glibc x32 autopkgtest fails in hirsute with binutils and audit in -proposed

Bug #1907472 reported by Balint Reczey
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
audit (Ubuntu)
New
Undecided
Unassigned
binutils (Ubuntu)
New
Undecided
Unassigned
glibc (Ubuntu)
New
Undecided
Unassigned

Bug Description

https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-hirsute/hirsute/amd64/g/glibc/20201208_160143_17354@/log.gz

----------
FAIL: math/test-double-libmvec-sincos
original exit status 1
Didn't expect signal from child: got `Illegal instruction'
----------
----------
FAIL: math/test-double-vlen2-sincos
original exit status 132
----------
----------
FAIL: math/test-float-libmvec-sincosf
original exit status 1
Didn't expect signal from child: got `Illegal instruction'
----------
----------
FAIL: math/test-float-vlen4-sincos
original exit status 132
----------

Balint Reczey (rbalint)
summary: - x32 autopkgtest fails in hirsute with binutils/2.35.50.20201207-0ubuntu1
+ glibc x32 autopkgtest fails in hirsute with binutils and audit in
+ -proposed
Revision history for this message
Balint Reczey (rbalint) wrote :

It did not fail locally for me on an i5 CPU.

CPU in the autopkgtest infra:

https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-hirsute/hirsute/amd64/c/cpuinfo/20201114_005620_e8a1d@/log.gz

autopkgtest [00:55:54]: test command2: /usr/bin/cpu-info
autopkgtest [00:55:54]: test command2: [-----------------------
Packages:
 0: AMD Opteron 62xx class
Microarchitectures:
 1x Bulldozer
Cores:
 0: 1 processor (0), AMD Bulldozer
Logical processors (System ID):
 0 (0): APIC ID 0x00000000

...

Local CPU:
autopkgtest [17:16:25]: test command2: /usr/bin/cpu-info
autopkgtest [17:16:25]: test command2: [-----------------------
Packages:
 0: Intel Core i5-4670
Microarchitectures:
 4x Haswell
Cores:
 0: 1 processor (0), Intel Haswell
 1: 1 processor (1), Intel Haswell
 2: 1 processor (2), Intel Haswell
 3: 1 processor (3), Intel Haswell
Logical processors (System ID):
 0 (0): APIC ID 0x00000000
 1 (1): APIC ID 0x00000002
 2 (2): APIC ID 0x00000004
 3 (3): APIC ID 0x00000006

Revision history for this message
Balint Reczey (rbalint) wrote :

$ diff bulldozer-isa-info i5-isa-info
6,7c6,7
< BMI: no
< BMI2: no
---
> BMI: yes
> BMI2: yes
10,12c10,12
< MOVBE: no
< PREFETCH: yes
< PREFETCHW: yes
---
> MOVBE: yes
> PREFETCH: no
> PREFETCHW: no
28,29c28,29
< SSE4a: yes
< Misaligned SSE: yes
---
> SSE4a: no
> Misaligned SSE: no
31,35c31,35
< FMA3: no
< FMA4: yes
< XOP: yes
< F16C: no
< AVX2: no
---
> FMA3: yes
> FMA4: no
> XOP: no
> F16C: yes
> AVX2: yes
54c54
< MONITOR/MWAIT: no
---
> MONITOR/MWAIT: yes
67c67
< RDRAND: no
---
> RDRAND: yes
71c71
< RDTSCP: no
---
> RDTSCP: yes

Revision history for this message
Balint Reczey (rbalint) wrote :

I hoped to reproduce the issue with the help of qemu-x86_64 but it does not want to run x32 ELF.

apt build-dep glibc
sed -i 's/backports/proposed/' /etc/apt/sources.list
apt update
apt install libctf-nobfd0 libctf0 binutils-x86-64-linux-gnu libbinutils binutils binutils-common
x86_64-linux-gnu-gcc-10 -mx32 test-double-libmvec-sincos-main.c -std=gnu11 -fgnu89-inline -pipe -O2 -g -Wwrite-strings -Wundef -Werror -fmerge-all-constants -frounding-math -fstack-protector-strong -Wstrict-prototypes -Wold-style-definition -fmath-errno -fpie -fcf-protection -U_FORTIFY_SOURCE -lm
./a.out

 qemu-x86_64-static ./a.out
./a.out: Invalid ELF image for this architecture

Will continue investigation tomorrow.

Revision history for this message
Matthias Klose (doko) wrote :

removed the update-excuse tag. that's unrelated to the current 2.33 migration.

tags: removed: update-excuse
Revision history for this message
Balint Reczey (rbalint) wrote :

I've observed the failure in the i386 build of glibc_2.33-0ubuntu4:

Linux lcy01-amd64-014 4.15.0-136-generic #140-Ubuntu SMP Thu Jan 28 05:20:47 UTC 2021 i686 athlon i686 GNU/Linux
...
if [ -f /proc/cpuinfo ] ; then cat /proc/cpuinfo ; fi
processor : 0
vendor_id : AuthenticAMD
cpu family : 16
model : 2
model name : AMD Opteron 23xx (Gen 3 Class Opteron)
stepping : 3
microcode : 0x1000065
cpu MHz : 2500.048
cache size : 512 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 5
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt pdpe1gb lm 3dnowext 3dnow rep_good nopl cpuid extd_apicid tsc_known_freq pni cx16 x2apic popcnt hypervisor lahf_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw vmmcall
bugs : tlb_mmatch fxsave_leak sysret_ss_attrs null_seg spectre_v1 spectre_v2
bogomips : 5000.09
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:
...

+---------------------------------------------------------------------+
| Encountered regressions that don't match expected failures. |
+---------------------------------------------------------------------+
FAIL: math/test-double-libmvec-sincos
FAIL: math/test-double-vlen2-sincos
FAIL: math/test-float-libmvec-sincosf
FAIL: math/test-float-vlen4-sincos
touch /<<PKGBUILDDIR>>/stamp-dir/check_x32
CHECK SUMMARY
check for check_libc passed
check for check_amd64 passed
check for check_x32 failed

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.