From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751484AbdISSCl (ORCPT ); Tue, 19 Sep 2017 14:02:41 -0400 Received: from mail-pf0-f195.google.com ([209.85.192.195]:34570 "EHLO mail-pf0-f195.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750733AbdISSCh (ORCPT ); Tue, 19 Sep 2017 14:02:37 -0400 X-Google-Smtp-Source: AOwi7QDQi9JmPm+GezGKTzBf9q7c710nanKKw9a4uBwm7McNop83yKrs6S8V+I+3++8RCZ3s6kWj2g== Date: Tue, 19 Sep 2017 11:02:33 -0700 From: Guenter Roeck To: Lorenzo Pieralisi Cc: "linux-kernel@vger.kernel.org" , Bjorn Helgaas , Richard Henderson , Ivan Kokshaysky , Matt Turner , alpha Subject: Re: alpha runtime warnings due to commit 0e4c2eeb758 ("alpha/PCI: Replace pci_fixup_irqs() call with host ...") Message-ID: <20170919180233.GA7018@roeck-us.net> References: <32ec730f-c1b0-5584-cd35-f8a809122b96@roeck-us.net> <20170919091232.GA19596@red-moon> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170919091232.GA19596@red-moon> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Sep 19, 2017 at 10:12:32AM +0100, Lorenzo Pieralisi wrote: > On Mon, Sep 18, 2017 at 07:00:55PM -0700, Guenter Roeck wrote: > > Hi, > > > > I see the following runtime warnings in mainline when running alpha images in qemu. > > > > > > Floppy drive(s): fd0 is 2.88M > > ide0: disabled, no IRQ > > ide0: failed to initialize IDE interface > > ide0: disabling port > > cmd64x 0000:00:02.0: IDE controller (0x1095:0x0646 rev 0x07) > > CMD64x_IDE 0000:00:02.0: BAR 0: can't reserve [io 0x8050-0x8057] > > cmd64x 0000:00:02.0: can't reserve resources > > CMD64x_IDE: probe of 0000:00:02.0 failed with error -16 > > ide_generic: please use "probe_mask=0x3f" module parameter for probing all legacy ISA IDE ports > > ------------[ cut here ]------------ > > WARNING: CPU: 0 PID: 1 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x94/0xd0 > > sysfs: cannot create duplicate filename '/class/ide_port/ide0' > > ... > > > > Trace: > > [] __warn+0x160/0x190 > > [] sysfs_warn_dup+0x94/0xd0 > > [] warn_slowpath_fmt+0x58/0x70 > > [] sysfs_warn_dup+0x94/0xd0 > > [] kernfs_path_from_node+0x30/0x60 > > [] kernfs_put+0x16c/0x2c0 > > [] kernfs_put+0x16c/0x2c0 > > [] sysfs_do_create_link_sd.isra.2+0x100/0x120 > > [] device_add+0x2a4/0x7c0 > > [] device_create_groups_vargs+0x14c/0x170 > > [] device_create_groups_vargs+0x98/0x170 > > [] device_create+0x50/0x70 > > [] ide_host_register+0x48c/0xa00 > > [] ide_host_register+0x450/0xa00 > > [] device_register+0x20/0x50 > > [] ide_host_register+0x450/0xa00 > > [] ide_host_add+0x64/0xe0 > > [] kobject_uevent_env+0x16c/0x710 > > [] do_one_initcall+0x68/0x260 > > [] kernel_init+0x1c/0x1a0 > > [] kernel_init+0x0/0x1a0 > > [] ret_from_kernel_thread+0x18/0x20 > > [] kernel_init+0x0/0x1a0 > > > > ---[ end trace 24a70433c3e4d374 ]--- > > ide0: disabling port > > > > [ multiple times ] > > > > A complete log is available at http://kerneltests.org/builders/qemu-alpha-master. > > > > Prior to the offending commit, the kernel log looks as follows. > > > > ... > > Uniform Multi-Platform E-IDE driver > > cmd64x 0000:00:02.0: IDE controller (0x1095:0x0646 rev 0x07) > > cmd64x 0000:00:02.0: IDE port disabled > > cmd64x 0000:00:02.0: 100% native mode on irq 28 > > PCI: Setting latency timer of device 0000:00:02.0 to 64 > > ide0: BM-DMA at 0x8040-0x8047 > > Floppy drive(s): fd0 is 2.88M > > ide0 at 0x8050-0x8057,0x8062 on irq 28 (serialized) > > ide_generic: please use "probe_mask=0x3f" module parameter for probing all legacy ISA IDE ports > > ide1 at 0x1f0-0x1f7,0x3f6 on irq 14 > > ide2 at 0x170-0x177,0x376 on irq 15 > > ide-gd driver 1.18 > > ide-cd driver 5.00 > > ... > > > > Reverting the commit is not possible due to context changes. > > > > Bisect log is attached. > > Ok thanks. Can you please check if the diff below fixes the issue ? > It does. Feel free to add Tested-by: Guenter Roeck to the actual patch. > I think the problem is that now we allocate the IRQ at device_probe > instead of device_add time, if the patch below fixes the issue the > question is whether we want to move pci_assign_irq() to pci_device_add() > for ALL PCI devices or just fix this for IDE subsytem. > That I can not answer ... Guenter > Lorenzo > > -- >8 -- > diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c > index 112d2fe..94ca9cc 100644 > --- a/drivers/ide/setup-pci.c > +++ b/drivers/ide/setup-pci.c > @@ -502,6 +502,8 @@ static int do_ide_setup_pci_device(struct pci_dev *dev, > { > int pciirq, ret; > > + pci_assign_irq(dev); > + > /* > * Can we trust the reported IRQ? > */