Re: NAMD 2.7b1 on Athlon64 Gigabit MPI - Compile Error "src/Settle.C:153: error: ‘_mm_cvtsd_f64’ was not declared in this scope"

From: Greg Guthe (gguthe_at_gmail.com)
Date: Thu Apr 02 2009 - 21:02:44 CDT

gcc 4.0.3 and Opteron processors apparently, guess I'll try the
charm++ amd64 version.

Thanks much,
Greg

~/NAMD_2.7b1_Source/arch$ gcc -dumpversion
4.0.3
~/NAMD_2.7b1_Source/arch$ cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 15
model : 33
model name : Dual Core AMD Opteron(tm) Processor 270
stepping : 2
cpu MHz : 2004.603
cache size : 1024 KB
physical id : 0
siblings : 2
core id : 0
cpu cores : 2
fpu : yes
fpu_exception : yes
cpuid level : 1
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 mmxext fxsr_opt lm
3dnowext 3dnow pni lahf_lm cmp_legacy
bogomips : 4017.33
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

processor : 1
vendor_id : AuthenticAMD
cpu family : 15
model : 33
model name : Dual Core AMD Opteron(tm) Processor 270
stepping : 2
cpu MHz : 2004.603
cache size : 1024 KB
physical id : 0
siblings : 2
core id : 1
cpu cores : 2
fpu : yes
fpu_exception : yes
cpuid level : 1
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 mmxext fxsr_opt lm
3dnowext 3dnow pni lahf_lm cmp_legacy
bogomips : 4009.39
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

processor : 2
vendor_id : AuthenticAMD
cpu family : 15
model : 33
model name : Dual Core AMD Opteron(tm) Processor 270
stepping : 2
cpu MHz : 2004.603
cache size : 1024 KB
physical id : 1
siblings : 2
core id : 0
cpu cores : 2
fpu : yes
fpu_exception : yes
cpuid level : 1
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 mmxext fxsr_opt lm
3dnowext 3dnow pni lahf_lm cmp_legacy
bogomips : 4009.39
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

processor : 3
vendor_id : AuthenticAMD
cpu family : 15
model : 33
model name : Dual Core AMD Opteron(tm) Processor 270
stepping : 2
cpu MHz : 2004.603
cache size : 1024 KB
physical id : 1
siblings : 2
core id : 1
cpu cores : 2
fpu : yes
fpu_exception : yes
cpuid level : 1
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 mmxext fxsr_opt lm
3dnowext 3dnow pni lahf_lm cmp_legacy
bogomips : 4009.39
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management: ts fid vid ttp

On Thu, Apr 2, 2009 at 9:39 PM, Peter Freddolino <petefred_at_ks.uiuc.edu> wrote:
> In the short term, you should try adding
> -DMISSING_mm_cvtsd_f64
> to the compile flags in the .arch file that you're using, which ought to fix
> things. So that we can have a better idea of what's going on with your case,
> could you let us know what version of gcc you're using, and the exact model
> of processor (cat /proc/cpuinfo)?
>
> Best,
> Peter
>
> Greg Guthe wrote:
>>
>> Hi,
>>
>> I'm trying to compile NAMD 2.7b1 on a 5 node 8 ppn Athlon64 Gigabit
>> cluster using MPI messaging with OpenMPI 4.0.3.
>>
>> $ uname -a
>> Linux babbage 2.6.15-52-amd64-generic #1 SMP PREEMPT Wed Oct 22
>> 18:50:59 UTC 2008 x86_64 GNU/Linux
>> ~/NAMD_2.7b1_Source/Linux-x86_64-g++$ make
>> ..
>> g++ -I.rootdir/charm-6.1/mpi-linux-x86_64-mpicxx/include
>> -DCMK_OPTIMIZE=1 -Isrc -Iinc   -Iplugins/include -DSTATIC_PLUGIN
>> -I/n/home/gguthe/NAMD_2.7b1_Source/linux-x86_64/include -DNAMD_TCL
>> -I/n/home/gguthe/NAMD_2.7b1_Source/linux-x86_64/include -DNAMD_FFTW
>> -DNAMD_VERSION=\"2.7b1\" -DNAMD_PLATFORM=\"Linux-x86_64\"
>> -DREMOVE_PROXYRESULTMSG_EXTRACOPY   -O3 -m64 -fexpensive-optimizations
>> -ffast-math  -o obj/Settle.o -c src/Settle.C
>> src/Settle.C: In function ‘int settle1(const Vector*, Vector*,
>> Vector*, BigReal)’:
>> src/Settle.C:153: error: ‘_mm_cvtsd_f64’ was not declared in this scope
>> make: *** [obj/Settle.o] Error 1
>>
>> charm++ built and passed local megatest on archs:
>> mpi-linux-x86_64-SMP-mpicxx
>> mpi-linux-x86_64-mpicxx (smart builds)
>>
>> mpi-linux-x86_64
>> mpi-linux-amd64 (for Opteron but I built 2.6 using amd-64 thought it
>> was worth a shot)
>> net-linux-x86_64
>>
>> Tried those with the corresponding NAMD archs Linux-x86_64-MPI-pathcc
>> and Linux-x86_64-g++ and tried specifying a new arch
>> Linux-x86_64-MPI.arch using mpi compilers:
>>
>> ~/NAMD_2.7b1_Source/arch$ cat Linux-x86_64-MPI.arch
>> NAMD_ARCH = Linux-x86_64
>> CHARMARCH = mpi-linux-x86_64-mpicxx
>>
>> CXX = mpiCC
>> CXXOPTS = -O3 -m64 -fexpensive-optimizations -ffast-math
>> CC = mpicc
>> COPTS = -O3 -m64 -fexpensive-optimizations -ffast-math
>>
>> I tried commenting line 198   of the makefile to "#
>> $(DSTDIR)/Settle.o \" but I'm not sure how NAMD links Settle.o and
>> then Mom said it was past my 8pm bedtime and we just run the cluster
>> to keep the dog warm so ...
>>
>> Anywho,
>> Greg
>>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:33 CST