netdev - Re: [PATCH] tcp_input: move out condition check from tcp_data

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] [day] [month] [year] [list]

Message-ID: <1502082283.25002.56.camel@edumazet-glaptop3.roam.corp.google.com>
Date:   Sun, 06 Aug 2017 22:04:43 -0700
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Ilya Matveychikov <matvejchikov@...il.com>
Cc:     netdev@...r.kernel.org
Subject: Re: [PATCH] tcp_input: move out condition check from
 tcp_data_queue()

On Sun, 2017-08-06 at 21:57 +0400, Ilya Matveychikov wrote:
> > On Aug 6, 2017, at 9:07 PM, Eric Dumazet <eric.dumazet@...il.com> wrote:
> > 
> > On Sun, 2017-08-06 at 13:51 +0400, Ilya Matveychikov wrote:
> >> As tcp_data_queue() function is used just only twice it's better
> >> to move out the first check and wrap it with inline. It saves a
> >> single call in case the condition evaluated as true.
> >>

...

> > We wont accept such a change, because this code does not need to be
> > inlined in the callers, ( and btw inline in .c files are discouraged
> > these days )
> 
> Not sure that I understand you point. What’s the reason for that code
> not need to be inlined in the callers?

You sent a patch, you have to explain why it is needed.

Your changelog is absolutely not giving a compelling reason.

TCP stack is already complex, no need to add yet another obfuscation
unless there is a strong reason.