date:20170505

Re: [gem5-users] Build system failures with gem5.opt and libgem5.so

2017-05-05 Thread Gabe Black

This is the fix I mentioned:

https://gem5-review.googlesource.com/#/c/3083/

On Fri, May 5, 2017 at 5:54 PM, Gabe Black  wrote:

> From what I can tell, there are two fundamental problems here. First,
> scons isn't recognizing that the actual content of enums/AddrMap.cc depends
> on env['USE_PYTHON'], so it thinks everything is up to date even though the
> files contents aren't what you'd want/expect.
>
> Second, we're using the same .cc file in both cases. That means that we'll
> have .cc files which are correct for one build or the other, but just
> whatever the most recent build was. This isn't totally broken as long as
> you build one thing at a time and you don't expect to go back to the other
> and get a totally NULL build, but it still seems less than ideal.
>
> I'm working on a fix for the first problem since it's easier, but it
> wouldn't be a bad idea to consider how to fix the second.
>
> Gabe
>
> On Fri, May 5, 2017 at 4:03 PM, Paul Rosenfeld (prosenfeld) <
> prosenf...@micron.com> wrote:
>
>> Hello all,
>>
>> I'm encountering a somewhat strange issue with the latest gem5 master. In
>> a fresh build directory, I can build either gem5.opt or libgem5_opt.so, but
>> not one after the other. That is, when I run:
>>
>> rm -rf build
>> scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
>>
>> Everything compiles fine. However, when I run:
>>
>> rm -rf build
>> scons -j32 build/ARM/gem5.opt
>> scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
>>
>> Then I get this error:
>>
>> scons: Building targets ...
>>  [NEW DEPS] ARM/arch/arm/generated/inc.d -> arm-deps
>>  [ENVIRONS] arm-deps -> arm-environs
>> [   SHCXX] ARM/enums/AddrMap.cc -> .os
>> In file included from ext/pybind11/include/pybind11/pytypes.h:12:0,
>>  from ext/pybind11/include/pybind11/cast.h:13,
>>  from ext/pybind11/include/pybind11/attr.h:13,
>>  from ext/pybind11/include/pybind11/pybind11.h:36,
>>  from build/ARM/enums/AddrMap.cc:10:
>> ext/pybind11/include/pybind11/common.h:70:20: fatal error: Python.h: No
>> such file or directory
>>  #include 
>> ^
>> compilation terminated.
>>
>> Now since we aren't supposed to have python in the mix in this library, I
>> suspect that the configure step of gem5.opt is leaving behind some things
>> that are getting picked up by the subsequent library build.
>>
>> if I reverse the build steps:
>> rm -rf build
>> scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
>> scons -j32 build/ARM/gem5.opt
>>
>> Both the library and binary build OK, but when I try to run gem5.opt with
>> fs.py I get:
>>
>> Traceback (most recent call last):
>>   File "", line 1, in 
>>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/importer.py",
>> line 80, in load_module
>> exec code in mod.__dict__
>>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/m5/__init__.py",
>> line 54, in 
>> from simulate import *
>>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/importer.py",
>> line 80, in load_module
>> exec code in mod.__dict__
>>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/m5/simulate.py",
>> line 65, in 
>> "atomic" : objects.params.atomic,
>> AttributeError: 'module' object has no attribute 'atomic'
>>
>> (This does not happen with a clean build of gem5.opt; the sim works fine
>> if I first remove the build folder)
>>
>> I'm afraid my scons skills aren't nearly good enough to figuring out
>> where to look. Does anyone have any ideas?
>>
>> Thanks,
>> Paul
>> ___
>> gem5-users mailing list
>> gem5-users@gem5.org
>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
>
>
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] Build system failures with gem5.opt and libgem5.so

2017-05-05 Thread Gabe Black

>From what I can tell, there are two fundamental problems here. First, scons
isn't recognizing that the actual content of enums/AddrMap.cc depends on
env['USE_PYTHON'], so it thinks everything is up to date even though the
files contents aren't what you'd want/expect.

Second, we're using the same .cc file in both cases. That means that we'll
have .cc files which are correct for one build or the other, but just
whatever the most recent build was. This isn't totally broken as long as
you build one thing at a time and you don't expect to go back to the other
and get a totally NULL build, but it still seems less than ideal.

