Discussion:
gem5-users Digest, Vol 147, Issue 16
(too old to reply)
Idris M. Umar
2018-10-23 21:22:03 UTC
Permalink
I really appreciated the fact that a member of the Garnet development team responded to my inquiry. Thanks a lot.

Further, I really need to get an XYZ routing for the 3D topology because my target is to control traffic flow at the bridge links by using suitable buffer occupancy threshold settings.

I attempted to used gdb debugger to check the variables in the first five lines of codes in the RoutingUnit::outportComputeXY() function, but received error message as shown beneath the command lines below.
gdb ./build/Garnet_standalone/gem5.opt configs/example/garnet_synth_traffic.py --network=garnet2.0 --num-cpus=16 --num-dirs=16 --topology=Prism_Mesh --mesh-rows=4 --sim-cycles=5000 --injectionrate=0.02 --synthetic=uniform_random --routing-algorithm=1
gdb: unrecognized option '--network=garnet2.0'
Use `gdb --help' for a complete list of options.

How should I use gdb to recognize the optional parameters?

In addition, I suspect if one can split the m_router along the z-axis and treat each level individually, a working XYZ routing can be obtained from the XY routing.

Can you help with that?

It is a pleasure to interact with great minds.

----- Original Message -----
From: "gem5-users-request" <gem5-users-***@gem5.org>
To: "gem5-users" <gem5-***@gem5.org>
Sent: Tuesday, 23 October, 2018 17:00:01
Subject: gem5-users Digest, Vol 147, Issue 16

Send gem5-users mailing list submissions to
gem5-***@gem5.org

To subscribe or unsubscribe via the World Wide Web, visit
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
or, via email, send a message with subject or body 'help' to
gem5-users-***@gem5.org

You can reach the person managing the list at
gem5-users-***@gem5.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of gem5-users digest..."


Today's Topics:

1. M5ops could not dump stats as documented (Yi Shen)
2. 转发: M5ops could not dump stats as documented (Yi Shen)
3. Re: Gem5 and McPAT error (Weizheng Xu)
4. Re: Some problem with gem5-spm on full system mode (王宥憲)
5. Re: Gem5 and McPAT error (Andreas Brokalakis)
6. Re: Some problem with gem5-spm on full system mode (Ciro Santilli)
7. 3D MESH TOPOLOGY DEVELOMENT IN GARNET2.0 (Idris M. Umar)
8. Re: 3D MESH TOPOLOGY DEVELOMENT IN GARNET2.0 (Krishna, Tushar)


----------------------------------------------------------------------

Message: 1
Date: Mon, 22 Oct 2018 17:24:35 +0000
From: Yi Shen <***@mail.mcgill.ca>
To: "gem5-***@gem5.org" <gem5-***@gem5.org>
Subject: [gem5-users] M5ops could not dump stats as documented
Message-ID:
<***@YQBPR0101MB0929.CANPRD01.PROD.OUTLOOK.COM>

Content-Type: text/plain; charset="windows-1252"

Hi,

I have been using the kernel and image provided in the gem5 download page. When I put “/sbin/m5 dumpresetstats 0 1”, I was expecting a periodic dump every 1000 tick. But stats.txt shows “sim_ticks 1198037940 # Number of ticks s”. How can I fix it or is it a known bug?

Details:

Gem5 version: freshly checked out at GitHub.
Protocol: X86_MOESI_hammer
With Ruby: Yes
Details












Yi Shen

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/d56c4f78/attachment-0001.html>

------------------------------

Message: 2
Date: Mon, 22 Oct 2018 17:35:15 +0000
From: Yi Shen <***@mail.mcgill.ca>
To: "gem5-***@gem5.org" <gem5-***@gem5.org>
Subject: [gem5-users] 转发: M5ops could not dump stats as
documented
Message-ID:
<***@YQBPR0101MB0929.CANPRD01.PROD.OUTLOOK.COM>

Content-Type: text/plain; charset="gb2312"



Yi Shen

发件人: Yi Shen<mailto:***@mail.mcgill.ca>
发送时间: October 22, 2018 1:32 PM
收件人: Yi Shen<mailto:***@mail.mcgill.ca>
主题: [gem5-users] M5ops could not dump stats as documented


Details



Gem5 version:

Freshly checked out at GitHub

Protocol:

X86_MOESI_hammer

Ruby:

Yes

System script:

Fs.py

rcS:

hack_back_ckpt.rcS + self made rcS

Simulation content:

Parsec Benchmark 2.1










Self made rcS:
#!/bin/sh
# File to run the fluidanimate benchmark
cd /parsec/install/bin
/sbin/m5 dumpresetstats 0 1
./fluidanimate 2 5 /parsec/install/inputs/fluidanimate/in_35K.fluid /parsec/install/inputs/fluidanimate/out.fluid
echo "Done :D"
/sbin/m5 exit

Options for fs.py:
#!/bin/bash
echo "Hello, World!"
build/X86_MOESI_hammer/gem5.opt \
--outdir=./m5/ \
configs/example/fs.py \
--kernel=/data/yi/ToolChain/gem5-stable/full_system_images/x86-system/binaries/x86_64-vmlinux-2.6.22.9.smp \
--disk-image=/data/yi/ToolChain/gem5-stable/full_system_images/x86-system/disks/x86root-parsec.img \
--cpu-type=DerivO3CPU \
--sys-clock='3.7GHz' \
--cpu-clock='3.7GHz' \
--sys-voltage='0.8V' \
--l1i_size=32kB \
--l1d_size=32kB \
--l2_assoc=8 \
--caches \
--l2cache \
--l2_size=2MB \
--mem-size='2GB' \
--num-cpus=2 \
--checkpoint-dir=$x865/m5/ \
--script=/data/yi/ToolChain/arboretum/rcS/fluidanimate_2c_simsmall.rcS \
--checkpoint-restore=1 \
--ruby \

