Empathy List Archives

BN

Beser, Nicholas D.

Wed, Feb 26, 2025 2:32 PM

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=
gfx902 -c gem5-resources/src/gpu/square/bin/square
gem5 Simulator System. https://www.gem5.org
gem5 is copyrighted software; use the --copyright option for details.

I have setup a github codespace for teaching my advanced computer architecture class. I tried to run the example square for the AMD GPU, and got an error (gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory

Is this due to the compilation of square, or is this a common problem with codespace?

Nick

gem5 version 24.0.0.0
gem5 compiled Jul 25 2024 20:34:41
gem5 started Feb 26 2025 14:08:28
gem5 executing on codespaces-e992b2, pid 15064
command line: gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square

Num SQC = 1 Num scalar caches = 1 Num CU = 4
Global frequency set at 1000000000000 ticks per second
warn: system.ruby.network adopting orphan SimObject param 'ext_links'
warn: system.ruby.network adopting orphan SimObject param 'int_links'
src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) does not match the address range assigned (512 Mbytes)
src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated.
src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated.
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated.
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
system.remote_gdb: Listening for connections on port 7000
src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting simulation...
src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates recently became the responsibility of SLICC state machines. Make sure to setMRU() near callbacks in .sm files!
src/sim/mem_state.cc:448: info: Increasing stack size by one page.
gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory
breaking loop due to: exiting with last active thread context.
Ticks: 234627000
Exiting because exiting with last active thread context
root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version= gfx902 -c gem5-resources/src/gpu/square/bin/square gem5 Simulator System. https://www.gem5.org gem5 is copyrighted software; use the --copyright option for details. I have setup a github codespace for teaching my advanced computer architecture class. I tried to run the example square for the AMD GPU, and got an error (gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory Is this due to the compilation of square, or is this a common problem with codespace? Nick gem5 version 24.0.0.0 gem5 compiled Jul 25 2024 20:34:41 gem5 started Feb 26 2025 14:08:28 gem5 executing on codespaces-e992b2, pid 15064 command line: gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square Num SQC = 1 Num scalar caches = 1 Num CU = 4 Global frequency set at 1000000000000 ticks per second warn: system.ruby.network adopting orphan SimObject param 'ext_links' warn: system.ruby.network adopting orphan SimObject param 'int_links' src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) does not match the address range assigned (512 Mbytes) src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated. src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated. Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated. Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) system.remote_gdb: Listening for connections on port 7000 src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting simulation... src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates recently became the responsibility of SLICC state machines. Make sure to setMRU() near callbacks in .sm files! src/sim/mem_state.cc:448: info: Increasing stack size by one page. gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory breaking loop due to: exiting with last active thread context. Ticks: 234627000 Exiting because exiting with last active thread context root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#

MS

Matt Sinclair

Sat, Mar 1, 2025 3:12 AM

Hi Nicholas,

Sorry for my delayed response, but very cool to hear you are using these.
Jason (CC'd) might need to help about codespace specific issues though.
Before we get there though, there are some things to try. Ultimately, the
problem is that it can't find the libamdhip library. I'll need some more
information to help you triage, but the most obvious thing: what steps did
you take before this one? Specifically, did you run the docker pull
command like the bootcamp requires? My recollection is that this needs to
be done first, before anything else. And in the command at the top of your
email, you don't appear to be using the docker. I will also note you (I
think) changed the example to use gfx902 instead of gfx900. From looking
at my slides from the bootcamp, it does not appear I talked about gfx902 at
all. So, just making sure that change is intentional?

Jason, I'm having a hard time finding the right pointer to where in the
bootcamp repo the GPU instructions are. Is it this:
https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md
?

Thanks,
Matt

On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users <
gem5-users@gem5.org> wrote:

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#
gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=

gfx902 -c gem5-resources/src/gpu/square/bin/square

gem5 Simulator System. https://www.gem5.org

gem5 is copyrighted software; use the --copyright option for details.

I have setup a github codespace for teaching my advanced computer
architecture class. I tried to run the example square for the AMD GPU, and
got an error (gem5-resources/src/gpu/square/bin/square: error while loading
shared libraries: libamdhip64.so.4: cannot open shared object file: No such
file or directory

Is this due to the compilation of square, or is this a common problem with
codespace?

Nick

gem5 version 24.0.0.0

gem5 compiled Jul 25 2024 20:34:41

gem5 started Feb 26 2025 14:08:28

gem5 executing on codespaces-e992b2, pid 15064

command line: gem5-vega gem5/configs/example/apu_se.py -n 3
--gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square

Num SQC = 1 Num scalar caches = 1 Num CU = 4

Global frequency set at 1000000000000 ticks per second

warn: system.ruby.network adopting orphan SimObject param 'ext_links'

warn: system.ruby.network adopting orphan SimObject param 'int_links'

src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes)
does not match the address range assigned (512 Mbytes)

src/base/statistics.hh:279: warn: One of the stats is a legacy stat.
Legacy stat is a stat that does not belong to any statistics::Group. Legacy
stat is deprecated.

Forcing maxCoalescedReqs to 32 (TLB assoc.)

src/base/statistics.hh:279: warn: One of the stats is a legacy stat.
Legacy stat is a stat that does not belong to any statistics::Group. Legacy
stat is deprecated.

Forcing maxCoalescedReqs to 32 (TLB assoc.)

system.remote_gdb: Listening for connections on port 7000

src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting
simulation...

src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates
recently became the responsibility of SLICC state machines. Make sure to
setMRU() near callbacks in .sm files!

src/sim/mem_state.cc:448: info: Increasing stack size by one page.

gem5-resources/src/gpu/square/bin/square: error while loading shared
libraries: libamdhip64.so.4: cannot open shared object file: No such file
or directory

breaking loop due to: exiting with last active thread context.

