Re: QCA9074 problem on rk3568

From: Kalle Valo <kvalo@kernel.org>
To: Doug Brewer <brewer.doug@gmail.com>
Cc: ath11k@lists.infradead.org
Subject: Re: QCA9074 problem on rk3568
Date: Wed, 26 Apr 2023 10:28:21 +0300	[thread overview]
Message-ID: <87edo7t84q.fsf@kernel.org> (raw)
In-Reply-To: <CAG0V13SNRwhyyKMBxL71-u5mGH6kiagXecebage7xqCq7N3q3w@mail.gmail.com> (Doug Brewer's message of "Fri, 21 Apr 2023 09:35:34 +0800")

Doug Brewer <brewer.doug@gmail.com> writes:

> On Thu, Apr 20, 2023 at 8:16 PM Kalle Valo <kvalo@kernel.org> wrote:
>
>>
>> Doug Brewer <brewer.doug@gmail.com> writes:
>>
>> > I tried to install linux kernel 6.3-rc7 on my rk3568 evb.  It seems there's
>> > a problem about MHI:
>> >
>> > [   11.340558] ath11k_pci 0002:01:00.0: BAR 0: assigned [mem 0x380000000-0x3801f
>> > ffff 64bit]
>> > [   11.341399] ath11k_pci 0002:01:00.0: enabling device (0000 -> 0002)
>> > [   11.422849] ath11k_pci 0002:01:00.0: MSI vectors: 16
>> > [   11.423335] ath11k_pci 0002:01:00.0: qcn9074 hw1.0
>> > [   11.660846] mhi mhi0: Requested to power ON
>> > [   11.661271] mhi mhi0: Power on setup success
>> > [   31.716289] mhi mhi0: MHI did not load image over BHI, ret: -5
>> > [   60.551942] mhi mhi0: Device failed to enter READY state
>> > [   60.552544] mhi mhi0: Error moving from PM state: Firmware Download Error to:
>> > DISABLE
>> > [   60.553376] ath11k_pci 0002:01:00.0: failed to power up mhi: -110
>> > [   60.553972] ath11k_pci 0002:01:00.0: failed to start mhi: -110
>> > [   60.554532] ath11k_pci 0002:01:00.0: failed to power up :-110
>> > [   60.567907] ath11k_pci 0002:01:00.0: failed to create soc core:
>> > -110
>> > [   60.568488] ath11k_pci 0002:01:00.0: failed to init core: -110
>> >
>> > Any idea how to fix it?  Thanks.
>>
>> I would first check that the firmware files are not corrupted.
>
> # sha256sum amss.bin board-2.bin m3.bin
> 5ee1b7b204541b5f99984f21d694ececaec08fbce1b520ffe6fe740b02a4afd7  amss.bin
> dbf0ca14aa1229eccd48f26f1026901b9718b143bd30b51b8ea67c84ba6207f1  board-2.bin
> b6d957f335073a15a8de809398e1506f0200a08747eaf7189c843cf519ffc1de  m3.bin
>
> The firmware files look good, thanks.

Indeed, the firmware files don't seem to be corrupted. This is a weird
problem, I don't recall seeing anything like this before. Some other
ideas to try out:

* if you have access to the Qualcomm in-house QSDK driver (don't know
  what they call it nowadays) trying that would be good for ruling out
  hardware problems

* try older kernel

* try older firmware

* try the QCN9074 on a different platform

But these are just shots in the dark, I can't really guess what the
problem could be. The error comes from the very early during driver
initialisation so most likely a low level issue. I don't have QCN9074 in
my testbed so of course this could be a simple regression somewhere as
well.

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

-- 
ath11k mailing list
ath11k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath11k