Yi Shen

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/e151978c/attachment-0001.html>

------------------------------

Message: 3
Date: Mon, 22 Oct 2018 14:10:55 -0500
From: Weizheng Xu <***@u.northwestern.edu>
To: gem5 users mailing list <gem5-***@gem5.org>
Subject: Re: [gem5-users] Gem5 and McPAT error
Message-ID: <F4230707-A3BA-4403-8DCC-***@u.northwestern.edu>
Content-Type: text/plain; charset="utf-8"

Dear all,

I simulated the gem5 with full system mode and the command is "sudo build/ALPHA/gem5.opt configs/example/fs.py --script=./configs/boot/canneal_8c_simsmall.rcS --cpu-type=detailed --num-cpu=8 --caches --l2cache --num-l2caches=2 --l1d_size=32kB --l1i_size=32kB --l2_size=8MB --l1d_assoc=4 --l1i_assoc=4 --l2_assoc=16 --cacheline_size=64 --mem-type=NVMainMemory --nvmain-config=../nvmain/Config/RRAM_ISSCC_2012_4GB.config -F 5000000000 -I 2000000000” and get the stats.txt like below.

Then I use "python GEM5ToMcPAT.py stats.txt config.json template-x.xml” to transfer the output files of gem5 to the available input of McPAT, named as mcat-out.xml (the input of McPAT) here. But there have some warning here like this"Reading GEM5 stats from: stats.txt
Warning (stats): system.iocache.tags.avg_refs is nan. Setting it to 0
Warning (stats): system.iocache.avg_blocked_cycles::no_mshrs is nan. Setting it to 0
Warning (stats): system.iocache.avg_blocked_cycles::no_targets is nan. Setting it to 0
Warning (stats): system.cpu0.kern.mode_switch_good::user is nan. Setting it to 0
Warning (stats): system.cpu0.kern.mode_switch_good::idle is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedSwi is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedRxIdle is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedRxOk is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedRxDesc is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedTxOk is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedTxIdle is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedTxDesc is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedRxOrn is nan. Setting it to 0
Warning (stats): system.tsunami.ethernet.coalescedTotal is nan. Setting it to 0
Warning (stats): system.l2.avg_blocked_cycles::no_mshrs is nan. Setting it to 0
Warning (stats): system.l2.avg_blocked_cycles::no_targets is nan. Setting it to 0
…...
I don’t know how to solve this warning and guess this is not the problem here which lead to the error below. So I ignore them.

Then I use"./mcpat -infile mcpat-out.xml “ to run the McPAT, and can not get the output of McPAT. The error also shown as "McPAT (version 0.8 of Aug, 2010) is computing the target processor...

some value(s) of number_of_cores/number_of_L2s/number_of_L3s/number_of_NoCs is/are not correct!”

Ps: I use the ubuntu14.04 LTS system and the McPAT in gem5/ext/mcpat file

All helps appreciate!

The output file of gem5 is too big to put here. You can get it with this link https://drive.google.com/drive/folders/13yTj7gVtvi-mUjFUB16jgg2sAqmwSe3-?usp=sharing <https://drive.google.com/drive/folders/13yTj7gVtvi-mUjFUB16jgg2sAqmwSe3-?usp=sharing>. The template is used for transfer and the result of transferred(input of McPAT tools) are listed here. Sorry to bother you guys.



Best wishes,
weizheng
Hi,
The template seems good.
Possibly the problem is after the parser.
The best way is to create your own parser.
Share the output file in order to check what it is wrong.
Best Regards,
Georgios Bousdras
PhD Researcher — Embedded Systems
Dear all,
I have add the template I used for McPAT as the attachment. But I still have the same problem here. I have check the repository of the person who help me before https://github.com/H2020-COSSIM/cMcPAT <https://github.com/H2020-COSSIM/cMcPAT>, but I did not find the template in his files. I am not sure how can I change the template because when I change the value of number of cores and others, they have the same problem.
Any help appreciate.
Best wishes,
weizheng
<template-xeon.xml>
Dear Weizheng,
you are using a quite old version of McPAT. Current version is 1.3. It would be better if you used that one.
Now concerning your issues. You need to make sure that the system that you are trying to simulate in gem5 matches with the template that you are trying to use in McPAT. It is assumed that you need to make changes in the template xml file to match it with the gem5 system. Check the proper instructions on how to do that.
You can also check the modified McPAT and conversion scripts in my repository here: https://github.com/H2020-COSSIM/cMcPAT <https://github.com/H2020-COSSIM/cMcPAT>
Best regards,
Andreas
Dear all,
"McPAT (version 0.8 of Aug, 2010) is computing the target processor...
some value(s) of number_of_cores/number_of_L2s/number_of_L3s/number_of_NoCs is/are not correct!”
I tried to change these values in the template_xeon.xml but still failed with this problem.
I really don’t know how to solve this problem.
PS: I use ubuntu16.04 LTS and gem5. I use the McPAT provided in the gem5/ext file and with version of 0.8
Any help appreciated,
thanks
Weizheng
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users <http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users <http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/9bd1f211/attachment-0003.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: template-x.xml
Type: application/xml
Size: 56185 bytes
Desc: not available
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/9bd1f211/attachment-0002.xml>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/9bd1f211/attachment-0004.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mcpat-out.xml
Type: application/xml
Size: 29847 bytes
Desc: not available
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/9bd1f211/attachment-0003.xml>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181022/9bd1f211/attachment-0005.html>

------------------------------

Message: 4
Date: Tue, 23 Oct 2018 16:33:12 +0800
From: 王宥憲 <***@gmail.com>
To: ***@gmail.com
Cc: gem5-***@gem5.org
Subject: Re: [gem5-users] Some problem with gem5-spm on full system
mode
Message-ID:
<CAB=Ksphu-***@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hello Ciro Santilli,

I can't reproduce on master, because I want to simulate a ScratchPad Memory
in gem5.

