[gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

2021-01-12 Thread Poremba, Matthew via gem5-dev
[AMD Public Use]

The defines causing an error seem to be set in the HSA header file itself but 
only for x86 type hosts. I'm hesitant to change this since it is pulled 
directly, unmodified, from the AMD HSA runtime repo including license, etc.

However, we might be able to do some scons magic to define one of these by 
asking the host OS what the endianness is. For example use 'sys.byteorder' in 
python to determine endianness and then define one or the other.


-Matt

From: Bobby Bruce 
Sent: Tuesday, January 12, 2021 1:10 PM
To: gem5 Developer List 
Cc: Dutu, Alexandru ; Poremba, Matthew 
; Kyle Roarty ; Nathanael Premillieu 

Subject: Re: [gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

[CAUTION: External Email]
I'm guessing you're correct and this is because you're testing on an ARM 
machine. GCN3_X86 should compile fine, though we don't test on ARM hosts (we'd 
like to change this soon).

I've added a Jira ticket covering this bug: 
https://gem5.atlassian.net/browse/GEM5-881<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgem5.atlassian.net%2Fbrowse%2FGEM5-881=04%7C01%7CMatthew.Poremba%40amd.com%7Cfa37036afc77400d6c5e08d8b73e66d8%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637460825936359718%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000=w1ugJOlay4h9FWQ7DTnrafp0JAAhRK9nGUcSU7Z%2FDxc%3D=0>.

--
Dr. Bobby R. Bruce
Room 2235,
Kemper Hall, UC Davis
Davis,
CA, 95616

web: 
https://www.bobbybruce.net<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.bobbybruce.net%2F=04%7C01%7CMatthew.Poremba%40amd.com%7Cfa37036afc77400d6c5e08d8b73e66d8%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637460825936369717%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000=0gRK0blvE7WURELXXljJDKwsFdCoxqSGaAJzs7mgrDs%3D=0>

Please consider making a submission to GI@ICSE '21: 
http://geneticimprovementofsoftware.com/gi2021icse.html<https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fgeneticimprovementofsoftware.com%2Fgi2021icse.html=04%7C01%7CMatthew.Poremba%40amd.com%7Cfa37036afc77400d6c5e08d8b73e66d8%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637460825936369717%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000=Sb7lYmr%2B4%2BioZMqLVEB7wVDlPLswUKysOc4OB08ZsI0%3D=0>


On Tue, Jan 12, 2021 at 12:44 AM Nathanael Premillieu via gem5-dev 
mailto:gem5-dev@gem5.org>> wrote:
Hi,

Thanks Jason and Matt for your answers.
I'm not really interested in GCN3, it's just that it is tested by default when 
running all the tests.
And as I'm working on a change that is not specific to a particular 
architecture, I wanted to test it on everything.
For the moment I have relaunched the tests without the "GCN3_X86" isa.
But I'll also test on a x86 machine with everything.

Thanks,
Nathanael

From: Jason Lowe-Power via gem5-dev 
[mailto:gem5-dev@gem5.org<mailto:gem5-dev@gem5.org>]
Sent: Tuesday, January 12, 2021 12:17 AM
To: gem5 Developer List mailto:gem5-dev@gem5.org>>
Cc: Dutu, Alexandru mailto:alexandru.d...@amd.com>>; 
Poremba, Matthew mailto:matthew.pore...@amd.com>>; 
Kyle Roarty mailto:kroa...@wisc.edu>>; Jason Lowe-Power 
mailto:ja...@lowepower.com>>
Subject: [gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

Hi all,

It doesn't surprise me at all that GCN3 doesn't build on Arm. Nathanael, you 
should be able to ignore those errors. We don't test that combination of 
host-target and I have not heard of anyone trying to use that combination of 
host-target.

If you're trying to use the GPU model, I'd suggest moving to an x86 host. 
Unfortunately, right now the GPU model is intricately tied to the host OS. 
Hence why the docker is all but required to get the GPU model working.

