From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rene Herman Subject: Re: [patch 2/2] PNP: don't check disabled PCI BARs for conflicts in quirk_system_pci_resources() Date: Tue, 30 Sep 2008 22:48:26 +0200 Message-ID: <48E2909A.8090703@keyaccess.nl> References: <200809290953.56565.bjorn.helgaas@hp.com> <200809290957.59813.bjorn.helgaas@hp.com> <48E11EFA.8010402@keyaccess.nl> <48E1EF0E.8030006@keyaccess.nl> <48E24C6F.3030903@keyaccess.nl> <48E27E23.20400@keyaccess.nl> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="------------040206040806010003050400" Return-path: In-Reply-To: Sender: linux-pci-owner@vger.kernel.org To: Linus Torvalds Cc: Bjorn Helgaas , Jesse Barnes , Len Brown , Frans Pop , "Rafael J. Wysocki" , Linux Kernel Mailing List , linux-pci@vger.kernel.org, linux-acpi@vger.kernel.org, Adam Belay , Avuton Olrich , Karl Bellve , Willem Riede , Matthew Hall , Sam Ravnborg List-Id: linux-acpi@vger.kernel.org This is a multi-part message in MIME format. --------------040206040806010003050400 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit On 30-09-08 21:44, Linus Torvalds wrote: > On Tue, 30 Sep 2008, Rene Herman wrote: >> Yes, I also get that oops but other than that, both link order versions you >> sent out work -- ie, booting with acpi=noirq gets me to a functional system >> with the quirk having run for PNP0c02 (acpi=off disables all of PNP0c02) and >> doing its job. > > Ok. But that means that the last patch I sent out - the one that _only_ > changes the order for PnP itself, and moves pnpacpi_init and pnpbios_init > to be fs_initcalls - should also work, and have none of he other > interactions. Yes? Yes. I am fine on current mainline and with this seem to still be fine, with or without the quirk changes (*) applied. (*) http://marc.info/?l=linux-kernel&m=122280330516865&w=2 Frans Pop will need something like those quirk changes on top to have his machine stop yelling at him -- assuming it actually works for him that is (which it should I guess, but it's not been tested by him yet). The pci_start == 0 version, attached for convenience and also still available from the bugzilla: http://bugzilla.kernel.org/show_bug.cgi?id=11550 is still the minimal version for Frans' issue. (I see there are multiple copies of messages that I sent in that marc archive. Seem to again be experiencing severe email trouble since I'm also not getting back most messages that I see there. Anyways, if you get multiple copies, sorry, can't help it it seems, and I need to be away after this). Rene. --------------040206040806010003050400 Content-Type: text/plain; name="0001-PNP-avoid-checking-unitialized-BARs-for-conflicts.patch" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename*0="0001-PNP-avoid-checking-unitialized-BARs-for-conflicts.patch" >>From 6ba1072ef110f8977832592c092501b81439da4b Mon Sep 17 00:00:00 2001 From: Rene Herman Date: Tue, 30 Sep 2008 22:33:42 +0200 Subject: [PATCH] PNP: avoid checking unitialized BARs for conflicts Avoid checking a PCI BAR for conflicts if the BIOS left it unitialized. Reported-by: Frans Pop Tested-by: Frans Pop Signed-off-by: Rene Herman --- drivers/pnp/quirks.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/drivers/pnp/quirks.c b/drivers/pnp/quirks.c index 0bdf9b8..d0120a5 100644 --- a/drivers/pnp/quirks.c +++ b/drivers/pnp/quirks.c @@ -253,6 +253,9 @@ static void quirk_system_pci_resources(struct pnp_dev *dev) continue; pci_start = pci_resource_start(pdev, i); + if (!pci_start) + continue; + pci_end = pci_resource_end(pdev, i); for (j = 0; (res = pnp_get_resource(dev, type, j)); j++) { -- 1.6.0.2 --------------040206040806010003050400--