[gem5-users] Re: Pseudo Instruction - m5_reset_stats() - Body Modification

2021-10-12 Thread Sampad Mohapatra via gem5-users
I see. Thanks for the clarification. Will see if DPRINTF works, otherwise will write to a file, which should be ok as I am calling m5_stat_reset only once. On Tue, Oct 12, 2021 at 10:57 AM Jason Lowe-Power wrote: > I would suggest using DPRINTF instead of cout. It's possible that some > print

[gem5-users] Re: Pseudo Instruction - m5_reset_stats() - Body Modification

2021-10-12 Thread Jason Lowe-Power via gem5-users
I would suggest using DPRINTF instead of cout. It's possible that some print statements aren't being flushed. Jason On Mon, Oct 11, 2021 at 7:16 AM Sampad Mohapatra wrote: > Hi Jason, > > I have added a std::cout statement to the resetstats()'s body and I am > calling m5_reset_stats from my

[gem5-users] Re: Pseudo Instruction - m5_reset_stats() - Body Modification

2021-10-11 Thread Sampad Mohapatra via gem5-users
Hi Jason, I have added a std::cout statement to the resetstats()'s body and I am calling m5_reset_stats from my GPU benchmarks. The GPU kernels are launched right after reset is called. I pipe the output of simulations to a file. But, strangely enough some outputs show the std::cout statements

[gem5-users] Re: Pseudo Instruction - m5_reset_stats() - Body Modification

2021-10-04 Thread Jason Lowe-Power via gem5-users
Hi Sampad, Here is where m5_reset_stats is implemented in the simulator: https://gem5.googlesource.com/public/gem5/+/refs/heads/stable/src/sim/pseudo_inst.cc#303 There are a large number of steps between when the guest code calls m5_reset_stats and when the above function executes, but this