linux-kernel - Re: [PATCH 6/6] lib/scatterlist: Drop order argument from sgl_free_n

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]

Date:   Thu, 8 Mar 2018 07:59:33 +0000
From:   Tvrtko Ursulin <tursulin@...ulin.net>
To:     James Bottomley <James.Bottomley@...senPartnership.com>,
        linux-kernel@...r.kernel.org
Cc:     Tvrtko Ursulin <tvrtko.ursulin@...el.com>,
        Bart Van Assche <bart.vanassche@....com>,
        Hannes Reinecke <hare@...e.com>,
        Johannes Thumshirn <jthumshirn@...e.de>,
        Jens Axboe <axboe@...nel.dk>,
        "Nicholas A. Bellinger" <nab@...ux-iscsi.org>,
        linux-scsi@...r.kernel.org, target-devel@...r.kernel.org
Subject: Re: [PATCH 6/6] lib/scatterlist: Drop order argument from
 sgl_free_n_order

Hi,

On 07/03/18 18:30, James Bottomley wrote:
> On Wed, 2018-03-07 at 12:47 +0000, Tvrtko Ursulin wrote:
>> From: Tvrtko Ursulin <tvrtko.ursulin@...el.com>
> 
> Firstly, I don't see any justifiable benefit to churning this API, so
> why bother? but secondly this:

Primarily because I wanted to extend sgl_alloc_order slightly in order 
to be able to use it from i915. And then in the process noticed a couple 
of bugs in the implementation, type inconsistencies and unused exported 
symbols. That gave me a feeling API could actually use a bit of work.

>> We can derive the order from sg->length and so do not need to pass it
>> in explicitly.
> 
> Is wrong.  I can have a length 2 scatterlist that crosses a page
> boundary, but I can also have one within a single page, so the order
> cannot be deduced from the length.

sgl_alloc_order never does this.

However there is a different bug in my patch relating to the last entry 
which can have shorter length from the rest. So get_order on the last 
entry is incorrect - I have to store the deduced order and carry it over.

In which case it may even make sense to refactor sgl_alloc_order a bit 
more to avoid wastage on the last entry with high order allocations.

Regards,

Tvrtko