I download the gem5 from "https://github.com/markoshorro/gem5-spm" ,but I
don't know where files different from gem5 master.

I can tell you that my compiler is gcc 6 and OS version is ubuntu 16.04.5
LTS.

Before I use this gem5-spm, I have some experience to use gem5 master, and
can successfully build full system.

I will try to run through GDB, get the backtrace.


Respectfully yours, Yo-Xian,Wang
Hello everyone,
I want to simulate a ScratchPad Memory in gem5, so I download the gem5
from "https://github.com/markoshorro/gem5-spm".
But when build gem5 to use command "scons build/X86/gem5.opt", it will
show error .
[NEW DEPS] X86/arch/x86/generated/inc.d -> x86-deps
[ENVIRONS] x86-deps -> x86-environs
[ CXX] X86/proto/packet.pb.cc -> .o
build/X86/proto/packet.pb.cc: In function 'void
build/X86/proto/packet.pb.cc:61:6: error: no previous declaration for
'void InitDefaults_packet_2eproto()' [-Werror=missing-declarations]
void InitDefaults_packet_2eproto() {
^~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1plus: all warnings being treated as errors
scons: *** [build/X86/proto/packet.pb.o] Error 1
scons: building terminated because of errors.
My solution is add "static" before the error function.
After this error, the next error is "
[NEW DEPS] X86/arch/x86/generated/inc.d -> x86-deps
[ENVIRONS] x86-deps -> x86-environs
[ CXX] X86/proto/inst.pb.cc -> .o
build/X86/proto/inst.pb.cc: In function 'void
build/X86/proto/inst.pb.cc:81:7: error: no previous declaration for
'void InitDefaults_inst_2eproto()' [-Werror=missing-declarations]
void InitDefaults_inst_2eproto() {
^~~~~~~~~~~~~~~~~~~~~~~~~
cc1plus: all warnings being treated as errors
scons: *** [build/X86/proto/inst.pb.o] Error 1
scons: building terminated because of errors.
".
My solution is add " void InitDefaults_inst_2eproto() " in inst.pb.h.
After the two errors,gem5 is built successfully.
Can you reproduce on master? If yes, give us a commit and your
compiler and OS version.
When build full system mode to use command " build/X86/gem5.opt
configs/example/fs.py --kernel=x86_64-vmlinux-3.4.112
--disk-image=linux-x86.img " ,it will core dumped.
The main terminal show
gem5 Simulator System. http://gem5.org
gem5 is copyrighted software; use the --copyright option for details.
gem5 compiled Oct 17 2018 05:19:03
gem5 started Oct 17 2018 14:00:29
gem5 executing on lab-All-Series
command line: build/X86/gem5.opt configs/example/fs.py
--kernel=x86_64-vmlinux-3.4.112 --disk-image=linux-x86.img
Global frequency set at 1000000000000 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address
range assigned (512 Mbytes)
/home/lab/full_system_images/x86/binaries/x86_64-vmlinux-3.4.112
Listening for com_1 connection on port 3456
0: rtc: Real-time clock set to Sun Jan 1 00:00:00 2012
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
warn: Reading current count from inactive timer.
**** REAL SIMULATION ****
warn: Don't know what interrupt to clear for console.
warn: instruction 'wbinvd' unimplemented
4202367500: system.pc.com_1.terminal: attach terminal 0
*** stack smashing detected ***: build/X86/gem5.opt terminated
Program aborted at cycle 76227969500
已經終止 (core dumped)
Does it happen on gem5 master as well, or just with the branch?
Run through GDB, get the backtrace, give it to us.
Then possibly explore interesting variables on GDB, and try to figure
out from source a bit.
The m5term terminal show
==== m5 slave terminal: Terminal 0 ====
(Ubuntu 5.4.0-6ubuntu1~16.04.4) ) #1 SMP Tue Dec 12 03:12:57 CST 2017
Command line: earlyprintk=ttyS0 console=ttyS0 lpj=7999923 root=/dev/hda1
CPU: vendor_id 'M5 Simulator' unknown, using generic init.
CPU: Your system may be unstable.
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 0000000020000000 (usable)
BIOS-e820: 0000000020000000 - 00000000c0000000 (reserved)
BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
bootconsole [earlyser0] enabled
NX (Execute Disable) protection: active
SMBIOS 2.5 present.
No AGP bridge found
last_pfn = 0x20000 max_arch_pfn = 0x400000000
x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106
CPU MTRRs all blank - virtualized system.
found SMP MP-table at [ffff8800000f0050] f0050
init_memory_mapping: 0000000000000000-0000000020000000
DMA 0x00000010 -> 0x00001000
DMA32 0x00001000 -> 0x00100000
Normal empty
Movable zone start PFN for each node
Early memory PFN ranges
0: 0x00000010 -> 0x0000009f
0: 0x00000100 -> 0x00020000
Intel MultiProcessor Specification v1.4
MPTABLE: APIC at: 0xFEE00000
Processor #0 (Bootup-CPU)
IOAPIC[0]: apic_id 1, version 20, address 0xfec00000, GSI 0-23
Processors: 1
SMP: Allowing 1 CPUs, 0 hotplug CPUs
Allocating PCI resources starting at c0000000 (gap: c0000000:3fff0000)
setup_percpu: NR_CPUS:8 nr_cpumask_bits:8 nr_cpu_ids:1 nr_node_ids:1
u2097152
129165
Kernel command line: earlyprintk=ttyS0 console=ttyS0 lpj=7999923
root=/dev/hda1
PID hash table entries: 2048 (order: 2, 16384 bytes)
Dentry cache hash table entries: 65536 (order: 7, 524288 bytes)
Inode-cache hash table entries: 32768 (order: 6, 262144 bytes)
Can anyone help me?
look forward to hearing from you.
Respectfully yours, Yo-Xian,Wang
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181023/8c7984eb/attachment-0001.html>

