Received-SPF: pass (google.com: domain of linux-wireless-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <CAB3_1NtsSrka20wLocDkR2dz9vrY447+WK2P3hQ1UN9bynGGCg@mail.gmail.com>
In-Reply-To: <CAB3_1NtsSrka20wLocDkR2dz9vrY447+WK2P3hQ1UN9bynGGCg@mail.gmail.com>
From:   Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date:   Tue, 22 Nov 2022 23:27:48 +0100
Message-ID: <CAFBinCAvF5FgXRa-Bin6O6e-nua32NSOwsreJQ_EpNt84hZ2QQ@mail.gmail.com>
Subject: Re: Throughput related to RTL8822CS
To:     Nitin Gupta <nitin.gupta981@gmail.com>
Cc:     linux-wireless@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hello Nitin,

On Tue, Nov 15, 2022 at 3:26 PM Nitin Gupta <nitin.gupta981@gmail.com> wrote:
>
> Dear All
>
> I am new and Facing throughput issue for RTL8822CS chipset
> Same kind of issue observed in mail which is addressed by Martin .
Nitin is referring to my comment from [0] where I wrote:
> [...] it's not fast yet, in my tests I got ~6Mbit/s in either direction.

There is no major news on this issue. My first goal is to make it work
upstream, then we can improve performance.

> Please let me know if that patch is good to go for Improving performance
> also I came across below error which frequently coming any idea or hint which direction i need to check
>
> Error :-
> 3.734608] RTW: ERROR sdio_io: bSurpriseRemoved, skip write 0x00041, 1 bytes
> [   33.734609] RTW: ERROR _halmac_sdio_cmd52_write: I/O FAIL!
Nitin, I think the RTL8822CS driver which you are testing was written
by (or for) Realtek. It's an out of tree (meaning:
maintained/developed separately from the Linux kernel).
All of my efforts are for the "upstream" rtw88 driver (meaning: it's
part of the Linux kernel directly). That driver is also written by
Realtek employees but it's a completely different codebase.
Personally I don't know much about this out of tree driver - I'm only
using it to understand the hardware/firmware interface that the driver
communicates with.

If "surprise removed" kicks in you may have some SDIO communication
issues. Have you tried lowering the SDIO bus frequency (on systems
with device-tree: max-frequency property of the SDIO controller)?
Another reason could be some incorrectly set up CD (card detect) line.
Since you haven't mentioned what board and card this is it's
impossible to say for sure.


Best regards,
Martin


[0] https://lore.kernel.org/netdev/20211228211501.468981-1-martin.blumenstingl@googlemail.com/