Messages

46

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: July 30, 2022, 08:15:52 am »

I tried the Kate IDE, for which CMake has a generator, but the step debugger didn't work for me (but it could build). I took a look at Eclipse but only for Java (I maintain some legacy Java stuff at work) but I'm a C++ guy.

I didn't think VSCode was up-to-date? And I'll try Netbeans - I forget these Java-based IDEs have native builders.

47

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: July 30, 2022, 05:14:43 am »

I'm using the Talos II at work actually for work, and specifically for crunching numbers (lots of RAM, lots of cores).

Short-ish version of the story: I bought a machine partially to test the possibility of rolling them out to our Linux deep learning teams, and partially out of personal interest. The T2s are too finicky to give them out to others but it ticks a lot of boxes for me personally, plus getting workstation CPUs is difficult, waiting months to get Threadrippers (and supply has totally dried up here now) so I did the little work required to get my tools running on POWER and started using it. I've still not found a good IDE for native dev, so I'm working in Visual Studio on my Windows desktop and then running tasks on the T2.

As for the comparisons with the 64c/128t 3990X, Friday was a slow day ahead of the public holiday here and I thought I'd take a look. On a regular 3990X the code was taking 15s (and on the OC'd one 13s), vs the 19s on my T2 (which given I couldn't keep the cores at 100% is very good), but I spent some time to tune for the specific thread group requirement of the 3990X, taking the timing down to 8s and a very comfortable lead!

As for optimisations, to this code probably not. Once finished it'll run once to precalculate a series of numbers which will go into the end product (a texture tool).

A graph would've been better though.

48

GPU Compute / Accelerators / Re: Necessary firmware for Navi14

« on: July 29, 2022, 03:09:57 pm »

Hi! I haven't but I'm planning the same for a W5700 which should be identical, so I'm also interested.

49

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: July 29, 2022, 03:05:52 pm »

To add to the numbers, a MacBook Pro with M1 Max runs the same in 47s.

When we finally get some 5995X TRs at work I'll run the same on there (I have a feeling it won't beat the OC'd 3990X).

(If anyone has an OC'd Power or something else esoteric I can share the source - it's a project that'll be open sourced after it's shipped anyway, it's a graphics tool.)

50

General OpenPOWER Discussion / Re: Arctic Tern user manual posted

« on: July 29, 2022, 02:58:00 pm »

I have some wild speculation on why there are two module slots...

Wild speculation is always good!

51

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: July 29, 2022, 10:51:22 am »

Resurrecting this thread for anyone interested/curious... I mentioned forever ago I'd post some findings comparing the Power9 with other CPUs for my CPU-based number crunching needs.

For my latest work, the single-threaded initial implementation took about half an hour to run on my Xeon desktop, and any experimentation meant really thinking it through before a run.

Breaking down the calculations and into chunks then running on all threads, the slowest Threadripper at work, a 3060X, took 20 seconds, and the fastest, an overclocked water-cooled 3990X, took 8 seconds! The 24 core Xeon took 1m13s.

In comparison my 144 core Power9 takes 19 seconds, but, since the algorithm is broken into 256 chunks, it processes the first 128 batches followed by 112, with the processor showing 77% usage partway through, whereas the Threadrippers keep the CPU at 100%. Still, it's a good indication and the machine compares well with a 3060X.

All this said, the software was never optimised for Power (it has some SIMD for Intel), and I've found both Clang and GCC to be quite variable on Power. GGC 10 gave the best results, with Clang 15 the worst (27 seconds; I have a whole Clang/GCC rant for another day).

52

General OpenPOWER Discussion / Re: Arctic Tern user manual posted

« on: July 29, 2022, 10:11:26 am »

Yup, it's a big pile of cash, but for my tinkering machine I can kind of justify it to myself...

53

General OpenPOWER Discussion / Re: Arctic Tern user manual posted

« on: July 29, 2022, 07:15:02 am »

Ah, okay. I have space near the TPM connector, but the lowest PCEe slot is blocked by the graphics card.

