pacman has a dependency upon libarchive with a version recent enough to support zstd compression.

depends=(... 'libarchive>=3.3.3')

This is an issue when upgrading pacman to support the zstd packaging format.
pacman must depend upon libarchive>=3.3.3 for zstd compression support.
libarchive will then depend upon zstd>=1.3.6 for the symbol ZSTD_minCLevel.

Without these, upgrading an old Arch Linux system gives the incomplete - so unhelpful - error message:

error: could not open file /var/cache/pacman/pkg/something.pkg.tar.zst: Unrecognized archive format

How does pacman recognize an archive format? If the user does not already know the answer, they are in trouble. And, mistakenly circumventing the package compression to force an upgrade of pacman alone will lead to even more trouble, since pacman itself is not the source of the problem.

Again, we get into out-of-memory situations with rust:

pentium4
41:24.75   cargo:warning=src/vector_type.h:502:11: warning: ‘void* memcpy(void*, const void*, size_t)’ writing to an object of type ‘struct glsl::vec4’ with no trivial copy-assignment; use copy-assignment or copy-initialization instead [-Wclass-memaccess]
41:24.75   cargo:warning=  502 |     memcpy(&v, p, sizeof(v));
41:24.75   cargo:warning=      |     ~~~~~~^~~~~~~~~~~~~~~~~~
41:24.75   cargo:warning=In file included from src/gl.cc:78:
41:24.75   cargo:warning=src/glsl.h:1772:8: note: ‘struct glsl::vec4’ declared here
41:24.75   cargo:warning= 1772 | struct vec4 {
41:24.75   cargo:warning=      |        ^~~~
41:24.75   cargo:warning=
41:24.75   cargo:warning=cc1plus: out of memory allocating 21906724 bytes after a total of 213745664 bytes

i686
 6:42.75    Compiling geckoservo v0.0.1 (/build/firefox/src/firefox-87.0/servo/ports/geckolib)
Process Process-1:
Traceback (most recent call last):
  File "/usr/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
    self.run()
  File "/usr/lib/python3.9/multiprocessing/process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "/build/firefox/src/firefox-87.0/testing/mozbase/mozsystemmonitor/mozsystemmonitor/resourcemonitor.py", line 141, in _collect
    while not _poll(pipe, poll_interval=sleep_interval):
  File "/build/firefox/src/firefox-87.0/testing/mozbase/mozsystemmonitor/mozsystemmonitor/resourcemonitor.py", line 107, in _poll
    return pipe.poll(poll_interval)
  File "/usr/lib/python3.9/multiprocessing/connection.py", line 262, in poll
    return self._poll(timeout)
  File "/usr/lib/python3.9/multiprocessing/connection.py", line 429, in _poll
    r = wait([self], timeout)
  File "/usr/lib/python3.9/multiprocessing/connection.py", line 936, in wait
    ready = selector.select(timeout)
  File "/usr/lib/python3.9/selectors.py", line 416, in select
    fd_event_list = self._selector.poll(timeout)
KeyboardInterrupt
52:46.35 make[4]: *** [/build/firefox/src/firefox-87.0/config/makefiles/rust.mk:348: force-cargo-library-build] Interrupt
52:46.35 make[3]: *** [/build/firefox/src/firefox-87.0/config/recurse.mk:72: toolkit/library/rust/target] Interrupt
52:46.35 make[2]: *** [/build/firefox/src/firefox-87.0/config/recurse.mk:34: compile] Interrupt
52:46.35 make[1]: *** [/build/firefox/src/firefox-87.0/config/rules.mk:355: default] Interrupt

Rust is 1.51.0 on both pentium4 and i686.

this is a rather severe bug - it prevents GTK DMs, DEs, and programs from starting (lightdm, lxdm, lxde, leafpad)

the error is evident immediately in the pacman log - see: http://termbin.com/vezp

g_module_open() failed for /usr/lib/gdk-pixbuf-2.0/2.10.0/loaders/libpixbufloader-svg.so: /usr/lib/libpango-1.0.so.0: undefined symbol: g_memdup2

/usr/bin/gtk-query-immodules-3.0: symbol lookup error: /usr/lib/libpango-1.0.so.0: undefined symbol: g_memdup2

if you disable the DM service, you can get the same error by launching the DE directly

