| View previous topic :: View next topic |
| Author |
Message |
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
 Posted: Thu Sep 05, 2024 11:37 pm Post subject: A few Musl problems |
 |
|
LVM2 fail with this:
| Code: | clang -MT device/dev-lvm1-pool.o -MMD -MP -MF device/dev-lvm1-pool.d -c -I. -I.. -I../include -include configure.h -I/usr/include/blkid
-DHAVE_CONFIG_H -D_FILE_OFFSET_BITS=64 -O2 -pipe -fPIC device/dev-lvm1-pool.c -o device/dev-lvm1-pool.o
[CC] filesystem.c
[CC] online.c
[CC] parse_vpd.c
device/device_id.c:574:10: error: call to undeclared function 'basename'; ISO C99 and later do not support implicit
function declarations [-Wimplicit-function-declaration]
574 | base = basename(devname);
| ^
device/device_id.c:574:8: error: incompatible integer to pointer conversion assigning to 'const char *' from 'int'
[-Wint-conversion]
574 | base = basename(devname);
| ^ ~~~~~~~~~~~~~~~~~ |
elogind seem to fail with this:
| Code: | gind/sd-device -I../elogind-255.5/src/libelogind/sd-event -I../elogind-255.5/src/libelogind/sd-hwdb -I../elogind-255.5/src/libelogind/sd
-id128 -I../elogind-255.5/src/libelogind/sd-journal -Isrc/sleep -I../elogind-255.5/src/sleep -Isrc/update-utmp -I../elogind-255.5/src/up
date-utmp -Isrc/shared -I../elogind-255.5/src/shared -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -std=
gnu11 -O3 -Wno-missing-field-initializers -Wno-unused-parameter -Warray-bounds -Wdate-time -Wendif-labels -Werror=format=2 -Werror=impli
cit-function-declaration -Werror=implicit-int -Werror=incompatible-pointer-types -Werror=int-conversion -Werror=missing-declarations -We
rror=missing-prototypes -Werror=overflow -Werror=override-init -Werror=return-type -Werror=shift-count-overflow -Werror=undef -Wfloat-eq
ual -Winit-self -Wmissing-include-dirs -Wmissing-noreturn -Wnested-externs -Wold-style-definition -Wpointer-arith -Wredundant-decls -Wsh
adow -Wstrict-aliasing=2 -Wstrict-prototypes -Wunused-function -Wwrite-strings '-Wno-error=#warnings' -Wno-string-plus-int -fdiagnostics
-show-option -fno-common -fstack-protector -fstack-protector-strong --param=ssp-buffer-size=4 -Wno-typedef-redefinition -Wno-gnu-variabl
e-sized-type-not-at-end -ftrivial-auto-var-init=zero -Werror=shadow -O2 -pipe -fPIC -fno-strict-aliasing -fstrict-flex-arrays=1 -fvisibi
lity=hidden -ffunction-sections -fdata-sections -include config.h -pthread -fvisibility=default -MD -MQ src/libelogind/libelogind_static
.a.p/sd-journal_sd-journal.c.o -MF src/libelogind/libelogind_static.a.p/sd-journal_sd-journal.c.o.d -o src/libelogind/libelogind_static.
a.p/sd-journal_sd-journal.c.o -c ../elogind-255.5/src/libelogind/sd-journal/sd-journal.c
In file included from ../elogind-255.5/src/libelogind/sd-journal/sd-journal.c:32:
In file included from ../elogind-255.5/src/libelogind/sd-journal/journal-internal.h:13:
../elogind-255.5/src/libelogind/sd-journal/journal-file.h:80:21: error: field has incomplete type 'struct stat'
80 | struct stat last_stat;
| ^
../elogind-255.5/src/libelogind/sd-journal/journal-file.h:80:16: note: forward declaration of 'struct stat'
80 | struct stat last_stat;
| ^
1 error generated. |
ttf2pt1 seem to fail to a spectacular degree:
| Quote: | pt1.c:7069:72: warning: format specifies type 'unsigned int' but the argument has type 'struct gentry *' [-Wformat]
7069 | fprintf(stderr, "reverse path 0x%x <- 0x%x, 0x%x\n", ge, ge->prev, ge->bkwd);
| ~~ ^~~~~~~~
2 warnings generated.
