linux-kernel - Re: [PATCH] wifi: ath11k: fix layout of scan_flags in struct scan_req

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: <CYFAYRP5MWTZ.Q272WWLLE7MW@gmail.com>
Date: Mon, 15 Jan 2024 14:09:28 +0100
From: "Nicolas Escande" <nico.escande@...il.com>
To: "Jeff Johnson" <quic_jjohnson@...cinc.com>, "Kalle Valo"
 <kvalo@...nel.org>
Cc: <linux-wireless@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
 <ath11k@...ts.infradead.org>
Subject: Re: [PATCH] wifi: ath11k: fix layout of scan_flags in struct
 scan_req_params

On Thu Nov 30, 2023 at 9:24 AM CET, Nicolas Escande wrote:
> On Tue Nov 28, 2023 at 1:57 AM CET, Jeff Johnson wrote:
> > On 11/27/2023 2:54 PM, Nicolas Escande wrote:
> [...]
> > > So either we should not use WMI_SCAN_XXX with scan_req_params.scan_flags ever
> > > and only use the bitfield to set scan parameters or if we use WMI_SCAN_XXX with
> > > scan_req_params.scan_flags they need to match the corresponding bitfield.
> >
> > IMO the correct thing to do is to remove the unions from that struct and
> > only leave behind the bitfields and not use the WMI_SCAN_XXX masks
> > except when filling the firmware structure.
> >
> > But don't spin an update to your patches until Kalle has a chance to
> > give his opinion. I'm new to maintaining these drivers and Kalle may
> > have a different opinion on this.
> >
> > /jeff
>
> No problem, I'll wait for Kalle's input on this before doing anything.
> As soon as we decide which way is the right way, I'll work on this. I only care
> that this gets resolved.

Hi Kalle/Jeff,

Any new input on this so I can move forward on fixing this ?
Otherwise I think I'll end up going on with Jeff's proposal of only using the
bitfield for intra driver representation & then converting the bitfields to
their corresponding WMI_SCAN_XXX when transmiting the req to the hw with wmi.