$ startlxde
/usr/bin/lxsession: symbol lookup error: /usr/lib/libpango-1.0.so.0: undefined symbol: g_memdup2

   Compiling same-file v1.0.6
     |                        ^^^
     = note: `-D non-fmt-panic` implied by `-D warnings`
     = note: this is no longer accepted in Rust 2021
help: add a "{}" format string to Display the message
     |
1493 |                 panic!("{}", out);
     |                        ^^^^^
help: or use std::panic::panic_any instead
     |
1493 |                 std::panic::panic_any(out);
     |                 ^^^^^^^^^^^^^^^^^^^^^^

Brilliantly fast changing language..

$ gedit
gedit: error while loading shared libraries: libtepl-5.so.0: cannot open shared object file: No such file or directory

Exist extra/tepl 6.00.0-1.0 in the pacman.
I’m not sure if gedit is old or can only be used with an older version of tepl, which is not possible because i get tepl from 3 days.

- openexr missing header files?
/usr/include/openvdb/Types.h:10:10: fatal error: OpenEXR/half.h: No such file or directory
   10 | #include <OpenEXR/half.h>
      |          ^~~~~~~~~~~~~~~~
compilation terminated.

openexr 2->3 upgrade breaks blender in openvdb

openvdb needs numpy, which needs cythonize, which segfaults.

-fstack-clash-protection
-fcf-protection

We have to investiagate if one of them (or both) are
introducing illegal opcodes for i486..

#0  0xb79a0c6d in raise () from /usr/lib/libc.so.6
#1  0xb79892c6 in abort () from /usr/lib/libc.so.6
#2  0xb79e449c in __libc_message () from /usr/lib/libc.so.6
#3  0xb7a6ff4c in __fortify_fail () from /usr/lib/libc.so.6
#4  0xb7a6ff19 in __stack_chk_fail () from /usr/lib/libc.so.6
#5  0xb7182004 in ?? ()
   from /usr/lib/python3.9/site-packages/numpy/core/_multiarray_umath.cpython-39-i386-linux-gnu.so
#6  0xb7179c87 in ?? ()
   from /usr/lib/python3.9/site-packages/numpy/core/_multiarray_umath.cpython-39-i386-linux-gnu.so
#7  0xb6fd589a in PyInit__multiarray_umath ()
   from /usr/lib/python3.9/site-packages/numpy/core/_multiarray_umath.cpython-39-i386-linux-gnu.so
#8  0xb7d07f35 in ?? () from /usr/lib/libpython3.9.so.1.0
#9  0xb7c37301 in ?? () from /usr/lib/libpython3.9.so.1.0
#10 0xb7c46897 in PyObject_Call () from /usr/lib/libpython3.9.so.1.0
#11 0xb7c2c557 in _PyEval_EvalFrameDefault () from /usr/lib/libpython3.9.so.1.0
#12 0xb7c260b9 in ?? () from /usr/lib/libpython3.9.so.1.0
#13 0xb7c36b05 in _PyFunction_Vectorcall () from /usr/lib/libpython3.9.so.1.0
#14 0xb7c2b4e5 in _PyEval_EvalFrameDefault () from /usr/lib/libpython3.9.so.1.0
#15 0xb7c36938 in _PyFunction_Vectorcall () from /usr/lib/libpython3.9.so.1.0
#16 0xb7c2794c in _PyEval_EvalFrameDefault () from /usr/lib/libpython3.9.so.1.0
#17 0xb7c36938 in _PyFunction_Vectorcall () from /usr/lib/libpython3.9.so.1.0
#18 0xb7c2774a in _PyEval_EvalFrameDefault () from /usr/lib/libpython3.9.so.1.0
#19 0xb7c36938 in _PyFunction_Vectorcall () from /usr/lib/libpython3.9.so.1.0
#20 0xb7c2774a in _PyEval_EvalFrameDefault () from /usr/lib/libpython3.9.so.1.0

Builds fine on i686 and pentium4, so I suspect a i486 issue in numpy.

