Has anyone come across such an issue or can any one suggest me what is the appropriate way of finding out where the problem lies? Try using "valgrind" to check where the memory leak is coming from. It's a tremendously useful tool.
you'll want to run on a debug binary and for a limited number of cycles as well.. a command line such as "valgrind --tool=memcheck --leak-check=yes <cmd_line>" usually works for me. (it's a lot of output, be warned). > The GHB, prefetching mechanism works fine in all cases. > > Another problem that i am currently facing is as follow: > There are only integers and floating point numbers in the application that > i am trying to run. Check the mailing list archives but I'm not sure Floating Point support is all the way there for x86. I'll let someone else chime in. > build/X86_SE/cpu/simple/timing.cc:438: Fault TimingSimpleCPU::read(Addr, > T&, unsigned int) [with T = uint64_t]: Assertion `split_addr <= addr || > split_addr - addr < block_size' failed. > That looks like a error in the code trying to determine something with split accesses (accesses that span multiple cache blocks)... It could be something to do with your prefetcher...Or it could be something with the memory leak corrupting values. -- - Korey
_______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