------------------------------

Message: 5
Date: Tue, 23 Oct 2018 11:41:04 +0300
From: Andreas Brokalakis <***@gmail.com>
To: gem5 users mailing list <gem5-***@gem5.org>
Subject: Re: [gem5-users] Gem5 and McPAT error
Message-ID:
<CAEr7J9gxYDU2tTnniX5ZG+3Ho5aOkEUc-***@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Hi weizheng,

I used your mcpat-out.xml in our cMcPAT (and I guess that any McPATv1.3 is
equivalent) and it completed the process. Attached is the output at
print_level 5.

Best,
Andreas

On Mon, Oct 22, 2018 at 10:11 PM Weizheng Xu <
Dear all,
I simulated the gem5 with full system mode and the command is "sudo
build/ALPHA/gem5.opt configs/example/fs.py
--script=./configs/boot/canneal_8c_simsmall.rcS --cpu-type=detailed
--num-cpu=8 --caches --l2cache --num-l2caches=2 --l1d_size=32kB
--l1i_size=32kB --l2_size=8MB --l1d_assoc=4 --l1i_assoc=4 --l2_assoc=16
--cacheline_size=64 --mem-type=NVMainMemory
--nvmain-config=../nvmain/Config/RRAM_ISSCC_2012_4GB.config -F 5000000000
-I 2000000000” and get the stats.txt like below.
Then I use "python GEM5ToMcPAT.py stats.txt config.json template-x.xml” to
transfer the output files of gem5 to the available input of McPAT, named as
mcat-out.xml (the input of McPAT) here. But there have some warning here
like this"Reading GEM5 stats from: stats.txt
Warning (stats): system.iocache.tags.avg_refs is nan. Setting it to 0
Warning (stats): system.iocache.avg_blocked_cycles::no_mshrs is nan.
Setting it to 0
Warning (stats): system.iocache.avg_blocked_cycles::no_targets is nan.
Setting it to 0
Warning (stats): system.cpu0.kern.mode_switch_good::user is nan. Setting
it to 0
Warning (stats): system.cpu0.kern.mode_switch_good::idle is nan. Setting
it to 0
Warning (stats): system.tsunami.ethernet.coalescedSwi is nan. Setting it
to 0
Warning (stats): system.tsunami.ethernet.coalescedRxIdle is nan. Setting
it to 0
Warning (stats): system.tsunami.ethernet.coalescedRxOk is nan. Setting it
to 0
Warning (stats): system.tsunami.ethernet.coalescedRxDesc is nan. Setting
it to 0
Warning (stats): system.tsunami.ethernet.coalescedTxOk is nan. Setting it
to 0
Warning (stats): system.tsunami.ethernet.coalescedTxIdle is nan. Setting
it to 0
Warning (stats): system.tsunami.ethernet.coalescedTxDesc is nan. Setting
it to 0
Warning (stats): system.tsunami.ethernet.coalescedRxOrn is nan. Setting it
to 0
Warning (stats): system.tsunami.ethernet.coalescedTotal is nan. Setting it
to 0
Warning (stats): system.l2.avg_blocked_cycles::no_mshrs is nan. Setting it
to 0
Warning (stats): system.l2.avg_blocked_cycles::no_targets is nan. Setting
it to 0
…...
I don’t know how to solve this warning and guess this is not the problem
here which lead to the error below. So I ignore them.
Then I use"./mcpat -infile mcpat-out.xml “ to run the McPAT, and can not
get the output of McPAT. The error also shown as "McPAT (version 0.8 of
Aug, 2010) is computing the target processor...
some value(s) of
number_of_cores/number_of_L2s/number_of_L3s/number_of_NoCs is/are not
correct!”
Ps: I use the ubuntu14.04 LTS system and the McPAT in gem5/ext/mcpat file
All helps appreciate!
The output file of gem5 is too big to put here. You can get it with this
link
https://drive.google.com/drive/folders/13yTj7gVtvi-mUjFUB16jgg2sAqmwSe3-?usp=sharing.
The template is used for transfer and the result of transferred(input of
McPAT tools) are listed here. Sorry to bother you guys.
Best wishes,
weizheng
写道:
Hi,
The template seems good.
Possibly the problem is after the parser.
The best way is to create your own parser.
Share the output file in order to check what it is wrong.
Best Regards,
Georgios Bousdras
PhD Researcher — Embedded Systems
Dear all,
I have add the template I used for McPAT as the attachment. But I still
have the same problem here. I have check the repository of the person who
help me before https://github.com/H2020-COSSIM/cMcPAT, but I did not find
the template in his files. I am not sure how can I change the template
because when I change the value of number of cores and others, they have
the same problem.
Any help appreciate.
Best wishes,
weizheng
<template-xeon.xml>
Dear Weizheng,
you are using a quite old version of McPAT. Current version is 1.3. It
would be better if you used that one.
Now concerning your issues. You need to make sure that the system that you
are trying to simulate in gem5 matches with the template that you are
trying to use in McPAT. It is assumed that you need to make changes in the
template xml file to match it with the gem5 system. Check the proper
instructions on how to do that.
You can also check the modified McPAT and conversion scripts in my
repository here: https://github.com/H2020-COSSIM/cMcPAT
Best regards,
Andreas
On Mon, Sep 24, 2018 at 11:10 PM Weizheng Xu <
Dear all,
I use gem5 get the full system simulation results and want to use McPAT
to get the power model. I use the Gem5ToMcpat.py and template_xeon.xml
gotten from website to translate the stats.txt to the output.xml, which can
be read as input by McPAT. But when I use this output.xml as a input for
McPAT (I use command ./mcpat -infile
/home//Desktop/test/gem5/ext/mcpat/out.xml in the terminal to run the
"McPAT (version 0.8 of Aug, 2010) is computing the target processor...
some value(s) of
number_of_cores/number_of_L2s/number_of_L3s/number_of_NoCs is/are not
correct!”
I tried to change these values in the template_xeon.xml but still failed
with this problem.
I really don’t know how to solve this problem.
PS: I use ubuntu16.04 LTS and gem5. I use the McPAT provided in the
gem5/ext file and with version of 0.8
Any help appreciated,
thanks
Weizheng
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181023/33273d0e/attachment-0001.html>
-------------- next part --------------
McPAT (version 1.3 of Feb, 2015) is computing the target processor...


