linux-kernel - Re: [PATCH v2] pinctrl: samsung: add irq_set_affinity() for non wake up external gpio interrupt

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <ZWg5/rorgfSM/5UO@perf>
Date:   Thu, 30 Nov 2023 16:30:45 +0900
From:   Youngmin Nam <youngmin.nam@...sung.com>
To:     Sam Protsenko <semen.protsenko@...aro.org>
Cc:     tomasz.figa@...il.com, krzysztof.kozlowski@...aro.org,
        s.nawrocki@...sung.com, alim.akhtar@...sung.com,
        linus.walleij@...aro.org, linux-arm-kernel@...ts.infradead.org,
        linux-samsung-soc@...r.kernel.org, linux-gpio@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] pinctrl: samsung: add irq_set_affinity() for non
 wake up external gpio interrupt

On Wed, Nov 29, 2023 at 12:48:56PM -0600, Sam Protsenko wrote:
> On Wed, Nov 29, 2023 at 12:32 AM Youngmin Nam <youngmin.nam@...sung.com> wrote:
> > > I tried to test this patch on E850-96, and an attempt to write into
> > > smp_affinity (for some GPIO irq) also fails for me:
> > >
> > >     # echo f0 > smp_affinity
> > >     -bash: echo: write error: Input/output error
> > >
> > > When I add some pr_err() to exynos_irq_set_affinity(), I can't see
> > > those printed in dmesg. So I guess exynos_irq_set_affinity() doesn't
> > > get called at all. So the error probably happens before
> > > .irq_set_affinity callback gets called.
> > >
> > > Youngmin, can you please try and test this patch on E850-96? This
> > > board is already supported in upstream kernel. For example you can use
> > > "Volume Up" interrupt for the test, which is GPIO irq.
> > >
> >
> > I intened this affinity setting would work only on *Non* Wakeup External Interrupt.
> > The "Volume Up" on E850-96 board is connected with "gpa0-7" and
> > that is Wakeup External interrupt so that we can't test the callback.
> >
> 
> Oh no. It was really silly mistake on my part :) But please check my
> answer below for good news.
> 
> > I couldn't find out a pin for the test on E850-96 board yet.
> > We can test if there is a usage of *Non" Wake up External Interrupt of GPIO
> > on E850-96 board.
> >
> > Do you have any idea ?
> >
> 
> Yep, just managed to successfully test your patch on E850-96. My
> testing procedure below might appear messy, but I didn't want to do
> any extra soldering :)
> 
> Used GPG1[0] pin for testing. As you can see from schematics [1],
> GPG1[0] is connected to R196 resistor (which is not installed on the
> board).
> 
> I've modified E850-96 dts file like this:
> 
> 8<---------------------------------------------------------------------------->8
>         gpio-keys {
>                 compatible = "gpio-keys";
>                 pinctrl-names = "default";
> -               pinctrl-0 = <&key_voldown_pins &key_volup_pins>;
> +               pinctrl-0 = <&key_voldown_pins &key_volup_pins &key_test_pins>;
> 
>                 ...
> 
> +               test-key {
> +                       label = "Test Key";
> +                       linux,code = <KEY_RIGHTCTRL>;
> +                       gpios = <&gpg1 0 GPIO_ACTIVE_LOW>;
> +               };
>         };
> 
> ...
> 
> +&pinctrl_peri {
> +       key_test_pins: key-test-pins {
> +               samsung,pins = "gpg1-0";
> +               samsung,pin-function = <EXYNOS_PIN_FUNC_EINT>;
> +               samsung,pin-pud = <EXYNOS_PIN_PULL_NONE>;
> +               samsung,pin-drv = <EXYNOS5420_PIN_DRV_LV1>;
> +       };
> +};
> 8<---------------------------------------------------------------------------->8
> 
> It appeared in /proc/interrupts as follows:
> 
>     87:          2          0          0          0          0
>  0          0          0      gpg1   0 Edge      Test Key
> 
> Then I touched R196 resistor pads with my finger (emulating key press)
> and looked again at /proc/interrupts:
> 
>     87:        444          0          0          0          0
>  0          0          0      gpg1   0 Edge      Test Key
> 
> Then I reconfigured smp_affinity like so:
> 
>     # cat /proc/irq/87/smp_affinity
>     ff
>     # echo f0 > /proc/irq/87/smp_affinity
>     # cat /proc/irq/87/smp_affinity
>     f0
> 
> Then I messed with R196 resistor pads with my finger again, and
> re-checked /proc/interrupts:
> 
>                CPU0       CPU1       CPU2       CPU3       CPU4
> CPU5       CPU6       CPU7
>      87:        444          0          0          0        234
>   0          0          0      gpg1   0 Edge      Test Key
> 
> And without this patch that procedure above of course doesn't work.
> 
> So as far as I'm concerned, feel free to add:
> 
> Tested-by: Sam Protsenko <semen.protsenko@...aro.org>
> 
> [1] https://protect2.fireeye.com/v1/url?k=d3eebafe-8c7293d4-d3ef31b1-000babe598f7-06c0db00473e1bca&q=1&e=855a917d-d0f9-49cd-8e05-7cccf44d06a8&u=https%3A%2F%2Fwww.96boards.org%2Fdocumentation%2Fconsumer%2Fe850-96b%2Fhardware-docs%2Ffiles%2Fe850-96b-schematics.pdf
> 
> Thanks!
> 
Thanks for your test Sam. It's really great work.
I confirmed the patch did work by following your test sequence as below.

* Before
     CPU0       CPU1       CPU2       CPU3       CPU4 ...
87:    40          0          0          0          0 ... gpg1   0 Edge   Test Key

* After
87:    40          0          0          0         22 ... gpg1   0 Edge   Test Key

Krzysztof,
Sam and I tested this patch on e850-95 board.
Let me update commit message with test result and will update patchset.

Thanks.