netdev - Re: [PATCH 2/2] wifi: ath11k: use unique QRTR instance ID

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]

Date:   Wed, 26 Apr 2023 14:40:32 +0200
From:   Robert Marko <robimarko@...il.com>
To:     Kalle Valo <kvalo@...nel.org>
Cc:     Robert Marko <robert.marko@...tura.hr>,
        Manivannan Sadhasivam <mani@...nel.org>, davem@...emloft.net,
        edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com,
        gregkh@...uxfoundation.org, elder@...aro.org,
        hemantk@...eaurora.org, quic_jhugo@...cinc.com,
        quic_qianyu@...cinc.com, bbhatt@...eaurora.org,
        mhi@...ts.linux.dev, linux-arm-msm@...r.kernel.org,
        ath11k@...ts.infradead.org, linux-wireless@...r.kernel.org,
        netdev@...r.kernel.org, ansuelsmth@...il.com
Subject: Re: [PATCH 2/2] wifi: ath11k: use unique QRTR instance ID

> Still not sure what you mean. Are you saying that this patch under
> discussion ("wifi: ath11k: use unique QRTR instance ID") also works with
> QCA6390 and it's possible to connect two QCA6390 devices on the same
> host?
>
> Or are you referring to some other hack? Or have I totally
> misunderstood? :)

We probably have a misunderstanding, QCA6390 does not work with
("wifi: ath11k: use unique QRTR instance ID"), that is why we in OpenWrt
limited it to QCN9074 only so far.

>
> > so that is why its quite important for OpenWrt to have a generic
> > solution that works on all cards.
>
> I fully agree on importance of having a generic solution. It's just sad
> that it seems people who designed this didn't consider about having
> multiple devices on the same host. It looks like there's no easy way to
> implement a generic solution, we have only bad choices to choose from.
> Your solution[1] is racy and writing to a register which is marked as
> read-only in the spec.

I agree, this is purely a hack based on what QCA is doing downstream where
they hardcode the QRTR ID in DTS and write to the same register.

>
> Qualcomm's solution[2] needs changes in firmware and it's uncertain if
> I'm able to convince all firmware teams to implement the support.
> (Currently only QCN9074 firmware supports this.)
>
> Thoughts?

I mean, we need some kind of a solution cause trying to pitch using a QCA
AX SoC-s and PCI cards but then saying that they cannot use AHB+PCI
or multiple PCI cards at the same time are not viable.

Regards,
Robert
>
> [1] https://patchwork.kernel.org/project/linux-wireless/patch/20221105194943.826847-2-robimarko@gmail.com/
>
> [2] https://patchwork.kernel.org/project/linux-wireless/patch/20230111170033.32454-1-kvalo@kernel.org/
>
> --
> https://patchwork.kernel.org/project/linux-wireless/list/
>
> https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches