Live data from Hacker News

Open Hardware Ethernet Switch project, part 1

serd.es

31–38 of 38 posts

Re: Open Hardware Ethernet Switch project, part 1

#31
post #7

I wonder if the author considered at any point using Sparx-5 switch asic from Microchip? Those are available in single quantities for not too crazy price ($121 for 128G variant), and they are Linux based. Of course I understand that having custom switch engine is far more satisfying to do.

I really, really would love to see someone make Sparx-5 switch. It's weird that I can't even find any commercially available/closed source switches with them (anyone know of any?).

There was this: https://www.servethehome.com/insane-48-port-2-5gbe-2x-25gbe-...

But I have not been able to find a place to buy one.

Re: Open Hardware Ethernet Switch project, part 1

#32

Earlier quoted context omitted.

The code is at https://github.com/azonenberg/staticnet but I've intentionally avoided over-publicizing it since it hasn't had any kind of third party security review. As of now it's functional enough I'm willing to deploy it on a lab network but wouldn't trust it open to an untrusted network. I work in embedded security and have tried to avoid any of the most gross footguns, deliberately simplifying the implementatio…

For context of how lightweight this is, an -O3 release build of my entire firmware on the management processor right now (including the sshd, hardware drivers, TCP/IP stack, the CLI itself, all of the code to query the supported set of sensors, etc) uses 109 kB of flash and 84 kB of SRAM. The -Og debug build is smaller at 86 kB flash usage. It compiles in five seconds from a clean build tree on my workstation. Sure,…

I’m getting ready to turn 42yo, but I want to be like you when I grow up.

Seriously, impressive work. I’m a sysadmin who completed a two-year electronics engineering program (mostly for fun, although it’s become increasingly useful in my work). Since then, I have learned enough to design my own simple boards with BOMs that include maybe a few dozen parts. I can also hack on relatively complex projects that more experienced engineers design, but I find the bulk of my enjoyment in the iterative process of designing for myself.

My family doesn’t understand the appeal, and poke fun at my inability to finish a lot of the things I start, especially when it’s some gadget that I could have purchased off-the-shelf for less than I paid for the parts. But I tell them that, as a hobby, electronics is mostly about finding fulfillment within the journey, rather than the destination. Otherwise, it would be a very masochistic endeavor, as even successfully completed projects will have involved several failures along the way. Each time I fail, I learn something new and exciting to expand my horizons.

That unique experience of your own intuition clicking into place as the gaps in your knowledge are bridged, and you discover how fundamental physics work to produce the phenomena we use to create machines—it is something I have always struggled to describe. It’s sort of like when you learn how a magic trick works, but instead of being disappointed in the mundanity of seeing someone palm a playing card, you’re wonderstruck by the magical nature of physics, the elegance of simple laws that manifest complex systems when applied recursively, and the fractal nature of it all.

Pardon the self-indulgent musings, I have nothing productive to add, only that I very much enjoy exploring your work, and I look forward to following this project.

Re: Open Hardware Ethernet Switch project, part 1

#33
Wow, this is very impressive. This reminds me of my years of designing the analog front end of ethernet PHYs (5 generations of 1G base-T, 3 geberations of 2.5G/5G/10G base-T, 1 experinental 7.5G, the first gen 25G base-T, 2 genetation of 1G-baseT1 and the first generation of 10G base-T1 over single shielded pair for automotive). I got my hands dirty with board level testing and a lot of the firmeare!

I'm curious if there are any 10G base-T or 25G base-T PHYs out there that one can buy at lower quantities.

Re: Open Hardware Ethernet Switch project, part 1

#34
post #33

Wow, this is very impressive. This reminds me of my years of designing the analog front end of ethernet PHYs (5 generations of 1G base-T, 3 geberations of 2.5G/5G/10G base-T, 1 experinental 7.5G, the first gen 25G base-T, 2 genetation of 1G-baseT1 and the first generation of 10G base-T1 over single shielded pair for automotive). I got my hands dirty with board level testing and a lot of the firmeare! I'm curious if t…

No idea, the high power consumption and latency of >1000baseT were such that I've never had any interest in anything newer in baseT land.

At home I run 10/100/1000baseT, 10Gbase-SR, 40Gbase-SR4, and am just beginning to prepare for 25Gbase-SR and 100Gbase-SR4 deployments (I have some NICs and optics fielded but no switching yet).

I've worked with 100baseT1 and 1000baseT1 for automotive projects at work and am familiar with the line coding, and have written a protocol decode for 100baseT1 among others, but I don't use them in my own projects.

