netdev - Re: iscsi target regression due to "tcp: remove prequeue support" patch

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: <1516264893.24576.243.camel@haakon3.daterainc.com>
Date:   Thu, 18 Jan 2018 00:41:33 -0800
From:   "Nicholas A. Bellinger" <nab@...ux-iscsi.org>
To:     Mike Christie <mchristi@...hat.com>
Cc:     target-devel <target-devel@...r.kernel.org>, fw@...len.de,
        Linux-netdev <netdev@...r.kernel.org>,
        David Miller <davem@...emloft.net>
Subject: Re: iscsi target regression due to "tcp: remove prequeue support"
 patch

On Mon, 2018-01-15 at 14:33 -0600, Mike Christie wrote:
> On 01/15/2018 12:41 AM, Nicholas A. Bellinger wrote:
> > Hey MNC & Co,
> > 
> > Ping on the earlier iscsi-target authentication login failures atop
> > 4.14 + commit e7942d063 removing tcp prequeue support.
> > 
> > For reference, what is your pre 4.14 environment using for
> > sysctl_tcp_low_latency..?
> 
> tcp_low_latency=1.
> 
> I have tested the current kernel with 1 and 0, and it does not make a
> difference.
> 

Likewise, I've always used tcp_low_latency=1 on pre 4.14.y kernels and
never triggered this bug with multi login PDU exchanges.

What's strange from groking commit e7942d0 though, it appear the removal
of tcp prequeue logic should only effect tcp_low_latency=0 users,
right..?

But here, that doesn't seem to be the case..