I'm working on a fix for the first problem since it's easier, but it
wouldn't be a bad idea to consider how to fix the second.

Gabe

On Fri, May 5, 2017 at 4:03 PM, Paul Rosenfeld (prosenfeld) <
prosenf...@micron.com> wrote:

> Hello all,
>
> I'm encountering a somewhat strange issue with the latest gem5 master. In
> a fresh build directory, I can build either gem5.opt or libgem5_opt.so, but
> not one after the other. That is, when I run:
>
> rm -rf build
> scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
>
> Everything compiles fine. However, when I run:
>
> rm -rf build
> scons -j32 build/ARM/gem5.opt
> scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
>
> Then I get this error:
>
> scons: Building targets ...
>  [NEW DEPS] ARM/arch/arm/generated/inc.d -> arm-deps
>  [ENVIRONS] arm-deps -> arm-environs
> [   SHCXX] ARM/enums/AddrMap.cc -> .os
> In file included from ext/pybind11/include/pybind11/pytypes.h:12:0,
>  from ext/pybind11/include/pybind11/cast.h:13,
>  from ext/pybind11/include/pybind11/attr.h:13,
>  from ext/pybind11/include/pybind11/pybind11.h:36,
>  from build/ARM/enums/AddrMap.cc:10:
> ext/pybind11/include/pybind11/common.h:70:20: fatal error: Python.h: No
> such file or directory
>  #include 
> ^
> compilation terminated.
>
> Now since we aren't supposed to have python in the mix in this library, I
> suspect that the configure step of gem5.opt is leaving behind some things
> that are getting picked up by the subsequent library build.
>
> if I reverse the build steps:
> rm -rf build
> scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
> scons -j32 build/ARM/gem5.opt
>
> Both the library and binary build OK, but when I try to run gem5.opt with
> fs.py I get:
>
> Traceback (most recent call last):
>   File "", line 1, in 
>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/importer.py",
> line 80, in load_module
> exec code in mod.__dict__
>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/m5/__init__.py",
> line 54, in 
> from simulate import *
>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/importer.py",
> line 80, in load_module
> exec code in mod.__dict__
>   File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/m5/simulate.py",
> line 65, in 
> "atomic" : objects.params.atomic,
> AttributeError: 'module' object has no attribute 'atomic'
>
> (This does not happen with a clean build of gem5.opt; the sim works fine
> if I first remove the build folder)
>
> I'm afraid my scons skills aren't nearly good enough to figuring out where
> to look. Does anyone have any ideas?
>
> Thanks,
> Paul
> ___
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

[gem5-users] Build system failures with gem5.opt and libgem5.so

2017-05-05 Thread Paul Rosenfeld (prosenfeld)

Hello all, 

I'm encountering a somewhat strange issue with the latest gem5 master. In a 
fresh build directory, I can build either gem5.opt or libgem5_opt.so, but not 
one after the other. That is, when I run: 

rm -rf build
scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so

Everything compiles fine. However, when I run: 

rm -rf build
scons -j32 build/ARM/gem5.opt
scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so

Then I get this error: 

scons: Building targets ...
 [NEW DEPS] ARM/arch/arm/generated/inc.d -> arm-deps
 [ENVIRONS] arm-deps -> arm-environs
[   SHCXX] ARM/enums/AddrMap.cc -> .os
In file included from ext/pybind11/include/pybind11/pytypes.h:12:0,
 from ext/pybind11/include/pybind11/cast.h:13,
 from ext/pybind11/include/pybind11/attr.h:13,
 from ext/pybind11/include/pybind11/pybind11.h:36,
 from build/ARM/enums/AddrMap.cc:10:
ext/pybind11/include/pybind11/common.h:70:20: fatal error: Python.h: No such 
file or directory
 #include 
^
compilation terminated.

