Received-SPF: pass (google.com: domain of linux-kernel+bounces-85430-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Date: Thu, 29 Feb 2024 02:22:55 +0900
From: Krzysztof =?utf-8?Q?Wilczy=C5=84ski?= <kwilczynski@kernel.org>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: Saurabh Singh Sengar <ssengar@linux.microsoft.com>, bhelgaas@google.com,
	linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
	alexander.stein@ew.tq-group.com, decui@microsoft.com
Subject: Re: [PATCH] PCI/sysfs: Fix race in pci sysfs creation
Message-ID: <20240228172255.GA3579062@rocinante>
References: <20240227171458.GA16664@linuxonhyperv3.guj3yctzbm1etfxqx2vob5hsef.xx.internal.cloudapp.net>
 <20240228152222.GA272403@bhelgaas>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240228152222.GA272403@bhelgaas>

Hello,

Sorry for late reply.

[...]
> > > > Krzysztof has done a ton of work to convert these files to static
> > > > attributes, where the device model prevents most of these races:
> > > > 
> > > >   506140f9c06b ("PCI/sysfs: Convert "index", "acpi_index", "label" to static attributes")
> > > >   d93f8399053d ("PCI/sysfs: Convert "vpd" to static attribute")
> > > >   f42c35ea3b13 ("PCI/sysfs: Convert "reset" to static attribute")
> > > >   527139d738d7 ("PCI/sysfs: Convert "rom" to static attribute")
> > > >   e1d3f3268b0e ("PCI/sysfs: Convert "config" to static attribute")
> > > > 
> > > > and he even posted a series to do the same for the resource files:
> > > > 
> > > >   https://lore.kernel.org/linux-pci/20210910202623.2293708-1-kw@linux.com/
> > > > 
> > > > I can't remember why we didn't apply that at the time, and it no
> > > > longer applies cleanly, but I think that's the direction we should go.
> > > 
> > > Thanks for you review.
> > > 
> > > Please inform me if there's existing feedback explaining why this
> > > series hasn't been merged yet. I am willing to further improve it
> > > if necessary.
> > 
> > Let us know your opinion so that we can move ahead in fixing this
> > long pending bug.

I really thought you were asking me about your patch.  So, I didn't reply
since Bjorn added his review.

> There's no feedback on the mailing list (I checked the link above), so
> the way forward is to update the series so it applies cleanly again
> and post it as a v3.

Start with a review, if you have some time.  Perhaps we can make it better
before sending another revision.

There are two areas which this series decided not to tackle initially:

  - Support for the Alpha platform
  - Support for legacy PCI platforms

Feel free to have a look at the above.  Perhaps you will have some ideas on how
to best convert both of these to use static attributes, so that we could convert
everything at the same time.

> There's no need to wait for Krzysztof to refresh it, and if you have
> time to do it, it would be very welcomed!  The best base would be
> v6.8-rc1.

That I can do, perhaps this coming weekend.  Or even sooner when I find some
time this week.

	Krzysztof