2 warnings generated.
3 warnings generated.
6 warnings generated.
11 warnings generated.
42 warnings generated.
* ERROR: app-text/ttf2pt1-3.4.4-r2::gentoo failed (install phase):
* emake failedpt1.c:7069:72: warning: format specifies type 'unsigned int' but the argument has type 'struct gentry *' [-Wformat]
7069 | fprintf(stderr, "reverse path 0x%x <- 0x%x, 0x%x\n", ge, ge->prev, ge->bkwd);
| ~~ ^~~~~~~~
2 warnings generated.
2 warnings generated.
3 warnings generated.
6 warnings generated.
11 warnings generated.
42 warnings generated.
* ERROR: app-text/ttf2pt1-3.4.4-r2::gentoo failed (install phase):
* emake failed |
mesa-progs and wezterm ebuilds also fail, but seem less significant on my hardened musl setup that used to compile and run firefox just fine.
I removed firefox in attempts to solve this, but never showed up in this list.
And somehow I expect core applications to build on Gentoo Musl:
| Code: | *
* The following 5 packages have failed to build, install, or execute
* postinst:
*
* (sys-fs/lvm2-2.03.22-r5:0/0::gentoo, ebuild scheduled for merge), Log file:
* '/var/tmp/portage/sys-fs/lvm2-2.03.22-r5/temp/build.log'
* (sys-auth/elogind-255.5:0/0::gentoo, ebuild scheduled for merge), Log file:
* '/var/tmp/portage/sys-auth/elogind-255.5/temp/build.log'
* (app-text/ttf2pt1-3.4.4-r2:0/0::gentoo, ebuild scheduled for merge), Log file:
* '/var/tmp/portage/app-text/ttf2pt1-3.4.4-r2/temp/build.log'
* (x11-terms/wezterm-20240203.110809-r1:0/0::gentoo, ebuild scheduled for merge), Log file:
* '/var/tmp/portage/x11-terms/wezterm-20240203.110809-r1/temp/build.log'
* (x11-apps/mesa-progs-9.0.0:0/0::gentoo, ebuild scheduled for merge), Log file:
* '/var/tmp/portage/x11-apps/mesa-progs-9.0.0/temp/build.log'
* |
While mesa-progs-9.0.0 is my latest installed version, so some compiler or library is breaking things for me. |
|
| Back to top |
|
 |
Hu
Administrator
Joined: 06 Mar 2007
Posts: 22648
|
| Posted: Fri Sep 06, 2024 12:09 am Post subject: |
|
|
Are you seeking advice on how to correct these problems, or are you trying to get a Gentoo maintainer to correct them for you?
For the LVM2 problem, does that C file include the documented header for declaring basename? Does musl provide basename in the same header name as glibc does?
For the elogind problem, same question, but for the structure stat.
ttf2pt1 does not show any errors that can be addressed. That warning, while undesirable, is only a warning, and does not explain the failure. A quick search of bugs.gentoo.org reveals app-text/ttf2pt1-3.4.4-r2 fails to compile (CLANG-STRICTER-SYSTEM): ft.c:537:2: error: incompatible function pointer types initializing FT_Outline_MoveToFunc (aka int ()(const struct FT_Vector_ , void )) with an expression of type int (FT_Vector , void ). With no error message from your post, this is only a guess based on your apparent use of clang for LVM2. |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Fri Sep 06, 2024 12:26 am Post subject: |
|
|
| Hu wrote: | Are you seeking advice on how to correct these problems, or are you trying to get a Gentoo maintainer to correct them for you?
For the LVM2 problem, does that C file include the documented header for declaring basename? Does musl provide basename in the same header name as glibc does?
For the elogind problem, same question, but for the structure stat.
ttf2pt1 does not show any errors that can be addressed. That warning, while undesirable, is only a warning, and does not explain the failure. A quick search of bugs.gentoo.org reveals app-text/ttf2pt1-3.4.4-r2 fails to compile (CLANG-STRICTER-SYSTEM): ft.c:537:2: error: incompatible function pointer types initializing FT_Outline_MoveToFunc (aka int ()(const struct FT_Vector_ , void )) with an expression of type int (FT_Vector , void ). With no error message from your post, this is only a guess based on your apparent use of clang for LVM2. |
Thank you for not doing a "welcome back", as it is not that long ago I sucessfully recovered my account.
