[Bloat] XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[Bloat] XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[Frantisek Borsik]

Hello to all,

Looks interesting:
https://medium.com/@tom_84912/xdp2-this-changes-everything-at-least-for-ai-ml-infrastructure-850c1ba82771


All the best,

Frank

Frantisek (Frank) Borsik


*In loving memory of Dave Täht: *1965-2025

https://libreqos.io/2025/04/01/in-loving-memory-of-dave/


https://www.linkedin.com/in/frantisekborsik

Signal, Telegram, WhatsApp: +421919416714

iMessage, mobile: +420775230885

Skype: casioa5302ca

frantisek.borsik@gmail.com

[Bloat] Re: [Cake] XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[David P. Reed]

Hi Frank -
 
I think it is interesting as a concept. A project I am advising has been using DPDK very effectively to get rid of the huge path and locking delays in the current Linux network stack. XDP2 could be supported in a ring3 (user) address space, achieving a similar result.
 
But I don't think XDP2 is going that direction - so it may be stuckinto the mess of kernel space networking. Adding eBPF only has made this more of a mess, by the way (and adding a new "compiler" that needs to be veriried as safe for the kernel).

I will be watching how this evolves.
 
David
 
On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <frantisek.borsik@gmail.com> said:



> Hello to all,
> 
> Looks interesting:
> https://medium.com/@tom_84912/xdp2-this-changes-everything-at-least-for-ai-ml-infrastructure-850c1ba82771
> 
> 
> All the best,
> 
> Frank
> 
> Frantisek (Frank) Borsik
> 
> 
> *In loving memory of Dave Täht: *1965-2025
> 
> https://libreqos.io/2025/04/01/in-loving-memory-of-dave/
> 
> 
> https://www.linkedin.com/in/frantisekborsik
> 
> Signal, Telegram, WhatsApp: +421919416714
> 
> iMessage, mobile: +420775230885
> 
> Skype: casioa5302ca
> 
> frantisek.borsik@gmail.com
> _______________________________________________
> Cake mailing list -- cake@lists.bufferbloat.net
> To unsubscribe send an email to cake-leave@lists.bufferbloat.net
> 

[Bloat] Re: [Cake] XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[Frantisek Borsik]

Thanks a lot, David.

I have asked Tom if he wants to join us and he should be here to chat with
us now.

All the best,

Frank

Frantisek (Frank) Borsik


*In loving memory of Dave Täht: *1965-2025

https://libreqos.io/2025/04/01/in-loving-memory-of-dave/


https://www.linkedin.com/in/frantisekborsik

Signal, Telegram, WhatsApp: +421919416714

iMessage, mobile: +420775230885

Skype: casioa5302ca

frantisek.borsik@gmail.com


On Tue, Sep 9, 2025 at 10:25 PM David P. Reed <dpreed@deepplum.com> wrote:

> Hi Frank -
>
>
>
> I think it is interesting as a concept. A project I am advising has been
> using DPDK very effectively to get rid of the huge path and locking delays
> in the current Linux network stack. XDP2 could be supported in a ring3
> (user) address space, achieving a similar result.
>
>
>
> But I don't think XDP2 is going that direction - so it may be stuckinto
> the mess of kernel space networking. Adding eBPF only has made this more of
> a mess, by the way (and adding a new "compiler" that needs to be veriried
> as safe for the kernel).
>
> I will be watching how this evolves.
>
>
>
> David
>
>
>
> On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <
> frantisek.borsik@gmail.com> said:
>
> > Hello to all,
> >
> > Looks interesting:
> >
> https://medium.com/@tom_84912/xdp2-this-changes-everything-at-least-for-ai-ml-infrastructure-850c1ba82771
> >
> >
> > All the best,
> >
> > Frank
> >
> > Frantisek (Frank) Borsik
> >
> >
> > *In loving memory of Dave Täht: *1965-2025
> >
> > https://libreqos.io/2025/04/01/in-loving-memory-of-dave/
> >
> >
> > https://www.linkedin.com/in/frantisekborsik
> >
> > Signal, Telegram, WhatsApp: +421919416714
> >
> > iMessage, mobile: +420775230885
> >
> > Skype: casioa5302ca
> >
> > frantisek.borsik@gmail.com
> > _______________________________________________
> > Cake mailing list -- cake@lists.bufferbloat.net
> > To unsubscribe send an email to cake-leave@lists.bufferbloat.net
> >
>

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[Tom Herbert]

On Tue, Sep 9, 2025, 5:03 PM Frantisek Borsik <frantisek.borsik@gmail.com>
wrote:

> Thanks a lot, David.
>
> I have asked Tom if he wants to join us and he should be here to chat with
> us now.
>
> All the best,
>
> Frank
>
> Frantisek (Frank) Borsik
>
>
> *In loving memory of Dave Täht: *1965-2025
>
> https://libreqos.io/2025/04/01/in-loving-memory-of-dave/
>
>
> https://www.linkedin.com/in/frantisekborsik
>
> Signal, Telegram, WhatsApp: +421919416714
>
> iMessage, mobile: +420775230885
>
> Skype: casioa5302ca
>
> frantisek.borsik@gmail.com
>
>
> On Tue, Sep 9, 2025 at 10:25 PM David P. Reed <dpreed@deepplum.com> wrote:
>
> > Hi Frank -
> >
> >
> >
> > I think it is interesting as a concept. A project I am advising has been
> > using DPDK very effectively to get rid of the huge path and locking
> delays
> > in the current Linux network stack. XDP2 could be supported in a ring3
> > (user) address space, achieving a similar result.
>

HI David,

The idea is you could write the code in XDP2 and it would be compiled to
DPDK or eBPF and the compiler would handle the optimizations.


> >
> >
> >
> > But I don't think XDP2 is going that direction - so it may be stuckinto
> > the mess of kernel space networking. Adding eBPF only has made this more
> of
> > a mess, by the way (and adding a new "compiler" that needs to be veriried
> > as safe for the kernel).


