Madan, I can't say much about the pgi compiler. However, about ifort 11, the file sent to the list by José Torres with the changes to Siesta 3 rc1 states EXPLICITLY that ifort 11 does not work for compiling siesta, and that users should stick to ifort 10.x (I personally recommend 10.1). Also, if you search the list archives, you'll see that apparently the vast majority of users have had some sort of fatal problem using ifort 11. I don't remember anyone having sent an arch.make for compilation with ifort 11 saying is works perfectly, either serial or parallel. Unfortunately, "no compilation errors" and "a successful compilation" form no more than an oxymoron... So, try ifort 10.1 instead. It can be installed as non-root under your home directory, if needed be.
Marcos On Wed, Apr 28, 2010 at 2:52 PM, Madan Mithra. L. M <[email protected]>wrote: > Dear All, > > (1) I am trying to compile Siesta3 Release candidate1 using PGI > compiler (Version 10.4) and get this error : > > ---------- > pgf90 -c -g -O0 `FoX/FoX-config --fcflags` > /d/Comp/s339/Src/timestamp.f90 > pgf90 -c -g -O0 `FoX/FoX-config --fcflags` > /d/Comp/s339/Src/siesta_cmlsubs. > F90 > PGF90-S-0155-Could not resolve generic procedure str > (d:/Comp/s339/Src/siesta_cm > lsubs.F90: 64) > 0 inform, 0 warnings, 1 severes, 0 fatal for siesta_cml_init > make: *** [siesta_cmlsubs.o] Error 2 > ----------------------------- > > The line 64 in siesta_cmlsubs.F90 reads this : > > Call cmlAddMetadata(mainXML, name='siesta:Nodes', > content=str_fox(nodes)) > > Could someone please help me to figure out as to what went wrong with > this file ? > > Please note that I could successfully compile Siesta 3.0 Beta. The > problem arises when compiling the development versions and the Release > candidate 1. > > (2) I could successfully compile Siesta using Ifort Ver.11.1 (64 bit) > in linux. However, when I run the test job, the code generates few > lines of output and then stops giving the following error. Could it be > due to the buggy compiler ? > > ----------- > > *** glibc detected *** /home/dell-scratch/physunil/mit/siesbin/siesta: > malloc(): memory corruption: > 0x0000000004c390f0 *** > ======= Backtrace: ========= > /lib64/libc.so.6[0x7f073f0f2118] > /lib64/libc.so.6[0x7f073f0f4a5c] > /lib64/libc.so.6(__libc_malloc+0xa1)[0x7f073f0f6891] > /lib64/libc.so.6(posix_memalign+0x68)[0x7f073f0f6cf8] > /home/dell-scratch/physunil/mit/siesbin/siesta[0xaf36d8] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x7a9b27] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x7754e1] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x7685e7] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x762b2c] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x740064] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x5f67b1] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x519296] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x52b839] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x40c93c] > /lib64/libc.so.6(__libc_start_main+0xe6)[0x7f073f09c586] > /home/dell-scratch/physunil/mit/siesbin/siesta[0x40c839] > --------------------------------------------- > > Thank you. > > -- Madan > ---------------------------------------- > Madan Mithra .L.M > Senior Research Fellow > Dept.of Physics > Indian Institute of Science > Bangalore - 560 012 > INDIA > ......................................... > Phone : 91-80-22932313 (Lab) > FAX : (080) 360 2602 > ---------------------------------------- >