Second time around it seemed to involve another level of absurdity, given time perspectives involved.
I never got into programming and I somewhat concider it a personal fault, but my request is well within; a weirdo playing quake2 and quake3 on Gentoo in 2003, while helping people with 32bit chroots to run firefox.
My skills back then quickly went towards making ISDN work, and compiling a custom Linux kernel.
Following include lines and headers at a superificial level -- as my interest was always architecture:
Am I seeking help on how to learn C programming? Not very likely, but am I seeking interest?
The above stuff fail on my Gentoo Musl install, some fail spectacularly and parts might; or should be considered important.
https://forums.gentoo.org/search.php?search_author=zpiro
But that is in excess of 20 years, right? |
|
| Back to top |
|
|
GDH-gentoo
Veteran
Joined: 20 Jul 2019
Posts: 1699
Location: South America
|
| Posted: Fri Sep 06, 2024 5:50 pm Post subject: Re: A few Musl problems |
|
|
| zpiro wrote: | LVM2 fail with this:
| Code: | ...
device/device_id.c:574:10: error: call to undeclared function 'basename'; ISO C99 and later do not support implicit
function declarations [-Wimplicit-function-declaration]
574 | base = basename(devname);
| ^
device/device_id.c:574:8: error: incompatible integer to pointer conversion assigning to 'const char *' from 'int'
[-Wint-conversion]
574 | base = basename(devname);
| ^ ~~~~~~~~~~~~~~~~~ |
|
This is LVM2 using (I can't say if intentionally or unintentionally) the GNU version of basename() (available with GNU libc by including <string.h>, probably included indirectly in lib/device/device_id.c by other included headers), which takes an argument of type pointer to const char. Musl only provides the POSIX version (available by including <libgen.h>). Compounded by the fact that recent versions of GCC and Clang abort compilation when an undeclared function is called (see Modern C porting).
Requires a bug report.
| zpiro wrote: | elogind seem to fail with this:
| Code: | ...
In file included from ../elogind-255.5/src/libelogind/sd-journal/journal-internal.h:13:
../elogind-255.5/src/libelogind/sd-journal/journal-file.h:80:21: error: field has incomplete type 'struct stat'
80 | struct stat last_stat;
| ^
../elogind-255.5/src/libelogind/sd-journal/journal-file.h:80:16: note: forward declaration of 'struct stat'
80 | struct stat last_stat;
| ^
1 error generated. |
|
That's a bug, the declaration of struct stat is only guaranteed by POSIX to be available by including <sys/stat.h>, including <fcntl.h> may or may not make the declaration available. Seemingly, GNU libc makes it available if the program includes <fcntl.h>, but musl doesn't.
Requires a bug report.
| zpiro wrote: | ttf2pt1 seem to fail to a spectacular degree:
| Quote: | ...
* ERROR: app-text/ttf2pt1-3.4.4-r2::gentoo failed (install phase):
... |
|
This doesn't look like the verbatim output. The quoted line suggests that the ebuild is failing in the src_install() phase, not the src_compile() phase.
_________________
| NeddySeagoon wrote: | I'm not a witch, I'm a retired electronics engineer  |
| Ionen wrote: | | As a packager I just don't want things to get messier with weird build systems and multiple toolchains requirements though |
|
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Wed Sep 11, 2024 7:30 pm Post subject: |
|
|
Thank you!
Was exactly the kind of information I was looking for, and while I can attest to being able to follow include lines headers somewhat.
Now running what is probably yet again a fringe installation of Gentoo, where I was one of the first adopters of AMD64 branch by Opteraon 146 CPU system.
In an ecosystem of of Clang, GCC, glibc and Musl and what else is out there it is increasingly difficult to follow.
I am somewhat perplexed that Musl, and especially the hardened branch doesn't have more interest for its preference for embedded developers.
Where Gentoo historically has had many niche applications, like best distro for a fast DNS server.
On top of Musl being of interest in security research.
Musl on Gentoo isn't inherently a system administrator problem, it's more of paranoid professionals not wanting to show their hand.
And a personal anecdote, why did I play quake on Linux?
Back in the day the frames per second was 40% higher on OpenGL games and engines like those made by John Carmack.