Ticks: 234627000

Exiting because exiting with last active thread context

root@codespaces-e992b2
:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#

gem5-users mailing list -- gem5-users@gem5.org
To unsubscribe send an email to gem5-users-leave@gem5.org

Hi Nicholas, Sorry for my delayed response, but very cool to hear you are using these. Jason (CC'd) might need to help about codespace specific issues though. Before we get there though, there are some things to try. Ultimately, the problem is that it can't find the libamdhip library. I'll need some more information to help you triage, but the most obvious thing: what steps did you take before this one? Specifically, did you run the docker pull command like the bootcamp requires? My recollection is that this needs to be done first, before anything else. And in the command at the top of your email, you don't appear to be using the docker. I will also note you (I think) changed the example to use gfx902 instead of gfx900. From looking at my slides from the bootcamp, it does not appear I talked about gfx902 at all. So, just making sure that change is intentional? Jason, I'm having a hard time finding the right pointer to where in the bootcamp repo the GPU instructions are. Is it this: https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md ? Thanks, Matt On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users < gem5-users@gem5.org> wrote: > root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# > gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version= > > gfx902 -c gem5-resources/src/gpu/square/bin/square > > gem5 Simulator System. https://www.gem5.org > > gem5 is copyrighted software; use the --copyright option for details. > > > > I have setup a github codespace for teaching my advanced computer > architecture class. I tried to run the example square for the AMD GPU, and > got an error (gem5-resources/src/gpu/square/bin/square: error while loading > shared libraries: libamdhip64.so.4: cannot open shared object file: No such > file or directory > > > > Is this due to the compilation of square, or is this a common problem with > codespace? > > > > Nick > > > > gem5 version 24.0.0.0 > > gem5 compiled Jul 25 2024 20:34:41 > > gem5 started Feb 26 2025 14:08:28 > > gem5 executing on codespaces-e992b2, pid 15064 > > command line: gem5-vega gem5/configs/example/apu_se.py -n 3 > --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square > > > > Num SQC = 1 Num scalar caches = 1 Num CU = 4 > > Global frequency set at 1000000000000 ticks per second > > warn: system.ruby.network adopting orphan SimObject param 'ext_links' > > warn: system.ruby.network adopting orphan SimObject param 'int_links' > > src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) > does not match the address range assigned (512 Mbytes) > > src/base/statistics.hh:279: warn: One of the stats is a legacy stat. > Legacy stat is a stat that does not belong to any statistics::Group. Legacy > stat is deprecated. > > src/base/statistics.hh:279: warn: One of the stats is a legacy stat. > Legacy stat is a stat that does not belong to any statistics::Group. Legacy > stat is deprecated. > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > src/base/statistics.hh:279: warn: One of the stats is a legacy stat. > Legacy stat is a stat that does not belong to any statistics::Group. Legacy > stat is deprecated. > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > system.remote_gdb: Listening for connections on port 7000 > > src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting > simulation... > > src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates > recently became the responsibility of SLICC state machines. Make sure to > setMRU() near callbacks in .sm files! > > src/sim/mem_state.cc:448: info: Increasing stack size by one page. > > gem5-resources/src/gpu/square/bin/square: error while loading shared > libraries: libamdhip64.so.4: cannot open shared object file: No such file > or directory > > breaking loop due to: exiting with last active thread context. > > Ticks: 234627000 > > Exiting because exiting with last active thread context > > root@codespaces-e992b2 > :/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# > _______________________________________________ > gem5-users mailing list -- gem5-users@gem5.org > To unsubscribe send an email to gem5-users-leave@gem5.org >

BN

Beser, Nicholas D.

Sat, Mar 1, 2025 7:41 PM

Matt,

Thank you for getting back to me. If I include the docker command I get the following error:

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024/gem5# docker run --volume $(pwd):$(pwd) -w $(pwd) ghcr.io/gem5/gcn-gpu:v24-0 gem5-vega configs/example/apu_se.py -n 3 -c gem5-resources/src/gpu/square/bin/square

docker: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: exec: "gem5-vega": executable file not found in $PATH: unknown.
root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-us

If I exclude the docker command, I get the error that I posted.

The codespace was setup by the group that is setting up the payment for using github. Their instructions were:

Here is what I did, if you need to recreate this repo from scratch or after gem5 updates their codebase:

Initialize a new repo 'bootcamp2024' under my account with the + button on my top bar, so the .git directory is generated.
From cli; git clone <newly-created-repo> to a local directory on my PC: ./bootcamp24/
cd ..
git clone https://github.com/gem5bootcamp/2024.git https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgem5bootcamp%2F2024.git&data=05%7C02%7Cnbeser1%40jhu.edu%7Cbd426705445f4ec97dfd08dd567dc3ee%7C9fa4f438b1e6473b803f86f8aedf0dec%7C0%7C0%7C638761820328498774%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Wwv3b3yMN5T2Zh7znZgbGQv8hbTb6x1sNQFwffssaUA%3D&reserved=0
rm -dr gem5bootcamp/2024/.git
cd bootcamp24 && mv gem5bootcamp/2024/* .
mv gem5bootcamp/2024/.* .
cd bootcamp2024
git add *
git add .*
git commit
git push

really any way you need to copy the file tree of the official bootcamp repo, minus the parts of the git manifest directory that prevent you committing changes because you made some automatic test fail that gem5 put in there to make sure builds succeed.

I’m not sure how to follow their directions to setup a new repository.

I have setup a x86 VM using virtualbox and it is able to run the instructions for using the AMD GPU. If I have the 2024 directory with the gem5 code added using the –recursive command, is there a way of setting up the codespace from that? I am missing the step that guides the creation of the repository.

Nick

From: Matt Sinclair mattdsinclair.wisc@gmail.com
Sent: Friday, February 28, 2025 10:13 PM
To: The gem5 Users mailing list gem5-users@gem5.org
Cc: Beser, Nicholas D. Nick.Beser@jhuapl.edu; Jason Lowe-Power jason@lowepower.com
Subject: [EXT] Re: [gem5-users] Problem running gpu example with codespace

APL external email warning: Verify sender mattdsinclair.wisc@gmail.com mailto:mattdsinclair.wisc@gmail.com before clicking links or attachments

Hi Nicholas,

Sorry for my delayed response, but very cool to hear you are using these. Jason (CC'd) might need to help about codespace specific issues though. Before we get there though, there are some things to try. Ultimately, the problem is that it can't find the libamdhip library. I'll need some more information to help you triage, but the most obvious thing: what steps did you take before this one? Specifically, did you run the docker pull command like the bootcamp requires? My recollection is that this needs to be done first, before anything else. And in the command at the top of your email, you don't appear to be using the docker. I will also note you (I think) changed the example to use gfx902 instead of gfx900. From looking at my slides from the bootcamp, it does not appear I talked about gfx902 at all. So, just making sure that change is intentional?