McPAT (version 1.3 of Feb, 2015) results (current print level is 5)
*****************************************************************************************
Technology 65 nm
Interconnect metal projection= aggressive interconnect technology projection
Core clock Rate(MHz) 2000

*****************************************************************************************
Processor:
Area = 539.806 mm^2
Peak Power = 516.546 W
Total Leakage = 152.182 W
Peak Dynamic = 364.364 W
Subthreshold Leakage = 145.424 W
Subthreshold Leakage with power gating = 59.0703 W
Gate Leakage = 6.75844 W
Runtime Dynamic = 37.8719 W

Total Cores: 8 cores
Device Type= ITRS high performance device type
Area = 504.142 mm^2
Peak Dynamic = 359.336 W
Subthreshold Leakage = 140.705 W
Subthreshold Leakage with power gating = 56.3836 W
Gate Leakage = 6.72068 W
Runtime Dynamic = 37.8719 W

Total L2s:
Device Type= ITRS high performance device type
Area = 35.6639 mm^2
Peak Dynamic = 5.02798 W
Subthreshold Leakage = 4.71847 W
Subthreshold Leakage with power gating = 2.68667 W
Gate Leakage = 0.0377659 W
Runtime Dynamic = 2.73578e-09 W

*****************************************************************************************
Core:
Area = 63.0178 mm^2
Peak Dynamic = 44.917 W
Subthreshold Leakage = 17.5882 W
Subthreshold Leakage with power gating = 7.04795 W
Gate Leakage = 0.840085 W
Runtime Dynamic = 37.8719 W

Instruction Fetch Unit:
Area = 13.3026 mm^2
Peak Dynamic = 6.54787 W
Subthreshold Leakage = 2.15834 W
Subthreshold Leakage with power gating = 0.89003 W
Gate Leakage = 0.094432 W
Runtime Dynamic = 3.06877 W

Instruction Cache:
Area = 3.82901 mm^2
Peak Dynamic = 1.68753 W
Subthreshold Leakage = 0.665397 W
Subthreshold Leakage with power gating = 0.278253 W
Gate Leakage = 0.0274371 W
Runtime Dynamic = 0.0674506 W

Branch Target Buffer:
Area = 0.946863 mm^2
Peak Dynamic = 0.0843991 W
Subthreshold Leakage = 0.0901849 W
Subthreshold Leakage with power gating = 0.0485736 W
Gate Leakage = 0.00212441 W
Runtime Dynamic = 0.0064217 W

Branch Predictor:
Area = 0.279233 mm^2
Peak Dynamic = 0.0740148 W
Subthreshold Leakage = 0.0552507 W
Subthreshold Leakage with power gating = 0.0291987 W
Gate Leakage = 0.00144552 W
Runtime Dynamic = 0.0053149 W

Global Predictor:
Area = 0.0919946 mm^2
Peak Dynamic = 0.0191883 W
Subthreshold Leakage = 0.0197055 W
Subthreshold Leakage with power gating = 0.0104397 W
Gate Leakage = 0.000495847 W
Runtime Dynamic = 0.00158708 W

Local Predictor:
L1_Local Predictor:
Area = 0.0545159 mm^2
Peak Dynamic = 0.0163803 W
Subthreshold Leakage = 0.0100013 W
Subthreshold Leakage with power gating = 0.005277 W
Gate Leakage = 0.000280458 W
Runtime Dynamic = 0.00139101 W

L2_Local Predictor:
Area = 0.0279555 mm^2
Peak Dynamic = 0.00834983 W
Subthreshold Leakage = 0.0050708 W
Subthreshold Leakage with power gating = 0.00266635 W
Gate Leakage = 0.000147753 W
Runtime Dynamic = 0.000709093 W

Chooser:
Area = 0.0919946 mm^2
Peak Dynamic = 0.0191883 W
Subthreshold Leakage = 0.0197055 W
Subthreshold Leakage with power gating = 0.0104397 W
Gate Leakage = 0.000495847 W
Runtime Dynamic = 0.00158708 W

RAS:
Area = 0.0127724 mm^2
Peak Dynamic = 0.0109081 W
Subthreshold Leakage = 0.000767647 W
Subthreshold Leakage with power gating = 0.000375968 W
Gate Leakage = 2.56146e-05 W
Runtime Dynamic = 4.06415e-05 W

Instruction Buffer:
Area = 0.0791178 mm^2
Peak Dynamic = 1.67741 W
Subthreshold Leakage = 0.00576253 W
Subthreshold Leakage with power gating = 0.00267498 W
Gate Leakage = 0.00018516 W
Runtime Dynamic = 0.0389129 W

