Re: [gem5-users] Increasing TLB size not working for X86 with O3CPU

Hi, Jason Sorry for my unclear description before. For our workload, the switch_cpus.dtb's miss rate for 64 tlb entries is 154654 / 1589214 = 9.74%; the miss rate for 1048576 tlb entries is 154360 / 1583757 = 9.73%. Both are running for 20ms warm up in atomic mode and 2.5ms real simulation with

Re: [gem5-users] x86 floating point instruction

Thanks Jason. I also came across the same document earlier but I just wanted to ask about this in general. On Mon, May 28, 2018 at 11:09 AM, Jason Lowe-Power wrote: > Hi Tariq, > > It's up to you what you want the latency for SSE instructions to be. It > depends on what architecture you're

Re: [gem5-users] How to run a dynamically linked executable syscall emulation mode se.py in gem5?

Thanks, it does work on non native QEMU user mode with the -L option: https://github.com/cirosantilli/linux-kernel-module-cheat/tree/b60c6f1b9c8bb7c34cf2c7fbce6f035d11483d4c#qemu-user-mode On Mon, May 28, 2018 at 5:01 PM, Jason Lowe-Power wrote: > Hi Ciro, > > As you seemed to have figured out,

Re: [gem5-users] x86 floating point instruction

Hi Tariq, It's up to you what you want the latency for SSE instructions to be. It depends on what architecture you're simulating. Unfortunately, we currently don't have any "known good" configurations for x86 cores so you'll have to come up with your own :). Here's some examples of numbers you

Re: [gem5-users] Response for WritebackDirty packets (learning.gem5)

Hi Muhammad, Generally, if sendTimingReq fails, you have to save the packet so you can resend it. In my Learning gem5 code, I *try* to simplify the retry logic so that this is hidden. Instead of saving the packet in the cache code, the packet is saved in the port code. Also, the code in Learning

Re: [gem5-users] Increasing TLB size not working for X86 with O3CPU

Hi Da, "For size > 512, the whole stats.txt is identical." This isn't surprising. 512*4KB = 2MB. So, if your workload is only 1MB when you have at least 512 entries you are only seeing compulsory (cold) misses. Try running larger workloads and/or workloads with more reuse. Cheers, Jason On

Re: [gem5-users] How to run a dynamically linked executable syscall emulation mode se.py in gem5?

Hi Ciro, As you seemed to have figured out, running dynamically linked executables has only been tested for x86_64 native platforms. It *is supported* if your binary is x86 and your native machine is x86. I'm not sure what it would take to get this working for native ARM machines (e.g.,

Re: [gem5-users] RISCV ISA : "C" (compressed) extension supported?

Hi Marcelo, For future reference, if someone else has this issue... Another possibility is that the branch predictor is the problem. It looks like it could be predicting that instruction is a branch. I'm not sure if it's specifically because of the compressed format or not, though. It's another