You are not logged in.
Pages: 1
When I run "recoll", I recive ":3:../common/rclinit.cpp:410::Recoll 1.39.1 + Xapian 1.4.25 [/home/papas/.recoll]".
Has anyone come across this?
Last edited by kafigitis (2024-07-04 16:34:59)
Offline
What do you mean when you say that it does not work?
When I launch the program I also get the same output (and more), but the window opens as expected:
$ recoll
:3:../common/rclinit.cpp:410::Recoll 1.39.1 + Xapian 1.4.25 [/home/chris/.recoll]
Path override failed for key base::DIR_APP_DICTIONARIES and path '/usr/bin/qtwebengine_dictionaries'
Path override failed for key base::DIR_APP_DICTIONARIES and path '/usr/lib/qt6/qtwebengine_dictionaries'
Path override failed for key base::DIR_APP_DICTIONARIES and path '/usr/lib/qt6/qtwebengine_dictionaries'
:2:../rcldb/rcldb.cpp:1038::Db::open: exception while opening [/home/chris/.recoll/xapiandb]: Couldn't stat '/home/chris/.recoll/xapiandb'
:2:../rcldb/rclterms.cpp:598::Db::dirlist: exception while accessing index: Couldn't stat '/home/chris/.recoll/xapiandb'
:2:../rcldb/rcldb.cpp:1038::Db::open: exception while opening [/home/chris/.recoll/xapiandb]: Couldn't stat '/home/chris/.recoll/xapiandb'
:2:../rcldb/rcldb.cpp:1038::Db::open: exception while opening [/home/chris/.recoll/xapiandb]: Couldn't stat '/home/chris/.recoll/xapiandb'
:3:../common/rclconfig.cpp:1834::RclConfig: pid/lock file: /run/user/1001/recoll-1bd02995df75c9288772966339e62137-index.pidOffline
I am also affected:
recoll
qt.qpa.plugin: Could not find the Qt platform plugin "wayland" in ""
:3:../common/rclinit.cpp:410::Recoll 1.39.1 + Xapian 1.4.25 [/home/lavilao570/.recoll]
Illegal instruction (core dumped)Offline
Could one of you share the output of "cat /proc/cpuinfo | grep flags | sort --unique"?
It seems like the recoll app contains sse{,2} instructions:
$ bin-cpuflags-x86 /usr/bin/recoll
Format: Elf
Architecture: X86_64
Features: INTEL8086 INTEL186 INTEL386 INTEL486 X64 CET_IBT CMOV MULTIBYTENOP SSE SSE2 Offline
sure, here it is:
cat /proc/cpuinfo | grep flags | sort --unique
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq vmx ssse3 cx16 pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave rdrand hypervisor lahf_lm 3dnowprefetch cpuid_fault pti ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust smep erms mpx rdseed smap clflushopt sha_ni xsaveopt xsavec xgetbv1 xsaves arat vnmi umip rdpid md_clear arch_capabilities
vmx flags : vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid shadow_vmcs pml tsc_scaling Offline
I also have trouble with new recoil. Same error. Log after journalctl -r
lip 05 09:16:22 arch-server systemd-coredump[3333]: [?] Process 3331 (recollq) of user 1000 dumped core.
Stack trace of thread 3331:
#0 0x00007c64db1dcfe6 _ZNK7simdutf8internal29available_implementation_list21detect_best_supportedEv (librecoll.so.39 + 0x1dcfe6)
#1 0x00007c64db1dd40f _ZNK7simdutf8internal49detect_best_supported_implementation_on_first_use8set_bestEv (librecoll.so.39 + 0x1dd40f)
#2 0x00007c64db1dd4d7 n/a (librecoll.so.39 + 0x1dd4d7)
#3 0x00007c64db276f67 _Z9transcodeRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEERS4_S6_S6_Pi (librecoll.so.39 + 0x276f67)
#4 0x00007c64db152041 _Z7recollqPP9RclConfigiPPc (librecoll.so.39 + 0x152041)
#5 0x00007c64dae39c88 n/a (libc.so.6 + 0x25c88)
#6 0x00007c64dae39d4c __libc_start_main (libc.so.6 + 0x25d4c)
#7 0x00005a590a8c4085 n/a (recollq + 0x1085)
ELF object binary architecture: AMD x86-64
lip 05 09:16:22 arch-server systemd[1]: Started Process Core Dump (PID 3332/UID 0).
lip 05 09:16:22 arch-server systemd[1]: Created slice Slice /system/systemd-coredump.
lip 05 09:16:22 arch-server systemd-coredump[3332]: Process 3331 (recollq) of user 1000 terminated abnormally with signal 4/ILL, processing...
lip 05 09:16:22 arch-server kernel: traps: recollq[3331] trap invalid opcode ip:7c64db1dcfe6 sp:7ffd3e16f1c0 error:0 in librecoll.so.39.1.0[7c64db04d000+245000] Offline
What do you mean when you say that it does not work?