Instruction Decoder:
Area = 7.75308 mm^2
Peak Dynamic = 2.6634 W
Subthreshold Leakage = 1.1189 W
Subthreshold Leakage with power gating = 0.443085 W
Gate Leakage = 0.0381395 W
Runtime Dynamic = 0.0617862 W

Renaming Unit:
Area = 1.24 mm^2
Peak Dynamic = 8.25131 W
Subthreshold Leakage = 0.236699 W
Subthreshold Leakage with power gating = 0.0957199 W
Gate Leakage = 0.0255856 W
Runtime Dynamic = 11.6942 W

Int Front End RAT with 1 internal checkpoints:
Area = 0.185063 mm^2
Peak Dynamic = 5.1442 W
Subthreshold Leakage = 0.00273136 W
Subthreshold Leakage with power gating = 0.00138123 W
Gate Leakage = 0.000106646 W
Runtime Dynamic = 0.108134 W

FP Front End RAT with 1 internal checkpoints:
Area = 0.0262203 mm^2
Peak Dynamic = 1.25615 W
Subthreshold Leakage = 0.000877353 W
Subthreshold Leakage with power gating = 0.000453103 W
Gate Leakage = 5.24626e-05 W
Runtime Dynamic = 0.0181622 W

Free List:
Area = 0.481389 mm^2
Peak Dynamic = 0.25164 W
Subthreshold Leakage = 0.00659346 W
Subthreshold Leakage with power gating = 0.00366856 W
Gate Leakage = 0.000203822 W
Runtime Dynamic = 0.0111793 W

FP Free List:
Area = 0.131985 mm^2
Peak Dynamic = 0.154875 W
Subthreshold Leakage = 0.00365632 W
Subthreshold Leakage with power gating = 0.00197195 W
Gate Leakage = 0.000122342 W
Runtime Dynamic = 0.00118312 W

Load Store Unit:
Area = 16.0655 mm^2
Peak Dynamic = 6.91716 W
Subthreshold Leakage = 2.57734 W
Subthreshold Leakage with power gating = 1.03975 W
Gate Leakage = 0.127566 W
Runtime Dynamic = 3.08156 W

Data Cache:
Area = 13.0319 mm^2
Peak Dynamic = 6.04766 W
Subthreshold Leakage = 2.20454 W
Subthreshold Leakage with power gating = 0.892119 W
Gate Leakage = 0.095807 W
Runtime Dynamic = 0.124325 W

LoadQ:
Area = 0.597754 mm^2
Peak Dynamic = 0.254191 W
Subthreshold Leakage = 0.0749804 W
Subthreshold Leakage with power gating = 0.0296922 W
Gate Leakage = 0.00332956 W
Runtime Dynamic = 0.0227822 W

StoreQ:
Area = 0.597754 mm^2
Peak Dynamic = 0.254191 W
Subthreshold Leakage = 0.0749804 W
Subthreshold Leakage with power gating = 0.0296922 W
Gate Leakage = 0.00332956 W
Runtime Dynamic = 0.0455644 W

Memory Management Unit:
Area = 0.839485 mm^2
Peak Dynamic = 0.982217 W
Subthreshold Leakage = 0.279585 W
Subthreshold Leakage with power gating = 0.110716 W
Runtime Dynamic = 7.23068 W

Itlb:
Area = 0.0930895 mm^2
Peak Dynamic = 0.0167954 W
Subthreshold Leakage = 0.0144293 W
Subthreshold Leakage with power gating = 0.00571399 W
Gate Leakage = 0.000629867 W
Runtime Dynamic = 0.00298179 W

Dtlb:
Area = 0.331051 mm^2
Peak Dynamic = 0.062644 W
Subthreshold Leakage = 0.0423145 W
Subthreshold Leakage with power gating = 0.0167565 W
Gate Leakage = 0.00180917 W
Runtime Dynamic = 0.00547322 W

Execution Unit:
Area = 23.104 mm^2
Peak Dynamic = 22.2185 W
Subthreshold Leakage = 5.94238 W
Subthreshold Leakage with power gating = 2.37978 W
Runtime Dynamic = 12.7967 W

Register Files:
Area = 9.35713 mm^2
Peak Dynamic = 8.61495 W
Subthreshold Leakage = 0.183607 W
Subthreshold Leakage with power gating = 0.0927938 W
Gate Leakage = 0.00649419 W
Runtime Dynamic = 0.0680038 W

Integer RF:
Area = 4.67856 mm^2
Peak Dynamic = 6.92342 W
Subthreshold Leakage = 0.0918036 W
Subthreshold Leakage with power gating = 0.0463969 W
Gate Leakage = 0.00324709 W
Runtime Dynamic = 0.0643155 W

Floating Point RF:
Area = 4.67856 mm^2
Peak Dynamic = 1.69153 W
Subthreshold Leakage = 0.0918036 W
Subthreshold Leakage with power gating = 0.0463969 W
Gate Leakage = 0.00324709 W
Runtime Dynamic = 0.00368828 W

Instruction Scheduler:
Area = 2.50431 mm^2
Peak Dynamic = 4.84345 W
Subthreshold Leakage = 0.100503 W
Subthreshold Leakage with power gating = 0.0463108 W
Gate Leakage = 0.00390898 W
Runtime Dynamic = 0.426836 W

Instruction Window:
Area = 0.497673 mm^2
Peak Dynamic = 1.50883 W
Subthreshold Leakage = 0.0358961 W
Subthreshold Leakage with power gating = 0.0142149 W
Gate Leakage = 0.00164666 W
Runtime Dynamic = 0.136315 W

FP Instruction Window:
Area = 0.138043 mm^2
Peak Dynamic = 0.412527 W
Subthreshold Leakage = 0.0178333 W
Subthreshold Leakage with power gating = 0.00706199 W
Gate Leakage = 0.00084265 W
Runtime Dynamic = 0.0110503 W