Think of XDP2 as the generalization of XDP to go beyond just the kernel.
The idea is that the user writes their datapath code once and they compile
it to run in whatever targets they have-- DPDK, P4, other programmable
hardware, and yes XDP/eBPF. It's really not limited to kernel networking.

As for the name XDP2, when we created XDP, eXpress DataPath, my vision was
that it would be implementation agnostic. eBPF was the first instantiation
for practicality, but now ten years later I think we can realize the
initial vision.

Tom



>
> > I will be watching how this evolves.
> >
> >
> >
> > David
> >
> >
> >
> > On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <
> > frantisek.borsik@gmail.com> said:
> >
> > > Hello to all,
> > >
> > > Looks interesting:
> > >
> >
> https://medium.com/@tom_84912/xdp2-this-changes-everything-at-least-for-ai-ml-infrastructure-850c1ba82771
> > >
> > >
> > > All the best,
> > >
> > > Frank
> > >
> > > Frantisek (Frank) Borsik
> > >
> > >
> > > *In loving memory of Dave Täht: *1965-2025
> > >
> > > https://libreqos.io/2025/04/01/in-loving-memory-of-dave/
> > >
> > >
> > > https://www.linkedin.com/in/frantisekborsik
> > >
> > > Signal, Telegram, WhatsApp: +421919416714
> > >
> > > iMessage, mobile: +420775230885
> > >
> > > Skype: casioa5302ca
> > >
> > > frantisek.borsik@gmail.com
> > > _______________________________________________
> > > Cake mailing list -- cake@lists.bufferbloat.net
> > > To unsubscribe send an email to cake-leave@lists.bufferbloat.net
> > >
> >
> _______________________________________________
> Cake mailing list -- cake@lists.bufferbloat.net
> To unsubscribe send an email to cake-leave@lists.bufferbloat.net
>

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[BeckW]

Interesting work! One problem of P4 is that the networking hardware varies so much in number of resources (queues, schedulers, policers, counters, table memory) that the code inevitably becomes tied to a certain system.
It will be difficult to abstract the peculiarities of systems -- eg Broadcom 88800 vs linux kernel -- in a good way.

Wolfgang

-----Ursprüngliche Nachricht-----
Von: Tom Herbert via Bloat <bloat@lists.bufferbloat.net> 
Gesendet: Dienstag, 9. September 2025 23:37
An: Frantisek Borsik <frantisek.borsik@gmail.com>
Cc: David P. Reed <dpreed@deepplum.com>; Cake List <cake@lists.bufferbloat.net>; codel@lists.bufferbloat.net; bloat <bloat@lists.bufferbloat.net>; Jeremy Austin via Rpm <rpm@lists.bufferbloat.net>
Betreff: [Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

On Tue, Sep 9, 2025, 5:03 PM Frantisek Borsik <frantisek.borsik@gmail.com>
wrote:

> Thanks a lot, David.
>
> I have asked Tom if he wants to join us and he should be here to chat 
> with us now.
>
> All the best,
>
> Frank
>
> Frantisek (Frank) Borsik
>
>
> *In loving memory of Dave Täht: *1965-2025
>
> https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flibr
> eqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05%7C02%7C
> BeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604c
> f68b04a5eeb25f5c4f%7C0%7C0%7C638930853609276702%7CUnknown%7CTWFpbGZsb3
> d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoi
> TWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=TAA6NdEkB3KQQt8UhKqs3ZKuLj
> N7A9h9J9FAjNRDDuU%3D&reserved=0
>
>
> https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.
> linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40telekom.de%
> 7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%
> 7C0%7C0%7C638930853609297031%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiO
> nRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%
> 3D%3D%7C0%7C%7C%7C&sdata=ucL7U%2FQ81ks09nVEGUbe%2FFq1rhJNLaegicv%2FLCV
> cIlg%3D&reserved=0
>
> Signal, Telegram, WhatsApp: +421919416714
>
> iMessage, mobile: +420775230885
>
> Skype: casioa5302ca
>
> frantisek.borsik@gmail.com
>
>
> On Tue, Sep 9, 2025 at 10:25 PM David P. Reed <dpreed@deepplum.com> wrote:
>
> > Hi Frank -
> >
> >
> >
> > I think it is interesting as a concept. A project I am advising has 
> > been using DPDK very effectively to get rid of the huge path and 
> > locking
> delays
> > in the current Linux network stack. XDP2 could be supported in a 
> > ring3
> > (user) address space, achieving a similar result.
>

HI David,

The idea is you could write the code in XDP2 and it would be compiled to DPDK or eBPF and the compiler would handle the optimizations.


> >
> >
> >
> > But I don't think XDP2 is going that direction - so it may be 
> > stuckinto the mess of kernel space networking. Adding eBPF only has 
> > made this more
> of
> > a mess, by the way (and adding a new "compiler" that needs to be 
> > veriried as safe for the kernel).


Think of XDP2 as the generalization of XDP to go beyond just the kernel.
The idea is that the user writes their datapath code once and they compile it to run in whatever targets they have-- DPDK, P4, other programmable hardware, and yes XDP/eBPF. It's really not limited to kernel networking.

As for the name XDP2, when we created XDP, eXpress DataPath, my vision was that it would be implementation agnostic. eBPF was the first instantiation for practicality, but now ten years later I think we can realize the initial vision.