Cheers,
Jason

On Mon, Jan 11, 2021 at 10:59 AM Matt Sinclair via gem5-dev 
mailto:gem5-dev@gem5.org>> wrote:
Hi Nathanael,

I have personally never tested GCN3 on an ARM machine, so I can't say 
definitively if it will work there or not.  But if you are not, I strongly 
recommend testing anything you need to test with GCN3 inside the docker we 
created:

- 
https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/util/dockerfiles/gcn-gpu/README.md<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgem5.googlesource.com%2Fpublic%2Fgem5%2F%2B%2Frefs%2Fheads%2Fdevelop%2Futil%2Fdockerfiles%2Fgcn-gpu%2FREADME.md=04%7C01%7CMatthew.Poremba%40amd.com%7Cfa37036afc77400d6c5e08d8b73e66d8%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637460825936379706%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000=h%2BK77k%2FV626%2BoCuxRwWby4IoLZ41rChF%2B%2BLbj%2FqCIDk%3D=0>
- 
http://www.gem5.org/2020/05/27/modern-gpu-applications.html<https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.gem5.or

[gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

2021-01-12 Thread Bobby Bruce via gem5-dev
I'm guessing you're correct and this is because you're testing on an ARM
machine. GCN3_X86 should compile fine, though we don't test on ARM hosts
(we'd like to change this soon).

I've added a Jira ticket covering this bug:
https://gem5.atlassian.net/browse/GEM5-881.

--
Dr. Bobby R. Bruce
Room 2235,
Kemper Hall, UC Davis
Davis,
CA, 95616

web: https://www.bobbybruce.net

Please consider making a submission to GI@ICSE '21:
http://geneticimprovementofsoftware.com/gi2021icse.html


On Tue, Jan 12, 2021 at 12:44 AM Nathanael Premillieu via gem5-dev <
gem5-dev@gem5.org> wrote:

> Hi,
>
>
>
> Thanks Jason and Matt for your answers.
>
> I’m not really interested in GCN3, it’s just that it is tested by default
> when running all the tests.
>
> And as I’m working on a change that is not specific to a particular
> architecture, I wanted to test it on everything.
>
> For the moment I have relaunched the tests without the “GCN3_X86” isa.
>
> But I’ll also test on a x86 machine with everything.
>
>
>
> Thanks,
>
> Nathanael
>
>
>
> *From:* Jason Lowe-Power via gem5-dev [mailto:gem5-dev@gem5.org]
> *Sent:* Tuesday, January 12, 2021 12:17 AM
> *To:* gem5 Developer List 
> *Cc:* Dutu, Alexandru ; Poremba, Matthew <
> matthew.pore...@amd.com>; Kyle Roarty ; Jason
> Lowe-Power 
> *Subject:* [gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine
>
>
>
> Hi all,
>
>
>
> It doesn't surprise me at all that GCN3 doesn't build on Arm. Nathanael,
> you should be able to ignore those errors. We don't test that combination
> of host-target and I have not heard of anyone trying to use that
> combination of host-target.
>
>
>
> If you're trying to use the GPU model, I'd suggest moving to an x86 host.
> Unfortunately, right now the GPU model is intricately tied to the host OS.
> Hence why the docker is all but required to get the GPU model working.
>
>
>
> Cheers,
>
> Jason
>
>
>
> On Mon, Jan 11, 2021 at 10:59 AM Matt Sinclair via gem5-dev <
> gem5-dev@gem5.org> wrote:
>
> Hi Nathanael,
>
>
>
> I have personally never tested GCN3 on an ARM machine, so I can't say
> definitively if it will work there or not.  But if you are not, I strongly
> recommend testing anything you need to test with GCN3 inside the docker we
> created:
>
>
>
> -
> https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/util/dockerfiles/gcn-gpu/README.md
>
> - http://www.gem5.org/2020/05/27/modern-gpu-applications.html
>
>
>
> Thanks,
>
> Matt
>
>
>
> On Mon, Jan 11, 2021 at 8:42 AM Nathanael Premillieu via gem5-dev <
> gem5-dev@gem5.org> wrote:
>
> Hi,
>
>
>
> I’m currently working on some changes in gem5 and before submitting them,
> I’m using the testing framework as required.
>
> However, even without my changes, I get failures when building GCN3_X86
> (See the trace at the end).
>
> I guess it is linked to the fact that I’m working on an Arm machine. Do I
> need to set up something to make it work?
>
>
>
> Thanks,
>
> Nathanael Premillieu
>
>
>
> The trace:
>
> /scratch/npremill/gem5_public.git/build/GCN3_X86/gem5.opt
>
> You may want to run with only a single ISA(--isa=), use --skip-build, or
> use 'rerun'.
>
> MOESI_AMD_Base-dir.sm:220: Warning: Non-void return ignored, return type
> is 'bool'
>
> MOESI_AMD_Base-dir.sm:1052: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1056: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1060: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1064: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1068: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1072: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1076: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:586: Warning: Unused action: l_queueMemWBReq, Write
> WB data to memory
>
> MOESI_AMD_Base-dir.sm:953: Warning: Unused action:
> mwc_markSinkWriteCancel, Mark to sink impending VicDirty
>
> MOESI_AMD_Base-dir.sm:1051: Warning: Unused action: dl_deallocateL3,
> deallocate the L3 block
>
> MOESI_AMD_Base-dir.sm:1087: Warning: Unused action:
> yy_recycleResponseQueue, recycle response queue
>
> MOESI_AMD_Base-dma.sm:187: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dma.sm:191: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-CorePair.sm:325: Warning: Non-void 

[gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

2021-01-12 Thread Nathanael Premillieu via gem5-dev
Hi,

Thanks Jason and Matt for your answers.
I’m not really interested in GCN3, it’s just that it is tested by default when 
running all the tests.
And as I’m working on a change that is not specific to a particular 
architecture, I wanted to test it on everything.
For the moment I have relaunched the tests without the “GCN3_X86” isa.
But I’ll also test on a x86 machine with everything.

Thanks,
Nathanael

From: Jason Lowe-Power via gem5-dev [mailto:gem5-dev@gem5.org]
Sent: Tuesday, January 12, 2021 12:17 AM
To: gem5 Developer List 
Cc: Dutu, Alexandru ; Poremba, Matthew 
; Kyle Roarty ; Jason Lowe-Power 

Subject: [gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

Hi all,

It doesn't surprise me at all that GCN3 doesn't build on Arm. Nathanael, you 
should be able to ignore those errors. We don't test that combination of 
host-target and I have not heard of anyone trying to use that combination of 
host-target.

If you're trying to use the GPU model, I'd suggest moving to an x86 host. 
Unfortunately, right now the GPU model is intricately tied to the host OS. 
Hence why the docker is all but required to get the GPU model working.

Cheers,
Jason

On Mon, Jan 11, 2021 at 10:59 AM Matt Sinclair via gem5-dev 
mailto:gem5-dev@gem5.org>> wrote:
Hi Nathanael,

I have personally never tested GCN3 on an ARM machine, so I can't say 
definitively if it will work there or not.  But if you are not, I strongly 
recommend testing anything you need to test with GCN3 inside the docker we 
created:

- 
https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/util/dockerfiles/gcn-gpu/README.md
- http://www.gem5.org/2020/05/27/modern-gpu-applications.html

Thanks,
Matt

On Mon, Jan 11, 2021 at 8:42 AM Nathanael Premillieu via gem5-dev 
mailto:gem5-dev@gem5.org>> wrote:
Hi,

I’m currently working on some changes in gem5 and before submitting them, I’m 
using the testing framework as required.
However, even without my changes, I get failures when building GCN3_X86 (See 
the trace at the end).
I guess it is linked to the fact that I’m working on an Arm machine. Do I need 
to set up something to make it work?

