linux-kernel - Re: [PATCH 03/31] wifi: mwifiex: drop HostCmd_CMD_802_11_MAC

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: <Zs0FW-LPW0ShGXV5@google.com>
Date: Mon, 26 Aug 2024 15:44:43 -0700
From: Brian Norris <briannorris@...omium.org>
To: Sascha Hauer <s.hauer@...gutronix.de>
Cc: Kalle Valo <kvalo@...nel.org>, linux-wireless@...r.kernel.org,
	Francesco Dolcini <francesco@...cini.it>, kernel@...gutronix.de,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 03/31] wifi: mwifiex: drop HostCmd_CMD_802_11_MAC_ADDRESS
 response handling

Hi Sascha,

On Mon, Aug 26, 2024 at 11:07:03AM +0200, Sascha Hauer wrote:
> On Thu, Aug 22, 2024 at 11:07:35AM -0700, Brian Norris wrote:
> > Hi Sascha,
> > 
> > On Tue, Aug 20, 2024 at 01:55:28PM +0200, Sascha Hauer wrote:
> > > The command response handler copies the new MAC address over to
> > > priv->curr_addr. The same is done in the code issuing the call
> > > already, so drop the unnecessary HostCmd_CMD_802_11_MAC_ADDRESS
> > > handling.
> > 
> > It took a bit to figure out what you meant here -- I guess you're
> > referring to mwifiex_set_mac_address()? It could help to document what
> > you mean.
> 
> Ok, I can clarify this a bit when sending this next time.
> 
> Right now what we have is:
> 
> 1) mwifiex_set_mac_address() sets priv->curr_addr to the desired new MAC
>    address
> 2) mwifiex_cmd_802_11_mac_address() (called from mwifiex_send_cmd())
>    constructs the HostCmd_CMD_802_11_MAC_ADDRESS command, using the MAC
>    address in priv->curr_addr
> 3) mwifiex_ret_802_11_mac_address(), called from the response handler,
>    sets priv->curr_addr to the MAC address received with the command
>    response, which of course is the same as we initially copied there
>    in step 1), which makes 3) redundant and unnecessary

Ack, that's the understanding I got, but it took a bit of reading to get
there.

> > I'm also a bit torn; this command API ostensibly has a (unused so far,
> > for this command) HostCmd_ACT_GEN_GET mode, in which case this *is*
> > important.
> > 
> > If anything, I might consider dropping some of the handling in
> > mwifiex_set_mac_address(), because it seems to presume (and then has to
> > undo for failure) behavior of the underlying command.
> 
> What we could do instead of dropping 3) is:
> 
> 1) pass the new MAC address in the data_buf argument to
>    mwifiex_send_cmd()
> 2) instead of priv->curr_addr use data_buf in
>    mwifiex_cmd_802_11_mac_address()
> 
> With this the response handler would still set priv->curr_addr in case
> the command went through successfully. No need to undo priv->curr_addr
> to the previous MAC address in case the command failed.
> 
> Sounds good to me. Is that where you aiming at?

Yes, that seems about right.

Brian