TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Vramfs: Vram Based Filesystem for Linux

130 pointsby signa11about 2 months ago

12 comments

devitabout 2 months ago
That&#x27;s cool but I think the proper solution is to write a Linux kernel module that can reserve GPU RAM via DRM to create ramdisks, not create a userspace filesystem using OpenCL.<p>That would give proper caching, direct mmap support if desired, a reliable, correct and concurrent filesystem (as opposed to this author&#x27;s &quot;all of the FUSE callbacks share a mutex to ensure that only one thread is mutating the file system at a time&quot;), etc.
dcanelhasabout 2 months ago
On the topic of coercing bits into functioning as data storage: harder drive ( <a href="http:&#x2F;&#x2F;tom7.org&#x2F;harder&#x2F;" rel="nofollow">http:&#x2F;&#x2F;tom7.org&#x2F;harder&#x2F;</a> )
评论 #43518243 未加载
评论 #43518236 未加载
shadowphoabout 2 months ago
2 GB&#x2F;s is pretty crappy, that’s about the burst speed of many nvme SSDs.<p>Virtual disk should me more then 6 gb&#x2F;s at least with ddr5.
评论 #43518495 未加载
finnjohnsen2about 2 months ago
using precious vram to store files is a special kind of humor. especially since someone actually implemented it. kudos
评论 #43532475 未加载
12destroyer21about 2 months ago
What is the overhead on a FUSE filesystem compared to being implemented in the kernel? Could something like eBPF be used to make a faster FUSE-like filesystem driver?
评论 #43518651 未加载
评论 #43518477 未加载
评论 #43522644 未加载
评论 #43518614 未加载
fp64about 2 months ago
Somewhat related, there is NVIDIA CUDA Direct Storage[0] which provides an API for efficient “file transfer” between GPU and local filesystem. Always wanted to give it a try but haven’t yet<p>[0]: <a href="https:&#x2F;&#x2F;docs.nvidia.com&#x2F;gpudirect-storage&#x2F;index.html" rel="nofollow">https:&#x2F;&#x2F;docs.nvidia.com&#x2F;gpudirect-storage&#x2F;index.html</a>
dheeraabout 2 months ago
If you want a vramfs, why would you use GPU VRAM? CPU&lt;-&gt;GPU copy speeds are not great.<p>I have 192GB of CPU VRAM in my desktop and that was cheap to obtain. Absolute best build decision ever.
评论 #43518940 未加载
评论 #43518149 未加载
评论 #43519753 未加载
harha_about 2 months ago
Cool, I love that there are ways to utilize RAM and VRAM as filesystems. Sometimes you just don&#x27;t need all that pure RAM&#x2F;VRAM.
hinkleyabout 2 months ago
These days is it better to use an old video card or a few PCIE NVME multiplexer for those same lanes?
评论 #43519205 未加载
评论 #43522493 未加载
Dweditabout 2 months ago
Using something like this would keep the GPU powered on and unable to shut itself off.
评论 #43518464 未加载
ggmabout 2 months ago
Doesn&#x27;t the graphics processor of the pi double as bootstrap loader?
knomeabout 2 months ago
could be a good place to sequester a swap file, similar to zram
评论 #43517541 未加载