From: <dan.j.williams@intel.com>
To: "Sinyuk, Konstantin" <konstantin.sinyuk@intel.com>,
<ksummit@lists.linux.dev>
Subject: Re: [TECH TOPIC] UALink driver upstreaming
Date: Wed, 10 Sep 2025 16:58:50 -0700 [thread overview]
Message-ID: <68c210baeff65_75db10043@dwillia2-mobl4.notmuch> (raw)
In-Reply-To: <fc612440-af2b-4799-97b0-d5631380f0be@intel.com>
Sinyuk, Konstantin wrote:
> Hi All,
>
> The UALink Consortium is defining an open, vendor‑neutral interconnect aimed
> at scaling AI workloads with low‑latency, memory‑semantic communication
> beyond PCIe. Unlike proprietary solutions such as NVLink (NVIDIA) or
> Infinity Fabric (AMD), UALink is a cross‑vendor standard and was recently
> recognized at FMS 2025.
>
> I would like to present a proposal on what UALink support could look like in
> the upstream Linux kernel.
>
> Key areas for discussion:
> - Core driver design: proposed start under drivers/misc/ual/ for discovery,
> topology, and resource management.
> - Memory semantics: same‑OS and multi‑OS rack scenarios, leveraging dma_buf,
> HMM, and NUMA.
> - Control path: AUX bus for vendor extensions, offloading real‑time sequences
> to device microcontrollers, generic UALink interface.
> - Security: confidential compute support and a userspace daemon for topology
> and authentication.
> - Upstreaming strategy: begin with a minimal core driver, then incrementally
> extend toward MM integration, dma_buf support, security, and
> cross‑subsystem work.
>
> The goal is to decide how UALink should be represented as a first‑class
> interconnect in Linux, complementing CXL while remaining vendor‑neutral,
> ABI‑stable, and maintainable.
Hi Konstantin,
Please do also consider submitting this as a topic for the Device Memory
Microconference (https://lpc.events/event/19/contributions/2009/).
Many of folks with experience with CXL, NVLink, Infinity Fabric, RDMA,
HMM, NUMA, dma_buf, Confidential Computing etc... will also be in the
room for a focused topic like this.
next prev parent reply other threads:[~2025-09-10 23:58 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-10 19:37 [TECH TOPIC] UALink driver upstreaming Sinyuk, Konstantin
2025-09-10 23:58 ` dan.j.williams [this message]
2025-09-11 5:13 ` Greg KH
[not found] ` <a70edcba-0ef7-4d0d-bc00-0e8519a458e8@intel.com>
2025-09-11 11:13 ` Konstantin Sinyuk
2025-09-11 10:59 ` Mimi Zohar
[not found] ` <DM3PR11MB86833A11AD52C01C9063FFE1E309A@DM3PR11MB8683.namprd11.prod.outlook.com>
2025-09-11 11:09 ` FW: " Konstantin Sinyuk
2025-09-11 15:19 ` Linus Walleij
[not found] ` <a74382d8-a2bf-4534-b0ee-a97d8faabf16@intel.com>
2025-09-11 18:10 ` Konstantin Sinyuk
2025-09-11 19:02 ` Leon Romanovsky
2025-09-12 7:22 ` Linus Walleij
2025-09-12 7:47 ` Greg KH
[not found] ` <0f876c7c-566b-476a-b590-d490d41d605c@intel.com>
2025-09-12 12:07 ` Konstantin Sinyuk
2025-09-12 12:18 ` Linus Walleij
2025-09-12 13:49 ` Rodrigo Vivi
2025-09-12 16:13 ` Matthew Brost
2025-09-13 19:11 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=68c210baeff65_75db10043@dwillia2-mobl4.notmuch \
--to=dan.j.williams@intel.com \
--cc=konstantin.sinyuk@intel.com \
--cc=ksummit@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.