Bug 32102 - Digikam - fails on hardware without SSE2 instructions
Summary: Digikam - fails on hardware without SSE2 instructions
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 9
Hardware: All Linux
Priority: Normal normal
Target Milestone: ---
Assignee: QA Team
QA Contact:
URL:
Whiteboard: MGA9-32-OK MGA9-64-OK
Keywords: advisory, validated_update
Depends on:
Blocks:
 
Reported: 2023-07-15 15:57 CEST by Brian Rockwell
Modified: 2024-11-05 03:47 CET (History)
7 users (show)

See Also:
Source RPM: digikam-8.0.0-3.mga9.src.rpm
CVE:
Status comment:


Attachments
strace of digikam start (8.1) (225.46 KB, text/plain)
2023-07-17 01:32 CEST, Brian Rockwell
Details

Description Brian Rockwell 2023-07-15 15:57:14 CEST
Description of problem:  digikam fails:

$ digikam
Illegal instruction (core dumped)



Version-Release number of selected component (if applicable): Cauldron - 32bit (installed from classic)


How reproducible:  from command line or from menu.  Click digikam


$ digikam
Illegal instruction (core dumped)

$ uname -a
Linux localhost 6.4.3-desktop-1.mga9 #1 SMP PREEMPT_DYNAMIC Tue Jul 11 08:56:12 UTC 2023 i686 GNU/Linux

Nouveau graphics

Steps to Reproduce:
1.  
2.
3.
Comment 1 Lewis Smith 2023-07-15 20:14:43 CEST
Thank you for the report.
I have tried this on x86, Mate, and digikam seems to work OK. Unusually, from a terminal, it output no messages at all:
 $ digikam
 $

Can you please:
- confirm the digikam version I have put in SRPM.
- say what desktop you were using.
- post the terminal output.

CC'ing for confirmation Ben who has a 32-bit system, and can try this easily.

In spite of all these questions, assigning to DavidG who currently nurses this application. Wait for the answers!

CC: (none) => westel
Source RPM: (none) => digikam-8.0.0-3.mga9.src.rpm
Hardware: All => i586

Lewis Smith 2023-07-15 20:14:56 CEST

Assignee: bugsquad => geiger.david68210
CC: (none) => lewyssmith

Lewis Smith 2023-07-15 20:15:06 CEST

CC: lewyssmith => (none)

Comment 2 Ben McMonagle 2023-07-15 23:03:44 CEST
Hi Brian,

cant reproduce from a single Plasma DE i586 install, (intel 810 or later)

do you have any other relevant info?

output of:
inxi -Fxxxz

lspcidrake -v

report.bug.xz (can be found here => /root/drakx/)

your report from https://pad.riseup.net/p/mageia9-RC implies a multi DE install.

Will try a plasma/xfce install.
Comment 3 Jani Välimaa 2023-07-16 00:14:09 CEST
The output of the following would be interesting to see.

$ cat /proc/cpuinfo |grep ^flags | uniq

I think we need to apply https://invent.kde.org/graphics/digikam/-/commit/799953eef9f43fb38d781cff8bea8eb201bb78bb and perhaps also build digikam without sse{2,3,4.1,4.2} on %ix86.

See also https://bugs.kde.org/show_bug.cgi?id=465548.
Comment 4 Ben McMonagle 2023-07-16 01:47:20 CEST
(In reply to Ben McMonagle from comment #2)
 
> Will try a plasma/xfce install.

no issue here,

from lscpu:

Model name:Intel(R) Core(TM) i5 CPU       M 480  @ 2.67GHz
Comment 5 Ben McMonagle 2023-07-16 08:01:42 CEST
confirmed on real 32bit hardware:

