Hi Sharjeel,

If it crashes early on at boot where linux still executes using only one core 
you could try to enable the ExecNoTicks flag, it will give you an instruction 
trace that you could compare against the working simulation. That should give 
you an idea of where things start to go wrong.


Since this bug only affects the system where you use a mix of O3CPU and 
MinorCPU configurations, I would expect though that it happens after you power 
at least one more core. You could still use the ExecNoTicks,-ExecSymbol flags 
and have a look at the PC in the trace, you might be able to identify the point 
where you start fetching and executing instructions from weird addresses.


HTH,


Nikos

________________________________
From: gem5-users <[email protected]> on behalf of SHARJEEL KHILJI 
<[email protected]>
Sent: 25 August 2017 15:32:35
To: gem5 users mailing list
Subject: Re: [gem5-users] ex5_big issues with ruby memory system

Hi,

Thanks for the reply. Yes it crashes as the Linux starts  to boot. What 
particular flags can i enable?

best regards,
Sharjeel

On 25 August 2017 at 19:03, Nikos Nikoleris 
<[email protected]<mailto:[email protected]>> wrote:

Hi Sharjeel,


I haven't been working that much with MESI_Two_Level so I am not sure whether 
it should work or not. The memory map shouldn't be a problem. Does it crash 
early on while Linux is booting? Maybe you could try to enable the debug-flags 
to understand what is going on.


Nikos



________________________________
From: gem5-users 
<[email protected]<mailto:[email protected]>> on behalf of 
SHARJEEL KHILJI 
<[email protected]<mailto:[email protected]>>
Sent: 25 August 2017 13:41:32
To: gem5 users mailing list
Subject: Re: [gem5-users] ex5_big issues with ruby memory system

Hi,
I am attaching fs_het.py for reference.

best regards,
Sharjeel

On 25 August 2017 at 17:36, SHARJEEL KHILJI 
<[email protected]<mailto:[email protected]>> wrote:

Hi, thanks for the reply. I am ex5_big and ex5_LITTLE in the NoC. The protocol 
is MESI TWO LEVEL. The build statement is as follows.

./build/ARM/gem5.fast configs/example/fs_het.py --l1d_size=32kB --l1i_size=32kB 
--num-l2caches 4 --l2_size=1MB --cacheline_size=64 
--machine-type=VExpress_GEM5_V1 --kernel 
/home/khilji/gem5/m5/system/binaries/vmlinux-aarch32 --disk-image 
/home/khilji/gem5/m5/system/disks/arm-ubuntu-natty-headless.img --dtb-filename 
/home/khilji/gem5/m5/system/dtb/armv7_gem5_v1_big_little_2_2.dtb --num-cpus 4 
--cpu-type exynos --big-cpus 2 --little-cpus 2 --big-cpu-clock 2GHz 
--little-cpu-clock 1GHz --ruby --num-dirs=4 --network=garnet2.0 --topology 
Mesh_XY --mesh-rows 2 --mem-size 1GB

I think that ex5_big has some issues with ruby. Since I have used it separately 
in a simple NoC with all four cores as ex5_big and got the same issue with 
fs.py. The build statement for that is following

./build/ARM/gem5.fast configs/example/fs.py --l1d_size=32kB --l1i_size=32kB 
--num-l2caches 4 --l2_size=1MB --cacheline_size=64 
--machine-type=VExpress_GEM5_V1 --kernel 
/home/khilji/gem5/m5/system/binaries/vmlinux-aarch32 --disk-image 
/home/khilji/gem5/m5/system/disks/arm-ubuntu-natty-headless.img --dtb-filename 
/home/khilji/gem5/m5/system/dtb/armv7_gem5_v1_4cpu.dtb --num-cpus=4 --ruby 
--num-dirs=4 --cpu-type ex5_big --network=garnet2.0 --topology Mesh_XY 
--mesh-rows 2 --mem-size 1GB --cpu-clock 1GHz --routing_algorithm 1 
--work-begin-exit-count=1 --work-end-exit-count=1

Also does ex5_big supports memory map of VExpress_GEM5_V1? May be that is the 
issue. The ex5_LITTLE has no issue when it is used in the above NoC with four 
cores.Any suggestions ?

best regards,

Sharjeel



On 25 August 2017 at 16:58, Nikos Nikoleris 
<[email protected]<mailto:[email protected]>> wrote:

Hi Sharjeel,


Can you provide some more information about your simulation setup? What ruby 
protocol are you using and what is the workload you're running? It looks like 
the simulation is trying to load instructions from an invalid address.


Nikos



________________________________
From: gem5-users 
<[email protected]<mailto:[email protected]>> on behalf of 
SHARJEEL KHILJI 
<[email protected]<mailto:[email protected]>>
Sent: 24 August 2017 17:06:00
To: gem5 users mailing list
Subject: [gem5-users] ex5_big issues with ruby memory system

Hi,

I am getting following issue when I use ex5_big with ruby memory system.

warning : Address 0 is outside of physical memory, stopping fetch

which results in simulation limit reached. Can anyone tell me that does ex5_big 
really works with ruby memory system.

best regards,

Sharjeel


IMPORTANT NOTICE: The contents of this email and any attachments are 
confidential and may also be privileged. If you are not the intended recipient, 
please notify the sender immediately and do not disclose the contents to any 
other person, use it for any purpose, or store or copy the information in any 
medium. Thank you.

_______________________________________________
gem5-users mailing list
[email protected]<mailto:[email protected]>
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users


IMPORTANT NOTICE: The contents of this email and any attachments are 
confidential and may also be privileged. If you are not the intended recipient, 
please notify the sender immediately and do not disclose the contents to any 
other person, use it for any purpose, or store or copy the information in any 
medium. Thank you.

_______________________________________________
gem5-users mailing list
[email protected]<mailto:[email protected]>
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

IMPORTANT NOTICE: The contents of this email and any attachments are 
confidential and may also be privileged. If you are not the intended recipient, 
please notify the sender immediately and do not disclose the contents to any 
other person, use it for any purpose, or store or copy the information in any 
medium. Thank you.
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to