 50/105 mutter:clutter+clutter/conform / actor-clone                                                   FAIL            0.13s   killed by signal 11 SIGSEGV
 55/105 mutter:clutter+clutter/conform / actor-layout                                                  FAIL            0.13s   killed by signal 11 SIGSEGV
 61/105 mutter:clutter+clutter/conform / actor-shader-effect                                           FAIL            0.13s   killed by signal 11 SIGSEGV
 70/105 mutter:clutter+clutter/conform / timeline                                                      FAIL            0.12s   killed by signal 11 SIGSEGV
 71/105 mutter:clutter+clutter/conform / timeline-interpolate                                          FAIL            1.63s   killed by signal 11 SIGSEGV
 73/105 mutter:clutter+clutter/conform / timeline-rewind                                               FAIL            0.12s   killed by signal 11 SIGSEGV
 83/105 mutter:core+mutter/stacking / closed-transient-no-input-parent-delayed-focus-default-cancelled FAIL            0.14s   killed by signal 11 SIGSEGV
 84/105 mutter:core+mutter/stacking / closed-transient-no-input-parents                                FAIL            0.13s   killed by signal 11 SIGSEGV
104/105 mutter:core+mutter/ref-test/sanity / ref-test-sanity                                           FAIL            0.13s   killed by signal 11 SIGSEGV

For now I’m just ignoring test results.

The go build fails with this error:

unsupported setting GO386=387. Consider using GO386=softfloat instead.

go tool dist: FAILED: /build/go/src/go/pkg/tool/linux_386/compile -std -pack -o /tmp/go-tool-dist-250139381/runtime/internal/atomic/_go_.a -p runtime/internal/atomic -importcfg /tmp/go-tool-dist-250139381/runtime/internal/atomic/importcfg -asmhdr /tmp/go-tool-dist-250139381/runtime/internal/atomic/go_asm.h -symabis /tmp/go-tool-dist-250139381/runtime/internal/atomic/symabis /build/go/src/go/src/runtime/internal/atomic/atomic_386.go /build/go/src/go/src/runtime/internal/atomic/stubs.go /build/go/src/go/src/runtime/internal/atomic/unaligned.go: exit status 1
go tool dist: open /tmp/go-tool-dist-250139381/runtime/internal/atomic/_go_.a: no such file or directory
go tool dist: open /tmp/go-tool-dist-250139381/runtime/internal/sys/_go_.a: no such file or directory
go tool dist: open /tmp/go-tool-dist-250139381/internal/cpu/_go_.a: no such file or directory
unsupported setting GO386=387. Consider using GO386=softfloat instead.

go tool dist: FAILED: /build/go/src/go/pkg/tool/linux_386/compile -std -pack -o /tmp/go-tool-dist-250139381/internal/cpu/_go_.a -p internal/cpu -importcfg /tmp/go-tool-dist-250139381/internal/cpu/importcfg -asmhdr /tmp/go-tool-dist-250139381/internal/cpu/go_asm.h -symabis /tmp/go-tool-dist-250139381/internal/cpu/symabis /build/go/src/go/src/internal/cpu/cpu.go /build/go/src/go/src/internal/cpu/cpu_386.go /build/go/src/go/src/internal/cpu/cpu_x86.go: exit status 1
unsupported setting GO386=387. Consider using GO386=softfloat instead.

go tool dist: FAILED: /build/go/src/go/pkg/tool/linux_386/compile -std -pack -o /tmp/go-tool-dist-250139381/runtime/internal/sys/_go_.a -p runtime/internal/sys -importcfg /tmp/go-tool-dist-250139381/runtime/internal/sys/importcfg -asmhdr /tmp/go-tool-dist-250139381/runtime/internal/sys/go_asm.h -symabis /tmp/go-tool-dist-250139381/runtime/internal/sys/symabis /build/go/src/go/src/runtime/internal/sys/arch.go /build/go/src/go/src/runtime/internal/sys/arch_386.go /build/go/src/go/src/runtime/internal/sys/intrinsics_common.go /build/go/src/go/src/runtime/internal/sys/intrinsics_stubs.go /build/go/src/go/src/runtime/internal/sys/stubs.go /build/go/src/go/src/runtime/internal/sys/sys.go /build/go/src/go/src/runtime/internal/sys/zgoarch_386.go /build/go/src/go/src/runtime/internal/sys/zgoos_linux.go /build/go/src/go/src/runtime/internal/sys/zversion.go: exit status 1
=⇒ ERROR: A failure occurred in build().