Thanks,
Nathanael Premillieu

The trace:
/scratch/npremill/gem5_public.git/build/GCN3_X86/gem5.opt
You may want to run with only a single ISA(--isa=), use --skip-build, or use 
'rerun'.
MOESI_AMD_Base-dir.sm:220: Warning: Non-void return ignored, return type is 
'bool'
MOESI_AMD_Base-dir.sm:1052: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:1056: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:1060: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:1064: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:1068: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:1072: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:1076: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dir.sm:586: Warning: Unused action: l_queueMemWBReq, Write WB 
data to memory
MOESI_AMD_Base-dir.sm:953: Warning: Unused action: mwc_markSinkWriteCancel, 
Mark to sink impending VicDirty
MOESI_AMD_Base-dir.sm:1051: Warning: Unused action: dl_deallocateL3, deallocate 
the L3 block
MOESI_AMD_Base-dir.sm:1087: Warning: Unused action: yy_recycleResponseQueue, 
recycle response queue
MOESI_AMD_Base-dma.sm:187: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-dma.sm:191: Warning: Non-void return ignored, return type is 
'Tick'
MOESI_AMD_Base-CorePair.sm:325: Warning: Non-void return ignored, return type 
is 'bool'
MOESI_AMD_Base-CorePair.sm:802: Warning: Non-void return ignored, return type 
is 'Tick'
MOESI_AMD_Base-CorePair.sm:806: Warning: Non-void return ignored, return type 
is 'Tick'
MOESI_AMD_Base-CorePair.sm:810: Warning: Non-void return ignored, return type 
is 'Tick'
MOESI_AMD_Base-CorePair.sm:814: Warning: Non-void return ignored, return type 
is 'Tick'
GPU_VIPER-TCP.sm:166: Warning: Non-void return ignored, return type is 'bool'
GPU_VIPER-TCP.sm:451: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-TCP.sm:455: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-TCP.sm:385: Warning: Unused action: norl_issueRdBlkOrloadDone, local 
load done
GPU_VIPER-SQC.sm:143: Warning: Non-void return ignored, return type is 'bool'
GPU_VIPER-SQC.sm:275: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-SQC.sm:279: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-TCC.sm:168: Warning: Non-void return ignored, return type is 'bool'
GPU_VIPER-TCC.sm:551: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-TCC.sm:555: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-TCC.sm:559: Warning: Non-void return ignored, return type is 'Tick'
GPU_VIPER-TCC.sm:583: Warning: Non-void return ignored, return type is 'Tick'
MOESI_AMD_Base-L3cache.sm:

[gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

2021-01-11 Thread Jason Lowe-Power via gem5-dev
Hi all,

It doesn't surprise me at all that GCN3 doesn't build on Arm. Nathanael,
you should be able to ignore those errors. We don't test that combination
of host-target and I have not heard of anyone trying to use that
combination of host-target.

If you're trying to use the GPU model, I'd suggest moving to an x86 host.
Unfortunately, right now the GPU model is intricately tied to the host OS.
Hence why the docker is all but required to get the GPU model working.

Cheers,
Jason

On Mon, Jan 11, 2021 at 10:59 AM Matt Sinclair via gem5-dev <
gem5-dev@gem5.org> wrote:

