Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
mstdn.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A general-purpose Mastodon server with a 500 character limit. All languages are welcome.

Administered by:

Server stats:

15K
active users

mstdn.social: AboutStatusProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.4

#rocm

0 posts0 participants0 posts today
Public
Mauve 👁💜

Makes me sad that #AMD #ROCm isn't officially supported on iGPUs. Like, it'll kinda run, but then it's likely to crash my window manager and freeze my machine. Sadly the stuff I want to use doesn't have vulkan support yet.

Public
Michael DiLeo

Last night I was up until 2AM trying to get #trunas #amd drivers installed inside of a #docker #container so that #ollama would actually use the #gpu. I was so close. It sees the gpu, it sees it has 16GB of ram, then it uses the #cpu.

Trunas locks down the file system at the root level, so if you want to do much of anything, you have to do it inside of a container. So I made a container for the #rocm drivers, which btw comes to like 40GB in size.

It's detecting, but I don't know if the ollama container has some missing commands, ie rocm or rocm-info, that it may need.

Another alternative is one I don't really want, and that's to install either #debian or windows as a VM - windows because I did a test on the application that runs locally in windows on this machine before and it was super fast. It isn't ideal from RAM usage, but I may be able to run the models more easily with the #windows drivers than the #linux ones.

But anyway, last night was too much of #onemoreturn for a weeknight.

Public
ℒӱḏɩę :blahaj:

The B-17 Bomber was amazing and helped win WWII. I flew on one in 2002 as a tourist - I have family members that were ball turret gunners - bad place to be.

This video was shot on Hi-8, and thankfully I digitized it (at 720x480) way back in that day. Now, I've up-scaled it with local AI (1408x954) and the improvement is astounding.

Sadly, this actual B17 crashed in 2019: en.wikipedia.org/wiki/2019_Boe

#localai
#stablediffusion
#rocm
#amd
#b17
#flyingfortress

Continued thread
Public
Giuseppe Bilotta

Even now, Thrust as a dependency is one of the main reason why we have a #CUDA backend, a #HIP / #ROCm backend and a pure #CPU backend in #GPUSPH, but not a #SYCL or #OneAPI backend (which would allow us to extend hardware support to #Intel GPUs). <doi.org/10.1002/cpe.8313>

This is also one of the reason why we implemented our own #BLAS routines when we introduced the semi-implicit integrator. A side-effect of this choice is that it allowed us to develop the improved #BiCGSTAB that I've had the opportunity to mention before <doi.org/10.1016/j.jcp.2022.111>. Sometimes I do wonder if it would be appropriate to “excorporate” it into its own library for general use, since it's something that would benefit others. OTOH, this one was developed specifically for GPUSPH and it's tightly integrated with the rest of it (including its support for multi-GPU), and refactoring to turn it into a library like cuBLAS is

a. too much effort
b. probably not worth it.

Again, following @eniko's original thread, it's really not that hard to roll your own, and probably less time consuming than trying to wrangle your way through an API that may or may not fit your needs.

6/

Public
Fishd

AI rabbit hole ... I've been playing with Ollama and some stability diffusion tools on my MacBook Pro M2 Max and my Linux desktop ... the desktop is way faster and only has an RX6800 in it, so of course I'm now thinking about an Rx7900XTX ... (I don't do Nvidia cards) ...

Anyone have experience with this upgrade? Is going from 16gb of VRAM to 24gb going to make a massive difference?

Using radeontop I can see it's using all 16gb at some points, but not consistently ... and I'm not sure if that's an issue or a feature. I believe #rocm still has some issues.

#selfhosting#ai#sdxl
Public
Ryan He

在Ryzen 7 #8845HS w/ Radeon #780M 用 #ComfyUI 生圖（ #Linux ）
https://blog.pastwind.org/2025/02/ryzen-7-8845hs-w-radeon-780mcomfyuilinux.html

試了很久才發現成功的方程式…這是因為每次安裝 #ROCm 都需要下載安裝超過30GB的檔案！！！

tl;dr 直接說結論

OS: Ubuntu 22.04（因為ROCm 6.1只支援此以下的版本）

ROCm: <= 6.1.2，6.2跟6.3都沒辦法正常運行

PyTorch: <= 2.4.1，2.5.1版會顯示不支援硬體的警告，圖片有時候無法正確產生。

UserWarning: Attempting to use hipBLASLt on an unsupported architecture! Overriding blas backend to hipblas
2.6以上則完全無法正常運行。使用PyTorch官網的版本而不是AMD提供的。

https://pytorch.org/get-started/previous-versions/

ComfyUI: 當前版本v0.3.14可正常運行。
blog.pastwind.org在Ryzen 7 8845HS w/ Radeon 780M用ComfyUI生圖（Linux）試了很久才發現成功的方程式…這是因為每次安裝ROCm都需要下載安裝超過30GB的檔案！！！   tl;dr 直接說結論 OS: Ubuntu 22.04（因為ROCm 6.1只支援此以下的版本） ROCm:  <= 6.1.2，6.2跟6.3都沒辦法正常運行 PyTorch: <...
ExploreLive feeds
About