  Aborting...

Upstream discussion about this bug: https://github.com/golang/go/issues/44500

Someone mentions a fix, but it doesn’t apply for non SSE2 devices.

SPAM

SPAM

SPAM

SPAM

SPAM

The bug report has been sent upstream to GDBM directly. This is too important to just ignore the failure.. (make check || true)

no task description

no task description

make and guile made it half (and broken) to stable.
breaking the cycle required a shim package libffi33 in build-support.
packages depending of libffi are rescheduled and need to be rebuilt first, pushing libffi.so.8 to stable now would make
‘make/guile’ work again, but break a lot of other stuff!

 2:52.15 error: `[u32; 2]` is forbidden as the type of a const generic parameter
 2:52.15   --> /build/firefox/src/firefox-93.0/third_party/rust/packed_simd/src/codegen/llvm.rs:20:44
 2:52.15    |
 2:52.15 20 | pub unsafe fn __shuffle_vector2<const IDX: [u32; 2], T, U>(x: T, y: T) -> U
 2:52.15    |                                            ^^^^^^^^
 2:52.15    |
 2:52.15    = note: the only supported types are integers, `bool` and `char`
 2:52.15    = help: more complex types are supported with `#![feature(adt_const_params)]`
 2:52.15 error: `[u32; 4]` is forbidden as the type of a const generic parameter
 2:52.15   --> /build/firefox/src/firefox-93.0/third_party/rust/packed_simd/src/codegen/llvm.rs:30:44
 2:52.15    |
 2:52.15 30 | pub unsafe fn __shuffle_vector4<const IDX: [u32; 4], T, U>(x: T, y: T) -> U
 2:52.15    |                                            ^^^^^^^^
 2:52.15    |
 2:52.15    = note: the only supported types are integers, `bool` and `char`
 2:52.15    = help: more complex types are supported with `#![feature(adt_const_params)]`
 2:52.15 error: `[u32; 8]` is forbidden as the type of a const generic parameter
 2:52.15   --> /build/firefox/src/firefox-93.0/third_party/rust/packed_simd/src/codegen/llvm.rs:40:44
 2:52.15    |
 2:52.15 40 | pub unsafe fn __shuffle_vector8<const IDX: [u32; 8], T, U>(x: T, y: T) -> U
 2:52.15    |                                            ^^^^^^^^
 2:52.15    |
 2:52.15    = note: the only supported types are integers, `bool` and `char`
 2:52.15    = help: more complex types are supported with `#![feature(adt_const_params)]`
 2:52.15 error: `[u32; 16]` is forbidden as the type of a const generic parameter
 2:52.15   --> /build/firefox/src/firefox-93.0/third_party/rust/packed_simd/src/codegen/llvm.rs:50:45
 2:52.15    |
 2:52.15 50 | pub unsafe fn __shuffle_vector16<const IDX: [u32; 16], T, U>(x: T, y: T) -> U
 2:52.15    |                                             ^^^^^^^^^
 2:52.15    |
 2:52.15    = note: the only supported types are integers, `bool` and `char`
 2:52.15    = help: more complex types are supported with `#![feature(adt_const_params)]`
 2:52.15 error: `[u32; 32]` is forbidden as the type of a const generic parameter
 2:52.15   --> /build/firefox/src/firefox-93.0/third_party/rust/packed_simd/src/codegen/llvm.rs:60:45
 2:52.15    |
 2:52.15 60 | pub unsafe fn __shuffle_vector32<const IDX: [u32; 32], T, U>(x: T, y: T) -> U
 2:52.15    |                                             ^^^^^^^^^
 2:52.15    |
 2:52.15    = note: the only supported types are integers, `bool` and `char`
 2:52.15    = help: more complex types are supported with `#![feature(adt_const_params)]`
 2:52.15 error: `[u32; 64]` is forbidden as the type of a const generic parameter
 2:52.15   --> /build/firefox/src/firefox-93.0/third_party/rust/packed_simd/src/codegen/llvm.rs:70:45
 2:52.15    |
 2:52.15 70 | pub unsafe fn __shuffle_vector64<const IDX: [u32; 64], T, U>(x: T, y: T) -> U
 2:52.15    |                                             ^^^^^^^^^
 2:52.15    |
 2:52.15    = note: the only supported types are integers, `bool` and `char`
 2:52.15    = help: more complex types are supported with `#![feature(adt_const_params)]`
 2:52.35    Compiling num_cpus v1.13.0
 2:53.13    Compiling fluent-langneg v0.13.0
 2:53.15    Compiling intl_pluralrules v7.0.1
 2:53.59    Compiling intl-memoizer v0.5.1
 2:55.32    Compiling gl_generator v0.14.0
 2:55.37    Compiling cose-c v0.1.5
 2:55.96    Compiling time v0.1.43
 2:56.01    Compiling iovec v0.1.4
 2:56.05    Compiling atty v0.2.14
 2:56.21 For more information about this error, try `rustc --explain E0557`.
 2:56.23    Compiling net2 v0.2.37
 2:56.24 The following warnings were emitted during compilation:
 2:56.25 warning: Cannot set `RUSTC_BOOTSTRAP=1` from build script of `packed_simd v0.3.4 (https://github.com/hsivonen/packed_simd?rev=6a16f954950401b92b4e220fbf2dfaf6f00e1fb2#6a16f954)`.
 2:56.25 note: Crates cannot set `RUSTC_BOOTSTRAP` themselves, as doing so would subvert the stability guarantees of Rust for your project.
 2:56.25 error: could not compile `packed_simd` due to 13 previous errors
 2:56.25 warning: build failed, waiting for other jobs to finish...
 3:17.66 error: build failed