Tom



>
> > I will be watching how this evolves.
> >
> >
> >
> > David
> >
> >
> >
> > On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" < 
> > frantisek.borsik@gmail.com> said:
> >
> > > Hello to all,
> > >
> > > Looks interesting:
> > >
> >
> https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmedi
> um.com%2F%40tom_84912%2Fxdp2-this-changes-everything-at-least-for-ai-m
> l-infrastructure-850c1ba82771&data=05%7C02%7CBeckW%40telekom.de%7C299d
> 64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C
> 0%7C638930853609308950%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWU
> sIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%
> 7C0%7C%7C%7C&sdata=RTLHTVFR540C8Psr73uiuRvkx1sSyjmlUIICEHFj0HA%3D&rese
> rved=0
> > >
> > >
> > > All the best,
> > >
> > > Frank
> > >
> > > Frantisek (Frank) Borsik
> > >
> > >
> > > *In loving memory of Dave Täht: *1965-2025
> > >
> > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > libreqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05
> > > %7C02%7CBeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbd
> > > e4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C0%7C638930853609323027%7CUnkn
> > > own%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAi
> > > OiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=98
> > > musUCMTZR4ID%2Bo6GYxWOX99aiYBspBUdh%2BNV1fzwc%3D&reserved=0
> > >
> > >
> > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > www.linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40tel
> > > ekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a
> > > 5eeb25f5c4f%7C0%7C0%7C638930853609334239%7CUnknown%7CTWFpbGZsb3d8e
> > > yJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjo
> > > iTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=0rD47zFeLVV1YDqMc5SO0
> > > xGWEOTrE3FOYv0mOqGF%2FW4%3D&reserved=0
> > >
> > > Signal, Telegram, WhatsApp: +421919416714
> > >
> > > iMessage, mobile: +420775230885
> > >
> > > Skype: casioa5302ca
> > >
> > > frantisek.borsik@gmail.com
> > > _______________________________________________
> > > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe 
> > > send an email to cake-leave@lists.bufferbloat.net
> > >
> >
> _______________________________________________
> Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe send an 
> email to cake-leave@lists.bufferbloat.net
>
_______________________________________________
Bloat mailing list -- bloat@lists.bufferbloat.net To unsubscribe send an email to bloat-leave@lists.bufferbloat.net

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[Tom Herbert]

On Wed, Sep 10, 2025, 1:54 AM <BeckW@telekom.de> wrote:

> Interesting work! One problem of P4 is that the networking hardware varies
> so much in number of resources (queues, schedulers, policers, counters,
> table memory) that the code inevitably becomes tied to a certain system.
> It will be difficult to abstract the peculiarities of systems -- eg
> Broadcom 88800 vs linux kernel -- in a good way.
>

Hi Wolfgang

>
Yes, the non-portability of P4 code between different architectures has
been raised as an issue. I believe this is subset of the general problem
that we need to deal with differing resource constraints across targets
(like different memory resources, hardware accelerators, table sizes, etc
). It should be a problem of resource constraints, and not differences in
core functionality supported by the targets

I think there's three possibilities:

1) We create an image and attempt to resolve the resource constraints at
runtime (load time). Determine the resources and run with those. It's
possible that a HW configuration doesn't have sufficient resources to meet
the programmer's requirements in which case we may have to inform the user
the program can't run (e.g.if the programmer wants line rate encryption,
but there's no HW accelerators then the program requirements can't be met
and the program can't run on that target).
2) Recompile the backend to the different targets. This is needed when the
executable is incompatible between targets. The goal is to still be
transparent to the user, but we still need to meet the requirement of the
application.
3) Change the source code if the hardware peculiarities are not transparent
to program source. Obviously, this is the least preferred option, but if
there's no alternative then the goal here would be to isolate the target
specific code as much as possible. Hopefully, the vast majority of program
code is target agnostic and there's just a little glue code for each target.