Jason, I'm having a hard time finding the right pointer to where in the bootcamp repo the GPU instructions are. Is it this: https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md?

Thanks,
Matt

On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users <gem5-users@gem5.org mailto:gem5-users@gem5.org> wrote:
root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=
gfx902 -c gem5-resources/src/gpu/square/bin/square
gem5 Simulator System. https://www.gem5.org
gem5 is copyrighted software; use the --copyright option for details.

I have setup a github codespace for teaching my advanced computer architecture class. I tried to run the example square for the AMD GPU, and got an error (gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory

Is this due to the compilation of square, or is this a common problem with codespace?

Nick

gem5 version 24.0.0.0
gem5 compiled Jul 25 2024 20:34:41
gem5 started Feb 26 2025 14:08:28
gem5 executing on codespaces-e992b2, pid 15064
command line: gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square

Num SQC = 1 Num scalar caches = 1 Num CU = 4
Global frequency set at 1000000000000 ticks per second
warn: system.ruby.network adopting orphan SimObject param 'ext_links'
warn: system.ruby.network adopting orphan SimObject param 'int_links'
src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) does not match the address range assigned (512 Mbytes)
src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated.
src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated.
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated.
Forcing maxCoalescedReqs to 32 (TLB assoc.)
Forcing maxCoalescedReqs to 32 (TLB assoc.)
system.remote_gdb: Listening for connections on port 7000
src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting simulation...
src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates recently became the responsibility of SLICC state machines. Make sure to setMRU() near callbacks in .sm files!
src/sim/mem_state.cc:448: info: Increasing stack size by one page.
gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory
breaking loop due to: exiting with last active thread context.
Ticks: 234627000
Exiting because exiting with last active thread context
root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#

gem5-users mailing list -- gem5-users@gem5.org mailto:gem5-users@gem5.org
To unsubscribe send an email to gem5-users-leave@gem5.org mailto:gem5-users-leave@gem5.org

Matt, Thank you for getting back to me. If I include the docker command I get the following error: root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024/gem5# docker run --volume $(pwd):$(pwd) -w $(pwd) ghcr.io/gem5/gcn-gpu:v24-0 gem5-vega configs/example/apu_se.py -n 3 -c gem5-resources/src/gpu/square/bin/square docker: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: exec: "gem5-vega": executable file not found in $PATH: unknown. root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-us If I exclude the docker command, I get the error that I posted. The codespace was setup by the group that is setting up the payment for using github. Their instructions were: Here is what I did, if you need to recreate this repo from scratch or after gem5 updates their codebase: Initialize a new repo 'bootcamp2024' under my account with the + button on my top bar, so the .git directory is generated. From cli; git clone <newly-created-repo> to a local directory on my PC: ./bootcamp24/ cd .. git clone https://github.com/gem5bootcamp/2024.git<https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgem5bootcamp%2F2024.git&data=05%7C02%7Cnbeser1%40jhu.edu%7Cbd426705445f4ec97dfd08dd567dc3ee%7C9fa4f438b1e6473b803f86f8aedf0dec%7C0%7C0%7C638761820328498774%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Wwv3b3yMN5T2Zh7znZgbGQv8hbTb6x1sNQFwffssaUA%3D&reserved=0> rm -dr gem5bootcamp/2024/.git cd bootcamp24 && mv gem5bootcamp/2024/* . mv gem5bootcamp/2024/.* . cd bootcamp2024 git add * git add .* git commit git push really any way you need to copy the file tree of the official bootcamp repo, minus the parts of the git manifest directory that prevent you committing changes because you made some automatic test fail that gem5 put in there to make sure builds succeed. I’m not sure how to follow their directions to setup a new repository. I have setup a x86 VM using virtualbox and it is able to run the instructions for using the AMD GPU. If I have the 2024 directory with the gem5 code added using the –recursive command, is there a way of setting up the codespace from that? I am missing the step that guides the creation of the repository. Nick From: Matt Sinclair <mattdsinclair.wisc@gmail.com> Sent: Friday, February 28, 2025 10:13 PM To: The gem5 Users mailing list <gem5-users@gem5.org> Cc: Beser, Nicholas D. <Nick.Beser@jhuapl.edu>; Jason Lowe-Power <jason@lowepower.com> Subject: [EXT] Re: [gem5-users] Problem running gpu example with codespace APL external email warning: Verify sender mattdsinclair.wisc@gmail.com<mailto:mattdsinclair.wisc@gmail.com> before clicking links or attachments Hi Nicholas, Sorry for my delayed response, but very cool to hear you are using these. Jason (CC'd) might need to help about codespace specific issues though. Before we get there though, there are some things to try. Ultimately, the problem is that it can't find the libamdhip library. I'll need some more information to help you triage, but the most obvious thing: what steps did you take before this one? Specifically, did you run the docker pull command like the bootcamp requires? My recollection is that this needs to be done first, before anything else. And in the command at the top of your email, you don't appear to be using the docker. I will also note you (I think) changed the example to use gfx902 instead of gfx900. From looking at my slides from the bootcamp, it does not appear I talked about gfx902 at all. So, just making sure that change is intentional? Jason, I'm having a hard time finding the right pointer to where in the bootcamp repo the GPU instructions are. Is it this: https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md? Thanks, Matt On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users <gem5-users@gem5.org<mailto:gem5-users@gem5.org>> wrote: root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version= gfx902 -c gem5-resources/src/gpu/square/bin/square gem5 Simulator System. https://www.gem5.org gem5 is copyrighted software; use the --copyright option for details. I have setup a github codespace for teaching my advanced computer architecture class. I tried to run the example square for the AMD GPU, and got an error (gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory Is this due to the compilation of square, or is this a common problem with codespace? Nick gem5 version 24.0.0.0 gem5 compiled Jul 25 2024 20:34:41 gem5 started Feb 26 2025 14:08:28 gem5 executing on codespaces-e992b2, pid 15064 command line: gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square Num SQC = 1 Num scalar caches = 1 Num CU = 4 Global frequency set at 1000000000000 ticks per second warn: system.ruby.network adopting orphan SimObject param 'ext_links' warn: system.ruby.network adopting orphan SimObject param 'int_links' src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) does not match the address range assigned (512 Mbytes) src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated. src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated. Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) src/base/statistics.hh:279: warn: One of the stats is a legacy stat. Legacy stat is a stat that does not belong to any statistics::Group. Legacy stat is deprecated. Forcing maxCoalescedReqs to 32 (TLB assoc.) Forcing maxCoalescedReqs to 32 (TLB assoc.) system.remote_gdb: Listening for connections on port 7000 src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting simulation... src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates recently became the responsibility of SLICC state machines. Make sure to setMRU() near callbacks in .sm files! src/sim/mem_state.cc:448: info: Increasing stack size by one page. gem5-resources/src/gpu/square/bin/square: error while loading shared libraries: libamdhip64.so.4: cannot open shared object file: No such file or directory breaking loop due to: exiting with last active thread context. Ticks: 234627000 Exiting because exiting with last active thread context root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# _______________________________________________ gem5-users mailing list -- gem5-users@gem5.org<mailto:gem5-users@gem5.org> To unsubscribe send an email to gem5-users-leave@gem5.org<mailto:gem5-users-leave@gem5.org>