librsvg/rust fails because llvm is now 13 and we don’t have the llvm12-libs around anymore.
Will make a shim package.

subprocess.CalledProcessError: Command ‘[’/build/nss/src/dist/Release/bin/shlibsign’, ‘-v’, ‘-i’, ‘/build/nss/src/dist/Release/lib/libfreebl3.so’]’ died with <Signals.SIGSEGV: 11>.

After my most recent sysupgrade, postgresql refuses to start. The main postgres binary appears to be linked to multiple versions of libicu:

$  ldd /usr/bin/postgres | grep icu
	libicui18n.so.69 => /usr/lib/libicui18n.so.69 (0xb6c00000)
	libicuuc.so.69 => /usr/lib/libicuuc.so.69 (0xb6a0c000)
	libicuuc.so.70 => /usr/lib/libicuuc.so.70 (0xb6582000)
	libicudata.so.69 => /usr/lib/libicudata.so.69 (0xb4601000)
	libicudata.so.70 => /usr/lib/libicudata.so.70 

(0xb27a4000)

Installing the icu69 compatibility package made it work again, but it doesn’t look right that it should link against multiple versions.

I’m getting a “BadAccess” Xorg error. I think this is happening because the packaged version has been compiled against libraries which have meanwhile been upgraded.

Just to get to this point I had to install “libldap24” and “libffi7”.

It needs to be updated to reflect the switch to gitlab.archlinux.org

On i686, with testing and without testing claws mail is broken:
$ claws-mail
claws-mail: error while loading shared libraries: liblber-2.4.so.2: cannot open shared object file: No such file or directory

While I run Parabola i686, the repositories being used come from Archlinux 32:
$ pacman -sS claws-mail
extra/claws-mail 4.0.0-2.0 [installed]

  A GTK+ based e-mail client

$ pacman -Q -o /usr/lib/liblber*
/usr/lib/liblber.so is owned by libldap 2.6.0-2.4
/usr/lib/liblber.so.2 is owned by libldap 2.6.0-2.4
/usr/lib/liblber.so.2.0.200 is owned by libldap 2.6.0-2.4
$ pacman -sS libldap
core/libldap 2.6.0-2.4 [installed]

  Lightweight Directory Access Protocol (LDAP) client libraries

$ pacman -Q -l libldap | grep /usr/lib/liblber
libldap /usr/lib/liblber.so
libldap /usr/lib/liblber.so.2
libldap /usr/lib/liblber.so.2.0.200

So it’s most probably broken on Arch Linux 32 as well.

