Re: [gem5-users] DRAM Controller (gem5 vs DRAMSim2)

Hi, You can use Ramulator with gem5. It already has SALP and is also integrated On Jun 22, 2017 12:08 AM, "Muzamil Rafique" wrote: Hi All, I have a doubt on my mind for a while if DRAM Controller model in gem5 is more detailed and accurate or the DRAM Controller

Re: [gem5-users] DRAM Controller (gem5 vs DRAMSim2)

SALP is not available in either of them. You will have to implement it yourself. I have experimented with both of them. I think Dramsim2 is more suitable for your work. Although Dramsim2 is slower than gem5 built in controller. You can just start with looking at the code of both of them and then

[gem5-users] DRAM Controller (gem5 vs DRAMSim2)

Hi All, I have a doubt on my mind for a while if DRAM Controller model in gem5 is more detailed and accurate or the DRAM Controller model of DRAMSim2? Do we have support like sub-array level parallelism (SALP) and refreshing certain selected number of rows (instead of a whole bank) available in

Re: [gem5-users] Parsec 2.1 + Garnet 2.0: Too Slow!!!

Hi Faisal, I see you are running a 256 threaded vips on the 16 core system. Maybe try a 16 threaded version? Also, can you try running without garnet -- I'm curious to see how long the benchmarks take in that case. Cheers, Tushar On Jun 21, 2017, at 3:07 AM, F. A. Faisal

Re: [gem5-users] Invalid transition DMA controller in NoC

Hi Muhammad, I am afraid, I haven't looked much in the code for MESI_Two_Level, I remember I run into the same issue as you and I figured what was the problem but I haven't been able to find some time to fix it. >From what I can tell you're looking the right direction. It seems like the

Re: [gem5-users] Invalid transition DMA controller in NoC

Hi Nikos, I have tried to add the support for multiple DMA controllers to MESI Two Level by using the Machine ID of the DMA controllers. The Node ID in machine ID is the version of the machine. As per config.ini i get three different DMA controllers with versions 0 1 2. Now the Node ID is based

Re: [gem5-users] SystemC and GEM5 SE Mode

Hi Christian, I will start looking at the address mappings in SE mode as a starting point for this and hope to get a working system. Thank you for your help and suggestions. Regards Yasir -Original Message- From: Christian Menard [mailto:christian.men...@tu-dresden.de] Sent: 20 June

[gem5-users] Parsec 2.1 + Garnet 2.0: Too Slow!!!

Dear All, I tried to simulate the Parsec 2.1 benchmarks with garnet 2.0. I found that other than Blackscholes, rest of benchmarks are taking several days (and still executing) and even I tried with increasing the num of benchmarks, which actually don't have too much effect. However, the actual