Mouse interaction was also not equivalent, and felt different and important as a target dummy for Scandinavian pro gamers.
And 20 years ago, -CA-Pttern lost to All*Nirvana 40 to negative something.
I lost to -CA-Pettern as zpiro 40 to negative something.
These bug reports are more along the line of just fix it, users and lusers shouldn't see it as it gives an impression that nobody use Gentoo where it matters or that the code base isn't stable on behalf of build scripts.
All*Nirvana being the newphew of Simen Agdesten(IIRC), the chess player whom I visited at his home with cxz from Tromsø, confirmation of this account being mine is possible.
Patches exist for all of my build issues, somewhere in the world.
Unfortunately, it may be professional with a gag order for peripheral reasoning.
As programmers, or wannabe profesionals; if you can just follow trends and argue standards as some git, do it!
Last edited by zpiro on Wed Sep 11, 2024 9:23 pm; edited 1 time in total |
|
| Back to top |
|
|
Hu
Administrator
Joined: 06 Mar 2007
Posts: 22648
|
| Posted: Wed Sep 11, 2024 8:58 pm Post subject: |
|
|
| Although some Gentoo developers read the forums, and may fix problems as a result of what they read, the forums are in general not a good way of getting bugs fixed. We can help you determine whether something is a bug, and possibly how to fix it, but then it needs to be reported to have a good chance of it being fixed. |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Wed Sep 11, 2024 9:47 pm Post subject: |
|
|
| Hu wrote: | | Although some Gentoo developers read the forums, and may fix problems as a result of what they read, the forums are in general not a good way of getting bugs fixed. We can help you determine whether something is a bug, and possibly how to fix it, but then it needs to be reported to have a good chance of it being fixed. |
Albeit, this record will not be broken:
https://forums.gentoo.org/search.php?search_author=zpiro |
|
| Back to top |
|
|
pjp
Administrator
Joined: 16 Apr 2002
Posts: 20485
|
| Posted: Wed Sep 11, 2024 9:56 pm Post subject: |
|
|
This would be the correct link:
https://bugs.gentoo.org
_________________
Quis separabit? Quo animo? |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Wed Sep 11, 2024 10:24 pm Post subject: |
|
|
I'm no longer a profesional at CERN, whocm were put into a diffcultut position after discorvering a root kit.
And it didn't involve gentoo, but put me into a difficult position.
But convient down to public research and the more profilic comprises being an oscipiloscope running windows.
I did not work at CERN for my skills in language.
Everyone else worked at CERN for their linguistic abilities in maths and languages.
Do you expect that me or anyone I worked with would take the time out of usual acitivty to use that link?
As these things happen in the real world, and does not invite compilers problems and intersting programming standards.
No ill intent, but; in public research it isn't a big deal:
It may be a big deal that I argue this here, maybe you pin this thread?
But from student to CERN, I can attest to that; we got a gag order from the principal investagor.
I didn't really disagre, it was removed from our production systems.
And while many involved were able to correspond on compiler and programming problems.
While I could not care less, I wouldn't have used that link of yours for any lessoned learned. |
|
| Back to top |
|
|
Hu
Administrator
Joined: 06 Mar 2007
Posts: 22648
|
| Posted: Wed Sep 11, 2024 11:03 pm Post subject: |
|
|
Gentoo developers are more responsive to bug reports delivered via bugs.gentoo.org than to reports delivered in forum threads. If you would rather not report bugs in a way that gets them fixed, that is your choice, but then you are hoping that someone will spot the forum thread and fix it anyway.
I don't see how a root kit on a system at a place you no longer work has any relevance to our request that you use the proper venue for reporting a problem. |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Wed Sep 11, 2024 11:21 pm Post subject: |
|
|
| Hu wrote: | Gentoo developers are more responsive to bug reports delivered via bugs.gentoo.org than to reports delivered in forum threads. If you would rather not report bugs in a way that gets them fixed, that is your choice, but then you are hoping that someone will spot the forum thread and fix it anyway.
I don't see how a root kit on a system at a place you no longer work has any relevance to our request that you use the proper venue for reporting a problem. |
Gentoo has a history as old as I am around network security.
And seeing senior forum adminstators adressing this thread, I am assured and arrested.
D20 critical, fix the things as it is guaranteed to be reviewed.
