Patchwork [Oneiric,1/1] UBUNTU: SAUCE: x86: reboot: Make Dell Latitude E6520 use reboot=pci

login
register
mail settings
Submitter Leann Ogasawara
Date Aug. 31, 2011, 5:25 p.m.
Message ID <1314811527.1809.34.camel@adamo>
Download mbox | patch
Permalink /patch/112627/
State New
Headers show

Comments

Leann Ogasawara - Aug. 31, 2011, 5:25 p.m.
Hi All,

http://bugs.launchpad.net/bugs/833705

It's been reported that the Dell Latitude E6520 doesn't reboot unless
using reboot=pci.  The following patch resolves this issue and had been
successfully tested by the bug reporter.  I additionally sent this patch
upstream but was given some push back from one of the maintainers
because "we are starting to have something close to a full Dell product
catalog in the system.  I think we need to come up with something more
systematic..."

So with Kernel Freeze looming for Oneiric, the patch being system
specific (ie low risk of regression), precedence for similar quirk
patches already existing, and having received positive test results, I
feel it should be carried as a SAUCE patch until a more systematic
approach can be found which upstream will accept.

Thanks,
Leann

From 88f224f473a6eceb4f961ccd8f32067aa9f5bcb4 Mon Sep 17 00:00:00 2001
From: Leann Ogasawara <leann.ogasawara@canonical.com>
Date: Fri, 26 Aug 2011 07:18:46 -0700
Subject: [PATCH] UBUNTU: SAUCE: x86: reboot: Make Dell Latitude E6520 use reboot=pci

The Dell Latitude E6520 doesn't reboot unless reboot=pci is set.

BugLink: http://bugs.launchpad.net/bugs/833705
Cc: <stable@kernel.org>
Signed-off-by: Leann Ogasawara <leann.ogasawara@canonical.com>
---
 arch/x86/kernel/reboot.c |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)
Tim Gardner - Aug. 31, 2011, 6:38 p.m.
On 08/31/2011 11:25 AM, Leann Ogasawara wrote:
> Hi All,
>
> http://bugs.launchpad.net/bugs/833705
>
> It's been reported that the Dell Latitude E6520 doesn't reboot unless
> using reboot=pci.  The following patch resolves this issue and had been
> successfully tested by the bug reporter.  I additionally sent this patch
> upstream but was given some push back from one of the maintainers
> because "we are starting to have something close to a full Dell product
> catalog in the system.  I think we need to come up with something more
> systematic..."
>
> So with Kernel Freeze looming for Oneiric, the patch being system
> specific (ie low risk of regression), precedence for similar quirk
> patches already existing, and having received positive test results, I
> feel it should be carried as a SAUCE patch until a more systematic
> approach can be found which upstream will accept.
>
> Thanks,
> Leann
>

Did upstream have any suggestions ? I used to think device tree was the 
answer, but now I'm not as sure. Perhaps a sysfs interface and an 
application to send the right bits based on DMI info is whats needed.

rtg
Leann Ogasawara - Aug. 31, 2011, 7:17 p.m.
On Wed, 2011-08-31 at 12:38 -0600, Tim Gardner wrote:
> On 08/31/2011 11:25 AM, Leann Ogasawara wrote:
> > Hi All,
> >
> > http://bugs.launchpad.net/bugs/833705
> >
> > It's been reported that the Dell Latitude E6520 doesn't reboot unless
> > using reboot=pci.  The following patch resolves this issue and had been
> > successfully tested by the bug reporter.  I additionally sent this patch
> > upstream but was given some push back from one of the maintainers
> > because "we are starting to have something close to a full Dell product
> > catalog in the system.  I think we need to come up with something more
> > systematic..."
> >
> > So with Kernel Freeze looming for Oneiric, the patch being system
> > specific (ie low risk of regression), precedence for similar quirk
> > patches already existing, and having received positive test results, I
> > feel it should be carried as a SAUCE patch until a more systematic
> > approach can be found which upstream will accept.
> >
> > Thanks,
> > Leann
> >
> 
> Did upstream have any suggestions ? I used to think device tree was the 
> answer, but now I'm not as sure. Perhaps a sysfs interface and an 
> application to send the right bits based on DMI info is whats needed.

Nope, no further suggestions, only what I quoted.

It has been brought to my attention that we're also starting to see this
reboot=pci quirk needed in some of the Dell Optiplex systems as well,
see LP bugs 800660 [1], 768039 [2], 818933 [3].

Do we have any contacts at Dell we should raise this with?

Thanks,
Leann

[1] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/800660
[2] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/768039
[3] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/818933

Patch

diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 9242436..6ac6e12 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -443,6 +443,14 @@  static struct dmi_system_id __initdata pci_reboot_dmi_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "Latitude E6420"),
 		},
 	},
+	{	/* Handle problems with rebooting on the Latitude E6520. */
+		.callback = set_pci_reboot,
+		.ident = "Dell Latitude E6520",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Latitude E6520"),
+		},
+	},
 	{ }
 };