When I launch the program I also get the same output (and more), but the window opens as expected:
$ recoll :3:../common/rclinit.cpp:410::Recoll 1.39.1 + Xapian 1.4.25 [/home/chris/.recoll] Path override failed for key base::DIR_APP_DICTIONARIES and path '/usr/bin/qtwebengine_dictionaries' Path override failed for key base::DIR_APP_DICTIONARIES and path '/usr/lib/qt6/qtwebengine_dictionaries' Path override failed for key base::DIR_APP_DICTIONARIES and path '/usr/lib/qt6/qtwebengine_dictionaries' :2:../rcldb/rcldb.cpp:1038::Db::open: exception while opening [/home/chris/.recoll/xapiandb]: Couldn't stat '/home/chris/.recoll/xapiandb' :2:../rcldb/rclterms.cpp:598::Db::dirlist: exception while accessing index: Couldn't stat '/home/chris/.recoll/xapiandb' :2:../rcldb/rcldb.cpp:1038::Db::open: exception while opening [/home/chris/.recoll/xapiandb]: Couldn't stat '/home/chris/.recoll/xapiandb' :2:../rcldb/rcldb.cpp:1038::Db::open: exception while opening [/home/chris/.recoll/xapiandb]: Couldn't stat '/home/chris/.recoll/xapiandb' :3:../common/rclconfig.cpp:1834::RclConfig: pid/lock file: /run/user/1001/recoll-1bd02995df75c9288772966339e62137-index.pid
In my case, the "window" does not open.
Offline
июл 05 12:53:58 nuc systemd[1]: systemd-coredump@1-4331-0.service: Deactivated successfully.
июл 05 12:53:58 nuc systemd-coredump[4332]: [?] Process 4320 (recoll) of user 1000 dumped core.
Stack trace of thread 4320:
#0 0x0000736271c00188 unac_set_except_translations (librecoll.so.39 + 0x200188)
#1 0x0000736271acf535 _Z10recollinitiPFvvEPFviERNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEEPKS8_ (librecoll.so.39 + 0xcf535)
#2 0x000058bfb7ca733d n/a (recoll + 0x5933d)
#3 0x0000736265839c88 n/a (libc.so.6 + 0x25c88)
#4 0x0000736265839d4c __libc_start_main (libc.so.6 + 0x25d4c)
#5 0x000058bfb7caa405 n/a (recoll + 0x5c405)
Stack trace of thread 4325:
#0 0x000073626591c39d __poll (libc.so.6 + 0x10839d)
#1 0x00007362647768fd n/a (libglib-2.0.so.0 + 0xbe8fd)
#2 0x0000736264715787 g_main_loop_run (libglib-2.0.so.0 + 0x5d787)
#3 0x000073626459e494 n/a (libgio-2.0.so.0 + 0x112494)
#4 0x0000736264744236 n/a (libglib-2.0.so.0 + 0x8c236)
#5 0x00007362658a6ded n/a (libc.so.6 + 0x92ded)
#6 0x000073626592a0dc n/a (libc.so.6 + 0x1160dc)
Stack trace of thread 4321:
#0 0x000073626591c39d __poll (libc.so.6 + 0x10839d)
#1 0x00007362647768fd n/a (libglib-2.0.so.0 + 0xbe8fd)
#2 0x0000736264713f95 g_main_context_iteration (libglib-2.0.so.0 + 0x5bf95)
#3 0x00007362661a2cbd _ZN20QEventDispatcherGlib13processEventsE6QFlagsIN10QEventLoop17ProcessEventsFlagEE (libQt6Core.so.6 + 0x3a2cbd)
#4 0x0000736265f4f01e _ZN10QEventLoop4execE6QFlagsINS_17ProcessEventsFlagEE (libQt6Core.so.6 + 0x14f01e)
#5 0x000073626603a410 _ZN7QThread4execEv (libQt6Core.so.6 + 0x23a410)
#6 0x0000736260775e4e n/a (libQt6DBus.so.6 + 0x2de4e)
#7 0x00007362660c9747 n/a (libQt6Core.so.6 + 0x2c9747)
#8 0x00007362658a6ded n/a (libc.so.6 + 0x92ded)
#9 0x000073626592a0dc n/a (libc.so.6 + 0x1160dc)
Stack trace of thread 4322:
#0 0x000073626591c39d __poll (libc.so.6 + 0x10839d)
#1 0x000073626433020b n/a (libxcb.so.1 + 0xd20b)
#2 0x0000736264331f3d xcb_wait_for_event (libxcb.so.1 + 0xef3d)
#3 0x0000736250c70761 n/a (libQt6XcbQpa.so.6 + 0x4e761)
#4 0x00007362660c9747 n/a (libQt6Core.so.6 + 0x2c9747)
#5 0x00007362658a6ded n/a (libc.so.6 + 0x92ded)
#6 0x000073626592a0dc n/a (libc.so.6 + 0x1160dc)
Stack trace of thread 4323:
#0 0x0000736265927e9d syscall (libc.so.6 + 0x113e9d)
#1 0x000073626476f9e0 g_cond_wait (libglib-2.0.so.0 + 0xb79e0)
#2 0x00007362646dd8dc n/a (libglib-2.0.so.0 + 0x258dc)
#3 0x0000736264749687 n/a (libglib-2.0.so.0 + 0x91687)
#4 0x0000736264744236 n/a (libglib-2.0.so.0 + 0x8c236)
Offline
Hmm, so whats different from your system to mine? I don't have anything configured for recoll, I just installed it to test, so maybe more configuration is needed to actually make it crash?
