in docker, train.sh error

[wangsky2001]

i have updated newest paddle images. When i run train.sh followed 'Quick Start Tutorial', get an error:
root@c79251c143f8:~/paddle/demo/quick_start# ./train.sh
/usr/local/bin/paddle: line 81: 322 Illegal instruction (core dumped) ${DEBUGGER} $MYDIR/../opt/paddle/bin/paddle_trainer ${@:2}

[reyoung]

It seems that your CPU don't contains AVX instructions and PaddlePaddle's Docker image only support CPU with AVX instructions.

The images without avx will be added soon.

[wangsky2001]

thx for your reply. But my cpu is ' Intel(R) Xeon(R) CPU E5-2630', support avx2.0 instructions.

[reyoung]

But in docker environment, what's your CPU info? Please cat /proc/cpuinfo in your docker container. Thanks.

[wangsky2001]

root@c79251c143f8:~/paddle/demo/quick_start# cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 63
model name : Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz
stepping : 2
microcode : 0x36
cpu MHz : 2400.078
cache size : 20480 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
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 mmx fxsr sse sse2 ht syscall nx rdtscp lm constant_tsc rep_good nopl pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes f16c rdrand hypervisor lahf_lm abm fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid
bogomips : 4800.15
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:

processor : 1
vendor_id : GenuineIntel
cpu family : 6
model : 63
model name : Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz
stepping : 2
microcode : 0x36
cpu MHz : 2400.078
cache size : 20480 KB
physical id : 1
siblings : 1
core id : 0
cpu cores : 1
apicid : 2
initial apicid : 2
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 mmx fxsr sse sse2 ht syscall nx rdtscp lm constant_tsc rep_good nopl pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes f16c rdrand hypervisor lahf_lm abm fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid
bogomips : 4808.70
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:

processor : 2
vendor_id : GenuineIntel
cpu family : 6
model : 63
model name : Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz
stepping : 2
microcode : 0x36
cpu MHz : 2400.078
cache size : 20480 KB
physical id : 2
siblings : 1
core id : 0
cpu cores : 1
apicid : 4
initial apicid : 4
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 mmx fxsr sse sse2 ht syscall nx rdtscp lm constant_tsc rep_good nopl pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes f16c rdrand hypervisor lahf_lm abm fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid
bogomips : 4812.70
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:

processor : 3
vendor_id : GenuineIntel
cpu family : 6
model : 63
model name : Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz
stepping : 2
microcode : 0x36
cpu MHz : 2400.078
cache size : 20480 KB
physical id : 3
siblings : 1
core id : 0
cpu cores : 1
apicid : 6
initial apicid : 6
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 mmx fxsr sse sse2 ht syscall nx rdtscp lm constant_tsc rep_good nopl pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes f16c rdrand hypervisor lahf_lm abm fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid
bogomips : 4813.47
clflush size : 64
cache_alignment : 64
address sizes : 46 bits physical, 48 bits virtual
power management:

[wangsky2001]

about how long i can get an new docker image?

[reyoung]

It is very strange that your CPU support AVX2 instructions but not contains AVX instructions.

We will add docker image without AVX instructions as soon as possible. Maybe in 2-3 days.

But we also want to fix this error in E5-2630 v3 CPU. Are you running PaddlePaddle in any cloud service, such as aws? Or you just use your machine?

[reyoung]

The E5-2676 v3 cpuinfo in AWS. It shows that both AVX and AVX2 instructions are supported.

Is there any virtual machine you used to run PaddlePaddle?

processor   : 1
vendor_id   : GenuineIntel
cpu family  : 6
model       : 63
model name  : Intel(R) Xeon(R) CPU E5-2676 v3 @ 2.40GHz
stepping    : 2
microcode   : 0x2b
cpu MHz     : 2400.068
cache size  : 30720 KB
physical id : 0
siblings    : 2
core id     : 0
cpu cores   : 1
apicid      : 1
initial apicid  : 1
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 mmx fxsr sse sse2 ht syscall nx rdtscp lm constant_tsc rep_good nopl xtopology eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm xsaveopt fsgsbase bmi1 avx2 smep bmi2 erms invpcid
bogomips    : 4800.13
clflush size    : 64
cache_alignment : 64
address sizes   : 46 bits physical, 48 bits virtual
power management:

[wangsky2001]

yes, i used virtual machine on pc server to run PaddlePaddle.

[wangsky2001]

thx, i will try and post the test result later. thx again.

[backyes]

@wangsky2001 How about your progress

[reyoung]

Too long without response