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.

Show HN: vGPU and SR-IOV on consumer GPUs

203 pointsby ArcVRArthurover 3 years ago

19 comments

csdvrxover 3 years ago
That&#x27;s very impressive!<p>You may want to do the same for NVMe: creating several namespaces is not supported on most consumer drives, while laptops can rarely have more than 1 NVMe (same problem as with the GPUs: a passthrough requires having 2 of them)<p>Being able to split the NVMe drive not by partition but by namespace would let each OS see a &quot;full drive&quot;.
评论 #28951986 未加载
评论 #28955441 未加载
gh123manover 3 years ago
This is insanely impressive. Having tried set up GPU passthrough in proxmox a few years ago, it was an absolute disaster. I would love to see this kind of approach more widely supported by other hypervisors!<p>It&#x27;s a real shame consumer GPUs are arbitrary locked down when the enterprise counterparts (often with the exact same chip) have much better support for virtualization.
评论 #28946505 未加载
评论 #28945957 未加载
rektideover 3 years ago
Really hoping AMD eventually does the right thing here. Not that it particularly matters seeing as how decent AMD video cards have been unpurchaseable for 18 months now.<p>Consumers should have the ability to use their hardware well too. Selling the same thing at 2X the price differentiating only on virtualization capabilities is not a moral path.<p>&gt; <i>We remain hopeful that AMD will recognize forthcoming changes in GPU virtualization with the creation of open standards such as Auxiliary Domains (AUX Domains), Mdev (VFIO-Mdev developed by Nvidia, RedHat, and Intel), and Alternative Routing-ID Interpretation (ARI) especially in light of Intel&#x27;s market entrance with their ARC line of GPUs supporting Intel Graphics Virtualization Technology (GVT-g).</i><p>Really cool to hear there are a bunch of vGPU-related efforts underway! That&#x27;s so great.
评论 #28947164 未加载
评论 #28953257 未加载
uberduperover 3 years ago
I&#x27;m familiar with linux virtualization, gpu passthrough, etc. I&#x27;ve never heard of arcd and they&#x27;ve made no attempt in this doc or on their git to explain what it is or why it exists as, I assume, a replacement (or wrapper?) for qemu.<p>My past experience with looking-glass is that it falls on its face at anything &gt; 1440p@60Hz. I&#x27;m interested in vGPU for my linux VMs (spice is slow and sdl&#x2F;gtk display is flakey) but for gaming, I don&#x27;t want looking-glass and prefer to just do the passthrough thing with a KVM switch.
评论 #28946160 未加载
评论 #28947133 未加载
评论 #28946931 未加载
评论 #28947525 未加载
liuliuover 3 years ago
Great! Still requires vGPU support and the merged driver approach last time I tried won&#x27;t support CUDA on host (I was probably the first one tried the merged driver thing with vgpu_unlocked?).<p>Looking forward someone write a Vulkan driver on Windows that just shuttles down to the Linux host. virgl used to be a promising project ...
评论 #28946120 未加载
awesnvadsomeover 3 years ago
This seems awesome! I have a passthrough setup with a very old card and a much newer one for games in a windows vm, it&#x27;ll be nice to look into getting this set up and I can reduce the power draw on my system which was causing some problems...<p>Is this something that could work with&#x2F;be integrated with libvirt for easy configuration? Itd be neat to set it up with my current install, although not at all a real problem.
评论 #28946720 未加载
DiabloD3over 3 years ago
I tried doing this years ago, and never quite go it to work.<p>Some of the software involved in that article simply didn&#x27;t exist yet, and GPUs weren&#x27;t shipping with SR-IOV support yet (instead, I did Intel iGPU for Linux fbcon, real AMD GPU fed directly to the Windows VM with PCI-E Passthrough). In the end, I bailed on that dream and moved the Linux install to its own smaller machine, and ran Windows bare on the big machine.<p>The problem was, if the GPU locked up hard, and GPUs back then would not respond to PCI Device Reset, if it wasn&#x27;t something that merely re-initializing it on VM restart would fix... I had to restart the <i>entire</i> machine, thus defeating the purpose of having Windows in the VM in the first place!<p>All my long-lived processes now run on the stand-alone Linux machine, and anything that is free to explode runs on my Windows machine. Windows gets wonky? Restart, ssh back into my screen sessions, reopen the browser, restart a bunch of cloud slaved apps, tada.
评论 #28946904 未加载
danbmil99over 3 years ago
Very cool approach. It&#x27;s going to be a fight I suspect; vendors lock things up specifically so they can have price differentials in different markets. It may end up being like the fight between workstations versus PCS back in the 90s
评论 #28959059 未加载
scaryglidersover 3 years ago
Very, very cool!<p>I&#x27;ve set up a dual-GPU system in the past using two nvidia GPUs and whilst I found the trek towards PCI passthrough to other virtual machines rewarding when it finally worked, I also found the arrangement to be inconvenient.<p>What you&#x27;ve achieved here, seems the ideal. Well done :)<p>I will either patiently wait for an Arch Linux version of the install, or I&#x27;ll eventually end up impatient and see if I can rustle up something - an install script is an install script, it should be just a matter (famous last words) of altering the install script&#x2F;procedures to suit.
ArtWombover 3 years ago
Whoa! Multi GPU is so hard. This could be the start of something. Looks like it makes use of nvidia&#x27;s capture api?
评论 #28951929 未加载
compsciphdover 3 years ago
isn&#x27;t this natively supported by nvidia?<p>i.e. you have a vgpu card (or a consumer card you an map to the equivalent vgpu card) and nvidia drivers and tools lets you load drivers that essetially partition it into XGB (all partitions being the same X) and then you just gpu passthrough the newly created device that maps to a single partition into the vm?<p>the whole trick being the ability to trick nvidia&#x27;s drivers into thinking that the consumer gpu is really the server model, but otherwise it becomes just normal nvidia usage?
评论 #28956781 未加载
评论 #28956705 未加载
评论 #28958787 未加载
AaronFrielover 3 years ago
This article almost got my hopes up too much! I&#x27;m curious whether Intel&#x27;s &quot;Ark&quot; GPUs will support the same or if they&#x27;ll go the same path of Nvidia in locking down virtual function support.
评论 #28952192 未加载
COGloryover 3 years ago
Very impressive. You should also post this to the Level1Techs forum
评论 #28946304 未加载
belvalover 3 years ago
Just to clarify because I&#x27;ve failed at doing pretty much the setup that you are describing with my 1080Ti. Does this still require the vgpu_unlock changes for Nvidia cards or is this something that bypasses the need for it entirely?
评论 #28951846 未加载
shmerlover 3 years ago
What&#x27;s the story with SR-IOV for consumer AMD GPUs on Linux? Last time I looked into it, it was impossible to use or more like AMD didn&#x27;t want to support it.
Kaze404over 3 years ago
Once this is running on NixOS I&#x27;ll be all over it. Great news!
评论 #29082918 未加载
ameliusover 3 years ago
Does this also work for CUDA (e.g. deep learning jobs)?
评论 #28959071 未加载
omreaderhnover 3 years ago
This looks really nice. Does this work with Ampere GPUs?
评论 #28947324 未加载
kfprtover 3 years ago
Note Intel GVT-g is only available on &lt;=Gen9 GPU&#x27;s. Gen11 and 12 are not supported.
评论 #28946059 未加载
评论 #28945575 未加载