So why? The response that wasn't an adminstaror addressed it directl.
So why doesn't he or any or any othter developer fix it?
I'm sure that lusers and users can contribute in some sense or other.
Someone that taught IBM System Z once told, for every option there is a user.
There is no way the user space change for unknown and known users.
Not sure about pipewire and all that stuff, but back in the day with OSS.
Recording was <, and plauong audio was >, to /dev/snd.
Gentoo breaks user space!
So why do I have the problems that I do?
Also, this forum thread follows as my first post in more than 20 years.
So why doesnt things change for Gentoo? It's not like end users of appliances will complain -- just me.
Do you expect me to run quake2 and quake3?
John Carmack may want to have a word, and well I agree completely if is about OSS versus Alsa.
I miss being able to pipe audio from unix shell, or bash, to /dev/snd and record in reverse.
Obviously, I run OpenRC given all of this.
And a minority of users that expect John Carmack to win for OSS vs Alsa, with personal experience.
And somehow, I suspect him to be running Gentoo around himself with a shared belief that that breaking user space is inherently supported and paid for.,
The ability to pipe adiuio, where > is recording, and > is output that back then always accepted wav format.
And Carmack and the qwuake franchine was upset, and widely covered on slashdot.
How is the systemd libraries covering and replacing the functionality?
My account is old enough, and everyone agrees!
That it is well within reason that that senionr and up and coming developers read it. |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Thu Sep 12, 2024 2:59 am Post subject: |
|
|
I am ranting on and on, but I had a bad feeling about this.
It grew somewhat less with qualifed input.
And I have serious qualms that come with limits of the absurdity possible.
Since 20 years ago:
https://ieeexplore.ieee.org/author/37088682708
About portage and programming after ontopic response, we can discuss topical relevance.
I leave you with this sentiment on behalf of my 20 year old forum account. |
|
| Back to top |
|
|
pjp
Administrator
Joined: 16 Apr 2002
Posts: 20485
|
| Posted: Thu Sep 12, 2024 5:23 am Post subject: |
|
|
| zpiro wrote: | | Do you expect that me or anyone I worked with would take the time out of usual acitivty to use that link? |
Since you posted here, yes, why not? The person that experiences the problem is most suited to helping resolve the issue. If you're not interested, then maybe it will have a similar affect as burning joss paper.
_________________
Quis separabit? Quo animo? |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Thu Sep 12, 2024 7:12 am Post subject: |
|
|
| pjp wrote: | | zpiro wrote: | | Do you expect that me or anyone I worked with would take the time out of usual acitivty to use that link? |
Since you posted here, yes, why not? The person that experiences the problem is most suited to helping resolve the issue. If you're not interested, then maybe it will have a similar affect as burning joss paper. |
I'm glad you asked!
Is there any way I could make this worse?
One thing is deduction, but the important part in mathematics is inductions and the really simple.
I hate the number line, the better definiton:
Point in one dometions, we define an alphabet usally.
Lets move on, so a point in a dimension definiton and alphabet.
Negative numbers are not what they seem, sometimes...
Point in one dimension, logic available to you is negative and positive.
Having learned how to count, what is left are numbers iummune to being negative and positive.
Sometimes it involve logic like addition and substraction, but its advanced.
Let's assume apples are apples, oranges are oranges and that I am not owed an explanation there.
