Way back when I was first adding SPARC support, I had hoped to get around providing FP support by making our implementation pretend there was no FPU and to fault on the instructions that required it like the spec said I could. That didn't really work out, and most of the instructions that just faulted were replaced by functional equivalents. It seems your workload is trying to use one that was never properly implemented. If you can identify what instruction it is you could implement it yourself, or if you're not in a hurry you can give me a way to reproduce the problem and I can give it a shot. I can't make any strong guarantees about when I could take care of it though.
Gabe Gustavo Henrique Nihei wrote: > Hi, > > This fatal error is happening during the simulation of a SPARC_SE > platform. > > M5 Simulator System > > Copyright (c) 2001-2008 > The Regents of The University of Michigan > All Rights Reserved > > > M5 compiled May 5 2010 14:24:46 > M5 revision 94c016415053+ 6283+ default tip > M5 started May 11 2010 15:11:44 > M5 executing on nime-desktop > command line: ../../build/SPARC_SE/m5.fast run.py -n 4 -t > Global frequency set at 1000000000000 ticks per second > 0: system.remote_gdb.listener: listening for remote gdb on port 7002 > 0: system.remote_gdb.listener: listening for remote gdb on port 7003 > 0: system.remote_gdb.listener: listening for remote gdb on port 7000 > 0: system.remote_gdb.listener: listening for remote gdb on port 7001 > info: Entering event queue @ 0. Starting simulation... > fatal: fault (fp_disabled) detected @ PC 0x10b194 > @ cycle 37023056000 > [invoke:build/SPARC_SE/sim/faults.cc, line 43] > Memory Usage: 220052 KBytes > For more information see: http://www.m5sim.org/fatal/f78c96bc > > I'm trying to run the MPGenc software in the platform. > I suppose the fault "fp_disabled" is related to a floating point > operation, right? > Is there some kind of workaround for this problem? > > -- > Gustavo Henrique Nihei > ------------------------------------------------------------------------ > > _______________________________________________ > 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