When I last looked into it, which was admittedly several years ago, 10GbaseT PHYs were unobtainium in the "I want to buy just one, without an NDA" space. Your best option for putting 10Gbase-T on a DIY design is to put a SFP+ cage on there and slap in a baseT SFP+ module that somebody else designed.

Re: Open Hardware Ethernet Switch project, part 1

#35
post #7

I wonder if the author considered at any point using Sparx-5 switch asic from Microchip? Those are available in single quantities for not too crazy price ($121 for 128G variant), and they are Linux based. Of course I understand that having custom switch engine is far more satisfying to do.

I really, really would love to see someone make Sparx-5 switch. It's weird that I can't even find any commercially available/closed source switches with them (anyone know of any?).

Yep, really strange indeed.

I posted the link for the ServeTheHome article last year back in 2024 although it's really year 2023 article [1].

Layer 3 switch for examples from Cisco or Juniper used to cost more than USD100K only ten years ago but now they can be dirt cheap with advent of the networking switch SoC like SparX-5.

High end SparX-5-200 (VSC7558) supports up to 200 Gbps of switch core bandwidth, with flexible interface options including 1/2.5/5/10/25 GbE, and the unit cost is only USD165 [2].

Imagine having this kind of really affordable powerful enterprise layer 3 switch with Linux Foundation SONiC or DENT OS [4], [5].

Someone just need to crowd source this SparX-5 SoC based layer 3 switch and will be selling like hot cakes.

[1] Managed Chinese E-Sports Cafe and Hotel Switch: 48-Port 2.5GbE 2x 25GbE 2x 10GbE:

https://news.ycombinator.com/item?id=40883384

[2] SparX-5:

https://ww1.microchip.com/downloads/en/DeviceDoc/SparX-5_sel...

[3] Sparx5:

https://microchip-ung.github.io/bsp-doc/bsp/2024.03/supporte...

[4] SONiC: Software for open networking in the cloud.

https://sonicfoundation.dev/

[5] DENT: SwitchDev based NOS:

https://dent.dev/

Re: Open Hardware Ethernet Switch project, part 1

#36
post #33

Wow, this is very impressive. This reminds me of my years of designing the analog front end of ethernet PHYs (5 generations of 1G base-T, 3 geberations of 2.5G/5G/10G base-T, 1 experinental 7.5G, the first gen 25G base-T, 2 genetation of 1G-baseT1 and the first generation of 10G base-T1 over single shielded pair for automotive). I got my hands dirty with board level testing and a lot of the firmeare! I'm curious if t…

No idea, the high power consumption and latency of >1000baseT were such that I've never had any interest in anything newer in baseT land. At home I run 10/100/1000baseT, 10Gbase-SR, 40Gbase-SR4, and am just beginning to prepare for 25Gbase-SR and 100Gbase-SR4 deployments (I have some NICs and optics fielded but no switching yet). I've worked with 100baseT1 and 1000baseT1 for automotive projects at work and am familia…

> When I last looked into it, which was admittedly several years ago, 10GbaseT PHYs were unobtainium in the "I want to buy just one, without an NDA" space. Your best option for putting 10Gbase-T on a DIY design is to put a SFP+ cage on there and slap in a baseT SFP+ module that somebody else designed.

Indeed this seems to be the case. I hate this. There's no reason to make hardware inaccessible.

Re: Open Hardware Ethernet Switch project, part 1

#37

Earlier quoted context omitted.

But not a cheap open source commodity

This is not going to be cheap. That was never the goal. In low volume when you combine several custom multilayer boards, custom powder-coated sheet metal work, etc even if you allow for the practically-free recycled FPGA this is going to probably end up being a $5-10K project. The last custom one-off 1U I did from ProtoCase was like $800ish just for the enclosure, and that was before the recent tariff hikes.

Oh fair, maybe just “open source commodity”

Re: Open Hardware Ethernet Switch project, part 1

#38

There are 48+2 port switches with OpenWRT support. Re: initial specs for the (4 port) OpenWRT One, which is built on Banana Pi's, which supports U-boot: https://www.cnx-software.com/2024/01/12/openwrt-one-ap-24-xy... .. https://openwrt.org/toh/openwrt/one : > The non-open-source components include the 2.5GbE PHY and WiFi firmware with blobs running on separate cores that are independent of the main SoC where OpenWrt…

There are also xilinx (now AMD) FPGA modules in hwtlib:

hwtLib.xilinx package: https://hwtlib.readthedocs.io/en/latest/hwtLib.xilinx.html#

Post reply on HN