From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id D2EF2C433EF for ; Wed, 12 Jan 2022 20:08:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S242167AbiALUI6 (ORCPT ); Wed, 12 Jan 2022 15:08:58 -0500 Received: from mxout02.lancloud.ru ([45.84.86.82]:43256 "EHLO mxout02.lancloud.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236213AbiALUI5 (ORCPT ); Wed, 12 Jan 2022 15:08:57 -0500 Received: from LanCloud DKIM-Filter: OpenDKIM Filter v2.11.0 mxout02.lancloud.ru 104C0209B125 Received: from LanCloud Received: from LanCloud Received: from LanCloud Subject: Re: [PATCH] platform: finally disallow IRQ0 in platform_get_irq() and its ilk To: Marc Zyngier CC: Greg Kroah-Hartman , "Rafael J. Wysocki" , , Thomas Gleixner , Geert Uytterhoeven References: <5e001ec1-d3f1-bcb8-7f30-a6301fd9930c@omp.ru> <87pmp7volh.wl-maz@kernel.org> <9c3f9a6e-89b6-b73b-032f-e4b4ee6f72ac@omp.ru> <58caf8b54828bcc25cfc65220d29b68f@kernel.org> From: Sergey Shtylyov Organization: Open Mobile Platform Message-ID: Date: Wed, 12 Jan 2022 23:08:53 +0300 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.10.1 MIME-Version: 1.0 In-Reply-To: <58caf8b54828bcc25cfc65220d29b68f@kernel.org> Content-Type: text/plain; charset="utf-8" Content-Language: en-US Content-Transfer-Encoding: 8bit X-Originating-IP: [192.168.11.198] X-ClientProxiedBy: LFEXT02.lancloud.ru (fd00:f066::142) To LFEX1907.lancloud.ru (fd00:f066::207) Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 1/12/22 9:08 PM, Marc Zyngier wrote: [...] >>>> The commit a85a6c86c25b ("driver core: platform: Clarify that IRQ 0 is >>>> invalid") only calls WARN() when IRQ0 is about to be returned, however >>>> using IRQ0 is considered invalid (according to Linus) outside the arch/ >>>> code where it's used by the i8253 drivers. Many driver subsystems treat >>>> 0 specially (e.g. as an indication of the polling mode by libata), so >>>> the users of platform_get_irq[_byname]() in them would have to filter >>>> out IRQ0 explicitly and this (quite obviously) doesn't scale... >>>> Let's finally get this straight and return -EINVAL instead of IRQ0! >>>> >>>> Fixes: a85a6c86c25b ("driver core: platform: Clarify that IRQ 0 is invalid") >>>> Signed-off-by: Sergey Shtylyov >>>> >>>> --- >>>> The patch is against the 'driver-core-linus' branch of Greg Kroah-Hartman's >>>> 'driver-core.git' repo. >>>> >>>>  drivers/base/platform.c |    6 ++++-- >>>>  1 file changed, 4 insertions(+), 2 deletions(-) >>>> >>>> Index: driver-core/drivers/base/platform.c >>>> =================================================================== >>>> --- driver-core.orig/drivers/base/platform.c >>>> +++ driver-core/drivers/base/platform.c >>>> @@ -231,7 +231,8 @@ int platform_get_irq_optional(struct pla >>>>  out_not_found: >>>>      ret = -ENXIO; >>>>  out: >>>> -    WARN(ret == 0, "0 is an invalid IRQ number\n"); >>>> +    if (WARN(!ret, "0 is an invalid IRQ number\n")) >>>> +        return -EINVAL; >>>>      return ret; >>>>  } >>>>  EXPORT_SYMBOL_GPL(platform_get_irq_optional); >>>> @@ -445,7 +446,8 @@ static int __platform_get_irq_byname(str >>>> >>>>      r = platform_get_resource_byname(dev, IORESOURCE_IRQ, name); >>>>      if (r) { >>>> -        WARN(r->start == 0, "0 is an invalid IRQ number\n"); >>>> +        if (WARN(!r->start, "0 is an invalid IRQ number\n")) >>>> +            return -EINVAL; >>>>          return r->start; >>>>      } >>> >>> Geert recently mentioned that a few architectures (such as sh?) still >>> use IRQ0 as something valid in limited cases. >>> >>> From my PoV, this patch is fine, but please be prepared to fix things >>> in a couple of years when someone decides to boot a recent kernel on >>> their pet dinosaur. With that in mind: >>> >>> Acked-by: Marc Zyngier >> >>    Greg, so would that ACK be enough? Is there a chance this patch >> gets finally included >> into 5.17-rc1? Or should I look into fixing the recently found >> arch/sh/ issue 1st (as you >> can see, just WARN()'ing about IRQ0 wasn't enough to get this fixed)? > > Fixing SH would be a good thing. Who argues with that? :-) However, I don't think it should be a pre-requisite for this patch, so that we have extra time until 5.17 final... actually, I had couple quick workarounds in mind; the problem however is that we don't seem to have the targets for testing... :-( > Thanks, > >         M. MBR, Sergey