Re: [snipersim] parsec-2.1 Benchmark Suite

165 views
Skip to first unread message

Trevor E. Carlson

unread,
Apr 11, 2024, 9:04:15 PM4/11/24
to snip...@googlegroups.com, prajakt...@gmail.com
Prajakta,

     Unfortunately, we can’t help unless we know the commands use to run the scripts. It looks like you might have typed the commands in incorrectly.

      Can you walk us through the exact procedure you are using so we can verify it from our side?

Thanks,
Trevor

On Apr 12, 2024, at 08:58, Prajakta <prajakt...@gmail.com> wrote:

Hello Trevor,

Thanks for the docker image. Splash2 benchmarks are running with this docker, but I am still not able to run parsec benchmarks with it. 

I am getting below error:
'raise ValueError("Invalid benchmark %s" % program)'

Is anybody else getting this error? How to resolve this?

Thanks,
Prajakta
IIT Bombay

On Saturday, February 10, 2024 at 3:04:41 PM UTC+5:30 Trevor E. Carlson wrote:
Upasna,

For SPEC CPU2006/2017, we are not allowed to release the sources due to licensing issues. So, you could either purchase a license, or try an alternative, like Pinballs [1,2], which you can download and run without compilation.

In addition, for your previous Parsec questions, why not use the docker image with the pre-compiled binaries [3] instead of compiling them yourself?

I’ve pushed some updates recently [4] which should allow this to work again (downloading from working locations). If others want to try this out transparently, they’ll need the updated version to get the sources from the new location. One issue with Parsec 2.1 is that this benchmark suite requires specific library versions, which has been tested in our Docker image [1], but not on modern systems. This might not work on modern Linux systems.

Moving to Parsec 3.0 could be warranted if you want to use newer OSes (although, I haven’t tested this myself). While the Princeton server is now down, the Wayback Machine still has a copy [5]. There also seems to be a copy on GitHub [6], but it doesn’t look like they have all of the input datasets there.

Trevor

On Feb 10, 2024, at 2:41 PM, 'Upasna _' via Sniper simulator <snip...@googlegroups.com> wrote:

Thank you Trevor. This issue was resolved. Currently I am stuck in making cpu2006 work on Sniper. It shows:
SPEC CPU2006 sources not found. To build the CPU2006 benchmarks for Sniper, first run
benchmarks$ ./install_cpu2006 PATH_TO_CPU2006_DVD

Screenshot from 2024-02-10 11-58-10.png

As shown in the error, Makefile:19: recipe for target 'CPU2006/999.specrand/src/specrand.c' failed, there is currently no directory named 999.specrand in CPU2006.

Do we need to purchase SPEC CPU 2006? Can we pass iso file also to PATH_TO_CPU2006_DVD? Can we make SPEC CPU 2017 also working on Sniper, if we've its iso file? Would be grateful for your help.

Best Regards
Upasna
On Saturday, February 10, 2024 at 10:28:44 AM UTC+5:30 Trevor E. Carlson wrote:
Upasna,

     I’ve made a copy of the files, and patches, as a mirror on the Sniper website. All files for Parsec 2.1 should now mirrored to a directory similar to this one [1]. Can you try to download the files and let me know if they work for you?

Trevor


On Feb 10, 2024, at 03:22, 'Upasna _' via Sniper simulator <snip...@googlegroups.com> wrote:


Dear All

All the parsec download links provided in the benchmarks/parsec/Makefile are broken, such as http://parsec.cs.princeton.edu/download/2.1/parsec-2.1-core.tar.gz
If anyone of you has installed parsec benchmark in your system, can you please share the same or these tar.gz files. I would be so grateful.

Best Regards
Upasna
PhD Scholar
IIT Ropar

CONFIDENTIALITY NOTICE: The contents of this email message and any attachments are intended solely for the addressee(s) and may contain confidential and/or privileged information and may be legally protected from disclosure. If you are not the intended recipient of this message or their agent, or if this message has been addressed to you in error, please immediately alert the sender by reply email and then delete this message and any attachments. If you are not the intended recipient, you are hereby notified that any use, dissemination, copying, or storage of this message or its attachments is strictly prohibited.

--
--
--
You received this message because you are subscribed to the Google
Groups "Sniper simulator" group.
To post to this group, send email to snip...@googlegroups.com
To unsubscribe from this group, send email to
snipersim+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/snipersim?hl=en

---
You received this message because you are subscribed to the Google Groups "Sniper simulator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to snipersim+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/snipersim/076b1b93-b40c-4a19-9ece-449d65b24247n%40googlegroups.com.

CONFIDENTIALITY NOTICE: The contents of this email message and any attachments are intended solely for the addressee(s) and may contain confidential and/or privileged information and may be legally protected from disclosure. If you are not the intended recipient of this message or their agent, or if this message has been addressed to you in error, please immediately alert the sender by reply email and then delete this message and any attachments. If you are not the intended recipient, you are hereby notified that any use, dissemination, copying, or storage of this message or its attachments is strictly prohibited.

--
--
--
You received this message because you are subscribed to the Google
Groups "Sniper simulator" group.
To post to this group, send email to snip...@googlegroups.com
To unsubscribe from this group, send email to
snipersim+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/snipersim?hl=en

---
You received this message because you are subscribed to the Google Groups "Sniper simulator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to snipersim+...@googlegroups.com.

Screenshot from 2024-02-10 11-58-10.png

--
--
--
You received this message because you are subscribed to the Google
Groups "Sniper simulator" group.
To post to this group, send email to snip...@googlegroups.com
To unsubscribe from this group, send email to
snipersim+...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/snipersim?hl=en

---
You received this message because you are subscribed to the Google Groups "Sniper simulator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to snipersim+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/snipersim/f2b41d9e-1c34-4069-b423-e4c0b35a5381n%40googlegroups.com.

Prajakta

unread,
Apr 12, 2024, 6:16:18 AM4/12/24
to Sniper simulator
Sure, here are the steps I followed
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1. docker run -it --rm snipersim/snipersim
2. cd benchmarks/
3. ./run-sniper --benchmarks parsec-canneal-test-4 -n 4 
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
      After 3rd command, I am getting error: 
                Invalid benchmark canneal
                Traceback (most recent call last):
                    File "./run-sniper", line 289, in <module>
                        program = __import__(package).Program(programname, int(app_nthreads), inputsize, ['force_nthreads'])
                    File "/root/benchmarks/parsec/__init__.py", line 41, in __init__

                        raise ValueError("Invalid benchmark %s" % program)
ValueError: Invalid benchmark canneal

Thanks, 
Prajakta
IIT Bombay
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Reply all
Reply to author
Forward
0 new messages