Tom



> Wolfgang
>
> -----Ursprüngliche Nachricht-----
> Von: Tom Herbert via Bloat <bloat@lists.bufferbloat.net>
> Gesendet: Dienstag, 9. September 2025 23:37
> An: Frantisek Borsik <frantisek.borsik@gmail.com>
> Cc: David P. Reed <dpreed@deepplum.com>; Cake List <
> cake@lists.bufferbloat.net>; codel@lists.bufferbloat.net; bloat <
> bloat@lists.bufferbloat.net>; Jeremy Austin via Rpm <
> rpm@lists.bufferbloat.net>
> Betreff: [Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom
> Herbert (almost to the date, 10 years after XDP was released)
>
> On Tue, Sep 9, 2025, 5:03 PM Frantisek Borsik <frantisek.borsik@gmail.com>
> wrote:
>
> > Thanks a lot, David.
> >
> > I have asked Tom if he wants to join us and he should be here to chat
> > with us now.
> >
> > All the best,
> >
> > Frank
> >
> > Frantisek (Frank) Borsik
> >
> >
> > *In loving memory of Dave Täht: *1965-2025
> >
> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flibr
> > eqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05%7C02%7C
> > BeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604c
> > f68b04a5eeb25f5c4f%7C0%7C0%7C638930853609276702%7CUnknown%7CTWFpbGZsb3
> > d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoi
> > TWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=TAA6NdEkB3KQQt8UhKqs3ZKuLj
> > N7A9h9J9FAjNRDDuU%3D&reserved=0
> >
> >
> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.
> > linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40telekom.de%
> > 7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%
> > 7C0%7C0%7C638930853609297031%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiO
> > nRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%
> > 3D%3D%7C0%7C%7C%7C&sdata=ucL7U%2FQ81ks09nVEGUbe%2FFq1rhJNLaegicv%2FLCV
> > cIlg%3D&reserved=0
> >
> > Signal, Telegram, WhatsApp: +421919416714
> >
> > iMessage, mobile: +420775230885
> >
> > Skype: casioa5302ca
> >
> > frantisek.borsik@gmail.com
> >
> >
> > On Tue, Sep 9, 2025 at 10:25 PM David P. Reed <dpreed@deepplum.com>
> wrote:
> >
> > > Hi Frank -
> > >
> > >
> > >
> > > I think it is interesting as a concept. A project I am advising has
> > > been using DPDK very effectively to get rid of the huge path and
> > > locking
> > delays
> > > in the current Linux network stack. XDP2 could be supported in a
> > > ring3
> > > (user) address space, achieving a similar result.
> >
>
> HI David,
>
> The idea is you could write the code in XDP2 and it would be compiled to
> DPDK or eBPF and the compiler would handle the optimizations.
>
>
> > >
> > >
> > >
> > > But I don't think XDP2 is going that direction - so it may be
> > > stuckinto the mess of kernel space networking. Adding eBPF only has
> > > made this more
> > of
> > > a mess, by the way (and adding a new "compiler" that needs to be
> > > veriried as safe for the kernel).
>
>
> Think of XDP2 as the generalization of XDP to go beyond just the kernel.
> The idea is that the user writes their datapath code once and they compile
> it to run in whatever targets they have-- DPDK, P4, other programmable
> hardware, and yes XDP/eBPF. It's really not limited to kernel networking.
>
> As for the name XDP2, when we created XDP, eXpress DataPath, my vision was
> that it would be implementation agnostic. eBPF was the first instantiation
> for practicality, but now ten years later I think we can realize the
> initial vision.
>
> Tom
>
>
>
> >
> > > I will be watching how this evolves.
> > >
> > >
> > >
> > > David
> > >
> > >
> > >
> > > On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <
> > > frantisek.borsik@gmail.com> said:
> > >
> > > > Hello to all,
> > > >
> > > > Looks interesting:
> > > >
> > >
> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmedi
> > um.com%2F%40tom_84912%2Fxdp2-this-changes-everything-at-least-for-ai-m
> > l-infrastructure-850c1ba82771&data=05%7C02%7CBeckW%40telekom.de%7C299d
> > 64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C
> > 0%7C638930853609308950%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWU
> > sIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%
> > 7C0%7C%7C%7C&sdata=RTLHTVFR540C8Psr73uiuRvkx1sSyjmlUIICEHFj0HA%3D&rese
> > rved=0
> > > >
> > > >
> > > > All the best,
> > > >
> > > > Frank
> > > >
> > > > Frantisek (Frank) Borsik
> > > >
> > > >
> > > > *In loving memory of Dave Täht: *1965-2025
> > > >
> > > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > > libreqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05
> > > > %7C02%7CBeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbd
> > > > e4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C0%7C638930853609323027%7CUnkn
> > > > own%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAi
> > > > OiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=98
> > > > musUCMTZR4ID%2Bo6GYxWOX99aiYBspBUdh%2BNV1fzwc%3D&reserved=0
> > > >
> > > >
> > > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > > www.linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40tel
> > > > ekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a
> > > > 5eeb25f5c4f%7C0%7C0%7C638930853609334239%7CUnknown%7CTWFpbGZsb3d8e
> > > > yJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjo
> > > > iTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=0rD47zFeLVV1YDqMc5SO0
> > > > xGWEOTrE3FOYv0mOqGF%2FW4%3D&reserved=0
> > > >
> > > > Signal, Telegram, WhatsApp: +421919416714
> > > >
> > > > iMessage, mobile: +420775230885
> > > >
> > > > Skype: casioa5302ca
> > > >
> > > > frantisek.borsik@gmail.com
> > > > _______________________________________________
> > > > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe
> > > > send an email to cake-leave@lists.bufferbloat.net
> > > >
> > >
> > _______________________________________________
> > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe send an
> > email to cake-leave@lists.bufferbloat.net
> >
> _______________________________________________
> Bloat mailing list -- bloat@lists.bufferbloat.net To unsubscribe send an
> email to bloat-leave@lists.bufferbloat.net
>

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[Tom Herbert]

Cc XDP2 list

On Wed, Sep 10, 2025, 6:59 AM Tom Herbert <tom@herbertland.com> wrote:

>
>
> On Wed, Sep 10, 2025, 1:54 AM <BeckW@telekom.de> wrote:
>
>> Interesting work! One problem of P4 is that the networking hardware
>> varies so much in number of resources (queues, schedulers, policers,
>> counters, table memory) that the code inevitably becomes tied to a certain
>> system.
>> It will be difficult to abstract the peculiarities of systems -- eg
>> Broadcom 88800 vs linux kernel -- in a good way.
>>
>
> Hi Wolfgang
>
>>
> Yes, the non-portability of P4 code between different architectures has
> been raised as an issue. I believe this is subset of the general problem
> that we need to deal with differing resource constraints across targets
> (like different memory resources, hardware accelerators, table sizes, etc
> ). It should be a problem of resource constraints, and not differences in
> core functionality supported by the targets
>
> I think there's three possibilities:
>
> 1) We create an image and attempt to resolve the resource constraints at
> runtime (load time). Determine the resources and run with those. It's
> possible that a HW configuration doesn't have sufficient resources to meet
> the programmer's requirements in which case we may have to inform the user
> the program can't run (e.g.if the programmer wants line rate encryption,
> but there's no HW accelerators then the program requirements can't be met
> and the program can't run on that target).
> 2) Recompile the backend to the different targets. This is needed when the
> executable is incompatible between targets. The goal is to still be
> transparent to the user, but we still need to meet the requirement of the
> application.
> 3) Change the source code if the hardware peculiarities are not
> transparent to program source. Obviously, this is the least preferred
> option, but if there's no alternative then the goal here would be to
> isolate the target specific code as much as possible. Hopefully, the vast
> majority of program code is target agnostic and there's just a little glue
> code for each target.
>
> Tom
>
>
>
>> Wolfgang
>>
>> -----Ursprüngliche Nachricht-----
>> Von: Tom Herbert via Bloat <bloat@lists.bufferbloat.net>
>> Gesendet: Dienstag, 9. September 2025 23:37
>> An: Frantisek Borsik <frantisek.borsik@gmail.com>
>> Cc: David P. Reed <dpreed@deepplum.com>; Cake List <
>> cake@lists.bufferbloat.net>; codel@lists.bufferbloat.net; bloat <
>> bloat@lists.bufferbloat.net>; Jeremy Austin via Rpm <
>> rpm@lists.bufferbloat.net>
>> Betreff: [Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom
>> Herbert (almost to the date, 10 years after XDP was released)
>>
>> On Tue, Sep 9, 2025, 5:03 PM Frantisek Borsik <frantisek.borsik@gmail.com
>> >
>> wrote:
>>
>> > Thanks a lot, David.
>> >
>> > I have asked Tom if he wants to join us and he should be here to chat
>> > with us now.
>> >
>> > All the best,
>> >
>> > Frank
>> >
>> > Frantisek (Frank) Borsik
>> >
>> >
>> > *In loving memory of Dave Täht: *1965-2025
>> >
>> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flibr
>> > eqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05%7C02%7C
>> > BeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604c
>> > f68b04a5eeb25f5c4f%7C0%7C0%7C638930853609276702%7CUnknown%7CTWFpbGZsb3
>> > d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoi
>> > TWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=TAA6NdEkB3KQQt8UhKqs3ZKuLj
>> > N7A9h9J9FAjNRDDuU%3D&reserved=0
>> >
>> >
>> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.
>> > linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40telekom.de%
>> > 7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%
>> > 7C0%7C0%7C638930853609297031%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiO
>> > nRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%
>> > 3D%3D%7C0%7C%7C%7C&sdata=ucL7U%2FQ81ks09nVEGUbe%2FFq1rhJNLaegicv%2FLCV
>> > cIlg%3D&reserved=0
>> >
>> > Signal, Telegram, WhatsApp: +421919416714
>> >
>> > iMessage, mobile: +420775230885
>> >
>> > Skype: casioa5302ca
>> >
>> > frantisek.borsik@gmail.com
>> >
>> >
>> > On Tue, Sep 9, 2025 at 10:25 PM David P. Reed <dpreed@deepplum.com>
>> wrote:
>> >
>> > > Hi Frank -
>> > >
>> > >
>> > >
>> > > I think it is interesting as a concept. A project I am advising has
>> > > been using DPDK very effectively to get rid of the huge path and
>> > > locking
>> > delays
>> > > in the current Linux network stack. XDP2 could be supported in a
>> > > ring3
>> > > (user) address space, achieving a similar result.
>> >
>>
>> HI David,
>>
>> The idea is you could write the code in XDP2 and it would be compiled to
>> DPDK or eBPF and the compiler would handle the optimizations.
>>
>>
>> > >
>> > >
>> > >
>> > > But I don't think XDP2 is going that direction - so it may be
>> > > stuckinto the mess of kernel space networking. Adding eBPF only has
>> > > made this more
>> > of
>> > > a mess, by the way (and adding a new "compiler" that needs to be
>> > > veriried as safe for the kernel).
>>
>>
>> Think of XDP2 as the generalization of XDP to go beyond just the kernel.
>> The idea is that the user writes their datapath code once and they
>> compile it to run in whatever targets they have-- DPDK, P4, other
>> programmable hardware, and yes XDP/eBPF. It's really not limited to kernel
>> networking.
>>
>> As for the name XDP2, when we created XDP, eXpress DataPath, my vision
>> was that it would be implementation agnostic. eBPF was the first
>> instantiation for practicality, but now ten years later I think we can
>> realize the initial vision.
>>
>> Tom
>>
>>
>>
>> >
>> > > I will be watching how this evolves.
>> > >
>> > >
>> > >
>> > > David
>> > >
>> > >
>> > >
>> > > On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <
>> > > frantisek.borsik@gmail.com> said:
>> > >
>> > > > Hello to all,
>> > > >
>> > > > Looks interesting:
>> > > >
>> > >
>> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmedi
>> > um.com%2F%40tom_84912%2Fxdp2-this-changes-everything-at-least-for-ai-m
>> > l-infrastructure-850c1ba82771&data=05%7C02%7CBeckW%40telekom.de%7C299d
>> > 64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C
>> > 0%7C638930853609308950%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWU
>> > sIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%
>> > 7C0%7C%7C%7C&sdata=RTLHTVFR540C8Psr73uiuRvkx1sSyjmlUIICEHFj0HA%3D&rese
>> > rved=0
>> > > >
>> > > >
>> > > > All the best,
>> > > >
>> > > > Frank
>> > > >
>> > > > Frantisek (Frank) Borsik
>> > > >
>> > > >
>> > > > *In loving memory of Dave Täht: *1965-2025
>> > > >
>> > > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
>> > > > libreqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05
>> > > > %7C02%7CBeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbd
>> > > > e4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C0%7C638930853609323027%7CUnkn
>> > > > own%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAi
>> > > > OiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=98
>> > > > musUCMTZR4ID%2Bo6GYxWOX99aiYBspBUdh%2BNV1fzwc%3D&reserved=0
>> > > >
>> > > >
>> > > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
>> > > > www.linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40tel
>> > > > ekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a
>> > > > 5eeb25f5c4f%7C0%7C0%7C638930853609334239%7CUnknown%7CTWFpbGZsb3d8e
>> > > > yJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjo
>> > > > iTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=0rD47zFeLVV1YDqMc5SO0
>> > > > xGWEOTrE3FOYv0mOqGF%2FW4%3D&reserved=0
>> > > >
>> > > > Signal, Telegram, WhatsApp: +421919416714
>> > > >
>> > > > iMessage, mobile: +420775230885
>> > > >
>> > > > Skype: casioa5302ca
>> > > >
>> > > > frantisek.borsik@gmail.com
>> > > > _______________________________________________
>> > > > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe
>> > > > send an email to cake-leave@lists.bufferbloat.net
>> > > >
>> > >
>> > _______________________________________________
>> > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe send an
>> > email to cake-leave@lists.bufferbloat.net
>> >
>> _______________________________________________
>> Bloat mailing list -- bloat@lists.bufferbloat.net To unsubscribe send an
>> email to bloat-leave@lists.bufferbloat.net
>>
>

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[David P. Reed]

Tom -
 
An architecture-independent network framework independent of the OS kernel's peculiarities seems within reach (though a fair bit of work), and I think it would be a GOOD THING indeed. IMHO the Linux networking stack in the kernel is a horrific mess, and it doesn't have to be.
 