ROB:
Area = 1.86859 mm^2
Peak Dynamic = 2.92209 W
Subthreshold Leakage = 0.0467734 W
Subthreshold Leakage with power gating = 0.0250339 W
Gate Leakage = 0.00141967 W
Runtime Dynamic = 0.279471 W

Integer ALUs (Count: 6 ):
Area = 0.672672 mm^2
Peak Dynamic = 2.72645 W
Subthreshold Leakage = 1.01603 W
Subthreshold Leakage with power gating = 0.402349 W
Gate Leakage = 0.0454759 W
Runtime Dynamic = 0.187989 W

Floating Point Units (FPUs) (Count: 2 ):
Area = 9.71959 mm^2
Peak Dynamic = 0.883753 W
Subthreshold Leakage = 3.67022 W
Subthreshold Leakage with power gating = 1.45341 W
Gate Leakage = 0.164273 W
Runtime Dynamic = 0.338883 W

Complex ALUs (Mul/Div) (Count: 1 ):
Area = 0.336336 mm^2
Peak Dynamic = 0.327493 W
Subthreshold Leakage = 0.508016 W
Subthreshold Leakage with power gating = 0.201174 W
Gate Leakage = 0.0227379 W
Runtime Dynamic = 0.21939 W

Results Broadcast Bus:
Area Overhead = 0.0986178 mm^2
Peak Dynamic = 3.37793 W
Subthreshold Leakage = 0.24116 W
Subthreshold Leakage with power gating = 0.0954995 W
Gate Leakage = 0.0107939 W
Runtime Dynamic = 0 W

*****************************************************************************************
L2
Area = 17.8319 mm^2
Peak Dynamic = 2.51399 W
Subthreshold Leakage = 2.35924 W
Subthreshold Leakage with power gating = 1.34333 W
Gate Leakage = 0.018883 W
Runtime Dynamic = 2.73578e-09 W

*****************************************************************************************

------------------------------

Message: 6
Date: Tue, 23 Oct 2018 10:37:16 +0100
From: Ciro Santilli <***@gmail.com>
To: 王宥憲 <***@gmail.com>, gem5 users mailing list
<gem5-***@gem5.org>
Subject: Re: [gem5-users] Some problem with gem5-spm on full system
mode
Message-ID:
<CAFXrp_cr=XYK-GFhs1++Y68SAB7D_yOMPi1W5kPXDRzq2uw_-***@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"

My best advice is: do a git bisect to find the offending commit, and
report it to the autor (Mark?) directly, e.g. at
https://github.com/markoshorro/gem5-spm/issues
Hello Ciro Santilli,
I can't reproduce on master, because I want to simulate a ScratchPad Memory in gem5.
I download the gem5 from "https://github.com/markoshorro/gem5-spm" ,but I don't know where files different from gem5 master.
I can tell you that my compiler is gcc 6 and OS version is ubuntu 16.04.5 LTS.
Before I use this gem5-spm, I have some experience to use gem5 master, and can successfully build full system.
I will try to run through GDB, get the backtrace.
Respectfully yours, Yo-Xian,Wang
Hello everyone,
I want to simulate a ScratchPad Memory in gem5, so I download the gem5 from "https://github.com/markoshorro/gem5-spm".
But when build gem5 to use command "scons build/X86/gem5.opt", it will show error .
[NEW DEPS] X86/arch/x86/generated/inc.d -> x86-deps
[ENVIRONS] x86-deps -> x86-environs
[ CXX] X86/proto/packet.pb.cc -> .o
build/X86/proto/packet.pb.cc:61:6: error: no previous declaration for 'void InitDefaults_packet_2eproto()' [-Werror=missing-declarations]
void InitDefaults_packet_2eproto() {
^~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1plus: all warnings being treated as errors
scons: *** [build/X86/proto/packet.pb.o] Error 1
scons: building terminated because of errors.
My solution is add "static" before the error function.
After this error, the next error is "
[NEW DEPS] X86/arch/x86/generated/inc.d -> x86-deps
[ENVIRONS] x86-deps -> x86-environs
[ CXX] X86/proto/inst.pb.cc -> .o
build/X86/proto/inst.pb.cc:81:7: error: no previous declaration for 'void InitDefaults_inst_2eproto()' [-Werror=missing-declarations]
void InitDefaults_inst_2eproto() {
^~~~~~~~~~~~~~~~~~~~~~~~~
cc1plus: all warnings being treated as errors
scons: *** [build/X86/proto/inst.pb.o] Error 1
scons: building terminated because of errors.
".
My solution is add " void InitDefaults_inst_2eproto() " in inst.pb.h.
After the two errors,gem5 is built successfully.
Can you reproduce on master? If yes, give us a commit and your
compiler and OS version.
When build full system mode to use command " build/X86/gem5.opt configs/example/fs.py --kernel=x86_64-vmlinux-3.4.112 --disk-image=linux-x86.img " ,it will core dumped.
The main terminal show
gem5 Simulator System. http://gem5.org
gem5 is copyrighted software; use the --copyright option for details.
gem5 compiled Oct 17 2018 05:19:03
gem5 started Oct 17 2018 14:00:29
gem5 executing on lab-All-Series
command line: build/X86/gem5.opt configs/example/fs.py --kernel=x86_64-vmlinux-3.4.112 --disk-image=linux-x86.img
Global frequency set at 1000000000000 ticks per second
warn: DRAM device capacity (8192 Mbytes) does not match the address range assigned (512 Mbytes)
info: kernel located at: /home/lab/full_system_images/x86/binaries/x86_64-vmlinux-3.4.112
Listening for com_1 connection on port 3456
0: rtc: Real-time clock set to Sun Jan 1 00:00:00 2012
0: system.remote_gdb.listener: listening for remote gdb #0 on port 7000
warn: Reading current count from inactive timer.
**** REAL SIMULATION ****
warn: Don't know what interrupt to clear for console.
warn: instruction 'wbinvd' unimplemented
4202367500: system.pc.com_1.terminal: attach terminal 0
*** stack smashing detected ***: build/X86/gem5.opt terminated
Program aborted at cycle 76227969500
已經終止 (core dumped)
Does it happen on gem5 master as well, or just with the branch?
Run through GDB, get the backtrace, give it to us.
Then possibly explore interesting variables on GDB, and try to figure
out from source a bit.
The m5term terminal show
==== m5 slave terminal: Terminal 0 ====
Command line: earlyprintk=ttyS0 console=ttyS0 lpj=7999923 root=/dev/hda1
CPU: vendor_id 'M5 Simulator' unknown, using generic init.
CPU: Your system may be unstable.
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 0000000020000000 (usable)
BIOS-e820: 0000000020000000 - 00000000c0000000 (reserved)
BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
bootconsole [earlyser0] enabled
NX (Execute Disable) protection: active
SMBIOS 2.5 present.
No AGP bridge found
last_pfn = 0x20000 max_arch_pfn = 0x400000000
x86 PAT enabled: cpu 0, old 0x7040600070406, new 0x7010600070106
CPU MTRRs all blank - virtualized system.
found SMP MP-table at [ffff8800000f0050] f0050
init_memory_mapping: 0000000000000000-0000000020000000
DMA 0x00000010 -> 0x00001000
DMA32 0x00001000 -> 0x00100000
Normal empty
Movable zone start PFN for each node
Early memory PFN ranges
0: 0x00000010 -> 0x0000009f
0: 0x00000100 -> 0x00020000
Intel MultiProcessor Specification v1.4
MPTABLE: APIC at: 0xFEE00000
Processor #0 (Bootup-CPU)
IOAPIC[0]: apic_id 1, version 20, address 0xfec00000, GSI 0-23
Processors: 1
SMP: Allowing 1 CPUs, 0 hotplug CPUs
Allocating PCI resources starting at c0000000 (gap: c0000000:3fff0000)
setup_percpu: NR_CPUS:8 nr_cpumask_bits:8 nr_cpu_ids:1 nr_node_ids:1
Built 1 zonelists in Zone order, mobility grouping on. Total pages: 129165
Kernel command line: earlyprintk=ttyS0 console=ttyS0 lpj=7999923 root=/dev/hda1
PID hash table entries: 2048 (order: 2, 16384 bytes)
Dentry cache hash table entries: 65536 (order: 7, 524288 bytes)
Inode-cache hash table entries: 32768 (order: 6, 262144 bytes)
Can anyone help me?
look forward to hearing from you.
Respectfully yours, Yo-Xian,Wang
_______________________________________________
gem5-users mailing list
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
------------------------------

