I don't know the details, but I believe there is some special Fortran runtime code that C doesn't use, and perhaps merely linking with -static doesn't actually get that linked in.
I'm surprised you wouldn't get some error or warning about an unimplemented syscall or a file that's not found before you got the segfault though. Steve On Wed, Feb 18, 2009 at 3:44 AM, Nikolas Galanis <[email protected]>wrote: > Hi, > > I have compiled the SPECcpu2000 v1.2 natively on an Alpha machine > running Linux. All C benchmarks seem to be executing without any problem > in m5 SE mode. However, when I try to execute any fortran benchmark, I > get an error like the one below: > > M5 compiled Feb 16 2009 19:27:08 > M5 revision 5591:be93223fd0c3c8faabbd337db9e5e576c8905397 > M5 commit date Mon Dec 15 18:10:21 2008 +0100 > M5 started Feb 18 2009 12:36:33 > M5 executing on nikolas-dac > command line: ../build/ALPHA_SE/m5.debug ../tests/myrun.py > cpu2000/galgel/ref/alpha/linux/se_o3_detailed > Global frequency set at 1000000000000 ticks per second > 0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000 > warn: Increasing stack size by one page. > panic: Tried to access unmapped address 0x3ffc0096ce0. > @ cycle 317500 > [invoke:build/ALPHA_SE/arch/alpha/faults.cc, line 204] > Program aborted at cycle 317500 > Aborted > > I am using the "ref" inputs for all benchmarks. > > I have compiled the benchmarks using the script provided with the v1.2 > distribution adding only the -static flag. > > Am I missing something? > > Thanks in advance, > > Nikolas > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >
_______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
