linux-kernel - Re: [PATCH v2 1/2] Introduce maximum WQE size to check limits

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: <20191120000840.GQ4991@ziepe.ca>
Date:   Tue, 19 Nov 2019 20:08:40 -0400
From:   Jason Gunthorpe <jgg@...pe.ca>
To:     Rao Shoaib <rao.shoaib@...cle.com>
Cc:     monis@...lanox.com, dledford@...hat.com, sean.hefty@...el.com,
        hal.rosenstock@...il.com, linux-rdma@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 1/2] Introduce maximum WQE size to check limits

On Tue, Nov 19, 2019 at 03:55:35PM -0800, Rao Shoaib wrote:

> My intent is that we calculate and use the maximum buffer size using the
> maximum of, number of SGE's and inline data requested, not controlling the
> size of WQE buffer. If I was trying to limit WQE size I would agree with
> you. Defining MAX_WQE_SIZE based on MAX_SGE and recalculating MAX_SGE does
> not make sense to me. MAX_SGE and inline_data are independent variables and
> define the size of wqe size not the other wise around. I did make
> inline_dependent on MAX_SGE.

What you are trying to do is limit the size of the WQE to some maximum
and from there you can compute the upper limit on the SGE and the
inline data arrays, depending on how the WQE is being used.

If a limit must be had then the limit is the WQE size. It is also
reasonable to ask why rxe has a limit at all, or why the limit is so
small ie why can't it be 2k or something? But that is something else

Jason