Removing '-march=native' from

marss/ptlsim/SConstruct:    env.Append(CCFLAGS = '-O3 -march=native')
marss/qemu/SConstruct:    env.Append(CCFLAGS = '-O3 -march=native')

should do the trick.

BTW, I reposted this to the listserv in order to assist future users who
run into this issue. Please reply-all next time so that the message
remains on the listserv. :-)

Tyler

> Thanks Tyler. Yes I compiled it on another machine, as the server doesn't
> have any compiler instralled on it, but just runtime libraries. Anyway, I
> will try to tell the compiler not to use this instruction. Hopefully, it
> would work that way.
>
> Regards,
> Hamid.
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]]
> Sent: maandag 16 september 2013 17:28
> To: Hamid Mushtaq - EWI
> Cc: '[email protected]'
> Subject: RE: [marss86-devel] Unable to run marssx86 on Red Hat
>
> vxorps is an AVX instruction. You can see from the /proc/cpuinfo that your
> KVM CPU does not support AVX:
>
>> flags                : fpu de pse tsc msr pae mce cx8 apic mtrr pge mca cmov 
>> pat pse36
>> clflush mmx fxsr sse sse2 syscall nx lm unfair_spinlock pni cx16
>
> Did you compile MARSS on a machine other than the one you're running on?
> If so, this is likely the issue and you can just do `scons -c` and then
> execute scons again with your desired build settings and it should work.
>
> Tyler
>
>> I get the following by doing layout asm. The 4th instruction is the
>> highlighted one.
>>
>>    0x72b6e7 <_GLOBAL__I_bbcache()+2087>    mov    $0x1279120,%edx
>>
>>          
>>    0x72b6ec <_GLOBAL__I_bbcache()+2092>    mov    %edi,%ebx
>>
>>          
>>    0x72b6ee <_GLOBAL__I_bbcache()+2094>    mov    $0x4000,%esi
>>
>>          
>>
>> This is the highlighted instruction
>> ------------------------------------------------------------------------------------------------
>> 0x72b6f3 <_GLOBAL__I_bbcache()+2099>    vxorps %xmm0,%xmm0,%xmm0
>>
>>    
>> ----------------------------------------------------------------------
>> --------------------------
>>
>>  0x72b6f7 <_GLOBAL__I_bbcache()+2103>    nopw   0x0(%rax,%rax,1)
>>
>>      
>>    0x72b700 <_GLOBAL__I_bbcache()+2112>    (bad)
>>
>>          
>>    0x72b704 <_GLOBAL__I_bbcache()+2116>    retq   $0x103
>>
>>          
>>
>> And /proc/cpuinfo reveals the following.
>>
>> fpu_exception        : yes
>> cpuid level  : 4
>> wp           : yes
>> flags                : fpu de pse tsc msr pae mce cx8 apic mtrr pge mca cmov 
>> pat pse36
>> clflush mmx fxsr sse sse2 syscall nx lm unfair_spinlock pni cx16
>> hypervisor lahf_lm abm sse4a
>> bogomips     : 5600.04
>> TLB size     : 1024 4K pages
>> clflush size : 64
>> cache_alignment      : 64
>> address sizes        : 48 bits physical, 48 bits virtual
>> power management:
>>
>> processor    : 1
>> vendor_id    : AuthenticAMD
>> cpu family   : 6
>> model                : 13
>> model name   : QEMU Virtual CPU version (cpu64-rhel6)
>> stepping     : 3
>> cpu MHz              : 2800.024
>> cache size   : 512 KB
>> fpu          : yes
>> fpu_exception        : yes
>> cpuid level  : 4
>> wp           : yes
>> flags                : fpu de pse tsc msr pae mce cx8 apic mtrr pge mca cmov 
>> pat pse36
>> clflush mmx fxsr sse sse2 syscall nx lm unfair_spinlock pni cx16
>> hypervisor lahf_lm abm sse4a
>> bogomips     : 5600.04
>> TLB size     : 1024 4K pages
>> clflush size : 64
>> cache_alignment      : 64
>> address sizes        : 48 bits physical, 48 bits virtual
>> power management:
>>
>> -----Original Message-----
>> From: [email protected] [mailto:[email protected]]
>> Sent: maandag 16 september 2013 17:07
>> To: Hamid Mushtaq - EWI
>> Cc: [email protected]
>> Subject: RE: [marss86-devel] Unable to run marssx86 on Red Hat
>>
>>> This is what I get when I run it with gdb.
>>>
>>> Program received signal SIGILL, Illegal instruction.
>>> reset () at ptlsim/lib/superstl.h:2635
>>> 2635              foreach (i, setcount) { sets[i] = NULL; }
>>>
>>> And the processor 's info is the following. It has 8 cores.
>>>
>>> Architecture:          x86_64
>>> CPU op-mode(s):        32-bit, 64-bit
>>> Byte Order:            Little Endian
>>> CPU(s):                2
>>> On-line CPU(s) list:   0,1
>>> Thread(s) per core:    1
>>> Core(s) per socket:    1
>>> Socket(s):             2
>>> NUMA node(s):          1
>>> Vendor ID:             AuthenticAMD
>>> CPU family:            6
>>> Model:                 13
>>> Stepping:              3
>>> CPU MHz:               2800.024
>>> BogoMIPS:              5600.04
>>> Hypervisor vendor:     KVM
>>> Virtualization type:   full
>>> L1d cache:             64K
>>> L1i cache:             64K
>>> L2 cache:              512K
>>> NUMA node0 CPU(s):     0,1
>>>
>>> Regards,
>>> Hamid.
>>>
>>> -----Original Message-----
>>> From: [email protected] [mailto:[email protected]]
>>> Sent: maandag 16 september 2013 16:37
>>> To: Hamid Mushtaq - EWI
>>> Cc: [email protected]
>>> Subject: Re: [marss86-devel] Unable to run marssx86 on Red Hat
>>>
>>> 1) What processor is the machine running? Can you run MARSS through
>>> gdb and tell me the instruction that SIGILL gets raised on? My guess
>>> is that it's an SSE instruction and the machine in question doesn't
>>> support it...
>>>
>>> 2) Try using vnc instead of curses.
>>>
>>> Tyler
>>>
>>>> Hi,
>>>>
>>>> I am trying to run marssx86 on Red Hat 4.4.7-3, but whenever I do
>>>> so, I get "Illegal Instruction" error. Then I tried to compile it
>>>> with
>>>> debug=1 option, it does start to run, but says threaded-sim option
>>>> not supported.
>>>> Moreover if you give -curses as an option, it says "invalid option
>>>> -curses". I need curses because the computer in question is a remote
>>>> server.
>>>>
>>>> Regards,
>>>> Hamid.
>>>> _______________________________________________
>>>> http://www.marss86.org
>>>> Marss86-Devel mailing list
>>>> [email protected]
>>>> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>>>>
>>>
>>>
>>>
>>
>> When it crashes, can you type "layout asm" and paste the highlighted
>> line and the lines around it?
>>
>> Also can you give the dump of /proc/cpuinfo
>>
>> Thanks,
>> Tyler
>>
>>
>
>



_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to