From: David Miller <davem@davemloft.net>
To: alexei.starovoitov@gmail.com
Cc: therbert@google.com, jesse@nicira.com, gerlitz.or@gmail.com,
alexander.h.duyck@intel.com, john.r.fastabend@intel.com,
jeffrey.t.kirsher@intel.com, netdev@vger.kernel.org,
tgraf@suug.ch, pshelar@nicira.com, azhou@nicira.com
Subject: Re: [PATCH] net: Add ndo_gso_check
Date: Tue, 07 Oct 2014 16:48:02 -0400 (EDT) [thread overview]
Message-ID: <20141007.164802.624713385845633192.davem@davemloft.net> (raw)
In-Reply-To: <CAADnVQJDxPsMeuNmnR5iUwccOKG2A02dZ+3Ae6wGg-RNFPeeeQ@mail.gmail.com>
From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Date: Tue, 7 Oct 2014 13:43:09 -0700
> On Tue, Oct 7, 2014 at 1:32 PM, David Miller <davem@davemloft.net> wrote:
>> From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
>> Date: Tue, 7 Oct 2014 13:28:01 -0700
>>
>>> On Tue, Oct 7, 2014 at 11:47 AM, David Miller <davem@davemloft.net> wrote:
>>>>
>>>> I am totally against boolean "yes/no" protocol specific checksum
>>>> validation by HW.
>>>>
>>>> It's not faster. You have to look at the pseudo-header and bring it into
>>>> the CPU cache _anyways_, so negating it and 2's complementing it into
>>>> the CHECKSUM_COMPLETE value for validation is free.
>>>>
>>>> There is no performance advantage whatsoever to use another checksumming
>>>> scheme.
>>>
>>> ok, forget faster/slower argument for a second.
>>> Why is it a bad thing to have HW verifying checksums?
>>
>> Because you have to change the damn hardware and/or firmware for every
>> new protocol.
>
> It is true for existing NICs, but it is not true for upcoming devices.
> They're exposing packet parsers to users, so we will be able to
> program any protocol into the device without reflashing it.
> Some of the guys are even allowing reprogramming the parser
> while packets are flowing.
So we have to write new software in _EVERY_ driver to accomodate this.
That makes zero sense either, and it is unneeded complexity in the
hardware.
COMPLETE works everywhere, on everything, with no driver changes, and
is so much harder to get wrong.
Every protocol specific feature has major downsides whether you choose
to see them or not.
next prev parent reply other threads:[~2014-10-07 20:48 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-07 16:50 [PATCH] net: Add ndo_gso_check Alexei Starovoitov
2014-10-07 17:05 ` David Miller
2014-10-07 17:18 ` Alexei Starovoitov
2014-10-07 18:47 ` David Miller
2014-10-07 20:28 ` Alexei Starovoitov
2014-10-07 20:32 ` David Miller
2014-10-07 20:43 ` Alexei Starovoitov
2014-10-07 20:48 ` David Miller [this message]
2014-10-07 21:41 ` Thomas Graf
2014-10-07 17:23 ` Eric Dumazet
2014-10-07 18:15 ` Alexei Starovoitov
2014-10-07 18:50 ` Eric Dumazet
2014-10-07 18:51 ` David Miller
2014-10-07 19:10 ` Tom Herbert
-- strict thread matches above, loose matches on Subject: below --
2014-10-07 22:05 Alexei Starovoitov
2014-10-07 23:43 ` Tom Herbert
2014-09-29 3:50 Tom Herbert
2014-09-29 19:59 ` Or Gerlitz
2014-09-29 20:12 ` David Miller
2014-09-29 20:53 ` Tom Herbert
2014-09-29 21:10 ` Or Gerlitz
2014-09-29 21:38 ` Tom Herbert
2014-09-30 14:30 ` Or Gerlitz
2014-09-30 15:34 ` Tom Herbert
2014-09-30 21:37 ` Stephen Hemminger
2014-09-30 22:11 ` Eric Dumazet
2014-10-01 0:05 ` Tom Herbert
2014-10-01 0:18 ` Eric Dumazet
2014-10-01 6:12 ` Eric Dumazet
2014-10-01 20:58 ` Or Gerlitz
2014-10-01 21:12 ` Jesse Gross
2014-10-01 23:06 ` Tom Herbert
2014-10-05 14:04 ` Or Gerlitz
2014-10-05 18:49 ` Tom Herbert
2014-10-05 19:13 ` Or Gerlitz
2014-10-06 17:59 ` Tom Herbert
2014-10-06 20:22 ` Or Gerlitz
2014-10-06 21:28 ` Tom Herbert
2014-10-07 11:07 ` Or Gerlitz
2015-01-15 18:18 ` Or Gerlitz
2014-10-06 22:33 ` Jesse Gross
2014-10-07 0:17 ` Tom Herbert
2014-10-09 0:30 ` Jesse Gross
2014-10-09 1:46 ` Tom Herbert
2014-10-06 21:51 ` Jesse Gross
2014-09-29 20:13 ` Jesse Gross
2014-09-29 20:47 ` Tom Herbert
2014-09-30 0:33 ` Jesse Gross
2014-09-30 1:59 ` Tom Herbert
2014-10-07 18:13 ` Tom Herbert
2014-10-07 20:15 ` David Miller
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=20141007.164802.624713385845633192.davem@davemloft.net \
--to=davem@davemloft.net \
--cc=alexander.h.duyck@intel.com \
--cc=alexei.starovoitov@gmail.com \
--cc=azhou@nicira.com \
--cc=gerlitz.or@gmail.com \
--cc=jeffrey.t.kirsher@intel.com \
--cc=jesse@nicira.com \
--cc=john.r.fastabend@intel.com \
--cc=netdev@vger.kernel.org \
--cc=pshelar@nicira.com \
--cc=tgraf@suug.ch \
--cc=therbert@google.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).