$ lscpu
Architecture:           i686
  CPU op-mode(s):       32-bit
  Address sizes:        34 bits physical, 32 bits virtual
  Byte Order:           Little Endian
CPU(s):                 1
  On-line CPU(s) list:  0
Vendor ID:              AuthenticAMD
  Model name:           AMD Athlon(tm) XP 2400+
    CPU family:         6
    Model:              8
    Thread(s) per core: 1
    Core(s) per socket: 1
    Socket(s):          1
    Stepping:           1
    BogoMIPS:           3996.98
    Flags:              fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
                         cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dn
                        ow cpuid 3dnowprefetch vmmcall
Caches (sum of all):    
  L1d:                  64 KiB (1 instance)
  L1i:                  64 KiB (1 instance)
  L2:                   256 KiB (1 instance)
Vulnerabilities:        
  Itlb multihit:        Not affected
  L1tf:                 Not affected
  Mds:                  Not affected
  Meltdown:             Not affected
  Mmio stale data:      Not affected
  Retbleed:             Not affected
  Spec store bypass:    Vulnerable
  Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer 
                        sanitization
  Spectre v2:           Mitigation; Retpolines, STIBP disabled, RSB filling, PBR
                        SB-eIBRS Not affected
  Srbds:                Not affected
  Tsx async abort:      Not affected
Comment 6 Jani Välimaa 2023-07-16 11:56:03 CEST
(In reply to Jani Välimaa from comment #3)
> The output of the following would be interesting to see.
> 
> $ cat /proc/cpuinfo |grep ^flags | uniq
> 
> I think we need to apply
> https://invent.kde.org/graphics/digikam/-/commit/
> 799953eef9f43fb38d781cff8bea8eb201bb78bb and perhaps also build digikam
> without sse{2,3,4.1,4.2} on %ix86.
> 
> See also https://bugs.kde.org/show_bug.cgi?id=465548.
Should be fixed in SVN, but didn't bump the rel. It's out of my scope to decide whether the new release should be pushed now or after mga9 is released.

https://svnweb.mageia.org/packages?view=revision&revision=1963968
Comment 7 Giuseppe Ghibò 2023-07-16 13:30:25 CEST
Since(In reply to Jani Välimaa from comment #6)
> (In reply to Jani Välimaa from comment #3)
> > The output of the following would be interesting to see.
> > 
> > $ cat /proc/cpuinfo |grep ^flags | uniq
> > 
> > I think we need to apply
> > https://invent.kde.org/graphics/digikam/-/commit/
> > 799953eef9f43fb38d781cff8bea8eb201bb78bb and perhaps also build digikam
> > without sse{2,3,4.1,4.2} on %ix86.
> > 
> > See also https://bugs.kde.org/show_bug.cgi?id=465548.
> Should be fixed in SVN, but didn't bump the rel. It's out of my scope to
> decide whether the new release should be pushed now or after mga9 is
> released.
> 
> https://svnweb.mageia.org/packages?view=revision&revision=1963968

IMHO we should push 8.1.0. Actually that would bump the lib release to 8.1.0, but AFAIK the only application using that lib beyond digikam itself is showfoto which is produced by the same digikam SPEC file. I'm using 8.1.0 since some weeks and works flawlessly (though not tested with i586). Maybe you see with maintainer.

CC: (none) => anaselli, ghibomgx

Comment 8 Brian Rockwell 2023-07-16 16:18:01 CEST
Sorry for the delay.  I made a guess it may have been related to an old home folder.  So built a full fresh copy this morning.

still failing

--JANI

$ cat /proc/cpuinfo |grep ^flags | uniq
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nonstop_tsc cpuid extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt hw_pstate vmmcall npt lbrv svm_lock nrip_save