Now since we aren't supposed to have python in the mix in this library, I 
suspect that the configure step of gem5.opt is leaving behind some things that 
are getting picked up by the subsequent library build. 

if I reverse the build steps:
rm -rf build
scons -j32 --without-python --with-cxx-config build/ARM/libgem5_opt.so
scons -j32 build/ARM/gem5.opt

Both the library and binary build OK, but when I try to run gem5.opt with fs.py 
I get: 

Traceback (most recent call last):
  File "", line 1, in 
  File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/importer.py", line 
80, in load_module
exec code in mod.__dict__
  File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/m5/__init__.py", 
line 54, in 
from simulate import *
  File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/importer.py", line 
80, in load_module
exec code in mod.__dict__
  File "/proj/adg/REV/sim/prosenfeld/rhel7/gem5/src/python/m5/simulate.py", 
line 65, in 
"atomic" : objects.params.atomic,
AttributeError: 'module' object has no attribute 'atomic'

(This does not happen with a clean build of gem5.opt; the sim works fine if I 
first remove the build folder)

I'm afraid my scons skills aren't nearly good enough to figuring out where to 
look. Does anyone have any ideas?

Thanks, 
Paul
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] fatal: _pid

2017-05-05 Thread F. A. Faisal

try below link-

http://repo.gem5.org/gem5?cmd=changeset;node=0bf388858d1e

and download bz2 file...

I just tried garnet 2.0 now.

Thanks,

F. A. Faisal

On Sat, May 6, 2017 at 4:35 AM, raziye deylamsalehi <
raziye.deylamsal...@gmail.com> wrote:

> Hi
>
> I updated gem5 today.I could simulate n*n mesh network in gem5 but after
> updating, gem5 can not simulate  n*n mesh network.
>
> I visited this error:
>
> fatal: _pid 100 is already used
>
> what can I do?
>
> Thanks.
>
> ___
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

[gem5-users] fatal: _pid

2017-05-05 Thread raziye deylamsalehi

Hi

I updated gem5 today.I could simulate n*n mesh network in gem5 but after
updating, gem5 can not simulate  n*n mesh network.

I visited this error:

fatal: _pid 100 is already used

what can I do?

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

Re: [gem5-users] FS simulation with Garnet 2.0

2017-05-05 Thread Krishna, Tushar

Garnet_standalone is a different dummy protocol which consumes every message so 
it'll work. It's not realistic.

Try with other coherence protocols perhaps ...

Cheers,
Tushar

On May 5, 2017, at 2:21 PM, F. A. Faisal 
> wrote:

I tried simple network option. However, the result is the same. Please let me 
know, if I am missing any option to add (I used all the default settings).
However, I also run the garent_standalone protocol with the default settings, 
which runs properly.


./build/ALPHA/gem5.opt configs/example/fs.py --cpu-type=timing 
--script=script/blackscholes_4c_simsmall.rcS --num-cpus=4 --ruby --mesh-rows=2 
--network=simple --topology=Mesh_XY --caches --l2cache --l2_size=512kB 
--num-dirs=4 --num-l2=4 --l1d_size=32kB --l1i_size=32kB --l1d_assoc=2 
--l1i_assoc=2


panic: Invalid transition

system.ruby.dma_cntrl0 time: 4722258003 addr: 523845696 event: WriteRequest 
state: BUSY_WR

 @ tick 2361129001500

[doTransitionWorker:build/ALPHA/mem/protocol/DMA_Transitions.cc, line 135]

Memory Usage: 1015648 KBytes


Thanks a lot.


F. A. Faisal


On Sat, May 6, 2017 at 2:32 AM, Krishna, Tushar 
> wrote:
Hmm the coherence protocol seems to have received an invalid transaction. Not 
sure why. This does not seem to be a network issue. But for a sanity check, run 
with network=simple and see if it still fails.

Cheers,
Tushar

On May 5, 2017, at 11:46 AM, F. A. Faisal 
> wrote:

Thanks a lot.

I already figured out that problem by myself.
However, I encountered a new problem in running 4 core 4 threads blackscholes 
benchmarks. It shows system panic.
I also tried for 16 cpu and Mesh network, which is still running more than 8 
hours.

