linux-kernel - Re: [RFC/RFT] p54spi: Convert driver to use asynchronous firmware loading

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: <4F3858BD.3040707@lwfinger.net>
Date:	Sun, 12 Feb 2012 18:26:37 -0600
From:	Larry Finger <Larry.Finger@...inger.net>
To:	Max Filippov <jcmvbkbc@...il.com>
CC:	chunkeey@....de, m@...s.ch, linux-kernel@...r.kernel.org,
	devel@...verdev.osuosl.org, linux-wireless@...r.kernel.org
Subject: Re: [RFC/RFT] p54spi: Convert driver to use asynchronous firmware
 loading

On 02/12/2012 06:20 PM, Max Filippov wrote:
>> Drivers that load firmware from their probe routine have problems with the
>> latest versions of udev as they get timeouts while waiting for user
>> space to start. The problem is fixed by using request_firmware_nowait()
>> and delaying the start of mac80211 until the firmware is loaded.
>>
>> To prevent the possibility of the driver being unloaded while the firmware
>> loading callback is still active, a completion queue entry is used.
>>
>> Signed-off-by: Larry Finger<Larry.Finger@...inger.net>
>> ---
>>
>> This conversion of p54spi to use asynchronous firmware loading is based
>> on the method used in p54usb. As I do not have the hardware, it is only
>> compile tested. I would appreciate any feedback from people that have the
>> hardware.
>
> Hi, Larry.
>
> Are there any prerequisites for this patch?
> For now I'm applying it to the linux-omap ToT and having the following:

No, there are no prerequisites; however, as Michael Buesch noted, I missed the 
eeprom loading. This patch will not work.

I am currently testing an approach that is much simpler. Rather than using 
request_firmware_nowait(), I am having the probe routine start a work queue, and 
let that routine do a normal request_firmware(). This form of the patch is a lot 
less intrusive.

I hope to post a revised patch within a day.

Thanks for testing.

Larry




--
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/