--BEN

$ inxi -Fxxxz
System:
  Kernel: 6.4.3-desktop-1.mga9 arch: i686 bits: 32 compiler: gcc v: 12.3.0
    Desktop: KDE Plasma v: 5.27.5 tk: Qt v: 5.15.7 wm: kwin_x11 vt: 1 dm: SDDM
    Distro: Mageia 9
Machine:
  Type: Desktop Mobo: Gigabyte model: GA-78LMT-S2P v: x.x
    serial: <superuser required> BIOS: Award v: F2 date: 03/20/2012
CPU:
  Info: triple core model: AMD Athlon II X3 450 bits: 64 type: MCP
    smt: <unsupported> arch: K10 rev: 3 cache: L1: 384 KiB L2: 1.5 MiB
  Speed (MHz): avg: 800 min/max: 800/3200 boost: disabled cores: 1: 800
    2: 800 3: 800 bogomips: 19284
  Flags: ht lm nx pae sse sse2 sse3 sse4a svm
Graphics:
  Device-1: NVIDIA GF108 [GeForce GT 730] vendor: eVga.com. driver: nouveau
    v: kernel arch: Fermi pcie: speed: 2.5 GT/s lanes: 16 ports: active: VGA-1
    empty: DVI-I-1,HDMI-A-1 bus-ID: 01:00.0 chip-ID: 10de:0f02 class-ID: 0300
    temp: 47.0 C
  Display: x11 server: X.org v: 1.21.1.8 with: Xwayland v: 22.1.9
    compositor: kwin_x11 driver: X: loaded: nouveau,v4l dri: nouveau
    gpu: nouveau display-ID: :0 screens: 1
  Screen-1: 0 s-res: 1920x1080 s-dpi: 96 s-size: 507x285mm (19.96x11.22")
    s-diag: 582mm (22.9")
  Monitor-1: VGA-1 model: Samsung C24F390 serial: <filter> res: 1920x1080
    hz: 60 dpi: 94 size: 521x293mm (20.51x11.54") diag: 598mm (23.5") modes:
    max: 1920x1080 min: 720x400
  API: OpenGL v: 4.3 Mesa 23.1.3 renderer: NVC1 direct-render: Yes
Audio:
  Device-1: AMD SBx00 Azalia vendor: Gigabyte driver: snd_hda_intel v: kernel
    bus-ID: 00:14.2 chip-ID: 1002:4383 class-ID: 0403
  Device-2: NVIDIA GF108 High Definition Audio vendor: eVga.com.
    driver: snd_hda_intel v: kernel pcie: speed: 2.5 GT/s lanes: 16
    bus-ID: 01:00.1 chip-ID: 10de:0bea class-ID: 0403
  API: ALSA v: k6.4.3-desktop-1.mga9 status: kernel-api with: aoss
    type: oss-emulator
  Server-1: JACK v: 1.9.21 status: off
  Server-2: PipeWire v: 0.3.71 status: off with: pipewire-media-session
    status: off
  Server-3: PulseAudio v: 16.1 status: active
Network:
  Device-1: Qualcomm Atheros AR8151 v2.0 Gigabit Ethernet vendor: Gigabyte
    driver: atl1c v: kernel pcie: speed: 2.5 GT/s lanes: 1 port: df00
    bus-ID: 02:00.0 chip-ID: 1969:1083 class-ID: 0200
  IF: enp2s0 state: up speed: 1000 Mbps duplex: full mac: <filter>
  Device-2: Ralink RT2760 Wireless 802.11n 1T/2R vendor: Linksys WMP110 v2
    driver: rt2800pci v: 2.3.0 bus-ID: 03:06.0 chip-ID: 1814:0701 class-ID: 0280
  IF: wlp3s6 state: down mac: <filter>
Drives:
  Local Storage: total: 465.76 GiB used: 5.81 GiB (1.2%)
  ID-1: /dev/sda vendor: Seagate model: ST500LM030-1RK17D size: 465.76 GiB
    speed: 3.0 Gb/s type: HDD rpm: 5400 serial: <filter> rev: SDM1 scheme: GPT
Partition:
  ID-1: / size: 49.2 GiB used: 5.59 GiB (11.4%) fs: ext4 dev: /dev/sda2
  ID-2: /home size: 403.94 GiB used: 220.5 MiB (0.1%) fs: ext4
    dev: /dev/sda4
Swap:
  ID-1: swap-1 type: partition size: 4 GiB used: 0 KiB (0.0%) priority: -2
    dev: /dev/sda3
Sensors:
  System Temperatures: cpu: 25.5 C mobo: N/A gpu: nouveau temp: 47.0 C
  Fan Speeds (RPM): N/A
Info:
  Processes: 191 Uptime: 14m wakeups: 0 Memory: 7.85 GiB
  used: 2.23 GiB (28.4%) Init: systemd v: 253 target: graphical (5)
  default: runlevel5 Compilers: N/A Packages: pm: rpm pkgs: N/A
  note: see --rpm Shell: Bash v: 5.2.15 running-in: konsole inxi: 3.3.26