MS

Matt Sinclair

Sun, Mar 2, 2025 1:21 AM

Jason will have to chime in on setting up the codespace, as I am not
familiar with that.

But regarding the docker, did you try running "docker pull
ghcr.io/gem5/gcn-gpu:v24-0" (or "docker pull ghcr.io/gem5/gcn-gpu:latest")
first? I don't recall if Jason had the docker pre-downloaded and thus this
step was not needed or not, but this would be one way to check.

Matt

On Sat, Mar 1, 2025 at 1:41 PM Beser, Nicholas D. Nick.Beser@jhuapl.edu
wrote:

Matt,

Thank you for getting back to me. If I include the docker command I get
the following error:

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024/gem5#
docker run --volume $(pwd):$(pwd) -w $(pwd) ghcr.io/gem5/gcn-gpu:v24-0
gem5-vega configs/example/apu_se.py -n 3 -c
gem5-resources/src/gpu/square/bin/square

docker: Error response from daemon: failed to create task for container:
failed to create shim task: OCI runtime create failed: runc create failed:
unable to start container process: exec: "gem5-vega": executable file not
found in $PATH: unknown.

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-us

If I exclude the docker command, I get the error that I posted.

The codespace was setup by the group that is setting up the payment for
using github. Their instructions were:

Here is what I did, if you need to recreate this repo from scratch or
after gem5 updates their codebase:

Initialize a new repo 'bootcamp2024' under my account with the + button on
my top bar, so the .git directory is generated.
From cli; git clone <newly-created-repo> to a local directory on my PC:
./bootcamp24/
cd ..

git clone https://github.com/gem5bootcamp/2024.git
https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgem5bootcamp%2F2024.git&data=05%7C02%7Cnbeser1%40jhu.edu%7Cbd426705445f4ec97dfd08dd567dc3ee%7C9fa4f438b1e6473b803f86f8aedf0dec%7C0%7C0%7C638761820328498774%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Wwv3b3yMN5T2Zh7znZgbGQv8hbTb6x1sNQFwffssaUA%3D&reserved=0
rm -dr gem5bootcamp/2024/.git
cd bootcamp24 && mv gem5bootcamp/2024/* .
mv gem5bootcamp/2024/.* .
cd bootcamp2024
git add *

git add .*

git commit
git push

really any way you need to copy the file tree of the official bootcamp
repo, minus the parts of the git manifest directory that prevent you
committing changes because you made some automatic test fail that gem5 put
in there to make sure builds succeed.

I’m not sure how to follow their directions to setup a new repository.

I have setup a x86 VM using virtualbox and it is able to run the
instructions for using the AMD GPU. If I have the 2024 directory with the
gem5 code added using the –recursive command, is there a way of setting up
the codespace from that? I am missing the step that guides the creation of
the repository.

Nick

From: Matt Sinclair mattdsinclair.wisc@gmail.com
Sent: Friday, February 28, 2025 10:13 PM
To: The gem5 Users mailing list gem5-users@gem5.org
Cc: Beser, Nicholas D. Nick.Beser@jhuapl.edu; Jason Lowe-Power <
jason@lowepower.com>
Subject: [EXT] Re: [gem5-users] Problem running gpu example with
codespace

*APL external email warning: *Verify sender mattdsinclair.wisc@gmail.com
before clicking links or attachments

Hi Nicholas,

Sorry for my delayed response, but very cool to hear you are using these.
Jason (CC'd) might need to help about codespace specific issues though.
Before we get there though, there are some things to try. Ultimately, the
problem is that it can't find the libamdhip library. I'll need some more
information to help you triage, but the most obvious thing: what steps did
you take before this one? Specifically, did you run the docker pull
command like the bootcamp requires? My recollection is that this needs to
be done first, before anything else. And in the command at the top of your
email, you don't appear to be using the docker. I will also note you (I
think) changed the example to use gfx902 instead of gfx900. From looking
at my slides from the bootcamp, it does not appear I talked about gfx902 at
all. So, just making sure that change is intentional?

Jason, I'm having a hard time finding the right pointer to where in the
bootcamp repo the GPU instructions are. Is it this:
https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md
?

Thanks,

Matt

On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users <
gem5-users@gem5.org> wrote:

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#
gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=

gfx902 -c gem5-resources/src/gpu/square/bin/square

gem5 Simulator System. https://www.gem5.org

gem5 is copyrighted software; use the --copyright option for details.

