Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-04 Thread Guanglei Cui
I was following http://www.gromacs.org/Documentation/Installation_Instructions. The link to 4.6.3 regression test set isn't obvious. Following the pattern, I downloaded the 4.6.3 regression test tarball (which apparently unpacks to a folder named for 4.6.2). Now, GMX_CPU_ACCELERATION=None passes al

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Mark Abraham
Please test using the test set version that matches the code! On Sep 4, 2013 5:16 AM, "Guanglei Cui" wrote: > Hi Szilard, > > Thanks for your reply. I may try your suggestions tomorrow when I get back > to work. > > Feeling curious, I downloaded and compiled gmx 4.6.3 on my home computer > (gcc-4

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Guanglei Cui
Hi Szilard, Thanks for your reply. I may try your suggestions tomorrow when I get back to work. Feeling curious, I downloaded and compiled gmx 4.6.3 on my home computer (gcc-4.6.3 and ubuntu 12.04). Even with the default (below), kernel (38 out of 142) and freeenergy (2 out of 9) tests would stil

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Mark Abraham
On Tue, Sep 3, 2013 at 7:47 PM, Guanglei Cui wrote: > Dear GMX users, > > I'm attempting to compile gromacs 4.6.3 with an older Intel compiler (ver > 11.x). Here is how I compiled FFTW, > > ./configure CC=icc F77=ifort CFLAGS="-O3 -gcc" > --prefix=/tmp/gromacs-4.6.3/fftw-3.3.3/build-intel-threads

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Szilárd Páll
On Tue, Sep 3, 2013 at 9:50 PM, Guanglei Cui wrote: > Hi Mark, > > I agree with you and Justin, but let's just say there are things that are > out of my control ;-) I just tried SSE2 and NONE. Both failed the > regression check. That's alarming, with GMX_CPU_ACCELERATION=None only the plain C ker

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Mark Abraham
Can't be sure - we don't know what the problem with the compiler *is*. Mark On Tue, Sep 3, 2013 at 9:50 PM, Guanglei Cui wrote: > Hi Mark, > > I agree with you and Justin, but let's just say there are things that are > out of my control ;-) I just tried SSE2 and NONE. Both failed the > regressio

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Guanglei Cui
Hi Mark, I agree with you and Justin, but let's just say there are things that are out of my control ;-) I just tried SSE2 and NONE. Both failed the regression check. I think I've spent enough time on this, which justifies escalating this to someone with the control, but is failing regression chec

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Guanglei Cui
Hi Justin, Thanks for the response. Right now, I will just have to get around this by using a slower code. How do I switch off SSE4.1 and use a different CPU optimization scheme? Regards, Guanglei On Tue, Sep 3, 2013 at 2:39 PM, Justin Lemkul wrote: > > > On 9/3/13 1:47 PM, Guanglei Cui wrote

Re: [gmx-users] gromacs 4.6.3 and Intel compiiler 11.x

2013-09-03 Thread Justin Lemkul
On 9/3/13 1:47 PM, Guanglei Cui wrote: Dear GMX users, I'm attempting to compile gromacs 4.6.3 with an older Intel compiler (ver 11.x). Here is how I compiled FFTW, ./configure CC=icc F77=ifort CFLAGS="-O3 -gcc" --prefix=/tmp/gromacs-4.6.3/fftw-3.3.3/build-intel-threads --enable-threads --ena