lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Mon, 16 Oct 2023 18:37:39 -0700
From: Coco Li <lixiaoyan@...gle.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Eric Dumazet <edumazet@...gle.com>, Neal Cardwell <ncardwell@...gle.com>, 
	Mubashir Adnan Qureshi <mubashirq@...gle.com>, Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org, 
	Chao Wu <wwchao@...gle.com>, Wei Wang <weiwan@...gle.com>
Subject: Re: [PATCH v1 net-next 1/5] Documentations: Analyze heavily used
 Networking related structs

Thank you for the reviews, Jakub!

We opted to have separate documentation for future folks who are
interested in more organization to have a baseline of what was
considered fast path (i.e. UDP case mentioned in the other patch
series).

I have added a pointer to the relevant header files in hopes of
getting better attraction for the documentations.


On Wed, Oct 4, 2023 at 6:41 AM Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Sat, 16 Sep 2023 01:06:21 +0000 Coco Li wrote:
> > Analyzed a few structs in the networking stack by looking at variables
> > within them that are used in the TCP/IP fast path.
> >
> > Fast path is defined as TCP path where data is transferred from sender to
> > receiver unidirectionaly. It doesn't include phases other than
> > TCP_ESTABLISHED, nor does it look at error paths.
> >
> > We hope to re-organizing variables that span many cachelines whose fast
> > path variables are also spread out, and this document can help future
> > developers keep networking fast path cachelines small.
> >
> > Optimized_cacheline field is computed as
> > (Fastpath_Bytes/L3_cacheline_size_x86), and not the actual organized
> > results (see patches to come for these).
>
> Great work! I wonder if it's not better to drop the Documentation/
> files and just add the info from the "comments" inline in the struct?
> Is there precedent for such out-of-line documentation?
> The grouping in structures makes it clear what the category of the
> field is (and we can add comments where it isn't).
>
> Right now the "documentation" does not seem to be mentioned anywhere
> in the source code. Chances that anyone will know to look for it are
> close to zero :(
>
> The guidance on how the optimizations were performed OTOH would be
> quite useful to document.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