I have setup a github codespace for teaching my advanced computer
architecture class. I tried to run the example square for the AMD GPU, and
got an error (gem5-resources/src/gpu/square/bin/square: error while loading
shared libraries: libamdhip64.so.4: cannot open shared object file: No such
file or directory

Is this due to the compilation of square, or is this a common problem with
codespace?

Nick

gem5 version 24.0.0.0

gem5 compiled Jul 25 2024 20:34:41

gem5 started Feb 26 2025 14:08:28

gem5 executing on codespaces-e992b2, pid 15064

command line: gem5-vega gem5/configs/example/apu_se.py -n 3
--gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square

Num SQC = 1 Num scalar caches = 1 Num CU = 4

Global frequency set at 1000000000000 ticks per second

warn: system.ruby.network adopting orphan SimObject param 'ext_links'

warn: system.ruby.network adopting orphan SimObject param 'int_links'

src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes)
does not match the address range assigned (512 Mbytes)

src/base/statistics.hh:279: warn: One of the stats is a legacy stat.
Legacy stat is a stat that does not belong to any statistics::Group. Legacy
stat is deprecated.

Forcing maxCoalescedReqs to 32 (TLB assoc.)

src/base/statistics.hh:279: warn: One of the stats is a legacy stat.
Legacy stat is a stat that does not belong to any statistics::Group. Legacy
stat is deprecated.

Forcing maxCoalescedReqs to 32 (TLB assoc.)

system.remote_gdb: Listening for connections on port 7000

src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting
simulation...

src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates
recently became the responsibility of SLICC state machines. Make sure to
setMRU() near callbacks in .sm files!

src/sim/mem_state.cc:448: info: Increasing stack size by one page.

gem5-resources/src/gpu/square/bin/square: error while loading shared
libraries: libamdhip64.so.4: cannot open shared object file: No such file
or directory

breaking loop due to: exiting with last active thread context.

Ticks: 234627000

Exiting because exiting with last active thread context

root@codespaces-e992b2
:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#

gem5-users mailing list -- gem5-users@gem5.org
To unsubscribe send an email to gem5-users-leave@gem5.org

Jason will have to chime in on setting up the codespace, as I am not familiar with that. But regarding the docker, did you try running "docker pull ghcr.io/gem5/gcn-gpu:v24-0" (or "docker pull ghcr.io/gem5/gcn-gpu:latest") first? I don't recall if Jason had the docker pre-downloaded and thus this step was not needed or not, but this would be one way to check. Matt On Sat, Mar 1, 2025 at 1:41 PM Beser, Nicholas D. <Nick.Beser@jhuapl.edu> wrote: > Matt, > > > > Thank you for getting back to me. If I include the docker command I get > the following error: > > > > root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024/gem5# > docker run --volume $(pwd):$(pwd) -w $(pwd) ghcr.io/gem5/gcn-gpu:v24-0 > gem5-vega configs/example/apu_se.py -n 3 -c > gem5-resources/src/gpu/square/bin/square > > > > docker: Error response from daemon: failed to create task for container: > failed to create shim task: OCI runtime create failed: runc create failed: > unable to start container process: exec: "gem5-vega": executable file not > found in $PATH: unknown. > > root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-us > > > > If I exclude the docker command, I get the error that I posted. > > > The codespace was setup by the group that is setting up the payment for > using github. Their instructions were: > > > > Here is what I did, if you need to recreate this repo from scratch or > after gem5 updates their codebase: > > Initialize a new repo 'bootcamp2024' under my account with the + button on > my top bar, so the .git directory is generated. > From cli; git clone <newly-created-repo> to a local directory on my PC: > ./bootcamp24/ > cd .. > > git clone https://github.com/gem5bootcamp/2024.git > <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgem5bootcamp%2F2024.git&data=05%7C02%7Cnbeser1%40jhu.edu%7Cbd426705445f4ec97dfd08dd567dc3ee%7C9fa4f438b1e6473b803f86f8aedf0dec%7C0%7C0%7C638761820328498774%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Wwv3b3yMN5T2Zh7znZgbGQv8hbTb6x1sNQFwffssaUA%3D&reserved=0> > rm -dr gem5bootcamp/2024/.git > cd bootcamp24 && mv gem5bootcamp/2024/* . > mv gem5bootcamp/2024/.* . > cd bootcamp2024 > git add * > > git add .* > > git commit > git push > > really any way you need to copy the file tree of the official bootcamp > repo, minus the parts of the git manifest directory that prevent you > committing changes because you made some automatic test fail that gem5 put > in there to make sure builds succeed. > > > > > > I’m not sure how to follow their directions to setup a new repository. > > > > I have setup a x86 VM using virtualbox and it is able to run the > instructions for using the AMD GPU. If I have the 2024 directory with the > gem5 code added using the –recursive command, is there a way of setting up > the codespace from that? I am missing the step that guides the creation of > the repository. > > > > Nick > > > > *From:* Matt Sinclair <mattdsinclair.wisc@gmail.com> > *Sent:* Friday, February 28, 2025 10:13 PM > *To:* The gem5 Users mailing list <gem5-users@gem5.org> > *Cc:* Beser, Nicholas D. <Nick.Beser@jhuapl.edu>; Jason Lowe-Power < > jason@lowepower.com> > *Subject:* [EXT] Re: [gem5-users] Problem running gpu example with > codespace > > > > *APL external email warning: *Verify sender mattdsinclair.wisc@gmail.com > before clicking links or attachments > > > > Hi Nicholas, > > > > Sorry for my delayed response, but very cool to hear you are using these. > Jason (CC'd) might need to help about codespace specific issues though. > Before we get there though, there are some things to try. Ultimately, the > problem is that it can't find the libamdhip library. I'll need some more > information to help you triage, but the most obvious thing: what steps did > you take before this one? Specifically, did you run the docker pull > command like the bootcamp requires? My recollection is that this needs to > be done first, before anything else. And in the command at the top of your > email, you don't appear to be using the docker. I will also note you (I > think) changed the example to use gfx902 instead of gfx900. From looking > at my slides from the bootcamp, it does not appear I talked about gfx902 at > all. So, just making sure that change is intentional? > > > > Jason, I'm having a hard time finding the right pointer to where in the > bootcamp repo the GPU instructions are. Is it this: > https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md > ? > > > > Thanks, > > Matt > > > > On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users < > gem5-users@gem5.org> wrote: > > root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# > gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version= > > gfx902 -c gem5-resources/src/gpu/square/bin/square > > gem5 Simulator System. https://www.gem5.org > > gem5 is copyrighted software; use the --copyright option for details. > > > > I have setup a github codespace for teaching my advanced computer > architecture class. I tried to run the example square for the AMD GPU, and > got an error (gem5-resources/src/gpu/square/bin/square: error while loading > shared libraries: libamdhip64.so.4: cannot open shared object file: No such > file or directory > > > > Is this due to the compilation of square, or is this a common problem with > codespace? > > > > Nick > > > > gem5 version 24.0.0.0 > > gem5 compiled Jul 25 2024 20:34:41 > > gem5 started Feb 26 2025 14:08:28 > > gem5 executing on codespaces-e992b2, pid 15064 > > command line: gem5-vega gem5/configs/example/apu_se.py -n 3 > --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square > > > > Num SQC = 1 Num scalar caches = 1 Num CU = 4 > > Global frequency set at 1000000000000 ticks per second > > warn: system.ruby.network adopting orphan SimObject param 'ext_links' > > warn: system.ruby.network adopting orphan SimObject param 'int_links' > > src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) > does not match the address range assigned (512 Mbytes) > > src/base/statistics.hh:279: warn: One of the stats is a legacy stat. > Legacy stat is a stat that does not belong to any statistics::Group. Legacy > stat is deprecated. > > src/base/statistics.hh:279: warn: One of the stats is a legacy stat. > Legacy stat is a stat that does not belong to any statistics::Group. Legacy > stat is deprecated. > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > src/base/statistics.hh:279: warn: One of the stats is a legacy stat. > Legacy stat is a stat that does not belong to any statistics::Group. Legacy > stat is deprecated. > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > Forcing maxCoalescedReqs to 32 (TLB assoc.) > > system.remote_gdb: Listening for connections on port 7000 > > src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting > simulation... > > src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates > recently became the responsibility of SLICC state machines. Make sure to > setMRU() near callbacks in .sm files! > > src/sim/mem_state.cc:448: info: Increasing stack size by one page. > > gem5-resources/src/gpu/square/bin/square: error while loading shared > libraries: libamdhip64.so.4: cannot open shared object file: No such file > or directory > > breaking loop due to: exiting with last active thread context. > > Ticks: 234627000 > > Exiting because exiting with last active thread context > > root@codespaces-e992b2 > :/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# > > _______________________________________________ > gem5-users mailing list -- gem5-users@gem5.org > To unsubscribe send an email to gem5-users-leave@gem5.org > >