Pseudoscience often involes minus and minus is possitive, but some semantics teend to be borrowed. |
|
| Back to top |
|
|
Hu
Administrator
Joined: 06 Mar 2007
Posts: 22648
|
| Posted: Thu Sep 12, 2024 4:52 pm Post subject: |
|
|
| zpiro wrote: | | And seeing senior forum adminstators adressing this thread, I am assured and arrested. |
Don't read too much into that. Most of my activity is done in a technical capacity, but the administrator badge is on every post I make, even if I'm just providing simple technical help. | zpiro wrote: | | D20 critical, fix the things as it is guaranteed to be reviewed. |
I don't understand this line. | zpiro wrote: | | So why? The response that wasn't an adminstaror addressed it directl. |
The response from GDH-gentoo was more technical than mine, but also directed you to file a bug report. GDH-gentoo is a knowledgeable user, but as evidenced from the lack of a developer badge, is not at this time a registered Gentoo developer, and so cannot fix the problem in gentoo.git on your behalf. | zpiro wrote: | | So why doesn't he or any or any othter developer fix it? |
What developer knows about the problem? I don't see any registered developers active in this thread, and no one has reported this as a bug. | zpiro wrote: | | I'm sure that lusers and users can contribute in some sense or other. |
Yes. Non-developers routinely contribute by advising users on technical issues, including how best to write a bug report. | zpiro wrote: | | Gentoo breaks user space! |
Gentoo passes through upstream's code. Although possible, it's usually not the case that a Gentoo developer broke something that worked with the upstream code. | zpiro wrote: | | So why do I have the problems that I do? |
I don't know. Perhaps you're using a configuration no one tested, so no one else knows it's broken. | zpiro wrote: | | Also, this forum thread follows as my first post in more than 20 years. |
I see that. Is your 20 year hiatus significant? We have users with very old accounts who have been active the entire time. | zpiro wrote: | | So why doesnt things change for Gentoo? It's not like end users of appliances will complain -- just me. |
You are complaining to people who have no direct ability to fix your problem. That is why I and others have repeatedly directed you to report your problem to someone who can fix it. | zpiro wrote: | | Do you expect me to run quake2 and quake3? |
No. Why would I expect that?
I don't understand any of the rest of your post. I will note however that your most recent post has drawn a report that the thread is wandering off-topic. Please stick to the issue at hand: various packages failing to build on your system. If you need help determining whether this is a package bug versus a local error, please ask. If you need help writing a better bug report, please ask. If you need help answering questions that were put to you in a bug report, please link to the bug report and tell us what you need. If you only want to rant that your desired setup doesn't already work, you will get nowhere. |
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Fri Sep 13, 2024 5:47 pm Post subject: |
|
|
| Hu wrote: | | zpiro wrote: | | And seeing senior forum adminstators adressing this thread, I am assured and arrested. |
Don't read too much into that. Most of my activity is done in a technical capacity, but the administrator badge is on every post I make, even if I'm just providing simple technical help. | zpiro wrote: | | D20 critical, fix the things as it is guaranteed to be reviewed. |
I don't understand this line. | zpiro wrote: | | So why? The response that wasn't an adminstaror addressed it directl. |
The response from GDH-gentoo was more technical than mine, but also directed you to file a bug report. GDH-gentoo is a knowledgeable user, but as evidenced from the lack of a developer badge, is not at this time a registered Gentoo developer, and so cannot fix the problem in gentoo.git on your behalf. | zpiro wrote: | | So why doesn't he or any or any othter developer fix it? |
What developer knows about the problem? I don't see any registered developers active in this thread, and no one has reported this as a bug. | zpiro wrote: | | I'm sure that lusers and users can contribute in some sense or other. |
Yes. Non-developers routinely contribute by advising users on technical issues, including how best to write a bug report. | zpiro wrote: | | Gentoo breaks user space! |
Gentoo passes through upstream's code. Although possible, it's usually not the case that a Gentoo developer broke something that worked with the upstream code. | zpiro wrote: | | So why do I have the problems that I do? |
I don't know. Perhaps you're using a configuration no one tested, so no one else knows it's broken. | zpiro wrote: | | Also, this forum thread follows as my first post in more than 20 years. |
I see that. Is your 20 year hiatus significant? We have users with very old accounts who have been active the entire time. | zpiro wrote: | | So why doesnt things change for Gentoo? It's not like end users of appliances will complain -- just me. |
You are complaining to people who have no direct ability to fix your problem. That is why I and others have repeatedly directed you to report your problem to someone who can fix it. | zpiro wrote: | | Do you expect me to run quake2 and quake3? |
No. Why would I expect that?
I don't understand any of the rest of your post. I will note however that your most recent post has drawn a report that the thread is wandering off-topic. Please stick to the issue at hand: various packages failing to build on your system. If you need help determining whether this is a package bug versus a local error, please ask. If you need help writing a better bug report, please ask. If you need help answering questions that were put to you in a bug report, please link to the bug report and tell us what you need. If you only want to rant that your desired setup doesn't already work, you will get nowhere. |
Parts of this where written plastered drunk, and include expressions that otherwise wouldn't feel helpful.
It is somewhat helpful that my account is this old, and I have offered a chain of trust that I am the account owner.
