Core26 released to full FAH ?

[muziqaz]

Yeah, I said the same last year around this time. That did not go very well

[arisu]

AMD has HIP out. It is now down to us to implement it. Initially core25 was supposed to be hip. Due to issues with other stuff, core26 was born, which fixed some issues, but had other issues, again HIP integration to backseat. Core27 then was thought to have HIP. But it was pushed without hip to make sure that all the issues with core26 were fixed on 27. It seems we have majority of things fixed now, so I hope core28 or revision of core27 will have hip, as excuses and reasoning is running out now

AMD will listen if we, as in FAH, have something to show and tell them that things can be improved, etc. I cannot tell them anything right now because we have nothing to work on yet.

Is it as simple as rebuilding one of the cores with OpenMM 8.2.0 with HIP enabled and doing tests? Or are there more blockers?

[muziqaz]

Not that simple. If it was, this sh*t would have been built early last year
Since I'm trying not to pressure the Devs, I keep the question to once a month or two.

[Joe_H]

At the least a core has to be built, go through validation testing to show whether or not the HIP implementation gives the same results as OpenCL or CUDA. Then they have to figure out the packaging with the necessary libraries, etc. Then test to see if those install properly over at least a couple versions of Windows and several Linux distros.

[calxalot]

The next client is also needed, because the arguments for the core are different for hip.

[arisu]

The next client is also needed, because the arguments for the core are different for hip.

I think that code is already written for 8.4.10 and can be released at any time.

[calxalot]

The client is not even buildable right now because of changes in cbang that have not been applied to the client.

Nobody has tested the code you are talking about. But yes, it seems to have been committed.

[arisu]

It's buildable, you just have to checkout the 8.4.10 tag. I've tested the code with a dummy core on 8.4.10 and it passes the correct arguments when enabled. Of course the dummy core does nothing but print out the arguments it was given, it can't actually fold.

But at least the client part seems to be working.

[calxalot]

I’m not seeing any 8.4.10 tags in the repo.

[arisu]

Huh, I forget how I did it then. Maybe I just used 8.4.9 for cbang when building the 8.4.10 client? Either way I was able to build 8.4.10 a few months ago, or whenever the hip commit was made (it was shortly after that).

[boboviz]

2 days ago OpenMM released the beta of OpenMM 8.3

[muziqaz]

2 days ago OpenMM released the beta of OpenMM 8.3

Don't tell us that
Before you know it we will run out of digits for our core designation

[toTOW]

2 days ago OpenMM released the beta of OpenMM 8.3

FAH doesn't integrate beta version of OpenMM, but it might be the next core 28 once it is released as a stable version ...

[Frogging101]

I get an error with Core26 on Arch Linux using rusticl on RX 6900 XT. "Bad platformId size".

