linux-kernel - Re: rtl8821ae.

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:	Tue, 04 Feb 2014 17:05:59 -0600
From:	Larry Finger <Larry.Finger@...inger.net>
To:	andrea.merello@...il.com
CC:	Dan Carpenter <dan.carpenter@...cle.com>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Stefan Lippers-Hollmann <s.L-H@....de>,
	Dave Jones <davej@...hat.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Linux Wireless List <linux-wireless@...r.kernel.org>,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	Linux Driver Project <devel@...uxdriverproject.org>
Subject: Re: rtl8821ae.

On 02/04/2014 01:57 PM, Andrea Merello wrote:
> Sorry, my fault.. you had already reported me the iowrite16 vs
> iowrite8 for anaparam3 issue, but I forgot to fix it :(
>
> If you are reasonably sure that a 16bits-wide write is OK on the
> rtl8187b hardware (In reference code I have here only 8bits are
> written), then I agree applying your fix, and I will leave unchanged
> my modifications.
> Otherwise, if you are unsure, I will revert thing in the common
> rtl818x struct (switching back to 8bit type) and I simply make a cast
> on rtl8187se, or I will use an union if possible, so the rtl8187b
> driver remains unchanged.
>
> On the other hand, for the remaining issues with changed names for
> bitfield (the XXXX_SHIFT), I wil vote to fix the rtl8187 driver:
> That names IMHO are wrong (misleading) and while fixing them should
> help avoid future confusion, the fix is trivial enough that I'm really
> confident not to broke anything.
>
> I had already sent a patch for it:

Somehow I missed this patch. Do you know if John Linville picked it up?

I'm not sure if the ANAPARAM3 register could handle a 16-bit write on an 
RTL8187B. To be cautious, I wrote and have tested the attached patch using a 
union. The patch includes a fix for an undefined symbol 
(RTL818X_TX_CONF_HW_SEQNUM). With this patch my RTL8187SE and RTL8187B devices 
both work.

Larry


View attachment "fix_anaparam3" of type "text/plain" (1687 bytes)