MS

Matt Sinclair

Sun, Mar 2, 2025 1:23 AM

Sorry, hit send too soon. The other thing to note from the error you sent
is that gem5-vega is not found in the docker. Typically this happens when
the docker working directory and/or volume is not able to find the file. I
am not sure what directory you are running your command from, but this
would be another thing to check.

Matt

On Sat, Mar 1, 2025 at 7:21 PM Matt Sinclair mattdsinclair.wisc@gmail.com
wrote:

Jason will have to chime in on setting up the codespace, as I am not
familiar with that.

But regarding the docker, did you try running "docker pull
ghcr.io/gem5/gcn-gpu:v24-0" (or "docker pull ghcr.io/gem5/gcn-gpu:latest")
first? I don't recall if Jason had the docker pre-downloaded and thus this
step was not needed or not, but this would be one way to check.

Matt

On Sat, Mar 1, 2025 at 1:41 PM Beser, Nicholas D. Nick.Beser@jhuapl.edu
wrote:

Matt,

Thank you for getting back to me. If I include the docker command I get
the following error:

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024/gem5#
docker run --volume $(pwd):$(pwd) -w $(pwd) ghcr.io/gem5/gcn-gpu:v24-0
gem5-vega configs/example/apu_se.py -n 3 -c
gem5-resources/src/gpu/square/bin/square

docker: Error response from daemon: failed to create task for container:
failed to create shim task: OCI runtime create failed: runc create failed:
unable to start container process: exec: "gem5-vega": executable file not
found in $PATH: unknown.

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-us

If I exclude the docker command, I get the error that I posted.

The codespace was setup by the group that is setting up the payment for
using github. Their instructions were:

Here is what I did, if you need to recreate this repo from scratch or
after gem5 updates their codebase:

Initialize a new repo 'bootcamp2024' under my account with the + button
on my top bar, so the .git directory is generated.
From cli; git clone <newly-created-repo> to a local directory on my PC:
./bootcamp24/
cd ..

git clone https://github.com/gem5bootcamp/2024.git
https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgem5bootcamp%2F2024.git&data=05%7C02%7Cnbeser1%40jhu.edu%7Cbd426705445f4ec97dfd08dd567dc3ee%7C9fa4f438b1e6473b803f86f8aedf0dec%7C0%7C0%7C638761820328498774%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Wwv3b3yMN5T2Zh7znZgbGQv8hbTb6x1sNQFwffssaUA%3D&reserved=0
rm -dr gem5bootcamp/2024/.git
cd bootcamp24 && mv gem5bootcamp/2024/* .
mv gem5bootcamp/2024/.* .
cd bootcamp2024
git add *

git add .*

git commit
git push

really any way you need to copy the file tree of the official bootcamp
repo, minus the parts of the git manifest directory that prevent you
committing changes because you made some automatic test fail that gem5 put
in there to make sure builds succeed.

I’m not sure how to follow their directions to setup a new repository.

I have setup a x86 VM using virtualbox and it is able to run the
instructions for using the AMD GPU. If I have the 2024 directory with the
gem5 code added using the –recursive command, is there a way of setting up
the codespace from that? I am missing the step that guides the creation of
the repository.