> Hi Nathanael,
>
> I have personally never tested GCN3 on an ARM machine, so I can't say
> definitively if it will work there or not.  But if you are not, I strongly
> recommend testing anything you need to test with GCN3 inside the docker we
> created:
>
> -
> https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/util/dockerfiles/gcn-gpu/README.md
> - http://www.gem5.org/2020/05/27/modern-gpu-applications.html
>
> Thanks,
> Matt
>
> On Mon, Jan 11, 2021 at 8:42 AM Nathanael Premillieu via gem5-dev <
> gem5-dev@gem5.org> wrote:
>
>> Hi,
>>
>>
>>
>> I’m currently working on some changes in gem5 and before submitting them,
>> I’m using the testing framework as required.
>>
>> However, even without my changes, I get failures when building GCN3_X86
>> (See the trace at the end).
>>
>> I guess it is linked to the fact that I’m working on an Arm machine. Do I
>> need to set up something to make it work?
>>
>>
>>
>> Thanks,
>>
>> Nathanael Premillieu
>>
>>
>>
>> The trace:
>>
>> /scratch/npremill/gem5_public.git/build/GCN3_X86/gem5.opt
>>
>> You may want to run with only a single ISA(--isa=), use --skip-build, or
>> use 'rerun'.
>>
>> MOESI_AMD_Base-dir.sm:220: Warning: Non-void return ignored, return type
>> is 'bool'
>>
>> MOESI_AMD_Base-dir.sm:1052: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:1056: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:1060: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:1064: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:1068: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:1072: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:1076: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dir.sm:586: Warning: Unused action: l_queueMemWBReq, Write
>> WB data to memory
>>
>> MOESI_AMD_Base-dir.sm:953: Warning: Unused action:
>> mwc_markSinkWriteCancel, Mark to sink impending VicDirty
>>
>> MOESI_AMD_Base-dir.sm:1051: Warning: Unused action: dl_deallocateL3,
>> deallocate the L3 block
>>
>> MOESI_AMD_Base-dir.sm:1087: Warning: Unused action:
>> yy_recycleResponseQueue, recycle response queue
>>
>> MOESI_AMD_Base-dma.sm:187: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-dma.sm:191: Warning: Non-void return ignored, return type
>> is 'Tick'
>>
>> MOESI_AMD_Base-CorePair.sm:325: Warning: Non-void return ignored, return
>> type is 'bool'
>>
>> MOESI_AMD_Base-CorePair.sm:802: Warning: Non-void return ignored, return
>> type is 'Tick'
>>
>> MOESI_AMD_Base-CorePair.sm:806: Warning: Non-void return ignored, return
>> type is 'Tick'
>>
>> MOESI_AMD_Base-CorePair.sm:810: Warning: Non-void return ignored, return
>> type is 'Tick'
>>
>> MOESI_AMD_Base-CorePair.sm:814: Warning: Non-void return ignored, return
>> type is 'Tick'
>>
>> GPU_VIPER-TCP.sm:166: Warning: Non-void return ignored, return type is
>> 'bool'
>>
>> GPU_VIPER-TCP.sm:451: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-TCP.sm:455: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-TCP.sm:385: Warning: Unused action: norl_issueRdBlkOrloadDone,
>> local load done
>>
>> GPU_VIPER-SQC.sm:143: Warning: Non-void return ignored, return type is
>> 'bool'
>>
>> GPU_VIPER-SQC.sm:275: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-SQC.sm:279: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-TCC.sm:168: Warning: Non-void return ignored, return type is
>> 'bool'
>>
>> GPU_VIPER-TCC.sm:551: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-TCC.sm:555: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-TCC.sm:559: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> GPU_VIPER-TCC.sm:583: Warning: Non-void return ignored, return type is
>> 'Tick'
>>
>> MOESI_AMD_Base-L3cache.sm:196: Warning: Non-void return ignored, return
>> type is 'bool'
>>
>> MOESI_AMD_Base-L3cache.sm:611: Warning: Non-void return ignored, return
>> type is 'Tick'
>>
>> MOESI_AMD_Base-L3cache.sm:615: Warning: Non-void return ignored, return
>> type is 'Tick'
>>
>> MOESI_AMD_Base-L3cache.sm:619: Warning: Non-void return ignored, return
>> type is 

[gem5-dev] Re: Testing fail on GCN3_X86 on Arm machine

2021-01-11 Thread Matt Sinclair via gem5-dev
Hi Nathanael,

I have personally never tested GCN3 on an ARM machine, so I can't say
definitively if it will work there or not.  But if you are not, I strongly
recommend testing anything you need to test with GCN3 inside the docker we
created:

-
https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/util/dockerfiles/gcn-gpu/README.md
- http://www.gem5.org/2020/05/27/modern-gpu-applications.html

Thanks,
Matt

On Mon, Jan 11, 2021 at 8:42 AM Nathanael Premillieu via gem5-dev <
gem5-dev@gem5.org> wrote:

