From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jesse Barnes Subject: Re: [Bug #16161] [2.6.35-rc1 regression] sysfs: cannot create duplicate filename ... XVR-600 related? Date: Fri, 18 Jun 2010 13:43:16 -0700 Message-ID: <20100618134316.408cf922@virtuousgeek.org> References: <19475.45707.157935.788004@pilspetsen.it.uu.se> <201006122052.25204.rjw@sisk.pl> <20100618131049.462ca7ce@virtuousgeek.org> <20100618.132655.260090086.davem@davemloft.net> <4C1BD9CC.5060300@athenacr.com> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4C1BD9CC.5060300-FfUH/8kkONhWk0Htik3J/w@public.gmane.org> Sender: kernel-testers-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" To: Brian Bloniarz Cc: David Miller , rjw-KKrjLPT3xs0@public.gmane.org, mikpe-1zs4UD6AkMk@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, kernel-testers-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, maciej.rutecki-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, achiang-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org On Fri, 18 Jun 2010 16:40:44 -0400 Brian Bloniarz wrote: > On 06/18/2010 04:26 PM, David Miller wrote: > > From: Jesse Barnes > > Date: Fri, 18 Jun 2010 13:10:49 -0700 > > > >> I reverted the symlink patch that was causing the trouble. The root > >> cause is elsewhere though; it seems some firmwares report duplicate PCI > >> slot numbers... > > > > Instead of postulating, you can confirm or deny such a theory > > by taking a look at the repository of sparc openfirmware tree > > dumps maintained at: > > > > master.kernel.org:/pub/scm/linux/kernel/git/davem/prtconfs.git > > (Adding Alex Chiang to the cc list) > > I was actually under the impression that it was just an > issue with the reverted patch, not an actual problem with > hardware. Yeah, I think you're right. I reverted it at Alex's request and assumed it was a firmware or configuration problem. Looking at the thread again I see that was a bad assumption. > In the patch, 2 individual code paths were trying to > create the same symlinks: > pci-sysfs.c:pci_create_slot_links(struct pci_dev *dev) > and > slot.c:create_sysfs_files(struct pci_slot *slot). > I think some archs managed to call those both during > initialization, and some not. Well that would explain it too. I'm happy to take a fixed up patch if there's demand. Thanks, -- Jesse Barnes, Intel Open Source Technology Center