[gem5-users] warmup in SE + Ruby mode

Hi all, I adopt gem5 in SE plus Ruby mode. I want to warm up cache before data measurement. During my experiment, I found the '-W' option does not work. The simulation process is controlled in Ruby mode from scratch instead of after instructions as expected. I check the file /config/common/S

Re: [gem5-users] warmup in SE + Ruby mode

09:38 PM, Nilay Vaish wrote: On Tue, 27 Nov 2012, hanfeng QIN wrote: Hi all, I adopt gem5 in SE plus Ruby mode. I want to warm up cache before data measurement. During my experiment, I found the '-W' option does not work. The simulation process is controlled in Ruby mode from scrat

Re: [gem5-users] warmup in SE + Ruby mode

On 11/29/2012 01:07 AM, Nilay Vaish wrote: On Wed, 28 Nov 2012, hanfeng QIN wrote: I used mercurial to clone the latest version of gem5 in development repository. By the way, I used to work on the stable release, however, when I was trapped in a coherence protocol error then I switched to

Re: [gem5-users] warmup in SE + Ruby mode

On 11/30/2012 11:21 AM, Nilay Vaish wrote: On Fri, 30 Nov 2012, hanfeng QIN wrote: On 11/29/2012 01:07 AM, Nilay Vaish wrote: On Wed, 28 Nov 2012, hanfeng QIN wrote: I used mercurial to clone the latest version of gem5 in development repository. By the way, I used to work on the stable

Re: [gem5-users] warmup in SE + Ruby mode

On 11/30/2012 07:53 PM, Nilay Vaish wrote: On Fri, 30 Nov 2012, hanfeng QIN wrote: I apologize for my carelessness, Nilay. I am sorry to say that I still can not prevent the error even I revert the patch. To make you understand my simulation flow, I list my process as following. Step 1: I

Re: [gem5-users] warmup in SE + Ruby mode

On 11/30/2012 07:53 PM, Nilay Vaish wrote: On Fri, 30 Nov 2012, hanfeng QIN wrote: I apologize for my carelessness, Nilay. I am sorry to say that I still can not prevent the error even I revert the patch. To make you understand my simulation flow, I list my process as following. Step 1: I

[gem5-users] Run PARSEC with MESI protocol in FS mode

Hi, I download the PARSEC images with pre-compiled benchmark suite from http://www.cs.utexas.edu/~cart/parsec_m5/ . I extract the pre-compiled benchmarks into my own disk image made from util/gem5img.py. According to the document in the above URL,

Re: [gem5-users] Run PARSEC with MESI protocol in FS mode

Starting simulation... . Exiting @ tick 2585108236000 because m5_exit instruction encountered Obviously, MESI can support Ruby well. But I can not explain why gem5 hangs when run with --ruby only. Best, Hanfeng On 12/04/2012 09:27 AM, hanfeng QIN wrote: Hi, I download the PARSEC image

[gem5-users] simulation methodology

Hi all, I learn a common multi-programmed simulation methodology adopted by many architecture researchers. But I am not clear its implementation internals. I describe its idea in brief as following. For multi-programmed workload consists of M programs, this methodology firstly fast-forwards

Re: [gem5-users] simulation methodology

ogram (thus total instruction simulated will be (N3 * 2) * Nr_cores) but only dump the stats after the first N3 instructions? But I am not clear on the stats dump internals. Hanfeng On 12/13/2012 11:45 PM, Nilay Vaish wrote: On Wed, 12 Dec 2012, hanfeng QIN wrote: Hi all, I learn a comm

Re: [gem5-users] simulation methodology

e get stats related to c0 in the 1st dump and that to c2 in the 2nd dump and so on. Here we must note that the stats dump order mirrors the order in which every program finishes N3 instructions. Thanks, Hanfeng On 12/14/2012 06:52 PM, Nilay Vaish wrote: On Fri, 14 Dec 2012, hanfeng QIN wro

[gem5-users] run SPEC CPU2K6 in SE mode

I found the simulator will sleep when I run 433.milc and 437.leslie3d from SPEC CPU2K6 in SE mode. However, both of them run normally on native real machine. Compared with the native run, I guess this phenomenon may relate to how gem5 processes '-c' and '-o' options. I check the run time parame

[gem5-users] about simulation statistics

Hi, I collect a group of simulation statistics in ALPHA_SE mode with classic memory system. The workload is selected from SPEC CPU 2K6 and is run in detailed measurement with 10 billion instructions. I compare the CPI and computed MPKI with the one obtained from native run (in x86_64 architec

[gem5-users] discussion on modeling shared L3 cache hierarchy

Hi all, I model a shared L3 while private L2 cache hierarchy with gem5. During my experiments (running gem5 under classic memory model and SE mode), I control the main simulation parameters as following. 1 core private L1 dcache: 32KB/8-way; icache: 32KB/4-way private L2 cache: 256KB/8-way sh

Re: [gem5-users] discussion on modeling shared L3 cache, hierarchy

: Hi Hanfeng, Without having looked at the numbers, have you had a look at the generated config.dot.pdf in m5out to ensure that the system architecture ends up being what you intend? You need py-dot installed for the figure to be generated. Andreas On 13/03/2013 08:19, "hanfeng QIN" wr

[gem5-users] '-I' option in FS mode

Hi all, If I run gem5 in FS mode, I specify the maximum instruction count to be simulation by '-I' option. However, does the maximum instruction count contain instructions from the operating system? I am not sure. Can any one assert it? Thanks, Hanfeng ___

[gem5-users] use simpoint on multi-programmed workload

Hi all, To reduce simulation time, Simpoint approach is used in my simulation. However, I am not aware of how to perform simpoint on multi-programmed workloads. For example, suppose a workload consists of 429.mcf plus 470.lbm. I have got the simpoint for each of the two applications using Pi