linux-kernel - Re: [PATCH 2/2] DMAENGINE: generic channel status v2

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]

Date:	Sat, 27 Mar 2010 09:32:52 +0100
From:	Linus Walleij <linus.ml.walleij@...il.com>
To:	Dan Williams <dan.j.williams@...el.com>
Cc:	Linus Walleij <linus.walleij@...ricsson.com>,
	Guennadi Liakhovetski <g.liakhovetski@....de>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"Sosnowski, Maciej" <maciej.sosnowski@...el.com>,
	Nicolas Ferre <nicolas.ferre@...el.com>,
	Pavel Machek <pavel@....cz>, Li Yang <leoli@...escale.com>,
	Paul Mundt <lethal@...ux-sh.org>,
	Ralf Baechle <ralf@...ux-mips.org>,
	Haavard Skinnemoen <haavard.skinnemoen@...el.com>,
	Magnus Damm <damm@...nsource.se>,
	Liam Girdwood <lrg@...mlogic.co.uk>,
	Joe Perches <joe@...ches.com>,
	Roland Dreier <rdreier@...co.com>,
	Richard Röjfors <richard.rojfors@...agicore.com>,
	"Steven J. Magnani" <steve@...idescorp.com>
Subject: Re: [PATCH 2/2] DMAENGINE: generic channel status v2

2010/3/27 Dan Williams <dan.j.williams@...el.com>:

> Sidenote: one thing I notice via this exercise is that not every driver
> is triggering a descriptor cleanup in their device_tx_status()
> implementation.  It's really only needed for the drivers that will
> service requests from the async_tx api (now the minority), but something
> to keep in mind.  In some cases async_tx polls for completion on
> descriptor allocation failure, so either ->device_tx_status() or
> ->device_prep* needs to run descriptor reclaim.

Not that I designed this thing, but the idea of device_tx_status()
cleaning things up seems a bit kludgy, what about adding a new
control command to device_control() instead, like
DMA_CLEANUP_TX and require that you call this after a
polling async transfer has been found to be complete?

Is this thing supposed to be used for DMA hardware that cannot
trigger interrupts on completion by the way? That seems like the
natural reason...

Linus Walleij
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/