[gem5-users] how much clock cycles gem5 needs to boot ?

2012-10-01 Thread Ardalan Pouya
Hello everyone , I want to trace cache replacement requests when running benchmark on gem5 and for that I'm using debug-flag CacheRepl and i want the exact cycle that gem5 needs for booting to skip boot replacements traces. does anyone know how many is that ? ___

[gem5-users] segmentation fault while running fft splash2 benchmark

2012-10-01 Thread Pavlos Maniotis
Hello every one. I have followed the instructions here: https://docs.google.com/document/preview?id=1B7nZSqMLwkwoVNEj_58tMPTk4bKWvoEMbokOAjqeC-k&pli=1 and i am trying to run fft benchmark in ALPHA ISA full system mode. last lines of system.terminal output: Bridge firewalling registered 802.1

Re: [gem5-users] questions on running benchmark on gem5

2012-10-01 Thread Anthony Gutierrez
Are you copying the checkpoint to the node on which it's running? On Mon, Oct 1, 2012 at 12:11 AM, Fangfei Liu wrote: > Sorry I may use the wrong command when I run the executable with gdb to > get previous information. The problem I met is a little bit strange. I'm > running gem5 on a cluster.

Re: [gem5-users] Question on PARSEC+GARNET+RUBY

2012-10-01 Thread Pavan Poluri
Hello, Thanks for your reply. If the 1-flit packets of type HEAD_TAIL_ are control messages, I am unable to figure out why do we need to perform Routing Computation, Virtual Channel allocation etc computations on these HEAD_TAIL_ flits? Thanks for your time. Thanks, Pavan On Tue, Sep 25, 2012 a

Re: [gem5-users] Question on PARSEC+GARNET+RUBY

2012-10-01 Thread Tushar Krishna
Why not? HEAD_TAIL_ is equivalent to a HEAD_ right (it just so happens that there are no BODY and TAIL flits in this packet). It still needs to choose a route, reserve a VC (a 1-flit-deep VC in this case). - Tushar On Oct 1, 2012, at 1:58 PM, Pavan Poluri wrote: > Hello, > > Thanks for your

Re: [gem5-users] Question on PARSEC+GARNET+RUBY

2012-10-01 Thread Pavan Poluri
Yes HEAD_TAIL_ is equivalent to HEAD_ without BODY and TAIL flits. So a single control packet is like a path guider to a single data packet? Every time a router injects a data packet into the network, it injects a control packet to initiate the data packet injection? Thanks, Pavan On Mon, Oct 1,

Re: [gem5-users] questions on running benchmark on gem5

2012-10-01 Thread Fangfei Liu
Actually no. But I think the directory where checkpoint is located is shared by all nodes via GPFS. There are no compiler but only runtime environment installed on the nodes. Will this be a potential problem? Thanks! From: gem5-users-boun...@gem5.org [gem5-u

[gem5-users] A possible bug when forwarding between stq and ldt

2012-10-01 Thread Veydan Wu
Hi, all, I wonder if anyone encountered this before. I ran a SPEC2006 application (GemsFDTD), there are two instructions 1: stq r24, 16(r30) followed by 2: ldt f0, 16(r30), very close. In the unmodified OOO execution, 1 forwards the value (0x17) to 2 correctly. If I separated them, and make 1 com

Re: [gem5-users] questions on running benchmark on gem5

2012-10-01 Thread Fangfei Liu
I find that it may not be related to checkpoint but segmentation occurs for all the configuration using detailed/arm_detailed cpu type. Is it because of the version of runtime envorinment? From: gem5-users-boun...@gem5.org [gem5-users-boun...@gem5.org] on behal

Re: [gem5-users] questions on running benchmark on gem5

2012-10-01 Thread Ali Saidi
Without a back trace of where the segmentation fault is happening and perhaps the output of valgrind we're not going to be much help. Ali On 01.10.2012 15:51, Fangfei Liu wrote: > I find that it may not be related to checkpoint but segmentation occurs for all the configuration using detail

[gem5-users] Doing Switch CPU when running in FS mode and using Ruby

2012-10-01 Thread Mahshid Sedghi
Hello all, I am trying to run a benchmark in full system mode on X86 and I am using Ruby. I want to load the kernel with some simple CPU and then switch to the detailed cpu to run my benchmark. I tried two options: - I used "switchcpu" in my script, but gem5 exits after loading the kernel.

Re: [gem5-users] Doing Switch CPU when running in FS mode and using Ruby

2012-10-01 Thread Anthony Gutierrez
The "switchcpu" pseudo op only calls m5 exit, i.e., it's not fully implemented. You could add a call to m5 exit to your rcS script right before it launches your benchmark, then have the config script do a drain and switch out the cpus. -Tony On Mon, Oct 1, 2012 at 7:47 PM, Mahshid Sedghi wrote:

Re: [gem5-users] Doing Switch CPU when running in FS mode and using Ruby

2012-10-01 Thread Mahshid Sedghi
Thanks for the quick response. What about standard switch (-s)? This is not also working? Mahshid On Mon, Oct 1, 2012 at 7:50 PM, Anthony Gutierrez wrote: > The "switchcpu" pseudo op only calls m5 exit, i.e., it's not fully > implemented. You could add a call to m5 exit to your rcS script righ

Re: [gem5-users] Doing Switch CPU when running in FS mode and using Ruby

2012-10-01 Thread Anthony Gutierrez
If you look at configs/common/Simulation.py you can see what each options does. -Tony On Mon, Oct 1, 2012 at 8:23 PM, Mahshid Sedghi wrote: > Thanks for the quick response. What about standard switch (-s)? This is > not also working? > > > Mahshid > > > > On Mon, Oct 1, 2012 at 7:50 PM, Anthony

Re: [gem5-users] questions on running benchmark on gem5

2012-10-01 Thread Fangfei Liu
Hi, This is the back trace of the stack for the segmentation fault: command line: /tigress-hsm/fangfeil/gem5/build/ARM/gem5.opt --outdir=bbench2 configs/example/fs.py -b bbench-gb --kernel=vmlinux.smp.mouse.arm --frame-capture --caches --cpu-type=detailed Global frequency set at 1 t

Re: [gem5-users] questions on running benchmark on gem5

2012-10-01 Thread Ali Saidi
That is very odd... Could you run the debug binary and see if you get a better trace. Also could you run it under valgrind? Thanks, Ali On 01.10.2012 19:33, Fangfei Liu wrote: > Hi, > > This is the back trace of the stack for the segmentation fault: > > command line: /tigress-hsm/fangf