Possible race while masking IRQ on Allwinner A20

From: "Jérôme Pouiller" <jerome.pouiller@silabs.com>
To: linux-arm-kernel@lists.infradead.org
Cc: Marc Dorval <marc.dorval@silabs.com>,
	Chen-Yu Tsai <wens@csie.org>, Maxime Ripard <mripard@kernel.org>
Subject: Possible race while masking IRQ on Allwinner A20
Date: Tue, 19 May 2020 10:59:26 +0200	[thread overview]
Message-ID: <11042983.UNsANRFJuY@pc-42> (raw)

Hello arm developers,

I have some trouble with integration of the wfx driver[1] on Allwinner
A20 platform.

The chip WF200 is connected to the SDIO bus. At the beginning, I tried 
to use the IRQ provided by the SDIO bus. However, I have noticed I 
received some IRQs twice. Since the IRQ line is multiplexed with the 
data line, it is not very clear if it is a bug, or if the SDIO device 
has to support that.

The chip WF200 allows to use a dedicated line for the IRQ (aka
"Out-Of-Band" IRQ). So I have enabled this feature with a edge triggered 
IRQ. However, I missed some IRQs. Indeed, it seems that Allwinner use a 
32KHz clock to sample the IRQs. It is not fast enough for us. I think it 
explains why we miss some IRQs (using the attribute "input-debounce"[2], 
I tried to enable the 24Mhz clock, but without success).

Nevermind, I tried to use a level triggered IRQ (and my request is on 
this part). As you can see in the wfx driver (in  bus_sdio.c and bh.c), 
I use a threaded IRQ for that. Unfortunately, I receive some IRQs twice. 
I traced the problem, I get:

 QSGRenderThread-981   [000] d.h.   247.485524: irq_handler_entry: irq=80 name=wfx
 QSGRenderThread-981   [000] d.h.   247.485547: irq_handler_exit: irq=80 ret=handled
 QSGRenderThread-981   [000] d.h.   247.485600: irq_handler_entry: irq=80 name=wfx
 QSGRenderThread-981   [000] d.h.   247.485606: irq_handler_exit: irq=80 ret=handled
      irq/80-wfx-260   [001] ....   247.485828: io_read32: CONTROL: 0000f046
      irq/80-wfx-260   [001] ....   247.486072: io_read32: CONTROL: 0000f046
    kworker/1:1H-116   [001] ....   247.486214: io_read: QUEUE: 8b 00 84 18 00 00 00 00 01 00 15 82 2b 48 01 1e 88 42 30 00 08 6b d7 c3 53 e0 28 80 88 67 32 af ... (192 bytes)
    kworker/1:1H-116   [001] ....   247.493097: io_read: QUEUE: 00 00 00 00 00 00 00 00 06 06 00 6a 3f 95 00 60 00 00 00 00 08 62 00 00 01 00 5e 00 00 07 28 80 ... (192 bytes)
    [...]

On this trace, we can see:
  - the hard IRQ handler
  - the IRQ acknowledge from the thread irq/80-wfx-260
  - the access to the data from kworker/1:1H-116

As far as I understand, the first call to the IRQ handler (at 
247.485524) should mask the IRQ 80. So, the second IRQ (at 247.485600) 
should not happen and the thread irq/80 should be triggered only once.

Do you have any idea of what is going wrong with this IRQ?

Thank you,

[1] available in drivers/staging/wfx of the staging-next branch of the
Greg KH's repository. Be sure you get the up-to-date version.

[2] see commit 7c926492d38a3feef4b4b29c91b7c03eb1b8b546 for detail about 
the interactions between clock and debouncing.

-- 
Jérôme Pouiller

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel