[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <44c057f5-d229-86f1-adc4-5585334d3113@sifive.com>
Date: Wed, 19 Jul 2023 11:55:12 -0500
From: Samuel Holland <samuel.holland@...ive.com>
To: Andy Shevchenko <andy@...nel.org>
Cc: Linus Walleij <linus.walleij@...aro.org>,
Bartosz Golaszewski <brgl@...ev.pl>,
Emil Renner Berthing <emil.renner.berthing@...onical.com>,
Rob Herring <robh+dt@...nel.org>,
Frank Rowand <frowand.list@...il.com>,
Palmer Dabbelt <palmer@...belt.com>,
Paul Walmsley <paul.walmsley@...ive.com>,
linux-gpio@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-riscv@...ts.infradead.org
Subject: Re: [PATCH v2 2/4] gpio: sifive: Look up IRQs only once during probe
On 2023-07-19 11:48 AM, Andy Shevchenko wrote:
> On Wed, Jul 19, 2023 at 09:34:43AM -0700, Samuel Holland wrote:
>> of_irq_count(), or eqivalently platform_irq_count(), simply looks up
>> successively-numbered IRQs until that fails. Since this driver needs to
>> look up each IRQ anyway to get its virq number, use that existing loop
>> to count the IRQs at the same time.
>
> ...
>
>> - ngpio = of_irq_count(node);
>> - if (ngpio > SIFIVE_GPIO_MAX) {
>> - dev_err(dev, "Too many GPIO interrupts (max=%d)\n",
>> - SIFIVE_GPIO_MAX);
>> - return -ENXIO;
>
> Do we still need this check?
I don't think so. I think this check was only intended to prevent overflowing
the irq_number array, and that is now handled by the loop condition.
>> - }
>
> ...
>
>> + for (ngpio = 0; ngpio < SIFIVE_GPIO_MAX; ngpio++) {
>> + ret = platform_get_irq_optional(pdev, ngpio);
>> if (ret < 0)
>> - return ret;
>> - chip->irq_number[i] = ret;
>> + break;
>> + chip->irq_number[ngpio] = ret;
>> }
>
> If so, here we need something like
>
> ret = platform_get_irq_optional(pdev, ngpio);
> if (ret > 0) {
> dev_err(dev, "Too many GPIO interrupts (max=%d)\n",
> SIFIVE_GPIO_MAX);
> return -ENXIO;
> }
>
> Otherwise you need to mention this relaxation in the commit message.
OK, I will add something to the commit message in v3.
Powered by blists - more mailing lists