From: Simon Horman <horms@verge.net.au>
To: Sven Wegener <sven.wegener@stealer.net>,
Michael Ellerman <michael@ellerman.id.au>
Cc: linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
Jesse Barnes <jbarnes@virtuousgeek.org>
Subject: Re: [patch] PCI: check the return value of device_create_bin_file() in pci_create_bus()
Date: Tue, 5 Aug 2008 21:00:07 +1000 [thread overview]
Message-ID: <20080805110006.GA13409@verge.net.au> (raw)
In-Reply-To: <1217932416.7758.4.camel@localhost> <alpine.LNX.1.10.0808051224350.5671@titan.stealer.net>
On Tue, Aug 05, 2008 at 12:39:31PM +0200, Sven Wegener wrote:
> On Tue, 5 Aug 2008, Simon Horman wrote:
>
> > Check the return value of device_create_bin_file in pci_create_bus,
> > unwind if neccessary, and propogate any errors to the caller.
> >
> > Signed-off-by: Simon Horman <horms@verge.net.au>
> >
> > ---
> >
> > drivers/pci/probe.c: In function `pci_create_bus':
> > drivers/pci/probe.c:66: warning: ignoring return value of `device_create_bin_file', declared with attribute warn_unused_result
> > drivers/pci/probe.c:74: warning: ignoring return value of `device_create_bin_file', declared with attribute warn_unused_result
> >
> > # ia64-unknown-linux-gnu-gcc --version
> > ia64-unknown-linux-gnu-gcc (GCC) 3.4.5
> > Copyright (C) 2004 Free Software Foundation, Inc.
> > This is free software; see the source for copying conditions. There is NO
> > warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
> >
> > Index: linux-2.6/drivers/pci/probe.c
> > ===================================================================
> > --- linux-2.6.orig/drivers/pci/probe.c 2008-08-05 19:58:43.000000000 +1000
> > +++ linux-2.6/drivers/pci/probe.c 2008-08-05 19:59:15.000000000 +1000
> > @@ -53,26 +53,37 @@ EXPORT_SYMBOL(no_pci_devices);
> > * a per-bus basis. This routine creates the files and ties them into
> > * their associated read, write and mmap files from pci-sysfs.c
> > */
> > -static void pci_create_legacy_files(struct pci_bus *b)
> > +static int pci_create_legacy_files(struct pci_bus *b)
> > {
> > + int error;
> > +
> > b->legacy_io = kzalloc(sizeof(struct bin_attribute) * 2,
> > GFP_ATOMIC);
> > - if (b->legacy_io) {
> > - b->legacy_io->attr.name = "legacy_io";
> > - b->legacy_io->size = 0xffff;
> > - b->legacy_io->attr.mode = S_IRUSR | S_IWUSR;
> > - b->legacy_io->read = pci_read_legacy_io;
> > - b->legacy_io->write = pci_write_legacy_io;
> > - device_create_bin_file(&b->dev, b->legacy_io);
> > -
> > - /* Allocated above after the legacy_io struct */
> > - b->legacy_mem = b->legacy_io + 1;
> > - b->legacy_mem->attr.name = "legacy_mem";
> > - b->legacy_mem->size = 1024*1024;
> > - b->legacy_mem->attr.mode = S_IRUSR | S_IWUSR;
> > - b->legacy_mem->mmap = pci_mmap_legacy_mem;
> > - device_create_bin_file(&b->dev, b->legacy_mem);
> > + if (!b->legacy_io)
> > + return -ENOMEM;
> > +
> > + b->legacy_io->attr.name = "legacy_io";
> > + b->legacy_io->size = 0xffff;
> > + b->legacy_io->attr.mode = S_IRUSR | S_IWUSR;
> > + b->legacy_io->read = pci_read_legacy_io;
> > + b->legacy_io->write = pci_write_legacy_io;
> > + error = device_create_bin_file(&b->dev, b->legacy_io);
> > + if (error)
> > + return error;
>
> I'd release the memory here and NULLify legacy_io.
>
> > +
> > + /* Allocated above after the legacy_io struct */
> > + b->legacy_mem = b->legacy_io + 1;
> > + b->legacy_mem->attr.name = "legacy_mem";
> > + b->legacy_mem->size = 1024*1024;
> > + b->legacy_mem->attr.mode = S_IRUSR | S_IWUSR;
> > + b->legacy_mem->mmap = pci_mmap_legacy_mem;
> > + error = device_create_bin_file(&b->dev, b->legacy_mem);
> > + if (error) {
> > + device_remove_bin_file(&b->dev, b->legacy_io);
> > + return error;
>
> Here too.
>
> Reason: If we fail to create the legacy_io file, legacy_mem will still be
> NULL, because it has been not initialized at that point. But we will try
> to remove it in pci_remove_legacy_files and in sysfs_remove_bin_file
> we're going to derefence it and blow up.
Yes, sorry, my bad.
next prev parent reply other threads:[~2008-08-05 11:04 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-08-05 10:16 [patch] PCI: check the return value of device_create_bin_file() in pci_create_bus() Simon Horman
2008-08-05 10:33 ` Michael Ellerman
2008-08-05 10:39 ` Sven Wegener
2008-08-05 11:00 ` Simon Horman [this message]
2008-08-05 11:14 ` Simon Horman
2008-08-05 11:28 ` Matthew Wilcox
2008-08-05 12:15 ` Simon Horman
2008-08-05 12:26 ` Matthew Wilcox
2008-08-06 0:57 ` Simon Horman
2008-08-06 13:19 ` Matthew Wilcox
2008-08-06 23:25 ` Simon Horman
2008-08-06 23:30 ` Simon Horman
2008-08-07 0:30 ` Michael Ellerman
2008-08-07 1:06 ` Simon Horman
2008-08-07 4:56 ` Simon Horman
2008-08-07 16:50 ` Jesse Barnes
2008-08-07 22:59 ` Simon Horman
2008-08-05 13:21 ` Sven Wegener
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20080805110006.GA13409@verge.net.au \
--to=horms@verge.net.au \
--cc=jbarnes@virtuousgeek.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=michael@ellerman.id.au \
--cc=sven.wegener@stealer.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox