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, 4 Nov 2020 17:46:47 +0100
From:   Pali =?utf-8?B?Um9ow6Fy?= <pali@kernel.org>
To:     Bjorn Helgaas <helgaas@kernel.org>
Cc:     Bjorn Helgaas <bhelgaas@google.com>, linux-pci@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
        Gregory Clement <gregory.clement@bootlin.com>,
        Andrew Lunn <andrew@lunn.ch>,
        Krzysztof =?utf-8?Q?Wilczy=C5=84ski?= <kw@linux.com>,
        Yinghai Lu <yinghai@kernel.org>
Subject: Re: PCI: Race condition in pci_create_sysfs_dev_files
Message-ID: <20201104164647.pctgyk2cbhjcq65z@pali>
References: <20200909112850.hbtgkvwqy2rlixst@pali>
 <20201006222222.GA3221382@bjorn-Precision-5520>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20201006222222.GA3221382@bjorn-Precision-5520>
User-Agent: NeoMutt/20180716
Precedence: bulk

On Tuesday 06 October 2020 17:22:22 Bjorn Helgaas wrote:
> It's not obvious from the code why we need pci_sysfs_init(), but
> Yinghai hinted [1] that we need to create sysfs after assigning
> resources.  I experimented by removing pci_sysfs_init() and skipping
> the ROM BAR sizing.  In that case, we create sysfs files in
> pci_bus_add_device() and later assign space for the ROM BAR, so we
> fail to create the "rom" sysfs file.
> 
> The current solution to that is to delay the sysfs files until
> pci_sysfs_init(), a late_initcall(), which runs after resource
> assignments.  But I think it would be better if we could create the
> sysfs file when we assign the BAR.  Then we could get rid of the
> late_initcall() and that implicit ordering requirement.
> 
> But I haven't tried to code it up, so it's probably more complicated
> than this.  I guess ideally we would assign all the resources before
> pci_bus_add_device().  If we could do that, we could just remove
> pci_sysfs_init() and everything would just work, but I think that's a
> HUGE can of worms.
> 
> [1] https://lore.kernel.org/linux-pci/CAE9FiQWBXHgz-gWCmpWLaBOfQQJwtRZemV6Ut9GVw_KJ-dTGTA@mail.gmail.com/

I found out that pci_sysfs_init() function was introduced in kernel
version 2.6.10 by this historic commit:

https://git.kernel.org/pub/scm/linux/kernel/git/history/history.git/commit/?id=f6d553444da20cd1e44f2c4864c2d0c56c934e0a

So it was really due to PCI ROM BAR and accessing it from sysfs.