I used the mesh_xy topology along with below config-


scons FULL_SYSTEM=1 build/ALPHA/gem5.opt RUBY=true PROTOCOL=MOESI_CMP_directory


 ./build/ALPHA/gem5.opt configs/example/fs.py --cpu-type=timing 
--script=script/blackscholes_4c_simsmall.rcS --num-cpus=4 --ruby --mesh-rows=2 
--network=garnet2.0 --topology=Mesh_XY --caches --l2cache --l2_size=512kB 
--num-dirs=4 --num-l2=4 --l1d_size=32kB --l1i_size=32kB --l1d_assoc=2 
--l1i_assoc=2

It showed me the below error-
--

panic: Invalid transition

system.ruby.dma_cntrl0 time: 4816339893 addr: 525287488 event: WriteRequest 
state: BUSY_WR

 @ tick 2408169946500

[doTransitionWorker:build/ALPHA/mem/protocol/DMA_Transitions.cc, line 135]

Memory Usage: 996944 KBytes

Program aborted at tick 2408169946500

--- BEGIN LIBC BACKTRACE ---

./build/ALPHA/gem5.opt[0x829015]

./build/ALPHA/gem5.opt[0x839d96]

/lib64/libpthread.so.0[0x3685e0f730]

/lib64/libc.so.6(gsignal+0x3e)[0x3685a3266e]

/lib64/libc.so.6(abort+0x175)[0x3685a33ab5]

./build/ALPHA/gem5.opt[0x6867ac]

./build/ALPHA/gem5.opt[0x631c58]

./build/ALPHA/gem5.opt[0x631e83]

./build/ALPHA/gem5.opt[0x634aad]

./build/ALPHA/gem5.opt[0x8321d1]

./build/ALPHA/gem5.opt[0x852538]

./build/ALPHA/gem5.opt[0x852abb]

./build/ALPHA/gem5.opt[0x7d8d2d]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x4e26)[0x369ced55b6]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x54f2)[0x369ced5c82]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]

/usr/lib64/libpython2.6.so.1.0[0x369cef1b9c]

/usr/lib64/libpython2.6.so.1.0(PyRun_StringFlags+0x7a)[0x369cef1daa]

./build/ALPHA/gem5.opt[0x83967f]

./build/ALPHA/gem5.opt[0x415453]

/lib64/libc.so.6(__libc_start_main+0xfd)[0x3685a1f56d]

./build/ALPHA/gem5.opt[0x43b1b1]

--- END LIBC BACKTRACE ---

Aborted (core dumped)



Best regards,

F. A. Faisal


On Fri, May 5, 2017 at 11:22 PM, Krishna, Tushar 
> wrote:
Hi Faisal,
You can still add --topology with the fs.py file.
You need to also add --ruby to run ruby.

For running garnet, you can add a --network=garnet2.0
(Look at options in the garnet2.0 page on the wiki).

Cheers,
Tushar

On May 3, 2017, at 10:50 AM, F. A. Faisal 
> wrote:

Hi,

I like to invoke the garnet 2.0 for running the benchmark programs Parsec 2.1.
However, previous ruby_fs.py file located at configs/example/ folder is not 
available

Re: [gem5-users] FS simulation with Garnet 2.0

2017-05-05 Thread F. A. Faisal

I tried simple network option. However, the result is the same. Please let
me know, if I am missing any option to add (I used all the default
settings).
However, I also run the garent_standalone protocol with the default
settings, which runs properly.

./build/ALPHA/gem5.opt configs/example/fs.py --cpu-type=timing
--script=script/blackscholes_4c_simsmall.rcS --num-cpus=4 --ruby
--mesh-rows=2 --network=simple --topology=Mesh_XY --caches --l2cache
--l2_size=512kB --num-dirs=4 --num-l2=4 --l1d_size=32kB --l1i_size=32kB
--l1d_assoc=2 --l1i_assoc=2

panic: Invalid transition

system.ruby.dma_cntrl0 time: 4722258003 addr: 523845696 event: WriteRequest
state: BUSY_WR

 @ tick 2361129001500

[doTransitionWorker:build/ALPHA/mem/protocol/DMA_Transitions.cc, line 135]

Memory Usage: 1015648 KBytes


Thanks a lot.