> Hi,
>
>
>
> I’m currently working on some changes in gem5 and before submitting them,
> I’m using the testing framework as required.
>
> However, even without my changes, I get failures when building GCN3_X86
> (See the trace at the end).
>
> I guess it is linked to the fact that I’m working on an Arm machine. Do I
> need to set up something to make it work?
>
>
>
> Thanks,
>
> Nathanael Premillieu
>
>
>
> The trace:
>
> /scratch/npremill/gem5_public.git/build/GCN3_X86/gem5.opt
>
> You may want to run with only a single ISA(--isa=), use --skip-build, or
> use 'rerun'.
>
> MOESI_AMD_Base-dir.sm:220: Warning: Non-void return ignored, return type
> is 'bool'
>
> MOESI_AMD_Base-dir.sm:1052: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1056: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1060: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1064: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1068: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1072: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:1076: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dir.sm:586: Warning: Unused action: l_queueMemWBReq, Write
> WB data to memory
>
> MOESI_AMD_Base-dir.sm:953: Warning: Unused action:
> mwc_markSinkWriteCancel, Mark to sink impending VicDirty
>
> MOESI_AMD_Base-dir.sm:1051: Warning: Unused action: dl_deallocateL3,
> deallocate the L3 block
>
> MOESI_AMD_Base-dir.sm:1087: Warning: Unused action:
> yy_recycleResponseQueue, recycle response queue
>
> MOESI_AMD_Base-dma.sm:187: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-dma.sm:191: Warning: Non-void return ignored, return type
> is 'Tick'
>
> MOESI_AMD_Base-CorePair.sm:325: Warning: Non-void return ignored, return
> type is 'bool'
>
> MOESI_AMD_Base-CorePair.sm:802: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-CorePair.sm:806: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-CorePair.sm:810: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-CorePair.sm:814: Warning: Non-void return ignored, return
> type is 'Tick'
>
> GPU_VIPER-TCP.sm:166: Warning: Non-void return ignored, return type is
> 'bool'
>
> GPU_VIPER-TCP.sm:451: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-TCP.sm:455: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-TCP.sm:385: Warning: Unused action: norl_issueRdBlkOrloadDone,
> local load done
>
> GPU_VIPER-SQC.sm:143: Warning: Non-void return ignored, return type is
> 'bool'
>
> GPU_VIPER-SQC.sm:275: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-SQC.sm:279: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-TCC.sm:168: Warning: Non-void return ignored, return type is
> 'bool'
>
> GPU_VIPER-TCC.sm:551: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-TCC.sm:555: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-TCC.sm:559: Warning: Non-void return ignored, return type is
> 'Tick'
>
> GPU_VIPER-TCC.sm:583: Warning: Non-void return ignored, return type is
> 'Tick'
>
> MOESI_AMD_Base-L3cache.sm:196: Warning: Non-void return ignored, return
> type is 'bool'
>
> MOESI_AMD_Base-L3cache.sm:611: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-L3cache.sm:615: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-L3cache.sm:619: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-L3cache.sm:623: Warning: Non-void return ignored, return
> type is 'Tick'
>
> MOESI_AMD_Base-L3cache.sm:561: Warning: Unused action:
> rd_copyDataFromRequest, write data to L3
>
> In file included from build/GCN3_X86/dev/hsa/hsa_packet_processor.hh:44,
>
>  from build/GCN3_X86/dev/hsa/hsa_device.hh:43,
>
>  from
> build/GCN3_X86/gpu-compute/gpu_command_processor.hh:50,
>
>  from build/GCN3_X86/gpu-compute/dispatcher.cc:41:
>
> build/GCN3_X86/dev/hsa/hsa.h:93:2: error: #error "BIGENDIAN_CPU or
> LITTLEENDIAN_CPU must be defined"
>
>93 | #error "BIGENDIAN_CPU or LITTLEENDIAN_CPU must be defined"
>
>   |  ^
>
> In file included from