Nick

From: Matt Sinclair mattdsinclair.wisc@gmail.com
Sent: Friday, February 28, 2025 10:13 PM
To: The gem5 Users mailing list gem5-users@gem5.org
Cc: Beser, Nicholas D. Nick.Beser@jhuapl.edu; Jason Lowe-Power <
jason@lowepower.com>
Subject: [EXT] Re: [gem5-users] Problem running gpu example with
codespace

*APL external email warning: *Verify sender mattdsinclair.wisc@gmail.com
before clicking links or attachments

Hi Nicholas,

Sorry for my delayed response, but very cool to hear you are using
these. Jason (CC'd) might need to help about codespace specific issues
though. Before we get there though, there are some things to try.
Ultimately, the problem is that it can't find the libamdhip library. I'll
need some more information to help you triage, but the most obvious thing:
what steps did you take before this one? Specifically, did you run the
docker pull command like the bootcamp requires? My recollection is that
this needs to be done first, before anything else. And in the command at
the top of your email, you don't appear to be using the docker. I will
also note you (I think) changed the example to use gfx902 instead of
gfx900. From looking at my slides from the bootcamp, it does not appear I
talked about gfx902 at all. So, just making sure that change is
intentional?

Jason, I'm having a hard time finding the right pointer to where in the
bootcamp repo the GPU instructions are. Is it this:
https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md
?

Thanks,

Matt

On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users <
gem5-users@gem5.org> wrote:

root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#
gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version=

gfx902 -c gem5-resources/src/gpu/square/bin/square

gem5 Simulator System. https://www.gem5.org

gem5 is copyrighted software; use the --copyright option for details.

I have setup a github codespace for teaching my advanced computer
architecture class. I tried to run the example square for the AMD GPU, and
got an error (gem5-resources/src/gpu/square/bin/square: error while loading
shared libraries: libamdhip64.so.4: cannot open shared object file: No such
file or directory

Is this due to the compilation of square, or is this a common problem
with codespace?

Nick

gem5 version 24.0.0.0

gem5 compiled Jul 25 2024 20:34:41

gem5 started Feb 26 2025 14:08:28

gem5 executing on codespaces-e992b2, pid 15064

command line: gem5-vega gem5/configs/example/apu_se.py -n 3
--gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square

Num SQC = 1 Num scalar caches = 1 Num CU = 4

Global frequency set at 1000000000000 ticks per second

warn: system.ruby.network adopting orphan SimObject param 'ext_links'

warn: system.ruby.network adopting orphan SimObject param 'int_links'

src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes)
does not match the address range assigned (512 Mbytes)

src/base/statistics.hh:279: warn: One of the stats is a legacy stat.
Legacy stat is a stat that does not belong to any statistics::Group. Legacy
stat is deprecated.

Forcing maxCoalescedReqs to 32 (TLB assoc.)

src/base/statistics.hh:279: warn: One of the stats is a legacy stat.
Legacy stat is a stat that does not belong to any statistics::Group. Legacy
stat is deprecated.

Forcing maxCoalescedReqs to 32 (TLB assoc.)

system.remote_gdb: Listening for connections on port 7000

src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting
simulation...

src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates
recently became the responsibility of SLICC state machines. Make sure to
setMRU() near callbacks in .sm files!

src/sim/mem_state.cc:448: info: Increasing stack size by one page.

gem5-resources/src/gpu/square/bin/square: error while loading shared
libraries: libamdhip64.so.4: cannot open shared object file: No such file
or directory

breaking loop due to: exiting with last active thread context.

Ticks: 234627000

Exiting because exiting with last active thread context

root@codespaces-e992b2
:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024#

gem5-users mailing list -- gem5-users@gem5.org
To unsubscribe send an email to gem5-users-leave@gem5.org