F. A. Faisal


On Sat, May 6, 2017 at 2:32 AM, Krishna, Tushar 
wrote:

> Hmm the coherence protocol seems to have received an invalid transaction.
> Not sure why. This does not seem to be a network issue. But for a sanity
> check, run with network=simple and see if it still fails.
>
> Cheers,
> Tushar
>
> On May 5, 2017, at 11:46 AM, F. A. Faisal  wrote:
>
> Thanks a lot.
>
> I already figured out that problem by myself.
> However, I encountered a new problem in running 4 core 4 threads
> blackscholes benchmarks. It shows system panic.
> I also tried for 16 cpu and Mesh network, which is still running more than
> 8 hours.
>
> I used the mesh_xy topology along with below config-
>
> scons FULL_SYSTEM=1 build/ALPHA/gem5.opt RUBY=true
> PROTOCOL=MOESI_CMP_directory
>
>  ./build/ALPHA/gem5.opt configs/example/fs.py --cpu-type=timing
> --script=script/blackscholes_4c_simsmall.rcS --num-cpus=4 --ruby
> --mesh-rows=2 --network=garnet2.0 --topology=Mesh_XY --caches --l2cache
> --l2_size=512kB --num-dirs=4 --num-l2=4 --l1d_size=32kB --l1i_size=32kB
> --l1d_assoc=2 --l1i_assoc=2
>
> It showed me the below error-
> --
>
> panic: Invalid transition
>
> system.ruby.dma_cntrl0 time: 4816339893 addr: 525287488 event:
> WriteRequest state: BUSY_WR
>
>  @ tick 2408169946500
>
> [doTransitionWorker:build/ALPHA/mem/protocol/DMA_Transitions.cc, line 135]
>
> Memory Usage: 996944 KBytes
>
> Program aborted at tick 2408169946500
>
> --- BEGIN LIBC BACKTRACE ---
>
> ./build/ALPHA/gem5.opt[0x829015]
>
> ./build/ALPHA/gem5.opt[0x839d96]
>
> /lib64/libpthread.so.0[0x3685e0f730]
>
> /lib64/libc.so.6(gsignal+0x3e)[0x3685a3266e]
>
> /lib64/libc.so.6(abort+0x175)[0x3685a33ab5]
>
> ./build/ALPHA/gem5.opt[0x6867ac]
>
> ./build/ALPHA/gem5.opt[0x631c58]
>
> ./build/ALPHA/gem5.opt[0x631e83]
>
> ./build/ALPHA/gem5.opt[0x634aad]
>
> ./build/ALPHA/gem5.opt[0x8321d1]
>
> ./build/ALPHA/gem5.opt[0x852538]
>
> ./build/ALPHA/gem5.opt[0x852abb]
>
> ./build/ALPHA/gem5.opt[0x7d8d2d]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x4e26)[0x369ced55b6]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x54f2)[0x369ced5c82]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]
>
> /usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]
>
> /usr/lib64/libpython2.6.so.1.0[0x369cef1b9c]
>
> /usr/lib64/libpython2.6.so.1.0(PyRun_StringFlags+0x7a)[0x369cef1daa]
>
> ./build/ALPHA/gem5.opt[0x83967f]
>
> ./build/ALPHA/gem5.opt[0x415453]
>
> /lib64/libc.so.6(__libc_start_main+0xfd)[0x3685a1f56d]
>
> ./build/ALPHA/gem5.opt[0x43b1b1]
>
> --- END LIBC BACKTRACE ---
>
> Aborted (core dumped)
> 
>
> Best regards,
>
> F. A. Faisal
>
>
> On Fri, May 5, 2017 at 11:22 PM, Krishna, Tushar 
> wrote:
>
>> Hi Faisal,
>> You can still add --topology with the fs.py file.
>> You need to also add --ruby to run ruby.
>>
>> For running garnet, you can add a --network=garnet2.0
>> (Look at options in the garnet2.0 page on the wiki).
>>
>> Cheers,
>> Tushar
>>
>> On May 3, 2017, at 10:50 AM, F. A. Faisal  wrote:
>>
>> Hi,
>>
>> I like to invoke the garnet 2.0 for running the benchmark programs Parsec
>> 2.1.
>> However, previous ruby_fs.py file located at configs/example/ folder is
>> not available with the current gem5 source, which could invoke the
>> garnet 1.0. And now the replaced fs.py file don't have option like-
>> topology, garnet. Hence, can any one please let me

Re: [gem5-users] FS simulation with Garnet 2.0

2017-05-05 Thread Krishna, Tushar

Hmm the coherence protocol seems to have received an invalid transaction. Not 
sure why. This does not seem to be a network issue. But for a sanity check, run 
with network=simple and see if it still fails.