The reason it doesn't have to be is that there should be no reason it cannot run in ring3/userland, just like DPDK. And it should be built using "real-time" userland programming techniques. (avoiding the generic linux scheduler). The ONLY reason for involving the scheduler would be because there aren't enough cores. Linux was designed to be a uniprocessor Unix, and that just is no longer true at all. With hyperthreading, too, one need never abandon a processor's context in userspace to run some "userland" application.

This would rip a huge amount of kernel code out of the kernel. (at least 50%, and probably more). THe security issues of all those 3rd party network drivers would go away.

And the performance would be much higher for networking.  (running in ring 3, especially if you don't do system calls, is no performance penalty, and interprocessor communications using shared memory is much lower latency than Linux IPC or mutexes).
 
I like the idea of a compilation based network stack, at a slightly higher level than C. eBPF is NOT what I have in mind - it's an interpreter with high overhead. The language should support high-performance co-routining - shared memory, ideally. I don't thing GC is a good thing. Rust might be a good starting point because its memory management is safe.
To me, some of what the base of DPDK is like is good stuff. However, it isn't architecturally neutral.

To me, the network stack should not be entangled with interrupt handling at all. "polling" is far more performant under load. The only use for interrupts is when the network stack is completely idle. That would be, in userland, a "wait for interrupt" call (not a poll). Ideally, on recent Intel machines, a userspace version of MONITOR/MWAIT).

Now I know that Linus and his crew are really NOT gonna like this. Linus is still thinking like MINIX, a uniprocessor time-sharing system with rich OS functions in the kernel and doing "file" reads and writes to communicate with the kernel state. But it is a much more modern way to think of real-time IO in a modern operating system. (Windows and macOS are also Unix-like, uniprocessor monolithic kernel designs).

So, if XDP2 got away from the Linux kernel, it could be great.
BTW, io_uring, etc. are half-measures. They address getting away from interrupts toward polling, but they still make the mistake of keeping huge drivers in the kernel.
 
 
On Tuesday, September 9, 2025 17:36, "Tom Herbert" <tom@herbertland.com> said:









On Tue, Sep 9, 2025, 5:03 PM Frantisek Borsik <[ frantisek.borsik@gmail.com ]( mailto:frantisek.borsik@gmail.com )> wrote:Thanks a lot, David.

 I have asked Tom if he wants to join us and he should be here to chat with
 us now.

 All the best,

 Frank

 Frantisek (Frank) Borsik


 *In loving memory of Dave Täht: *1965-2025

[ https://libreqos.io/2025/04/01/in-loving-memory-of-dave/ ]( https://libreqos.io/2025/04/01/in-loving-memory-of-dave/ )


[ https://www.linkedin.com/in/frantisekborsik ]( https://www.linkedin.com/in/frantisekborsik )

 Signal, Telegram, WhatsApp: +421919416714

 iMessage, mobile: +420775230885

 Skype: casioa5302ca

[ frantisek.borsik@gmail.com ]( mailto:frantisek.borsik@gmail.com )


 On Tue, Sep 9, 2025 at 10:25 PM David P. Reed <[ dpreed@deepplum.com ]( mailto:dpreed@deepplum.com )> wrote:

 > Hi Frank -
 >
 >
 >
 > I think it is interesting as a concept. A project I am advising has been
 > using DPDK very effectively to get rid of the huge path and locking delays
 > in the current Linux network stack. XDP2 could be supported in a ring3
 > (user) address space, achieving a similar result.
HI David,
The idea is you could write the code in XDP2 and it would be compiled to DPDK or eBPF and the compiler would handle the optimizations.
 >
 >
 >
 > But I don't think XDP2 is going that direction - so it may be stuckinto
 > the mess of kernel space networking. Adding eBPF only has made this more of
 > a mess, by the way (and adding a new "compiler" that needs to be veriried
 > as safe for the kernel).
Think of XDP2 as the generalization of XDP to go beyond just the kernel. The idea is that the user writes their datapath code once and they compile it to run in whatever targets they have-- DPDK, P4, other programmable hardware, and yes XDP/eBPF. It's really not limited to kernel networking.
As for the name XDP2, when we created XDP, eXpress DataPath, my vision was that it would be implementation agnostic. eBPF was the first instantiation for practicality, but now ten years later I think we can realize the initial vision.
Tom

>
 > I will be watching how this evolves.
 >
 >
 >
 > David
 >
 >
 >
 > On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <
 > [ frantisek.borsik@gmail.com ]( mailto:frantisek.borsik@gmail.com )> said:
 >
 > > Hello to all,
 > >
 > > Looks interesting:
 > >
 > [ https://medium.com/@tom_84912/xdp2-this-changes-everything-at-least-for-ai-ml-infrastructure-850c1ba82771 ]( https://medium.com/@tom_84912/xdp2-this-changes-everything-at-least-for-ai-ml-infrastructure-850c1ba82771 )
 > >
 > >
 > > All the best,
 > >
 > > Frank
 > >
 > > Frantisek (Frank) Borsik
 > >
 > >
 > > *In loving memory of Dave Täht: *1965-2025
 > >
 > > [ https://libreqos.io/2025/04/01/in-loving-memory-of-dave/ ]( https://libreqos.io/2025/04/01/in-loving-memory-of-dave/ )
 > >
 > >
 > > [ https://www.linkedin.com/in/frantisekborsik ]( https://www.linkedin.com/in/frantisekborsik )
 > >
 > > Signal, Telegram, WhatsApp: +421919416714
 > >
 > > iMessage, mobile: +420775230885
 > >
 > > Skype: casioa5302ca
 > >
 > > [ frantisek.borsik@gmail.com ]( mailto:frantisek.borsik@gmail.com )
 > > _______________________________________________
 > > Cake mailing list -- [ cake@lists.bufferbloat.net ]( mailto:cake@lists.bufferbloat.net )
 > > To unsubscribe send an email to [ cake-leave@lists.bufferbloat.net ]( mailto:cake-leave@lists.bufferbloat.net )
 > >
 >
 _______________________________________________
 Cake mailing list -- [ cake@lists.bufferbloat.net ]( mailto:cake@lists.bufferbloat.net )
 To unsubscribe send an email to [ cake-leave@lists.bufferbloat.net ]( mailto:cake-leave@lists.bufferbloat.net )

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[David P. Reed]

Tom -
 
Well, we may have  to disagree on whether eBPF is a good language/system for writing networking stacks or interfacing to hardware devices.
 