Sorry, hit send too soon. The other thing to note from the error you sent is that gem5-vega is not found in the docker. Typically this happens when the docker working directory and/or volume is not able to find the file. I am not sure what directory you are running your command from, but this would be another thing to check. Matt On Sat, Mar 1, 2025 at 7:21 PM Matt Sinclair <mattdsinclair.wisc@gmail.com> wrote: > Jason will have to chime in on setting up the codespace, as I am not > familiar with that. > > But regarding the docker, did you try running "docker pull > ghcr.io/gem5/gcn-gpu:v24-0" (or "docker pull ghcr.io/gem5/gcn-gpu:latest") > first? I don't recall if Jason had the docker pre-downloaded and thus this > step was not needed or not, but this would be one way to check. > > Matt > > On Sat, Mar 1, 2025 at 1:41 PM Beser, Nicholas D. <Nick.Beser@jhuapl.edu> > wrote: > >> Matt, >> >> >> >> Thank you for getting back to me. If I include the docker command I get >> the following error: >> >> >> >> root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024/gem5# >> docker run --volume $(pwd):$(pwd) -w $(pwd) ghcr.io/gem5/gcn-gpu:v24-0 >> gem5-vega configs/example/apu_se.py -n 3 -c >> gem5-resources/src/gpu/square/bin/square >> >> >> >> docker: Error response from daemon: failed to create task for container: >> failed to create shim task: OCI runtime create failed: runc create failed: >> unable to start container process: exec: "gem5-vega": executable file not >> found in $PATH: unknown. >> >> root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-us >> >> >> >> If I exclude the docker command, I get the error that I posted. >> >> >> The codespace was setup by the group that is setting up the payment for >> using github. Their instructions were: >> >> >> >> Here is what I did, if you need to recreate this repo from scratch or >> after gem5 updates their codebase: >> >> Initialize a new repo 'bootcamp2024' under my account with the + button >> on my top bar, so the .git directory is generated. >> From cli; git clone <newly-created-repo> to a local directory on my PC: >> ./bootcamp24/ >> cd .. >> >> git clone https://github.com/gem5bootcamp/2024.git >> <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgem5bootcamp%2F2024.git&data=05%7C02%7Cnbeser1%40jhu.edu%7Cbd426705445f4ec97dfd08dd567dc3ee%7C9fa4f438b1e6473b803f86f8aedf0dec%7C0%7C0%7C638761820328498774%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Wwv3b3yMN5T2Zh7znZgbGQv8hbTb6x1sNQFwffssaUA%3D&reserved=0> >> rm -dr gem5bootcamp/2024/.git >> cd bootcamp24 && mv gem5bootcamp/2024/* . >> mv gem5bootcamp/2024/.* . >> cd bootcamp2024 >> git add * >> >> git add .* >> >> git commit >> git push >> >> really any way you need to copy the file tree of the official bootcamp >> repo, minus the parts of the git manifest directory that prevent you >> committing changes because you made some automatic test fail that gem5 put >> in there to make sure builds succeed. >> >> >> >> >> >> I’m not sure how to follow their directions to setup a new repository. >> >> >> >> I have setup a x86 VM using virtualbox and it is able to run the >> instructions for using the AMD GPU. If I have the 2024 directory with the >> gem5 code added using the –recursive command, is there a way of setting up >> the codespace from that? I am missing the step that guides the creation of >> the repository. >> >> >> >> Nick >> >> >> >> *From:* Matt Sinclair <mattdsinclair.wisc@gmail.com> >> *Sent:* Friday, February 28, 2025 10:13 PM >> *To:* The gem5 Users mailing list <gem5-users@gem5.org> >> *Cc:* Beser, Nicholas D. <Nick.Beser@jhuapl.edu>; Jason Lowe-Power < >> jason@lowepower.com> >> *Subject:* [EXT] Re: [gem5-users] Problem running gpu example with >> codespace >> >> >> >> *APL external email warning: *Verify sender mattdsinclair.wisc@gmail.com >> before clicking links or attachments >> >> >> >> Hi Nicholas, >> >> >> >> Sorry for my delayed response, but very cool to hear you are using >> these. Jason (CC'd) might need to help about codespace specific issues >> though. Before we get there though, there are some things to try. >> Ultimately, the problem is that it can't find the libamdhip library. I'll >> need some more information to help you triage, but the most obvious thing: >> what steps did you take before this one? Specifically, did you run the >> docker pull command like the bootcamp requires? My recollection is that >> this needs to be done first, before anything else. And in the command at >> the top of your email, you don't appear to be using the docker. I will >> also note you (I think) changed the example to use gfx902 instead of >> gfx900. From looking at my slides from the bootcamp, it does not appear I >> talked about gfx902 at all. So, just making sure that change is >> intentional? >> >> >> >> Jason, I'm having a hard time finding the right pointer to where in the >> bootcamp repo the GPU instructions are. Is it this: >> https://github.com/gem5bootcamp/gem5-bootcamp-env/blob/51590ae00b0e451c9b6a8854addbb94128ab4cac/materials/developing-gem5-models/11-gpu/README.md >> ? >> >> >> >> Thanks, >> >> Matt >> >> >> >> On Wed, Feb 26, 2025 at 8:34 AM Beser, Nicholas D. via gem5-users < >> gem5-users@gem5.org> wrote: >> >> root@codespaces-e992b2:/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# >> gem5-vega gem5/configs/example/apu_se.py -n 3 --gfx-version= >> >> gfx902 -c gem5-resources/src/gpu/square/bin/square >> >> gem5 Simulator System. https://www.gem5.org >> >> gem5 is copyrighted software; use the --copyright option for details. >> >> >> >> I have setup a github codespace for teaching my advanced computer >> architecture class. I tried to run the example square for the AMD GPU, and >> got an error (gem5-resources/src/gpu/square/bin/square: error while loading >> shared libraries: libamdhip64.so.4: cannot open shared object file: No such >> file or directory >> >> >> >> Is this due to the compilation of square, or is this a common problem >> with codespace? >> >> >> >> Nick >> >> >> >> gem5 version 24.0.0.0 >> >> gem5 compiled Jul 25 2024 20:34:41 >> >> gem5 started Feb 26 2025 14:08:28 >> >> gem5 executing on codespaces-e992b2, pid 15064 >> >> command line: gem5-vega gem5/configs/example/apu_se.py -n 3 >> --gfx-version=gfx902 -c gem5-resources/src/gpu/square/bin/square >> >> >> >> Num SQC = 1 Num scalar caches = 1 Num CU = 4 >> >> Global frequency set at 1000000000000 ticks per second >> >> warn: system.ruby.network adopting orphan SimObject param 'ext_links' >> >> warn: system.ruby.network adopting orphan SimObject param 'int_links' >> >> src/mem/dram_interface.cc:690: warn: DRAM device capacity (8192 Mbytes) >> does not match the address range assigned (512 Mbytes) >> >> src/base/statistics.hh:279: warn: One of the stats is a legacy stat. >> Legacy stat is a stat that does not belong to any statistics::Group. Legacy >> stat is deprecated. >> >> src/base/statistics.hh:279: warn: One of the stats is a legacy stat. >> Legacy stat is a stat that does not belong to any statistics::Group. Legacy >> stat is deprecated. >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> src/base/statistics.hh:279: warn: One of the stats is a legacy stat. >> Legacy stat is a stat that does not belong to any statistics::Group. Legacy >> stat is deprecated. >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> Forcing maxCoalescedReqs to 32 (TLB assoc.) >> >> system.remote_gdb: Listening for connections on port 7000 >> >> src/sim/simulate.cc:199: info: Entering event queue @ 0. Starting >> simulation... >> >> src/mem/ruby/system/Sequencer.cc:680: warn: Replacement policy updates >> recently became the responsibility of SLICC state machines. Make sure to >> setMRU() near callbacks in .sm files! >> >> src/sim/mem_state.cc:448: info: Increasing stack size by one page. >> >> gem5-resources/src/gpu/square/bin/square: error while loading shared >> libraries: libamdhip64.so.4: cannot open shared object file: No such file >> or directory >> >> breaking loop due to: exiting with last active thread context. >> >> Ticks: 234627000 >> >> Exiting because exiting with last active thread context >> >> root@codespaces-e992b2 >> :/workspaces/en525-712-81-sp25-amd-gpu-using-gem5-gem5bootcamp2024# >> >> _______________________________________________ >> gem5-users mailing list -- gem5-users@gem5.org >> To unsubscribe send an email to gem5-users-leave@gem5.org >> >>