Because its very old... And you can check and verify without starting with higher principles and standards?
lead to AVX2 at CERN and paid for by Intel.
In either case, I have the ver least named dropped Ludovic of NixOS as a Gentoo user I may be able to request do a bug report for me.
Or well, be a git on git and push request patches.
D20 is a reference to dungeons and dragons, and borrow from historic arguments of Gentoo being known as the better option for a fast DNS server, so even in environments under work pressure runder enterprise alternatives this may still be done. And a bug report from someone like me getting some joy from new and flashy, and nobody here can truly know if Musl replace Glibc or what its role will be, but interest from security research and embedded developers is tremendous.
It may be preferred that people who are aware and knowledgeable about programming standards write bur reports and request git commits.
Which lend itself, not so much to the pleas to the community after losing the Gentoo Wiki that upset much of the user base that wrote aboout Desktop and application integration for esoteric and marginal interests and largely went to Arch Linux because it turned out that nobody whom had scraped the old Gentoo Wiki offered their offline copy of it, I certainly did not have one.
When iproute and other basic program do not work with Musl, I have very good reason to rant and rave.
As there are two types of Gentoo users, maybe three; those that solve problems with code, those that solve problems with infrastructure and those that trust neither coide, hardware or infrastrucutre.
The meticulous user that lojally contribute back is a unicorn that doesnt fit in the broader stereotypes.
As the general user is among the three, and look sideways and lack a platform to express their perspectives as to why any of the other two types of users do not grasp the oportunity.
I was able to build Firefox with Musl on Gentoo up until recently before iproute2 and lvm2 show up in failures.
As one of the first adopters of amd64 branch before 2003 helping people with 32bit chroots to run Firefox, this is perplexing; around the time when the founder of NixOS was among 10 users on IRC at Gentoo-amd64 discussing ideas for Ph.D.
Where the central point here, is "oh, you will fix and patch the build scripts, if and only if you a bug reports is made for lvm2 and iproute, involving C standards?".
Alas, it come down to a trivial degrees, that NixOS founder and Gentoo user may admit it, review of Gentoo built binaries is easier if trusted systems are a requirement.
Maybe or maybe not, perhaps I have no need to be offended by Musl branch going from stable to unstable.
As circlar argument, NixOs may never happened from any other Gentoo users, than the type that trust code.
It is implicit that Ph.D authoris behind NixOs and also Ceph, can be requted if they owe communities by my proxy.
I must be the typye of Gentoo user that undeterstand machine and isn't afraid of parallel logic.
With a lack a basis to trust social structures and hierarchies.
And I have trust issues, beyound questions if OpenBSD would be right? Seem like they are doing just fine, with a focus on writing code and possibly not reviewing it as much in lieu of the wider society.
Nothwisthanding that it looks absolutely beuatful and is wet dream of a code base, and there are not valid buts; Musl is morewidely studied, It isn't upstream unless its OpenSSH.
While do not know the name of the developrs that brought about the amd64 branch, and I'm not that sure it was a struggle as GCC was advanced and new inctructions were studied.
The only lunatic that has replied to me in communication, over frustrations and disrepair have been Richard Stallman after I worked with AVX instructions set.
Standing on the shoulders of Giants, compiling core applications from source.
Failing that, why wouldn't my more than 20 year old account be tied to CERN and a IEEE article as a cogntitive science student with speciliaztion in philosophy be such a massive mess?
https://ieeexplore.ieee.org/author/37088682708
As awkward as all of this is, I met the founder Powerterch at TG99, and the worest part isnt that someone calling himself Enemey_X through aquaintances wking at major network hub asked if he could run a quake2 server off my machine at TG99, I wasn't yet all that into gaming back then. Which is as about as bad as close call as someming weed on the roof of the highest building at CERN with a astrophysicists working out of Japan from the west, whom I could not resist by wearing a Gentoo cap, and was the most useless git imaginaeable given context, all he he could do to imrpove upon the context was to point at the adromeda galaxy, with knowledge of the big dipper, and maybe Gentoo astrophysicists cant be expected to deliver beyound this. Where the worst part is that I had diplomatic attache status by working full time at CERN, and I used it once during a work meeting. I walked through Geneva Airport security with a body language telling them to get screwed while explaining my case and perspective uninterrupted over phone, and shockingly profesionally at that level from border aithorities as if non-verbal communication was more than sufficient. And to assure and arrest precepts and epctations, no, It was attache status and only my car with aopriate license plates and a suitcase provided by me by my group leader would be without review of border control. With an ID card that ward full legal immunity in France and Switzerland in exeuction of duties as system administrator at CERN for ALICE. Heard a rumour that there has been a murder at CERN in its history dating back to who knows what, but I am sure the director of the time with class B diplomatic status and a peer of European primer ministers were able to solve it, with group laders equivalent to ambassadors at class C as an obserer at UN aty qual grounds with the Red Cross or higher for peace in cold war over nuclear physics etc. "I'm on the phone as Norwegian at CERN". While it is true that CERN is nowhere land, able to enter and exit from France and Switzerland at will, expemt from taxes with observer status UN and elevated above embassies in France and Switzlerland.
