netdev - Re: [RFC, PATCH] net: page_pool: Don't use page->private to store dma_addr

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20190207213758.GA10662@Iliass-MBP.lan>
Date:   Thu, 7 Feb 2019 23:37:58 +0200
From:   Ilias Apalodimas <ilias.apalodimas@...aro.org>
To:     David Miller <davem@...emloft.net>
Cc:     willy@...radead.org, brouer@...hat.com, tariqt@...lanox.com,
        toke@...hat.com, netdev@...r.kernel.org,
        mgorman@...hsingularity.net, linux-mm@...ck.org
Subject: Re: [RFC, PATCH] net: page_pool: Don't use page->private to store
 dma_addr_t

Hi David,

On Thu, Feb 07, 2019 at 01:25I:19PM -0800, David Miller wrote:
> From: Ilias Apalodimas <ilias.apalodimas@...aro.org>
> Date: Thu, 7 Feb 2019 17:20:34 +0200
> 
> > Well updating struct page is the final goal, hence the comment. I am mostly
> > looking for opinions here since we are trying to store dma addresses which are
> > irrelevant to pages. Having dma_addr_t definitions in mm-related headers is a
> > bit controversial isn't it ? If we can add that, then yes the code would look
> > better
> 
> I fundamentally disagree.
> 
> One of the core operations performed on a page is mapping it so that a device
> and use it.
> 
> Why have ancillary data structure support for this all over the place, rather
> than in the common spot which is the page.

You are right on that. Moreover the intention of this change is to facilitate
the page recycling patches we proposed with Jesper. In that context we do need
the dma mapping information in a common spot since we'll need to access it from
drivers, networking code etc. The struct page *is* the best place for that.

> 
> A page really is not just a 'mm' structure, it is a system structure.

Well if you put it that way i completely agree (also it makes our life a *lot*
easier :))


Thanks
/Ilias