Are you on wayland maybe?
Offline
Sorry for my English!
What mean the last phrase?
Offline
Are you using the X Display server or are you already on wayland?
Offline
echo $XDG_SESSION_TYPE
x11
Offline
From the spot where it crashes, it seems to die on some optimized string manipulation. Looking at the source 40c77f909279acabdfda7653cd8c9160dc4933b0 incooperates a library for optimized UTF-8 validation, which seems to do incorrect checks for available instruction sets and ends up trying to use AVX or so which isn't present in the CPUs mentioned as having a crash so far.
If you rebuild the current recoll with -Dsimdutf=false added to the meson setup line, does it work (... or if you just rebuild the package on your machine, so that the compile time checks select the max instruction sets avaialble for your CPU)?
Last edited by V1del (2024-07-05 13:50:53)
Offline
[papas@nuc recoll-1.39.1]$ ninja -C build
ninja: Entering directory `build'
[64/71] Generating qtgui/recoll with a custom command
FAILED: qtgui/recoll
/home/papas/recoll-1.39.1/qtgui/qmkmk.sh qtgui/recoll.pro PREFIX=/usr/local
Info: creating stash file /home/papas/recoll-1.39.1/build/qtgui/.qmake.stash
Project MESSAGE: Prefix is /usr/local
Project ERROR: Unknown module(s) in QT: webkitwidgets
[66/71] Compiling C object python/pych...-linux-gnu.so.p/recollchm_swig_chm.c.o
../python/pychm/recollchm/swig_chm.c: В функции «_wrap_chm_retrieve_object»:
../python/pychm/recollchm/swig_chm.c:3714:7: предупреждение: неиспользуемая переменная «res3» [-Wunused-variable]
3714 | int res3 = SWIG_TMPOBJ ;
| ^~~~
[68/71] Compiling C++ object python/re...2-x86_64-linux-gnu.so.p/pyrecoll.cpp.o
ninja: build stopped: subcommand failed.
Offline
From the spot where it crashes, it seems to die on some optimized string manipulation. Looking at the source 40c77f909279acabdfda7653cd8c9160dc4933b0 incooperates a library for optimized UTF-8 validation, which seems to do incorrect checks for available instruction sets and ends up trying to use AVX or so which isn't present in the CPUs mentioned as having a crash so far.
If you rebuild the current recoll with -Dsimdutf=false added to the meson setup line, does it work (... or if you just rebuild the package on your machine, so that the compile time checks select the max instruction sets avaialble for your CPU)?
Good catch! Compiling with -Dsimdutf=false avoids the crash.
The underlying issue is that Recoll bundles simdutf 4.0.9, which has a bug in its code for detecting instruction sets. The crash goes away when updating the bundled version to 5.2.3.
I sent a bug report via email according to the instructions on the upstream support page. Based on the auto-reply I received, I would expect it to show up as issue 267 when it gets approved on the upstream tracker.
Offline
Now also reported as a bug: https://gitlab.archlinux.org/archlinux/ … -/issues/4
Offline
Found ninja-1.12.1 at /usr/bin/ninja
Cleaning... 0 files.
[papas@nuc recoll-1.39.1]$ cd build
[papas@nuc build]$ ninja
[89/95] Generating qtgui/recoll with a custom command
FAILED: qtgui/recoll
/home/papas/recoll-1.39.1/qtgui/qmkmk.sh qtgui/recoll.pro PREFIX=/usr
Project MESSAGE: Prefix is /usr
Project ERROR: Unknown module(s) in QT: webkitwidgets
[94/95] Compiling C++ object recollindex.p/index_webqueue.cpp.o
ninja: build stopped: subcommand failed.
Offline
You lack dependencies, you should be able to build it using the PKGBUILD with the mentioned adjustment and using makepkg.
Offline
1.39.2 was released upstream with a fix:
- https://framagit.org/medoc92/recoll/-/issues/267
- https://www.recoll.org/pages/release-1.39.html
The issue should be resolved once that gets packaged.
Offline
None of this helped me!
I have returned to the recoll version-1.37.5-1.
Offline
The new version got released today, try it out.
Offline
Pages: 1