Cheers,
Tushar

On May 5, 2017, at 11:46 AM, F. A. Faisal 
> wrote:

Thanks a lot.

I already figured out that problem by myself.
However, I encountered a new problem in running 4 core 4 threads blackscholes 
benchmarks. It shows system panic.
I also tried for 16 cpu and Mesh network, which is still running more than 8 
hours.

I used the mesh_xy topology along with below config-


scons FULL_SYSTEM=1 build/ALPHA/gem5.opt RUBY=true PROTOCOL=MOESI_CMP_directory


 ./build/ALPHA/gem5.opt configs/example/fs.py --cpu-type=timing 
--script=script/blackscholes_4c_simsmall.rcS --num-cpus=4 --ruby --mesh-rows=2 
--network=garnet2.0 --topology=Mesh_XY --caches --l2cache --l2_size=512kB 
--num-dirs=4 --num-l2=4 --l1d_size=32kB --l1i_size=32kB --l1d_assoc=2 
--l1i_assoc=2

It showed me the below error-
--

panic: Invalid transition

system.ruby.dma_cntrl0 time: 4816339893 addr: 525287488 event: WriteRequest 
state: BUSY_WR

 @ tick 2408169946500

[doTransitionWorker:build/ALPHA/mem/protocol/DMA_Transitions.cc, line 135]

Memory Usage: 996944 KBytes

Program aborted at tick 2408169946500

--- BEGIN LIBC BACKTRACE ---

./build/ALPHA/gem5.opt[0x829015]

./build/ALPHA/gem5.opt[0x839d96]

/lib64/libpthread.so.0[0x3685e0f730]

/lib64/libc.so.6(gsignal+0x3e)[0x3685a3266e]

/lib64/libc.so.6(abort+0x175)[0x3685a33ab5]

./build/ALPHA/gem5.opt[0x6867ac]

./build/ALPHA/gem5.opt[0x631c58]

./build/ALPHA/gem5.opt[0x631e83]

./build/ALPHA/gem5.opt[0x634aad]

./build/ALPHA/gem5.opt[0x8321d1]

./build/ALPHA/gem5.opt[0x852538]

./build/ALPHA/gem5.opt[0x852abb]

./build/ALPHA/gem5.opt[0x7d8d2d]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x4e26)[0x369ced55b6]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x54f2)[0x369ced5c82]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]

/usr/lib64/libpython2.6.so.1.0[0x369cef1b9c]

/usr/lib64/libpython2.6.so.1.0(PyRun_StringFlags+0x7a)[0x369cef1daa]

./build/ALPHA/gem5.opt[0x83967f]

./build/ALPHA/gem5.opt[0x415453]

/lib64/libc.so.6(__libc_start_main+0xfd)[0x3685a1f56d]

./build/ALPHA/gem5.opt[0x43b1b1]

--- END LIBC BACKTRACE ---

Aborted (core dumped)



Best regards,

F. A. Faisal


On Fri, May 5, 2017 at 11:22 PM, Krishna, Tushar 
> wrote:
Hi Faisal,
You can still add --topology with the fs.py file.
You need to also add --ruby to run ruby.

For running garnet, you can add a --network=garnet2.0
(Look at options in the garnet2.0 page on the wiki).

Cheers,
Tushar

On May 3, 2017, at 10:50 AM, F. A. Faisal 
> wrote:

Hi,

I like to invoke the garnet 2.0 for running the benchmark programs Parsec 2.1.
However, previous ruby_fs.py file located at configs/example/ folder is not 
available with the current gem5 source, which could invoke the garnet 1.0. And 
now the replaced fs.py file don't have option like- topology, garnet. Hence, 
can any one please let me know how to invoke the garnet 2.0 for full system 
simulation.

Command for previous gem5 with garnet 1.0-

scons build/ALPHA_FS/gem5.opt RUBY=true PROTOCOL=MOESI_CMP_directory


build/ALPHA_FS/gem5.opt configs/example/ruby_fs.py --num-cpus=4 --num-dirs=4 
--num-l2caches=4 --topology=Mesh --mesh-rows=2 --garnet=fixed 
--kernel=vmlinux_2.6.27-gcc_4.3.4 --script=scripts/runparsec1.rcS


Thanks in advance.