54

General OpenPOWER Discussion / Re: Arctic Tern user manual posted

« on: July 29, 2022, 03:13:04 am »

I saw the Twitter post from RCS and this is great news, but I didn't find that much info. I did read that the Kestrel BMC wasn't yet running on the TalosII, from here:

https://gitlab.raptorengineering.com/kestrel-collaboration/kestrel-litex/litex-boards/-/blob/master/README.md

"Similarly, the Talos II mainboard should be able to be utilized as the host, but Raptor has not yet tested Kestrel on Talos II."

And my understanding is: I'd need the entire BMC kit (PCIe carrier and Arctic Tern module) to have a BMC replacement? I guess I need to free up a PCIe slot...

55

Mod Zone / Separating CPU/HSF?

« on: May 29, 2022, 06:39:09 am »

I'm about start watercooling my Talos II but before doing so I would like to know if there's anything special required when separating the CPU from the HSF with the indium pad in place?

My existing setup is the with the HSFs and indium pads from RCS. Do they stick? Do they need cleaning up?

I've two waterblocks to try out, one from vikings.thum, and another I had custom made from a company here in France.

(I'll be documenting the build and the steps for anyone interested.)

56

Applications and Porting / Re: Google's Depot Tools (and eventually Chrome and anything else)

« on: February 17, 2022, 11:23:44 am »

For anyone else looking at this, it's possible by disabling the updates and not using the (missing) precompiled binaries to get Depot Tools builds to work, but since Dawn (and the tools around Dawn such as the Tint shader compiler, etc.) will build with CMake this is an easier route. I wrote up some instructions here:

https://gist.github.com/cwoffenden/0854e5d7732f0bdaf84b6aaeee09b826

57

Applications and Porting / Google's Depot Tools (and eventually Chrome and anything else)

« on: February 16, 2022, 07:36:01 am »

I'm looking at building Dawn (with the idea to try some of the 3D tools from work on Power) and I'm stumbling at even getting Depot Tools to work. Starting with:

    https://wiki.raptorcs.com/wiki/Porting/Chromium_(New)

Building gn is fine, but then bootstrapping gclient fails (failing to resolve prebuilt packages for ppc64le). Since folks are building Chromium, V8, etc., is this just me not able to get started?

58

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: February 02, 2022, 07:47:13 am »

Answering one of my own questions: multi-boot off the same disk is straightforward, I'm running both Void and Ubuntu on different partitions (I just need to get the video bios loaded now).

59

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: February 01, 2022, 01:59:43 pm »

Yes, I installed Ubuntu and tried Firefox. On the first day I only noticed the font issue (easily repro'd by opening the web dev tools, which just show white). Then today, second day, I noticed FF complains on launch about not being able to save the bookmarks.

I wanted to try a few OS variants, and Ubuntu was my go-to since we have it on lots of machines successfully at work (all x64).

I was going to look at Chrome, but perhaps some other Linux too. Ideally I'd like multiple installed (big and little endian, for testing, though more out of habit now since I've not shipped anything on BE for over 10 years), though I wasn't sure whether this would need multiple disks rather than multiple partitions.

60

GPU Compute / Accelerators / Re: AMD GPU at boot

« on: February 01, 2022, 09:44:09 am »

I'll post real findings but I found the 3990X doesn't scale out linearly for our use. For our highly parallelised task (imagine 65'000 blocks of the same size being processed) going from 64 to 128 cores yields a 37% speed increase. I think we get hit by the four memory channels.

Overclocking the Power9 is on my list to look at (I'm assuming the power stages to be the limit here though). OC'ing the 3990X is a complete power hog, and we're able to draw over 800W for the CPU/mobo alone (for the TR Pros we have here with 4x RTX6000 we struggle to keep in the 2kW power limit of the PSU and wall socket).