Re: [gem5-users] Question about multi-program simulation usingspec2006

Hi, Jason, Thanks for your reply, I tried a lot of time to set up the full system simulation environment based some forums on your blog but failed. There will be a lot issues when building the kernel for full system. Is that possible that you can give more guidance on building full system envir

Re: [gem5-users] Zero Compressed Data with Spec2017

Hi Debiprasanna, I can't answer your question about compression. However, if you'd like to contribute, please check out the CONTIRUBTING.md document. I'm not sure exactly what your script does, so it's hard to say where the best location would be. Possibly in configs/ if it is a configuration or

Re: [gem5-users] Question about multi-program simulation using spec2006

Hi Fransisco, It depends on what you mean by reliable. It's important to understand the tradeoffs between different fidelities of simulation. In SE mode, there is no OS, page table, etc. Any system-level effects will be either completely ignored or not modeled correctly. Additionally, while the fu

Re: [gem5-users] Question about multi-program simulation using spec2006

Hi Jason and Yuan, Is this also valid to multithreaded applications (using pthreads library) in SE mode? If so, the results that gem5 simulator generates, in SE mode for multithreaded applications, is not reliable. thanks in advance.

Re: [gem5-users] Question about multi-program simulation using spec2006

Hi Yuan, Generally, it is difficult to get multiprogrammed workloads running in SE mode. It may be possible, but it probably isn't very realistic anyway. I suggest using FS mode. Cheers, Jason On Mon, Sep 23, 2019 at 5:24 PM yuan wrote: > Hi, All, > > > > I am trying multi program simulation u

[gem5-users] Missing necessary X86 KVM patches from main repo

Hello All, I have found the following 2 patches necessary to get X86 KVM working properly. However, they don't seem to be part of the main repo. https://gem5-review.googlesource.com/c/public/gem5/+/12278 https://gem5-review.googlesource.com/c/public/gem5/+/13855/1 I was wondering if there is a r

Re: [gem5-users] Snoop filter panic when bypassing blocks to MM

Hi Esteban, hi Rich, Your modifications seem to confuse the snoop filter that does sanity checks for the implemented coherence protocol. I'm not sure about the specifics of your changes but I think a safe way to bypass a cache for a given writeback would be to modify BaseCache::access() https://g

[gem5-users] Zero Compressed Data with Spec2017

Hi All, I am trying to evaluate the percentage of Zero Compressed data with Spec2017 on gem5 revision-603f137. I observe that most of these benchmarks have extremely high percentage of zeros. Even some benchmarks like leela and deepsjeng show that 100% of the data is zero. The simulation was done