linux-kernel - Re: [RFC] dmaengine: Add metadat_ops for dma_async_tx

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: <20180731042903.GC16775@vkoul-mobl>
Date:   Tue, 31 Jul 2018 09:59:03 +0530
From:   Vinod <vkoul@...nel.org>
To:     Peter Ujfalusi <peter.ujfalusi@...com>
Cc:     radheys@...inx.com, vinod.koul@...el.com, lars@...afoo.de,
        michal.simek@...inx.com, linux-kernel@...r.kernel.org,
        dmaengine@...r.kernel.org, dan.j.williams@...el.com,
        appanad@...inx.com, linux-arm-kernel@...ts.infradead.org
Subject: Re: [RFC] dmaengine: Add metadat_ops for dma_async_tx_descriptor

On 30-07-18, 12:46, Peter Ujfalusi wrote:
> Vinod,
> 
> On 2018-07-24 14:14, Vinod wrote:
> >>>> Clients must not mix the two way of handling the metadata.
> >>>> The set_len() is intended to tell the DMA driver the client provided
> >>>> metadata size (in MEM_TO_DEV case mostly).
> >>>>
> >>>> MEM_TO_DEV flow on client side:
> >>>> get_ptr()
> >>>> fill in the metadata to the pointer (not exceeding max_len)
> >>>> set_len() to tell the DMA driver the amount of valid bytes written
> >>>>
> >>>> DEV_TO_MEM flow on client side:
> >>>> In the completion callback, get_ptr()
> >>>> the metadata is payload_len bytes and can be accessed in the return pointer.
> >>>
> >>> I would think to unify this..
> >>
> >> I have tried it, but the attach mode and the pointer mode is hard to
> >> handle with a generic API.
> >> I will try to find a way to unify things in a sane way.
> > 
> > Hmmm, looking from the description they will be for different methods,
> > so lets make them orthogonal and not allow driver to register both.
> 
> I would allow DMA drivers to register both, but somehow enforce that
> clients are not mixing the two distinct way of dealing with the metadata.
> 
> The reason for that is for example the attach mode is the simplest (I
> implemented it first and I have a client using it), but if the pointer
> mode is found to be more efficient and feasible for the DMA then the DMA
> driver can implement that mode and the client can move as well w/o
> breaking anything.

Sounds reasonable...

-- 
~Vinod