linux-kernel - Re: [PATCH v2 2/8] pinctrl: stm32: Introduce HDP driver

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: <929f6f1f-d47f-4954-8c8b-91c932127975@foss.st.com>
Date: Wed, 21 May 2025 12:14:08 +0200
From: Clement LE GOFFIC <clement.legoffic@...s.st.com>
To: Linus Walleij <linus.walleij@...aro.org>
CC: Rob Herring <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>,
        Conor Dooley <conor+dt@...nel.org>,
        Maxime Coquelin
	<mcoquelin.stm32@...il.com>,
        Alexandre Torgue <alexandre.torgue@...s.st.com>,
        Bartosz Golaszewski <brgl@...ev.pl>, <linux-kernel@...r.kernel.org>,
        <linux-gpio@...r.kernel.org>, <devicetree@...r.kernel.org>,
        <linux-stm32@...md-mailman.stormreply.com>,
        <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH v2 2/8] pinctrl: stm32: Introduce HDP driver

On 5/21/25 11:49, Clement LE GOFFIC wrote:
> On 5/21/25 00:34, Linus Walleij wrote:
>> Hi Clément,
>>
>> thanks for your patch!
>>
>> On Tue, May 20, 2025 at 5:04 PM Clément Le Goffic
>> <clement.legoffic@...s.st.com> wrote:
>>
>>> This patch introduce the driver for the Hardware Debug Port available on
>>> STM32MP platforms. The HDP allows the observation of internal SoC
>>> signals by using multiplexers. Each HDP port can provide up to 16
>>> internal signals (one of them can be software controlled as a GPO).
>>>
>>> Signed-off-by: Clément Le Goffic <clement.legoffic@...s.st.com>
>>
>> (...)
>>> +static int stm32_hdp_gpio_get_direction(struct gpio_chip *gc, 
>>> unsigned int offset)
>>> +{
>>> +       return GPIO_LINE_DIRECTION_OUT;
>>> +}
>>
>> That's reasonable.
>>
>>> +static int stm32_hdp_gpio_get(struct gpio_chip *gc, unsigned int 
>>> offset)
>>> +{
>>> +       struct stm32_hdp *hdp = gpiochip_get_data(gc);
>>> +
>>> +       if (((hdp->mux_conf & HDP_MUX_MASK(offset))) == 
>>> HDP_MUX_GPOVAL(offset))
>>> +               return !!(readl_relaxed(hdp->base + HDP_GPOVAL) & 
>>> BIT(offset));
>>> +       else
>>> +               return !!(readl_relaxed(hdp->base + HDP_VAL) & 
>>> BIT(offset));
>>> +}
>>
>> ...but you still make it possible to read the value of the line
>> if it's not muxed as GPO?
>>
>> Should it not stm32_hdp_gpio_get_direction() return
>> GPIO_LINE_DIRECTION_IN if HDP_MUX_MASK(offset))) != 
>> HDP_MUX_GPOVAL(offset)?
> 
> Hi, oops, you're right !

I've answer too quickly. All the pins are "out" in fact.
The hardware is not able to read input signal.
I wanted to implement a way to read the register value in any case by 
leaving the line direction unchanged.

>>
>>> +static void stm32_hdp_gpio_set(struct gpio_chip *gc, unsigned int 
>>> offset, int value)
>>> +{
>>> +       struct stm32_hdp *hdp = gpiochip_get_data(gc);
>>> +
>>> +       if (value)
>>> +               writel_relaxed(BIT(offset), hdp->base + HDP_GPOSET);
>>> +       else
>>> +               writel_relaxed(BIT(offset), hdp->base + HDP_GPOCLR);
>>> +}
>>
>> Can't you just use GPIO_GENERIC for this?
>>
>> bgpio_init(gc, dev, ARRAY_SIZE(stm32_hdp_pins), // == 8
>>      hdp->base + HDP_VAL,
>>      hdp->base + HDP_GPOSET,
>>      hdp->base + HDP_GPOCLR,
>>      NULL,
>>      NULL,
>>      0);
>>
>> The default behaviour of GPIO MMIO is to read the output register
>> for the value if the line is in output mode.
>>
>> You may wanna override the .get_direction() callback after bgpio_init()
>> and before registering the chip, either with what you have or what
>> I described above.
> 
> I didn't know about it, I'll take a look and provide a V3.
> 
> Thank you,
> 
> Clément
> 
>> Yours,
>> Linus Walleij
>