# journalctl | grep digik
Jul 16 09:08:59 localhost systemd[2946]: Started app-org.kde.digikam-3237fcc9cf184d7b8337d511ca1f0162.scope.
Jul 16 09:09:08 localhost kernel: traps: digikam[3664] trap invalid opcode ip:b6c149df sp:bf83e400 error:0 in libdigikamcore.so.8.0.0[b63be000+987000]
Jul 16 09:09:30 localhost kernel: traps: digikam[3823] trap invalid opcode ip:b6c149df sp:bf903de0 error:0 in libdigikamcore.so.8.0.0[b63be000+987000]

Any idea where it is doing the dump?
Comment 9 Giuseppe Ghibò 2023-07-16 16:32:26 CEST
BTW, just in the case I've the digikam 8.1.0 here, if you want to try: 

https://download.copr.fedorainfracloud.org/results/ghibo/mageia9-bonus/mageia-cauldron-i586/06150155-digikam/
Comment 10 Brian Rockwell 2023-07-17 01:32:37 CEST
Created attachment 13923 [details]
strace of digikam start (8.1)

Installed digikam 8.1.  Didn't fix it.  Attaching my attempt at an strace of digikam.

Let me know if this helps.  This isn't a show stopper in my opinion, just an annoyance on my part.
Comment 11 Morgan Leijström 2024-03-22 17:54:32 CET
How is this for you now Brian?

A user here tell it was recently working on i586, bot now it fail both i586 and x86_64.
https://forums.mageia.org/en/viewtopic.php?f=7&t=15297

I wonder what cause this.
Maybe same problem as you Brian see/saw, or not...

This bug was opened when Cauldron was mga9, so setting it as such.

mga9 is still on version 8.0.0
Upstream five days ago released 8.3.0 so IMO worth to update anyway and maybe we are lucky :)

CC: (none) => fri
Version: Cauldron => 9

Comment 12 katnatek 2024-03-23 20:07:11 CET
Still valid RH mageia i586

digikam 
Instrucción ilegal (`core' generado)
Comment 13 katnatek 2024-03-23 20:10:13 CET
Run gdb.minimal digikam

Program received signal SIGILL, Illegal instruction.
0xb6c149df in operator*(QMatrix4x4 const&, QMatrix4x4 const&) () from /lib/libdigikamcore.so.8.0.0
(gdb) bt
#0  0xb6c149df in operator*(QMatrix4x4 const&, QMatrix4x4 const&) ()
   from /lib/libdigikamcore.so.8.0.0
#1  0xb64d18b5 in __static_initialization_and_destruction_0(int, int) [clone .constprop.0] ()
   from /lib/libdigikamcore.so.8.0.0
#2  0xb7fd37b3 in call_init.part () from /lib/ld-linux.so.2
#3  0xb7fd388b in _dl_init () from /lib/ld-linux.so.2
#4  0xb7fe9600 in _dl_start_user () from /lib/ld-linux.so.2
Comment 14 papoteur 2024-11-03 21:50:45 CET
This is fixed upstream since 8.1.0. The problem was usage of SSE2 instructions even if not available for the processor.
Submitting
RPMS
digikam-8.4.0-1.mga9
lib64digikamcore8.4.0-8.4.0-1.mga9
lib64digikamgui8.4.0-8.4.0-1.mga9
lib64digikamdatabase8.4.0-8.4.0-1.mga9
showfoto-8.4.0-1.mga9
lib64digikam-devel-8.4.0-1.mga9
SRPMS
digikam-8.4.0-1.mga9