In case I wasn't particularly clear about eBPF, here's a summary of that concern.


Perhaps the biggest drawback (beyond the fact that it is a terrible language) is that it's basically "sugared kretprobes" 1) its "abstractions" are whatever the Linux (or other OS) kernel code design allows to be "exported" within the kernel. You can't, for example, do co-routines or IPC that would be appropriate for a clean network stack. Just what Linus and crew decide to export symbols for. 2) Sadly eBPF is primarily maintained by folks who merely want "hooks" in the kernel for performance analysis. (the original BPF was for programming packet-processing pipelines). Its use for describing the implementation of full network stacks in a clean way, down to and including the semantics of, say, 802.11 devices or 802.2 devices is a masterful hack, but very much tied to the Linux kernel's control structure quirks.
 
As a guy who's been developing operating systems since 1970 starting with Multics and networking protocol implementations since 1976, I would never have thought that eBPF or any language designed to kludge with random APIs produced by a group like the Linux Kernel developers as a basis. There are so many alternatives that are far better that what eBPF is.
 
Go ahead and do whatever experiment you have planned. You don't need my approval to base it on eBPF.
 
Now, understand that I'm not saying DPDK is the answer either (nor is io_uring). It's got a number of design drawbacks, too. The main benefit is that it runs all the code in isolation from the monolithic kernel called Linux, and allows real-time execution in ring3 with interprocessor communication. I would throw most of the rest of DPDK away. I would think that your goals would be best suited by stepping back a bit.

David
 

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[David Collier-Brown]

I agree with your criticisms of eBPF,  but a partial counter-example 
lies in LibreQoS, where Dave was the CTO.

In their case, the API and the language allowed them to do some small, 
difficult tasks in a Linux environment. I find the code non-obvious, but 
then I'm not very smart (:-))

We should loop them in.

--dave

On 9/14/25 14:00, David P. Reed wrote:
> Tom -
>   
> Well, we may have  to disagree on whether eBPF is a good language/system for writing networking stacks or interfacing to hardware devices.
>   
> In case I wasn't particularly clear about eBPF, here's a summary of that concern.
>
>
> Perhaps the biggest drawback (beyond the fact that it is a terrible language) is that it's basically "sugared kretprobes" 1) its "abstractions" are whatever the Linux (or other OS) kernel code design allows to be "exported" within the kernel. You can't, for example, do co-routines or IPC that would be appropriate for a clean network stack. Just what Linus and crew decide to export symbols for. 2) Sadly eBPF is primarily maintained by folks who merely want "hooks" in the kernel for performance analysis. (the original BPF was for programming packet-processing pipelines). Its use for describing the implementation of full network stacks in a clean way, down to and including the semantics of, say, 802.11 devices or 802.2 devices is a masterful hack, but very much tied to the Linux kernel's control structure quirks.
>   
> As a guy who's been developing operating systems since 1970 starting with Multics and networking protocol implementations since 1976, I would never have thought that eBPF or any language designed to kludge with random APIs produced by a group like the Linux Kernel developers as a basis. There are so many alternatives that are far better that what eBPF is.
>   
> Go ahead and do whatever experiment you have planned. You don't need my approval to base it on eBPF.
>   
> Now, understand that I'm not saying DPDK is the answer either (nor is io_uring). It's got a number of design drawbacks, too. The main benefit is that it runs all the code in isolation from the monolithic kernel called Linux, and allows real-time execution in ring3 with interprocessor communication. I would throw most of the rest of DPDK away. I would think that your goals would be best suited by stepping back a bit.
>
> David
>   
> _______________________________________________
> Bloat mailing list -- bloat@lists.bufferbloat.net
> To unsubscribe send an email to bloat-leave@lists.bufferbloat.net

-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
davecb@spamcop.net           |              -- Mark Twain

[Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert (almost to the date, 10 years after XDP was released)

[David P. Reed]

I agree. The best solution, I think, is to move all the device management outside the kernel, and compiling the device abstractions into code that runs in user-level isolation.
 
On Wednesday, September 10, 2025 04:54, BeckW@telekom.de said:



