linux-kernel - Re: [PATCH] slimbus: Linux driver framework for SLIMbus.

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: <20120604104233.GG7538@opensource.wolfsonmicro.com>
Date:	Mon, 4 Jun 2012 11:42:33 +0100
From:	Mark Brown <broonie@...nsource.wolfsonmicro.com>
To:	Sagar Dharia <sdharia@...eaurora.org>
Cc:	davidb@...eaurora.org, bryanh@...eaurora.org,
	kheitke@...eaurora.org, gclemson@...ience.com,
	linux-arm-msm@...r.kernel.org,
	linux-arm-kernel@...ts.infradead.org, rob@...dley.net,
	grant.likely@...retlab.ca, rob.herring@...xeda.com,
	ohad@...ery.com, gregkh@...uxfoundation.org,
	linus.walleij@...aro.org, rusty@...tcorp.com.au,
	joerg.roedel@....com, trenn@...e.de, ak@...ux.intel.com,
	linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
	devicetree-discuss@...ts.ozlabs.org
Subject: Re: [PATCH] slimbus: Linux driver framework for SLIMbus.

On Mon, Jun 04, 2012 at 03:36:55AM -0700, Sagar Dharia wrote:

> I would expect that the slim_device's driver will only power-on the device
> during probe (and use wait_for_completion during 1st transfer to get LA).
> Typically transfers are not done as part of probe. Even if transfers need

I'd really expect that we'll have devices that we want to interact with
during probe, for example to determine the device variant.  I'd also
expect that it'd be useful to defer things like registering the device
with higher level APIs until we've actually got it up and running, it
tends to make life simpler.

> to be done as part of probe, I expect wait_for_completion (with timeout to
> avoid potential HW problems causing linux probe to wait forever) will be
> better than polling for get_logical_addr.

I agree that a completion is better if we have to block in the device
driver, but the idea someone suggested of having the framework generate
a second callback when it's assigned a logical address seems even
better.

Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)