Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   andy.shevchenko@gmail.com
Date:   Wed, 10 May 2023 20:25:31 +0300
To:     Jean-Michel Hautbois <jeanmichel.hautbois@yoseli.org>
Cc:     linux-gpio@vger.kernel.org, linux-kernel@vger.kernel.org,
        brgl@bgdev.pl, linus.walleij@linaro.org
Subject: Re: pca953x issue when driving a DSI bridge
Message-ID: <ZFvTi3tQGUq2OCHi@surfacebook>
References: <0b3a257a-f1e5-ad86-4c69-93e038a33ce9@yoseli.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <0b3a257a-f1e5-ad86-4c69-93e038a33ce9@yoseli.org>
Precedence: bulk

Wed, May 10, 2023 at 06:12:19PM +0200, Jean-Michel Hautbois kirjoitti:
> Hello there !
> 
> I have a custom board, based on a i.MX8mm SoC which has a MIPI-DSI to eDP
> bridge (namely, a TI sn65dsi86). This bridge has a DSI enable pin, which is
> basically its reset pin, connected to my PCA9539 GPIO expander.
> 
> The issue is that this pin can't be sleeping, and it is tested in the
> gpiod_set_value() function.
> 
> Here is where it fails in my dmesg:

...

> [   11.273968]  gpiod_set_value+0x5c/0xcc
> [   11.277722]  ti_sn65dsi86_resume+0x4c/0x94 [ti_sn65dsi86]

Your problem even worse, i.e. ->resume() might sleep.

> [   11.283131]  __rpm_callback+0x48/0x19c
> [   11.286885]  rpm_callback+0x6c/0x80
> [   11.290375]  rpm_resume+0x3b0/0x660
> [   11.293864]  __pm_runtime_resume+0x4c/0x90
> [   11.297960]  __device_attach+0x90/0x1e4
> [   11.301797]  device_initial_probe+0x14/0x20
> [   11.305980]  bus_probe_device+0x9c/0xa4
> [   11.309817]  device_add+0x3d8/0x820
> [   11.313308]  __auxiliary_device_add+0x40/0xa0
> [   11.317668]  ti_sn65dsi86_add_aux_device.isra.0+0xb0/0xe0 [ti_sn65dsi86]
> [   11.324381]  ti_sn65dsi86_probe+0x20c/0x2ec [ti_sn65dsi86]
> [   11.329876]  i2c_device_probe+0x3b8/0x3f0
> [   11.333889]  really_probe+0xc0/0x3dc

...

> I suppose this is not a corner case and we may have other drivers and other
> boards connecting a GPIO which can sleep in a context where it should not ?
> 
> I would like to add one thing: on this board, the expander is routed in a
> way that makes it impossible to "sleep" as the reset is forced pulled-up and
> the power regulators are fixed and can't be stopped.

Can you elaborate why you think there is a problem?

> I don't know how to address this issue nicely and any thoughts is
> appreciated !

As a workaround you can consider the code around i2c_in_atomic_xfer_mode()
but since I have heard about i.MX8 so many negative remarks which makes me
think that hardware is a train wreck and shouldn't be used at all.

-- 
With Best Regards,
Andy Shevchenko