Arch Linux

mapintar

Member

Registered: 2010-04-17

Posts: 50

There is a bug in a i686 system?

I have some bugs only in 686 system.

Example:

xlinpack_xeon32
Input data or print help ? Type [data]/help :

Number of equations to solve (problem size): 1000
Leading dimension of array: 1000
Number of trials to run: 6
Data alignment value (in Kbytes): 4
Current date/time: Sun Jun  9 15:10:50 2013

CPU frequency:    3.690 GHz
Number of CPUs: 1
Number of cores: 4
Number of threads: 4

Parameters are set to:

Number of tests: 1
Number of equations to solve (problem size) : 1000
Leading dimension of array                  : 1000
Number of trials to run                     : 6   
Data alignment value (in Kbytes)            : 4   

Maximum memory requested that can be used=8024096, at the size=1000

=================== Timing linear equation system solver ===================

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check
1000   1000   4      0.012      56.0402  9.776248e-13 3.333952e-02   pass
1000   1000   4      0.012      56.7960  9.776248e-13 3.333952e-02   pass
1000   1000   4      0.012      56.7483  9.776248e-13 3.333952e-02   pass
1000   1000   4      0.063      10.6747  -nan       -nan         pass
1000   1000   4      0.063      10.6775  -nan       -nan         pass
1000   1000   4      0.063      10.6780  -nan       -nan         pass

Performance Summary (GFlops)

Size   LDA    Align.  Average  Maximal
1000   1000   4       33.6024  56.7960

Residual checks PASSED

End of tests

There is other bugs. Slow write and read in reiserfs partition, and problem in cpupower governor ondemand, not present in x86_64 distros (test in arch linux and ubuntu 13.04).

I run qemu in a x86_64 host and i686 guest, and the bugs is reproduced when rum qemu "-cpu host" flag on:

qemu-system-x86_64 -m 1G -enable-kvm -netdev user,id=usernet -device virtio-net-pci,netdev=usernet -boot d 'myimage.qcow2' -vga vmware -smp 4,sockets=4,cores=1,threads=1 -cpu host

But work fine when processor flags ssse3 and sse4.2 are disabled:

qemu-system-x86_64 -m 1G -enable-kvm -netdev user,id=usernet -device virtio-net-pci,netdev=usernet -boot d 'myimage.qcow2' -vga vmware -smp 4,sockets=4,cores=1,threads=1 -cpu host,-ssse3,-sse4_2

It's a bug present in gcc, glibc or other package, only in i686?

Tested in a clean install from archinstall and the result is the same (sory for my bad english).

Last edited by mapintar (2013-06-09 18:39:08)

jasonwryan

Anarchist

From: .nz

Registered: 2009-05-09

Posts: 30,424

Website

Re: There is a bug in a i686 system?

There is a bugtracker: have you looked and or reported there?

Not an Arch discussion, moving to Kernel & Hardware...

---

Arch + dwm   •   Mercurial repos  •   Surfraw

Registered Linux User #482438

mapintar

Member

Registered: 2010-04-17

Posts: 50

Re: There is a bug in a i686 system?

There is a bugtracker: have you looked and or reported there?

I look in a bugtracker but not found any similar problem. I do not know package is involved, and unsafe for post in bugtracker...

Not an Arch discussion, moving to Kernel & Hardware...

Sorry for incorret post.

Max-P

Member

Registered: 2011-06-11

Posts: 164

Re: There is a bug in a i686 system?

What's your CPU? Have you overclocked your CPU? (3.690 GHz seems a bit high for defaults + linpack is a benchmark tool)

The only ways I could see this to fail, it either a compiler bug (very, very unlikely), or a CPU problem on your side. Post your /proc/cpuinfo

mapintar

Member

Registered: 2010-04-17

Posts: 50

Re: There is a bug in a i686 system?

What's your CPU? Have you overclocked your CPU? (3.690 GHz seems a bit high for defaults + linpack is a benchmark tool)

Thanks for reply.

I tested in a 2 machine, i5-2500k (asrock motherboard, no overclock, office) and i5-750 (msi motherboard, overclocked, home), and the result is the same.

The high frequency of the core (3.690 GHz) is set by Turbo Boost Technology when only one core is used, and decreases for 3.4 in full.

The only ways I could see this to fail, it either a compiler bug (very, very unlikely), or a CPU problem on your side. Post your /proc/cpuinfo

My work machine cat /proc/cpuinfo:

cat /proc/cpuinfo
processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping	: 7
microcode	: 0x28
cpu MHz		: 2112.000
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 0
cpu cores	: 4
apicid		: 0
initial apicid	: 0
fdiv_bug	: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc arch_perfmon pebs bts xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid
bogomips	: 6587.43
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

processor	: 1
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping	: 7
microcode	: 0x28
cpu MHz		: 1584.000
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 1
cpu cores	: 4
apicid		: 2
initial apicid	: 2
fdiv_bug	: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc arch_perfmon pebs bts xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid
bogomips	: 6587.43
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

processor	: 2
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping	: 7
microcode	: 0x28
cpu MHz		: 2640.000
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 2
cpu cores	: 4
apicid		: 4
initial apicid	: 4
fdiv_bug	: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc arch_perfmon pebs bts xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid
bogomips	: 6587.43
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

processor	: 3
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping	: 7
microcode	: 0x28
cpu MHz		: 1716.000
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 3
cpu cores	: 4
apicid		: 6
initial apicid	: 6
fdiv_bug	: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx rdtscp lm constant_tsc arch_perfmon pebs bts xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid
bogomips	: 6587.43
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

This is very mysterious because old distros (compiled before 2013, march) work fine.