> Interesting work! One problem of P4 is that the networking hardware varies so much
> in number of resources (queues, schedulers, policers, counters, table memory) that
> the code inevitably becomes tied to a certain system.
> It will be difficult to abstract the peculiarities of systems -- eg Broadcom 88800
> vs linux kernel -- in a good way.
> 
> Wolfgang
> 
> -----Ursprüngliche Nachricht-----
> Von: Tom Herbert via Bloat <bloat@lists.bufferbloat.net>
> Gesendet: Dienstag, 9. September 2025 23:37
> An: Frantisek Borsik <frantisek.borsik@gmail.com>
> Cc: David P. Reed <dpreed@deepplum.com>; Cake List
> <cake@lists.bufferbloat.net>; codel@lists.bufferbloat.net; bloat
> <bloat@lists.bufferbloat.net>; Jeremy Austin via Rpm
> <rpm@lists.bufferbloat.net>
> Betreff: [Bloat] Re: [Cake] Re: XDP2 is here - from one and only Tom Herbert
> (almost to the date, 10 years after XDP was released)
> 
> On Tue, Sep 9, 2025, 5:03 PM Frantisek Borsik
> <frantisek.borsik@gmail.com>
> wrote:
> 
> > Thanks a lot, David.
> >
> > I have asked Tom if he wants to join us and he should be here to chat
> > with us now.
> >
> > All the best,
> >
> > Frank
> >
> > Frantisek (Frank) Borsik
> >
> >
> > *In loving memory of Dave Täht: *1965-2025
> >
> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flibr
> > eqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05%7C02%7C
> > BeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604c
> > f68b04a5eeb25f5c4f%7C0%7C0%7C638930853609276702%7CUnknown%7CTWFpbGZsb3
> > d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoi
> > TWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=TAA6NdEkB3KQQt8UhKqs3ZKuLj
> > N7A9h9J9FAjNRDDuU%3D&reserved=0
> >
> >
> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.
> > linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40telekom.de%
> > 7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%
> > 7C0%7C0%7C638930853609297031%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiO
> > nRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%
> > 3D%3D%7C0%7C%7C%7C&sdata=ucL7U%2FQ81ks09nVEGUbe%2FFq1rhJNLaegicv%2FLCV
> > cIlg%3D&reserved=0
> >
> > Signal, Telegram, WhatsApp: +421919416714
> >
> > iMessage, mobile: +420775230885
> >
> > Skype: casioa5302ca
> >
> > frantisek.borsik@gmail.com
> >
> >
> > On Tue, Sep 9, 2025 at 10:25 PM David P. Reed
> <dpreed@deepplum.com> wrote:
> >
> > > Hi Frank -
> > >
> > >
> > >
> > > I think it is interesting as a concept. A project I am advising has
> > > been using DPDK very effectively to get rid of the huge path and
> > > locking
> > delays
> > > in the current Linux network stack. XDP2 could be supported in a
> > > ring3
> > > (user) address space, achieving a similar result.
> >
> 
> HI David,
> 
> The idea is you could write the code in XDP2 and it would be compiled to DPDK or
> eBPF and the compiler would handle the optimizations.
> 
> 
> > >
> > >
> > >
> > > But I don't think XDP2 is going that direction - so it may be
> > > stuckinto the mess of kernel space networking. Adding eBPF only has
> > > made this more
> > of
> > > a mess, by the way (and adding a new "compiler" that needs to be
> > > veriried as safe for the kernel).
> 
> 
> Think of XDP2 as the generalization of XDP to go beyond just the kernel.
> The idea is that the user writes their datapath code once and they compile it to
> run in whatever targets they have-- DPDK, P4, other programmable hardware, and yes
> XDP/eBPF. It's really not limited to kernel networking.
> 
> As for the name XDP2, when we created XDP, eXpress DataPath, my vision was that it
> would be implementation agnostic. eBPF was the first instantiation for
> practicality, but now ten years later I think we can realize the initial vision.
> 
> Tom
> 
> 
> 
> >
> > > I will be watching how this evolves.
> > >
> > >
> > >
> > > David
> > >
> > >
> > >
> > > On Tuesday, September 9, 2025 06:32, "Frantisek Borsik" <
> > > frantisek.borsik@gmail.com> said:
> > >
> > > > Hello to all,
> > > >
> > > > Looks interesting:
> > > >
> > >
> > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmedi
> > um.com%2F%40tom_84912%2Fxdp2-this-changes-everything-at-least-for-ai-m
> > l-infrastructure-850c1ba82771&data=05%7C02%7CBeckW%40telekom.de%7C299d
> > 64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C
> > 0%7C638930853609308950%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWU
> > sIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%
> > 7C0%7C%7C%7C&sdata=RTLHTVFR540C8Psr73uiuRvkx1sSyjmlUIICEHFj0HA%3D&rese
> > rved=0
> > > >
> > > >
> > > > All the best,
> > > >
> > > > Frank
> > > >
> > > > Frantisek (Frank) Borsik
> > > >
> > > >
> > > > *In loving memory of Dave Täht: *1965-2025
> > > >
> > > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > > libreqos.io%2F2025%2F04%2F01%2Fin-loving-memory-of-dave%2F&data=05
> > > > %7C02%7CBeckW%40telekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbd
> > > > e4dffc4b604cf68b04a5eeb25f5c4f%7C0%7C0%7C638930853609323027%7CUnkn
> > > > own%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAi
> > > > OiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=98
> > > > musUCMTZR4ID%2Bo6GYxWOX99aiYBspBUdh%2BNV1fzwc%3D&reserved=0
> > > >
> > > >
> > > > https://deu01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > > www.linkedin.com%2Fin%2Ffrantisekborsik&data=05%7C02%7CBeckW%40tel
> > > > ekom.de%7C299d64b9b76b4d4cf88b08ddf039e105%7Cbde4dffc4b604cf68b04a
> > > > 5eeb25f5c4f%7C0%7C0%7C638930853609334239%7CUnknown%7CTWFpbGZsb3d8e
> > > > yJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjo
> > > > iTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=0rD47zFeLVV1YDqMc5SO0
> > > > xGWEOTrE3FOYv0mOqGF%2FW4%3D&reserved=0
> > > >
> > > > Signal, Telegram, WhatsApp: +421919416714
> > > >
> > > > iMessage, mobile: +420775230885
> > > >
> > > > Skype: casioa5302ca
> > > >
> > > > frantisek.borsik@gmail.com
> > > > _______________________________________________
> > > > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe
> > > > send an email to cake-leave@lists.bufferbloat.net
> > > >
> > >
> > _______________________________________________
> > Cake mailing list -- cake@lists.bufferbloat.net To unsubscribe send an
> > email to cake-leave@lists.bufferbloat.net
> >
> _______________________________________________
> Bloat mailing list -- bloat@lists.bufferbloat.net To unsubscribe send an email to
> bloat-leave@lists.bufferbloat.net
>