Hi Rafael,
Today's linux-next merge of the pm tree got a conflict in
arch/x86/pci/acpi.c between commit 3f385fa9edce ("x86/PCI: Ignore _SEG on
HP xw9300") from the pci tree and commit 97a7108a3c00 ("ACPI / x86: Add
quirk for "CheckPoint P-20-00" to not use bridge _CRS_ info") from the pm
tree.
I fixed it up (see below) and can carry the fix as necessary (no action
is required).
--
Cheers,
Stephen Rothwell [email protected]
diff --cc arch/x86/pci/acpi.c
index 49e5195,7010c19..0000000
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@@ -106,16 -98,16 +106,27 @@@ static const struct dmi_system_id pci_c
DMI_MATCH(DMI_BIOS_VERSION, "6JET85WW (1.43 )"),
},
},
+
+ /* https://bugzilla.kernel.org/show_bug.cgi?id=15362 */
+ {
+ .callback = set_ignore_seg,
+ .ident = "HP xw9300",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "HP xw9300 Workstation"),
+ },
+ },
++
+ /* https://bugzilla.kernel.org/show_bug.cgi?id=47981 */
+ {
+ .callback = set_nouse_crs,
+ .ident = "CheckPoint P-20-00",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "CheckPoint"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "P-20-00"),
+ DMI_MATCH(DMI_BOARD_NAME, "Bridgeport"),
+ },
+ },
{}
};
On Mon, Nov 5, 2012 at 7:48 PM, Stephen Rothwell <[email protected]> wrote:
> Hi Rafael,
>
> Today's linux-next merge of the pm tree got a conflict in
> arch/x86/pci/acpi.c between commit 3f385fa9edce ("x86/PCI: Ignore _SEG on
> HP xw9300") from the pci tree and commit 97a7108a3c00 ("ACPI / x86: Add
> quirk for "CheckPoint P-20-00" to not use bridge _CRS_ info") from the pm
> tree.
>
> I fixed it up (see below) and can carry the fix as necessary (no action
> is required).
My opinion was that we should just drop the CheckPoint quirk. See
https://bugzilla.kernel.org/show_bug.cgi?id=47981#c36 for my
rationale.
The changelog of the quirk leaves the wrong impression:
This is to fix a regression
https://bugzilla.kernel.org/show_bug.cgi?id=47981
The CheckPoint P-20-00 works ok before new machines (2008 and later) are
forced to use the bridge _CRS info by default in 2.6.34. Add this quirk
to restore its old way of working: not using bridge _CRS info.
This is NOT a regression, at least not in the usual sense. The
CheckPoint BIOS has a serious defect that corrupts part of the DSDT.
It happened that things "worked" when we ignored the _CRS that was
corrupted (other things are almost certainly broken, but we don't know
what they are yet). And it's not a question of being "forced" to use
the bridge _CRS. Every system SHOULD be using _CRS; that's the only
way we can correctly manage resources of ACPI devices.
But if you (Rafael) decide we should keep this, I won't object.
The conflict resolution looks fine to me.
> diff --cc arch/x86/pci/acpi.c
> index 49e5195,7010c19..0000000
> --- a/arch/x86/pci/acpi.c
> +++ b/arch/x86/pci/acpi.c
> @@@ -106,16 -98,16 +106,27 @@@ static const struct dmi_system_id pci_c
> DMI_MATCH(DMI_BIOS_VERSION, "6JET85WW (1.43 )"),
> },
> },
> +
> + /* https://bugzilla.kernel.org/show_bug.cgi?id=15362 */
> + {
> + .callback = set_ignore_seg,
> + .ident = "HP xw9300",
> + .matches = {
> + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
> + DMI_MATCH(DMI_PRODUCT_NAME, "HP xw9300 Workstation"),
> + },
> + },
> ++
> + /* https://bugzilla.kernel.org/show_bug.cgi?id=47981 */
> + {
> + .callback = set_nouse_crs,
> + .ident = "CheckPoint P-20-00",
> + .matches = {
> + DMI_MATCH(DMI_SYS_VENDOR, "CheckPoint"),
> + DMI_MATCH(DMI_PRODUCT_NAME, "P-20-00"),
> + DMI_MATCH(DMI_BOARD_NAME, "Bridgeport"),
> + },
> + },
> {}
> };
>
On Tuesday, November 06, 2012 09:52:40 AM Bjorn Helgaas wrote:
> On Mon, Nov 5, 2012 at 7:48 PM, Stephen Rothwell <[email protected]> wrote:
> > Hi Rafael,
> >
> > Today's linux-next merge of the pm tree got a conflict in
> > arch/x86/pci/acpi.c between commit 3f385fa9edce ("x86/PCI: Ignore _SEG on
> > HP xw9300") from the pci tree and commit 97a7108a3c00 ("ACPI / x86: Add
> > quirk for "CheckPoint P-20-00" to not use bridge _CRS_ info") from the pm
> > tree.
> >
> > I fixed it up (see below) and can carry the fix as necessary (no action
> > is required).
>
> My opinion was that we should just drop the CheckPoint quirk. See
> https://bugzilla.kernel.org/show_bug.cgi?id=47981#c36 for my
> rationale.
OK, I agree. I'll revert that commit, then.
Thanks,
Rafael
--
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.