Assignee: geiger.david68210 => qa-bugs
Hardware: i586 => All
CC: (none) => yvesbrungard

katnatek 2024-11-03 23:01:54 CET

Keywords: (none) => advisory

Comment 15 katnatek 2024-11-03 23:29:22 CET
RH i586

rpm -q digikam
digikam-8.0.0-3.mga9

digikam 
Instrucción ilegal (`core' generado)


 LC_ALL=C urpmi --auto --auto-update
medium "QA Testing (32-bit)" is up-to-date
medium "Core Release (distrib1)" is up-to-date
medium "Core Updates (distrib3)" is up-to-date
medium "Nonfree Release (distrib11)" is up-to-date
medium "Nonfree Updates (distrib13)" is up-to-date
medium "Tainted Release (distrib21)" is up-to-date
medium "Tainted Updates (distrib23)" is up-to-date

installing libdigikamcore8.4.0-8.4.0-1.mga9.i586.rpm digikam-8.4.0-1.mga9.i586.rpm libdigikamdatabase8.4.0-8.4.0-1.mga9.i586.rpm libdigikamgui8.4.0-8.4.0-1.mga9.i586.rpm from //home/katnatek/qa-testing/i586
Preparing...                     #######################################################################################
      1/4: libdigikamcore8.4.0   #######################################################################################
      2/4: libdigikamdatabase8.4.0
                                 #######################################################################################
      3/4: libdigikamgui8.4.0    #######################################################################################
      4/4: digikam               #######################################################################################
      1/1: removing digikam-1:8.0.0-3.mga9.i586
                                 #######################################################################################


digikam 
digikam.facedb: Cannot found faces engine model "shapepredictor.dat"
digikam.facedb: Faces recognition feature cannot be used!
digikam.facedb: Cannot found faces engine DNN model "openface_nn4.small2.v1.t7"
digikam.facedb: Faces recognition feature cannot be used!
../../3rdparty/chromium/sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall 0383
kf.xmlgui: Unhandled container to remove :  Digikam::DigikamApp
../../3rdparty/chromium/sandbox/linux/seccomp-bpf-helpers/sigsys_handlers.cc:**CRASHING**:seccomp-bpf failure in syscall 0383
digikam.general: parent is null
QThreadStorage: Thread 0xad35b80 exited after QThreadStorage 18 destroyed
QThreadStorage: Thread 0xad35b80 exited after QThreadStorage 17 destroyed

The application start
Follow the 1st time wizard until end

I not use this application so is all I can test, the main bug is fixed
papoteur 2024-11-04 09:25:43 CET

Summary: Digikam - plasma i686 (32bit) fails => Digikam - fails on hardware without SSE2 instructions

katnatek 2024-11-04 21:28:15 CET

Whiteboard: (none) => MGA9-32-OK

Comment 16 Thomas Andrews 2024-11-05 03:14:13 CET
Did a 64-bit test as well, and it seems OK. I don't normally use it either, but I managed to get it to give me a full screen slide show of a folder of my personal photos.

Validating.

CC: (none) => andrewsfarm, sysadmin-bugs
Keywords: (none) => validated_update
Whiteboard: MGA9-32-OK => MGA9-32-OK MGA9-64-OK

Comment 17 Mageia Robot 2024-11-05 03:47:49 CET
An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGAA-2024-0223.html

Status: NEW => RESOLVED
Resolution: (none) => FIXED


Note You need to log in before you can comment on or make changes to this bug.