Message: 7
Date: Tue, 23 Oct 2018 14:04:58 +0100 (WAT)
From: "Idris M. Umar" <***@abu.edu.ng>
To: gem5-***@gem5.org
Subject: [gem5-users] 3D MESH TOPOLOGY DEVELOMENT IN GARNET2.0
Message-ID: <***@abu.edu.ng>
Content-Type: text/plain; charset="utf-8"

I want to modify the Mesh_XY topology in Garnet2.0 to create a 3D topology. I used Mesh_XY as a base layer to which a replicated layer is added with the MakeTopology function. Then I established the necessary bridge links. It works fine with table routing, but always produced a segmentation error when the XY routing is applied. Can someone please help me out? I can attached my scripts when asked to.

Thanks for your expected cooperation.

Idris Umar
Ahmadu Bello Univeristy
Zaria - Nigeria.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181023/64dca541/attachment-0001.html>

------------------------------

Message: 8
Date: Tue, 23 Oct 2018 15:23:11 +0000
From: "Krishna, Tushar" <***@ece.gatech.edu>
To: "gem5-***@gem5.org" <gem5-***@gem5.org>
Subject: Re: [gem5-users] 3D MESH TOPOLOGY DEVELOMENT IN GARNET2.0
Message-ID: <ff1a83e3-407b-40d6-a19d-***@Spark>
Content-Type: text/plain; charset="utf-8"

Table routing by default does XY for a 2D Mesh - since all X links have a weight of 1 and Y links have a weight of 2. You can give weights of 3 for your Z links to get a XYZ routing algorithm for instance.
The specific XY routing function inside Garnet only works for a 2D mesh as it routes assuming link directions of N/S/E/W and will not work for your 3D topology. I only recommend using it to study adaptive routing algorithms in a 2D Mesh for instance.

Cheers,
Tushar

On Oct 23, 2018, 9:05 AM -0400, Idris M. Umar <***@abu.edu.ng>, wrote:
I want to modify the Mesh_XY topology in Garnet2.0 to create a 3D topology. I used Mesh_XY as a base layer to which a replicated layer is added with the MakeTopology function. Then I established the necessary bridge links. It works fine with table routing, but always produced a segmentation error when the XY routing is applied. Can someone please help me out? I can attached my scripts when asked to.

Thanks for your expected cooperation.

Idris Umar
Ahmadu Bello Univeristy
Zaria - Nigeria.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://m5sim.org/cgi-bin/mailman/private/gem5-users/attachments/20181023/33c31933/attachment-0001.html>

------------------------------

Subject: Digest Footer

_______________________________________________
gem5-users mailing list
gem5-***@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

------------------------------

End of gem5-users Digest, Vol 147, Issue 16
*******************************************

Loading...