netdev - Re: Question about LRO/GRO and TCP acknowledgements

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]

Message-ID: <20110612095131.6d924082@konijn>
Date:	Sun, 12 Jun 2011 09:51:31 +0200
From:	Joris van Rantwijk <joris@...isvr.nl>
To:	Ben Hutchings <bhutchings@...arflare.com>
Cc:	netdev@...r.kernel.org
Subject: Re: Question about LRO/GRO and TCP acknowledgements

On 2011-06-12, Ben Hutchings <bhutchings@...arflare.com> wrote:
> LRO implementations (and GRO) are expected to put the actual segment
> size in skb_shared_info(skb)->gso_size on the aggregated skb.  TCP
> will then use that rather than the aggregated payload size when
> deciding whether to defer an ACK.

Thanks. I see that indeed gso_size is being used for MSS calculations
instead of the total GRO size.

However, I'm not sure that this completely answers my question.
I am not so much concerned about quick ACK vs delayed ACK.
Instead, I'm looking at the total number of ACKs transmitted.
The sender depends on the _number_ of ACKs to update its congestion
window.

As far as I can see, current code will send just one ACK per coalesced
GRO bundle, while the sender expects one ACK per two segments.

Thanks,
Joris.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html