Denis.

affects: tracker-miners tracker3-miners gnome-documents gnome-music gnome-photos nautilus

no task description

no task description

After just having updated my trusty EeePC 1000H to be fully up to date again, I discovered that clementine is no longer working. It fails, stating that libprotobuf.so.27 cannot be found. Looking in /usr/lib, I can see libprotobuf.so.28, so I assume that the clementine package needs a rebuild.

I am currently unable to install smplayer in testing due to a broken dependency of mpv

[~]$ sudo pacman -S smplayer
resolving dependencies...
warning: cannot resolve "libplacebo.so=120-32", a dependency of "mpv"
warning: cannot resolve "mpv", a dependency of "smplayer"
:: The following package cannot be upgraded due to unresolvable dependencies:
      smplayer

:: Do you want to skip the above package for this upgrade? [y/N] y
looking for conflicting packages...
 there is nothing to do

no task description

no task description

Current version in /usr/lib is libboost_regex.so.1.78.0

Hi,

pango-list (and many packages depending on pango) fails to run with the following error:

$ pango-list 
pango-list: symbol lookup error: /usr/lib/libpangoft2-1.0.so.0: undefined symbol: hb_font_set_synthetic_slant

hb_font_set_synthetic_slant seems to be related to harfbuzz.

While I used Parabola i686 for the test, both pango and harfbuzz come from Arch Linux 32 i686, so Arch Linux 32 i686 should also be affected:

$ pacman -Q -o pango-list 
/usr/bin/pango-list is owned by pango 1:1.50.4-1.0
$ pacman -sS "^pango$"
extra/pango 1:1.50.4-1.0 [installed]
    A library for layout and rendering of text
$ pacman -sS "^harfbuzz$"
extra/harfbuzz 3.1.2-1.0 [installed]
    OpenType text shaping engine

Denis.

Linux and Linux-zen kernel 5.16.8.arch1-1.0 made with gcc 11.1 and cannot build dkms modules with gcc 11.2

remake linux and linux-zen kernel with gcc version 11.2

libswscale/x86/rgb2rgb_template.c: In function ‘interleaveBytes_sse2’:
libswscale/x86/rgb2rgb_template.c:1838:9: error: the register ‘xmm2’ cannot be clobbered in ‘asm’ for the current target
 1838 |         __asm__(
      |         ^

Also on i486 symbols are missing if not using LTO and probing of libraries like x264/x265 fail.

Hi,

If I run mpd, it fails to run due to a missing PW_LOG_TOPIC_DEFAULT symbol:
> $ mpd
> mpd: symbol lookup error: mpd: undefined symbol: PW_LOG_TOPIC_DEFAULT

While I run Parabola, both mpd and pipewire comes from the i686 repository of Arch Linux 32:
> extra/mpd 0.23.5-1.2 [installed]
> Flexible, powerful, server-side application for playing music
> extra/pipewire 1:0.3.36-1.0 [installed]
> Low-latency audio/video router and processor

If I look on a x86_64 Parabola installation instead, PW_LOG_TOPIC_DEFAULT can be found in libpipewire:
> $ readelf -s /usr/lib/libpipewire-0.3.so.0 | grep PW_LOG_
> 383: 00000000000d7128 8 OBJECT GLOBAL DEFAULT 21 > PW_LOG_TOPIC_DEFAULT

But that command returns nothing on Parabola i686.

Denis.

This is just the sad state of what people do to working software:

if(CMAKE_SYSTEM_PROCESSOR STREQUAL "i686" AND HAVE_STACKREALIGN_FLAG)
    add_compile_options(-mstackrealign)
endif()

which results in nasm to fail with “unknown option -m”

a) they rewrite half of the build system without apparently understanding fully the autoconf part
b) they don’t bother to delete to autoconf part
c) they write code for porting the software apparenty without testing (why bother then in the first place?)

# pacman -Sy ffmpeg
:: Synchronizing package databases… core is up to date
extra is up to date
community is up to date
resolving dependencies… warning: cannot resolve “libx264.so=163-32”, a dependency of “ffmpeg” :: The following package cannot be upgraded due to unresolvable dependencies:

    ffmpeg

This could be bad because our librsvg (the Rust one) lags behind and librsvg-og might not
implement new stuff (for i486).

scp -rC user@server:directory .
\033[?2004lD0711 0 directory

Enabling -vvv just makes scp hang.

No clue, what the problem could be here..