-user] where is 'mke2fs' ?

[ralfconn]

On 11/17/22 21:30, Philip Webb wrote:


I need to rewrite the filesystem on a USB stick after it got damaged.
My notes tell me to use 'mke2fs', but 'which' doesn't find it :
has it been replaced by something else ?
I have 'e2fsprogs' installed, but there's no 'man' file for it
& it doesn't seem to create filesystems.
There is a utility 'sys-fs/genext2fs', which I emerged & looked at,
but it's not clear whether it does the simple job I need.

Can anyone advise ?


You have the right package, mke2fs is inside it:

$ equery f sys-fs/e2fsprogs ... /sbin/mke2fs /sbin/mkfs.ext2 -> mke2fs 
/sbin/mkfs.ext3 -> mke2fs /sbin/mkfs.ext4 -> mke2fs


Here man page is installed
$ man mke2fs

And which/whereis also:

$ which mke2fs
/sbin/mke2fs

$ whereis mke2fs
mke2fs: /sbin/mke2fs /usr/share/man/man8/mke2fs.8.bz2

Re: [gentoo-user] Upgrading from 5.14 to 6.0 version

[ralfconn]

On 11/11/22 07:25, Dale wrote:

Howdy,

I been stuck on gentoo-sources 5.14.15 for a while.  I tried upgrading
to I think 5.16 and then more recently 5.18.  I upgraded like I always
do, copy .config over and run make oldconfig.  Once I get everything in
/boot, init thingy and all, I update grub.  When I get around to
rebooting, the new kernels always fail part way through booting.  I
can't recall the error since I last tried a newer kernel several months
ago.

I'm about to try to jump to version 6.0.5 which is latest in the tree.
Is there some major change that causes copying .config file from 5.14 to
5.18 or higher to break?  Do I need to configure a new kernel from
scratch in other words?  While I try to answer each question the best I
can, either I'm breaking something or something else breaks preventing
updating from older versions.  I just don't know which it is. Me or it.

I've had a similar experience recently. I was using LTS (5.15) due to 
old, no longer supported Nvidia video card, when I switched to nouveau I 
tried 6.0.x but no go, either USB3 or USB2 where working, but not both. 
After much fiddling with .config I ended up booting the PC from USB with 
a recent, working binary distribution - Ubuntu 22.04 - and compared its 
.config with mine, and found out the problem was in one of the kernel 
boot parameters, not in the .config!


If you find the binary distribution kernel works, lsmod will tell you 
which modules are actually used on your machine so you can pick just 
those in your config. For the .config comparison, meld works just fine.


raffaele

Re: [gentoo-user] x11-bas/xorg-x11 removal

[ralfconn]

On 10/29/22 12:09, David Haller wrote:

On Sat, 29 Oct 2022, ralfconn wrote:

Rather than adding the whole x11-base/xorg-apps and x11-base/xorg-fonts, I'd
like to build a list of the x11 utilities currently installed on my systems
to add them to world in preparation of the xorg-x11 package removal.

Is anybody aware of some ready to use tool for the purpose, without
re-inventing the wheel?

for c in x11-base/ x11-apps/; do
 eix '-I*' --format '' "$c"
done

Wonderfully simple, thank you

[gentoo-user] x11-bas/xorg-x11 removal

[ralfconn]

Rather than adding the whole x11-base/xorg-apps and x11-base/xorg-fonts, 
I'd like to build a list of the x11 utilities currently installed on my 
systems to add them to world in preparation of the xorg-x11 package removal.


Is anybody aware of some ready to use tool for the purpose, without 
re-inventing the wheel?


thanks,

raffaele

[gentoo-user] sandboxed Firefox

[ralfconn]

This weekend I tried Sakaki's excellent guide to running Firefox in a 
sandbox [1]. For the most part I religiously followed the guide, except:


- being on a desktop I was not too interested in setting up the bridge 
needed for WLAN; since it did not work out-of-the-box (possibly due to 
my config already running shorewall), I ditched that part and run 
'firejail --net=xxx ...'


- I'm running firefox-bin so I could not follow the USE=-gmp-autoupdate part

- minor stuff related to Xfce vs GNOME, and some no longer present USE flags

Seems to work fine, I'm just wondering if I should adapt my shorewall 
configuration to the different usage. My understanding is that Sakaki 
mainly uses it for the WLAN bridge and optionally to filter port 25 (for 
email spam prevention) so if I'm using a wired Ethernet I can safely 
skip this part, at least initially, anybody can confirm?


Anybody else has had experience with this way of sandboxing?

thanks,

raffaele

[1] 
https://wiki.gentoo.org/wiki/User:Sakaki/Sakaki's_EFI_Install_Guide/Sandboxing_the_Firefox_Browser_with_Firejail

[gentoo-user] ddclient 3.10 spamming?

