Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 25 Aug 2021 14:38:22 +0100
Message-ID: <87bl5lpru9.wl-maz@kernel.org>
From:   Marc Zyngier <maz@kernel.org>
To:     Barry Song <21cnbao@gmail.com>
Cc:     bhelgaas@google.com, tglx@linutronix.de,
        Jonathan.Cameron@huawei.com, bilbao@vt.edu, corbet@lwn.net,
        gregkh@linuxfoundation.org, leon@kernel.org,
        linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
        linuxarm@huawei.com, luzmaximilian@gmail.com,
        mchehab+huawei@kernel.org, schnelle@linux.ibm.com,
        jesse.brandeburg@intel.com, anthony.l.nguyen@intel.com,
        intel-wired-lan@lists.osuosl.org,
        Barry Song <song.bao.hua@hisilicon.com>
Subject: Re: [PATCH v3 3/3] PCI/MSI: remove msi_attrib.default_irq in msi_desc
In-Reply-To: <20210825102636.52757-4-21cnbao@gmail.com>
References: <20210825102636.52757-1-21cnbao@gmail.com>
        <20210825102636.52757-4-21cnbao@gmail.com>
User-Agent: Wanderlust/2.15.9 (Almost Unreal) SEMI-EPG/1.14.7 (Harue)
 FLIM-LB/1.14.9 (=?UTF-8?B?R29qxY0=?=) APEL-LB/10.8 EasyPG/1.0.0 Emacs/27.1
 (x86_64-pc-linux-gnu) MULE/6.0 (HANACHIRUSATO)
MIME-Version: 1.0 (generated by SEMI-EPG 1.14.7 - "Harue")
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Wed, 25 Aug 2021 11:26:36 +0100,
Barry Song <21cnbao@gmail.com> wrote:
> 
> From: Marc Zyngier <maz@kernel.org>
> 
> default_irq is hideous as it should be per-device but not per-desc.
> On the other hand, MSI-X case doesn't use it at all. Since sysfs
> IRQ has moved to use the msi_entry instead of pci_dev.irq, now it
> seems it is safe to remove msi_attrib.default_irq.

Thanks for doing the write-up. Maybe worth adding that drivers that
use dev->irq while having enabled MSI will break (INTx will be
disabled while MSI is enabled). That should give people a clue about
what to fix when they bisect the problem to this patch.

Also, a link to the discussion that lead to this patch would be useful
to give some context.

No need to respin this for now, let's give it a shake after 5.14.

Thanks,

	M.

-- 
Without deviation from the norm, progress is not possible.