Code: Select all

19:58:50:I1:WU633:*********************** Log Started 2025-05-09T19:58:49Z ***********************
19:58:50:I1:WU633:*************************** Core26 Folding@home Core ***************************
19:58:50:I1:WU633: Core: Core26
19:58:50:I1:WU633: Type: 0x26
19:58:50:I1:WU633: Version: 8.2.0
19:58:50:I1:WU633: Author: Joseph Coffland <joseph@cauldrondevelopment.com>
19:58:50:I1:WU633: Copyright: 2022 foldingathome.org
19:58:50:I1:WU633: Homepage: https://foldingathome.org/
19:58:50:I1:WU633: Date: Jan 7 2025
19:58:50:I1:WU633: Time: 00:35:47
19:58:50:I1:WU633: Revision: 4f149b599caa4725076ef2de3b47c8d7ce725787
19:58:50:I1:WU633: Branch: HEAD
19:58:50:I1:WU633: Compiler: GNU 7.5.0
19:58:50:I1:WU633: Options: -faligned-new -std=c++14 -fsigned-char -ffunction-sections
19:58:50:I1:WU633: -fdata-sections -O3 -funroll-loops -fno-pie
19:58:50:I1:WU633: -DOPENMM_VERSION="\"8.2.0\""
19:58:50:I1:WU633: Platform: linux 6.8.0-1017-azure
19:58:50:I1:WU633: Bits: 64
19:58:50:I1:WU633: Mode: Release
19:58:50:I1:WU633:Maintainers: John Chodera <john.chodera@choderalab.org> and Peter Eastman
19:58:50:I1:WU633: <peastman@stanford.edu>
19:58:50:I1:WU633: Args: -dir 34dgCtdSDDqxfmKFSAADNG-wuftvwCoirDRQSPKNepk -suffix 01
19:58:50:I1:WU633: -version 8.4.9 -lifeline 931 -gpu-uuid
19:58:50:I1:WU633: 00000000-0c00-0000-0000-000000000000 -gpu-platform opencl
19:58:50:I1:WU633: -gpu-vendor amd -opencl-platform 0 -opencl-device 0 -gpu 0
19:58:50:I1:WU633:************************************ libFAH ************************************
19:58:50:I1:WU633: Date: Jan 7 2025
19:58:50:I1:WU633: Time: 00:29:24
19:58:50:I1:WU633: Revision: c7d2824a47eb025fa8cda8968c7a5e971585d90c
19:58:50:I1:WU633: Branch: HEAD
19:58:50:I1:WU633: Compiler: GNU 7.5.0
19:58:50:I1:WU633: Options: -faligned-new -std=c++11 -fsigned-char -ffunction-sections
19:58:50:I1:WU633: -fdata-sections -O3 -funroll-loops -fno-pie
19:58:50:I1:WU633: Platform: linux 6.8.0-1017-azure
19:58:50:I1:WU633: Bits: 64
19:58:50:I1:WU633: Mode: Release
19:58:50:I1:WU633:************************************ CBang *************************************
19:58:50:I1:WU633: Version: 1.7.2
19:58:50:I1:WU633: Author: Joseph Coffland <joseph@cauldrondevelopment.com>
19:58:50:I1:WU633: Org: Cauldron Development LLC
19:58:50:I1:WU633: Copyright: Cauldron Development LLC, 2003-2024
19:58:50:I1:WU633: Homepage: https://cauldrondevelopment.com/
19:58:50:I1:WU633: License: LGPL-2.1-or-later
19:58:50:I1:WU633: Date: Jan 7 2025
19:58:50:I1:WU633: Time: 00:28:59
19:58:50:I1:WU633: Revision: f1cd4c791e8c40a35dcfeab3ab85d910949cc0cb
19:58:50:I1:WU633: Branch: HEAD
19:58:50:I1:WU633: Compiler: GNU 7.5.0
19:58:50:I1:WU633: Options: -faligned-new -std=c++11 -fsigned-char -ffunction-sections
19:58:50:I1:WU633: -fdata-sections -O3 -funroll-loops -fno-pie -fPIC
19:58:50:I1:WU633: Platform: linux 6.8.0-1017-azure
19:58:50:I1:WU633: Bits: 64
19:58:50:I1:WU633: Mode: Release
19:58:50:I1:WU633:************************************ System ************************************
19:58:50:I1:WU633: CPU: AMD Ryzen 9 5950X 16-Core Processor
19:58:50:I1:WU633: CPU ID: AuthenticAMD Family 25 Model 33 Stepping 2
19:58:50:I1:WU633: CPUs: 32
19:58:50:I1:WU633: Memory: 31.27GiB
19:58:50:I1:WU633:Free Memory: 8.94GiB
19:58:50:I1:WU633: OS Version: 6.14
19:58:50:I1:WU633:Has Battery: false
19:58:50:I1:WU633: On Battery: false
19:58:50:I1:WU633: Hostname: 
19:58:50:I1:WU633: UTC Offset: -4
19:58:50:I1:WU633: PID: 925706
19:58:50:I1:WU633: CWD: /var/lib/fah-client/work
19:58:50:I1:WU633: Exec: /var/lib/fah-client/cores/openmm-core-26/centos-7.9.2009-64bit/release/fahcore-26-centos-7.9.2009-64bit-release-8.2.0/FahCore_26
19:58:50:I1:WU633:************************************ OpenMM ************************************
19:58:50:I1:WU633: Version: 8.2.0
19:58:50:I1:WU633:********************************************************************************
19:58:50:I1:WU633:Project: 18243 (Run 388, Clone 2, Gen 16)
19:58:50:I1:WU633:Reading tar file core.xml
19:58:50:I1:WU633:Reading tar file integrator.xml
19:58:50:I1:WU633:Reading tar file state.xml.bz2
19:58:50:I1:WU633:Reading tar file system.xml.bz2
19:58:50:I1:WU633:Digital signatures verified
19:58:50:I1:WU633:Folding@home GPU Core26 Folding@home Core
19:58:50:I1:WU633:Version 8.2
19:58:50:I1:WU633: Checkpoint write interval: 50000 steps (2%) [50 total]
19:58:50:I1:WU633: JSON viewer frame write interval: 25000 steps (1%) [100 total]
19:58:50:I1:WU633: XTC frame write interval: 10000 steps (0.4%) [250 total]
19:58:50:I1:WU633: TRR frame write interval: disabled
19:58:50:I1:WU633: Global context and integrator variables write interval: disabled
19:58:50:I1:WU633:There are 3 platforms available.
19:58:50:I1:WU633:Platform 0: Reference
19:58:50:I1:WU633:Platform 1: CPU
19:58:50:I1:WU633:Platform 2: OpenCL
19:58:50:I1:WU633: opencl-device 0 specified
19:58:53:I1:WU633:Attempting to create OpenCL context:
19:58:53:I1:WU633: Configuring platform OpenCL
19:58:53:I1:WU633:ERROR:126: Bad platformId size.
19:58:53:I1:WU633:Saving result file ../logfile_01.txt
19:58:53:I1:WU633:Saving result file science.log
19:58:53:I1:WU633:Folding@home Core Shutdown: BAD_WORK_UNIT
19:58:53:E :WU633:Core returned BAD_WORK_UNIT (114)

[muziqaz]

Mesa is in experimental support. Core26/27 kinda work on clover, but clover has been yanked out of the Mesa, so will see. Use amdgpu with rocm opencl sdk