I am cross with you, and enough of my lawn turf to do this, and ensure that it involve oportunities towards a simulacra of a movie life.
Last edited by zpiro on Sat Sep 14, 2024 2:55 am; edited 25 times in total |
|
| Back to top |
|
|
GDH-gentoo
Veteran
Joined: 20 Jul 2019
Posts: 1699
Location: South America
|
| Posted: Fri Sep 13, 2024 8:04 pm Post subject: |
|
|
zpiro, answer this: do you or do you not understand that bugs have a higher chance of getting fixed if a bug report is filed at https://bugs.gentoo.org?
Only developers can fix stuff in Gentoo's official repository, and they prefer the bug tracker because, well, bugs can be tracked that way. It is preferred that people who experience the problems file the bugs, because it's them who can later test any proposed fixes, or provide more information about the failure if asked. There's now three people, including myself, who have tried to explain this.
Every post of yours after the first one is mostly incomprehensible at least for me, and the time that you spent writing them could have been used instead for filing bugs for at least sys-fs/lvm2 and sys-auth/elogind. Non-developers can help by adding comments once the bugs have been filed.
I fail to see how the fact that you registered here more that 20 years ago is relevant.
_________________
| NeddySeagoon wrote: | | I'm not a witch, I'm a retired electronics engineer |
| Ionen wrote: | | As a packager I just don't want things to get messier with weird build systems and multiple toolchains requirements though |
|
|
| Back to top |
|
|
zpiro
n00b
Joined: 26 Jun 2002
Posts: 26
Location: Norway
|
| Posted: Sat Sep 14, 2024 3:36 am Post subject: |
|
|
| GDH-gentoo wrote: | zpiro, answer this: do you or do you not understand that bugs have a higher chance of getting fixed if a bug report is filed at https://bugs.gentoo.org?
Only developers can fix stuff in Gentoo's official repository, and they prefer the bug tracker because, well, bugs can be tracked that way. It is preferred that people who experience the problems file the bugs, because it's them who can later test any proposed fixes, or provide more information about the failure if asked. There's now three people, including myself, who have tried to explain this.
Every post of yours after the first one is mostly incomprehensible at least for me, and the time that you spent writing them could have been used instead for filing bugs for at least sys-fs/lvm2 and sys-auth/elogind. Non-developers can help by adding comments once the bugs have been filed.
I fail to see how the fact that you registered here more that 20 years ago is relevant. |
Before I quit Warwick University after having my temporary contract changed into a permanent one agianst my will.
I achieved 2-by-2 Linpack effiency above 98% and well anove Lenovo official numbers.
I did not file a bug, from my hardware insight that the resulst where founded in running CentOS that had left its maintenance windows and were considered deprecated due to the generation of Infiniband involved and the advertised new interconnects for HPC.
And a message to be conveyed, is that my type of user will not, and will often fall between chairs.
So why do you not find any interest in progamming standards, and expect that it will be revealed through bug reports?
And to make it personal, who is goatface that smoked weed with me on the roof of CERN? |
|
| Back to top |
|
|
pjp
Administrator
Joined: 16 Apr 2002
Posts: 20485
|
| Posted: Sat Sep 14, 2024 4:05 am Post subject: |
|
|
This thread no longer seems to have a purpose.
Whatever is presented as an issue is here, and the posts seem to have little to do with resolving it.
_________________
Quis separabit? Quo animo? |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Portage & Programming
