Feature Request, get Zluda to work on iber Hydrologic and Hydraulic 2D modelling software

[Crones8]

So, when using the Hydrologic and Hydraulic modelling 2D software Iber, there are some modules that speed up processing time significantly, using GPU. These modules or plug-ins are called R-Iber and IberPlus (in Iber_Tools window), however, these plug-ins, only work with nVidia cards. Tried using Zluda to get it to work on my 6800 XT, on different versions of Iber, 3.2.2, 3.3, 3.1, not a lot of success there. On Iber 3.1 it won't even load, it will finish the calculations prematurely (even though GPU is identified), in 3.2.2 and 3.3, the model runs for a little while, but then it hangs, and it stops calculating.

So anyways, this would be a great help.

Cheers, and thank you for all the hard work.

[lshqqytiger]

The software does not seem to be powered by CUDA, but OpenGL.
If there is CUDA mode, please let me know how to reproduce the problems you had.

[Crones8]

Hello,

Actually, it uses both, I believe it uses OpenGL for Graphical purposes (maybe postprocessing also), however calculations in R-Iber (Rapid Iber) and Iberplus are made through CUDA, check these papers for further information:

https://doi.org/10.1016/j.envsoft.2023.105628
https://doi.org/10.3390/w10101459

Anyways, I get these hangs in Iber 3.1 for AMD 6800 XT (actually it doesn't hang, but it won't even start processing and finishes process early) but not for RTX 3050 Laptop (in Iber 3.2.2 Model hangs in both GPUs):

To get Iber 3.1 here is the link (free softwatre): https://iberaula.es/1152/iber-model/version-repository

Model which was used is in this link: https://1drv.ms/f/s!AnLXxRasc_tL2BPx8-7eUIRPvUr3?e=PQPRZk

To get it to calculate with GPU, in the software, after you open the model, you have to go to Iber_Tools>Plug-ins then on "Select GPU parallelization", check the R-Iber box.

Any other thing you need, please do tell.

Cheers and thank you.

To get it complete, I'all also add the hang in Iber 3.2.2, which happens in both GPUs, though this is a problem with changing versions of the software and not the GPUs, since it prompts errors loading the model on Iber 3.2.2, which doesn't happen in Iber 3.1 (the model was developed in 3.1):

[lshqqytiger]

Sorry for the late reply. I was quite busy. All I could reveal is that the magic number that the fatbin wrapper holds is being changed for unknown reasons. I need to compare how it's going in the original CUDA. Unfortunately, however, I don't have an NVIDIA card to do so. If you are willing to dump CUDA logs and kernels, please follow troubleshooting manual (see dumper) and upload the results.

[Crones8]

Hello, no problem, will get to it during this week possibly.

[Crones8]

Hello Ishqytiger,

So, I'm not sure I'm doing this correctly since it didn't work.

Set Environment variable like this:

then tried this command in cmd Powershel:

Program started, so I started running the simulation, and it started working correctly:

Did notice that simulation was going much slower than usual when using the GPU normally without Zluda.

Finally, stopped the process and closed the program, so as to return to the powershell, however it seems it didn't work, since I Saw no special lines written in PowerShell, the folder "C:\temp\zluda_dump" was not created, and obviusly, no log file was created.

Please help.

Cheers and thank you.

[lshqqytiger]

Did you create zluda_dump folder? I remember dumper does not create if the path doesn't exist.

[Crones8]

Ok,

So now it worked, after I created the folder, was confused, since tthe instructions said that if the folder wasn't created, it would be created:

Anyways, here are the files.

RIber.exe.zip

Cheers and thank you.

[lshqqytiger]

I found out the reason. Iber uses two different kind (version) of CUDA fatbin. One is supported, another one (known as old-style) is unsupported by ZLUDA currently. I'll try to add support of later one. However, as supporting both is not so easy, it might take some time.

[Crones8]

Well,

I'm very glad that the functionality is possible, this is ver good news. Thank you so much. Will be expecting the support, no matter how much time it takes, honestly I've been asking the developers to implement support for AMD GPUs for almost 3 years now, and they still haven't. So, knowing it will be available eventually is great news for me, and I'm sure many others as well.

Cheers, and thank you.

[lshqqytiger]

Good news, I succeeded to interpret old style fat binary.
The kernel map_f64_to_f32 was built by nvcc 3.2 14 years ago!
For now, I have to implement some assembly instructions to make it work.
I'll back with next release (hopefully) soon.

[Crones8]

This is amazing news, thank you so much for this incredible work.

Will be expecting the release to try it.

Cheers.

[lshqqytiger]

Unfortunately, although I did what I can, the application still does not work properly. What I can find out is that the inner process terminates with an error right after ZLUDA passes recompiled binary of the kernel map_f64_to_f32 that I referenced before. Because the application consists of two processes when we enable RIber, it is quite troublesome to debug. You can check the last build (version) that I tried on attempt_iber branch. (does not work) I'm planning to dig into the application more later, but not sure.

[Crones8]

Well, it's certainly a great advance, at least for me, considering no support has been given to this software for AMD for years.

So still, thank you so much for your time and effort, doesn't matter how long it takes, it's a relief to know that at least someone is trying.