F. A. Faisal

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

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

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

Re: [gem5-users] How to run gem5 ruby in ALPHA?

2017-05-05 Thread Andreas Hansson

Hi,

I’d argue that the classic memory system is more robust and performant compared 
to Ruby, and consequently I would say it’s a good starting point for most 
use-cases. The classic caches definitely support inclusive policies, it is 
simply a configuration option “clusivity”. Inclusive L2s, exclusive L3s, 
etc…tune it as you please.

Andreas

From: gem5-users 
> on behalf of 
liuchen <13630583...@163.com>
Reply-To: gem5 users mailing list 
>
Date: Friday, 5 May 2017 at 13:02
To: "gem5-users@gem5.org" 
>
Subject: [gem5-users] How to run gem5 ruby in ALPHA?

Hi:
I am using gem5 to test ALPHA memory system.I found that classic cache is not 
inclusive,so I move on to ruby.But I don't know how to use ruby system. Is it 
OK  just add an option"--ruby" to command line?And how to use 
configs/ruby/MESI_two/three_level.py?is ruby strictly inclusive or not?
Thanks!





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
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] FS simulation with Garnet 2.0

2017-05-05 Thread F. A. Faisal

Thanks a lot.

I already figured out that problem by myself.
However, I encountered a new problem in running 4 core 4 threads
blackscholes benchmarks. It shows system panic.
I also tried for 16 cpu and Mesh network, which is still running more than
8 hours.

I used the mesh_xy topology along with below config-

scons FULL_SYSTEM=1 build/ALPHA/gem5.opt RUBY=true
PROTOCOL=MOESI_CMP_directory

 ./build/ALPHA/gem5.opt configs/example/fs.py --cpu-type=timing
--script=script/blackscholes_4c_simsmall.rcS --num-cpus=4 --ruby
--mesh-rows=2 --network=garnet2.0 --topology=Mesh_XY --caches --l2cache
--l2_size=512kB --num-dirs=4 --num-l2=4 --l1d_size=32kB --l1i_size=32kB
--l1d_assoc=2 --l1i_assoc=2

It showed me the below error-
--

panic: Invalid transition

system.ruby.dma_cntrl0 time: 4816339893 addr: 525287488 event: WriteRequest
state: BUSY_WR

 @ tick 2408169946500

[doTransitionWorker:build/ALPHA/mem/protocol/DMA_Transitions.cc, line 135]

Memory Usage: 996944 KBytes

Program aborted at tick 2408169946500

--- BEGIN LIBC BACKTRACE ---

./build/ALPHA/gem5.opt[0x829015]

./build/ALPHA/gem5.opt[0x839d96]

/lib64/libpthread.so.0[0x3685e0f730]

/lib64/libc.so.6(gsignal+0x3e)[0x3685a3266e]

/lib64/libc.so.6(abort+0x175)[0x3685a33ab5]

./build/ALPHA/gem5.opt[0x6867ac]

./build/ALPHA/gem5.opt[0x631c58]

./build/ALPHA/gem5.opt[0x631e83]

./build/ALPHA/gem5.opt[0x634aad]

./build/ALPHA/gem5.opt[0x8321d1]

./build/ALPHA/gem5.opt[0x852538]

./build/ALPHA/gem5.opt[0x852abb]

./build/ALPHA/gem5.opt[0x7d8d2d]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x4e26)[0x369ced55b6]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x63ef)[0x369ced6b7f]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x54f2)[0x369ced5c82]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalFrameEx+0x5304)[0x369ced5a94]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCodeEx+0x927)[0x369ced7647]

/usr/lib64/libpython2.6.so.1.0(PyEval_EvalCode+0x32)[0x369ced7722]

/usr/lib64/libpython2.6.so.1.0[0x369cef1b9c]

/usr/lib64/libpython2.6.so.1.0(PyRun_StringFlags+0x7a)[0x369cef1daa]

./build/ALPHA/gem5.opt[0x83967f]

./build/ALPHA/gem5.opt[0x415453]