[ralfconn]

Hello,

after a recent update I have ddclient continuously spamming me with 
warning messages, has anybody seen the same?


[I] net-dns/ddclient
 Available versions:  3.9.1-r2 (~)3.10.0_rc2^t {examples iproute2 
selinux}
 Installed versions:  3.10.0_rc2^t(10:15:22 AM 
10/08/2022)(-examples -selinux)


The relevant part of /etc/ddclient.conf:

use=web, web=myip.dnsomatic.com   \
server=updates.dnsomatic.com, \
protocol=dyndns2, \
login=,  \
password=   \


The email warnings:

FAILED:   updating : unexpected status (12)
SUCCESS:  updating : good: IP address set to xxx.xxx.xxx.xxx
FAILED:   updating : unexpected status (0)
FAILED:was not updated because protocol  is not supported.

It looks like the IP address is correctly set, but maybe not. Some logging:

# ddclient -daemon=0 -debug -verbose -noquiet

DEBUG:    proxy    = 
DEBUG:    protocol = https
DEBUG:    server   = myip.dnsomatic.com
DEBUG:    url  =
DEBUG:    ip ver   =
CONNECT:  myip.dnsomatic.com
CONNECTED:  using SSL
SENDING:  GET / HTTP/1.1
SENDING:  Host: myip.dnsomatic.com
SENDING:  User-Agent: ddclient/3.10.0_2
SENDING:  Connection: close
SENDING:
RECEIVE:  HTTP/1.1 429 Too Many Requests
RECEIVE:  Server: nginx
RECEIVE:  Date: Sun, 09 Oct 2022 09:37:52 GMT
RECEIVE:  Content-Type: text/html
RECEIVE:  Content-Length: 162
RECEIVE:  Connection: close
RECEIVE:  Strict-Transport-Security: max-age=15768000; includeSubDomains
RECEIVE:
RECEIVE:  
RECEIVE:  429 Too Many Requests
RECEIVE:  
RECEIVE:  429 Too Many Requests
RECEIVE:  nginx
RECEIVE:  
RECEIVE:  
DEBUG:    get_ip: using web, myip.dnsomatic.com reports 
WARNING:  : unable to determine IP address with strategy use=web
DEBUG:
DEBUG:    nic_dyndns2_update ---
Use of uninitialized value $_[0] in sprintf at /usr/bin/ddclient line 2136.
INFO: setting IP address to  for 
UPDATE:   updating 
DEBUG:    proxy    = 
DEBUG:    protocol = https
DEBUG:    server   = updates.dnsomatic.com
DEBUG:    url  = nic/update?
DEBUG:    ip ver   =
CONNECT:  updates.dnsomatic.com
CONNECTED:  using SSL
SENDING:  GET /nic/update?system=dyndns== HTTP/1.1
SENDING:  Host: updates.dnsomatic.com
SENDING:  Authorization: Basic X
SENDING:  User-Agent: ddclient/3.10.0_2
SENDING:  Connection: close
SENDING:
RECEIVE:  HTTP/1.1 200 OK
RECEIVE:  date: Sun, 09 Oct 2022 09:37:52 GMT
RECEIVE:  server: opendns
RECEIVE:  vary: Accept-Encoding
RECEIVE:  content-type: text/html; charset=UTF-8
RECEIVE:  x-envoy-upstream-service-time: 204
RECEIVE:  x-xss-protection: 1; mode=block
RECEIVE:  x-ingress-point: mil1
RECEIVE:  strict-transport-security: max-age=31536000; includeSubDomains
RECEIVE:  connection: close
RECEIVE:  transfer-encoding: chunked
RECEIVE:
RECEIVE:  12
RECEIVE:  good xxx.xxx.xxx.xxx
RECEIVE:  0
RECEIVE:
FAILED:   updating : unexpected status (12)
Use of uninitialized value $h in hash element at /usr/bin/ddclient line 
4077.
Use of uninitialized value $h in hash element at /usr/bin/ddclient line 
4079.
Use of uninitialized value $h in hash element at /usr/bin/ddclient line 
4080.

Use of uninitialized value $_[0] in sprintf at /usr/bin/ddclient line 2136.
SUCCESS:  updating : good: IP address set to 151.48.99.132
Use of uninitialized value $h in hash element at /usr/bin/ddclient line 
4077.

Use of uninitialized value $_[0] in sprintf at /usr/bin/ddclient line 2136.
FAILED:   updating : unexpected status (0)
Use of uninitialized value in string ne at /usr/bin/ddclient line 1130.
FAILED:    was not updated because protocol  is not supported.