/lib64/libc.so.6(__libc_start_main+0xfd)[0x3685a1f56d]

./build/ALPHA/gem5.opt[0x43b1b1]

--- END LIBC BACKTRACE ---

Aborted (core dumped)


Best regards,

F. A. Faisal


On Fri, May 5, 2017 at 11:22 PM, Krishna, Tushar 
wrote:

> Hi Faisal,
> You can still add --topology with the fs.py file.
> You need to also add --ruby to run ruby.
>
> For running garnet, you can add a --network=garnet2.0
> (Look at options in the garnet2.0 page on the wiki).
>
> Cheers,
> Tushar
>
> On May 3, 2017, at 10:50 AM, F. A. Faisal  wrote:
>
> Hi,
>
> I like to invoke the garnet 2.0 for running the benchmark programs Parsec
> 2.1.
> However, previous ruby_fs.py file located at configs/example/ folder is
> not available with the current gem5 source, which could invoke the
> garnet 1.0. And now the replaced fs.py file don't have option like-
> topology, garnet. Hence, can any one please let me know how to invoke the
> garnet 2.0 for full system simulation.
>
> Command for previous gem5 with garnet 1.0-
>
> scons build/ALPHA_FS/gem5.opt RUBY=true PROTOCOL=MOESI_CMP_directory
>
> build/ALPHA_FS/gem5.opt configs/example/ruby_fs.py --num-cpus=4 --num-dirs=4 
> --num-l2caches=4 --topology=Mesh --mesh-rows=2 --garnet=fixed 
> --kernel=vmlinux_2.6.27-gcc_4.3.4 --script=scripts/runparsec1.rcS
>
>
> Thanks in advance.
>
>
> F. A. Faisal
>
> ___
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
>
> ___
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

[gem5-users] How to run gem5 ruby in ALPHA?

2017-05-05 Thread liuchen

Hi:
I am using gem5 to test ALPHA memory system.I found that classic cache is not 
inclusive,so I move on to ruby.But I don't know how to use ruby system. Is it 
OK  just add an option"--ruby" to command line?And how to use 
configs/ruby/MESI_two/three_level.py?is ruby strictly inclusive or not?
Thanks!
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] seek help for hybrid cache controller

2017-05-05 Thread BOZHI LIU

我感觉gem5的cache模型是比较抽象的没有analytical描述底层的，有的是抽象的结构、coherence的协议外加一系列延迟参数。我觉得你这个情况是不是考虑在延迟参数的地方做文章？类似于不同的访问类型（用到不同的physical
component）给出一个可变的延时返回值

On Fri, May 5, 2017 at 1:24 AM, 平凡就好 <2418270...@qq.com> wrote:

> hello everyone，
> I want to do a L2 cache structure based on the SRAM and NVM , but the
> idea of how to make a  hybrid cache controller is so hard for me. Is there
> any experience or tutorial can help me?Any help I would be grateful！
> Thanks，
> Micheal
>
> ___
> gem5-users mailing list
> gem5-users@gem5.org
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>
___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

[gem5-users] seek help for hybrid cache controller

2017-05-05 Thread ????????

hello everyone??
I want to do a L2 cache structure based on the SRAM and NVM , but the idea of 
how to make a  hybrid cache controller is so hard for me. Is there any 
experience or tutorial can help me?Any help I would be grateful??
Thanks??
Micheal___
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Re: [gem5-users] Build system failures with gem5.opt and libgem5.so

Re: [gem5-users] Build system failures with gem5.opt and libgem5.so

[gem5-users] Build system failures with gem5.opt and libgem5.so

Re: [gem5-users] fatal: _pid

[gem5-users] fatal: _pid

Re: [gem5-users] FS simulation with Garnet 2.0

Re: [gem5-users] FS simulation with Garnet 2.0

Re: [gem5-users] FS simulation with Garnet 2.0

Re: [gem5-users] How to run gem5 ruby in ALPHA?

Re: [gem5-users] FS simulation with Garnet 2.0

[gem5-users] How to run gem5 ruby in ALPHA?

Re: [gem5-users] seek help for hybrid cache controller

[gem5-users] seek help for hybrid cache controller

13 matches

Site Navigation

Mail list logo

Footer information