[PATCH] x86/PCI: use host bridge _CRS info on MSI MS-7253

[PATCH] x86/PCI: use host bridge _CRS info on MSI MS-7253

[Bjorn Helgaas]

From: Jonathan Nieder <jrnieder@gmail.com>

In the spirit of commit 29cf7a30f8a0 ("x86/PCI: use host bridge _CRS
info on ASUS M2V-MX SE"), this DMI quirk turns on "pci_use_crs" by
default on a board that needs it.

This fixes boot failures and oopses introduced in 3e3da00c01d0
("x86/pci: AMD one chain system to use pci read out res").  The quirk
is quite targetted (to a specific board and BIOS version) for two
reasons:

 (1) to emphasize that this method of tackling the problem one quirk
     at a time is a little insane

 (2) to give BIOS vendors an opportunity to use simpler tables and
     allow us to return to generic behavior (whatever that happens to
     be) with a later BIOS update

In other words, I am not at all happy with having quirks like this.
But it is even worse for the kernel not to work out of the box on
these machines, so...

Reference: https://bugzilla.kernel.org/show_bug.cgi?id=42619
Reported-by: Svante Signell <svante.signell@telia.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
---
 arch/x86/pci/acpi.c |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index a312e76..2c5cbdb 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -60,6 +60,17 @@ static const struct dmi_system_id pci_use_crs_table[] __initconst = {
 			DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."),
 		},
 	},
+	/* https://bugzilla.kernel.org/show_bug.cgi?id=42619 */
+	{
+		.callback = set_use_crs,
+		.ident = "MSI MS-7253",
+		.matches = {
+			DMI_MATCH(DMI_BOARD_VENDOR, "MICRO-STAR INTERNATIONAL CO., LTD"),
+			DMI_MATCH(DMI_BOARD_NAME, "MS-7253"),
+			DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"),
+			DMI_MATCH(DMI_BIOS_VERSION, "V1.6"),
+		},
+	},
 
 	/* Now for the blacklist.. */
 

Re: [PATCH] x86/PCI: use host bridge _CRS info on MSI MS-7253

[Bjorn Helgaas]

On Tue, Feb 28, 2012 at 11:51 AM, Bjorn Helgaas <bhelgaas@google.com> wrote:
> From: Jonathan Nieder <jrnieder@gmail.com>
>
> In the spirit of commit 29cf7a30f8a0 ("x86/PCI: use host bridge _CRS
> info on ASUS M2V-MX SE"), this DMI quirk turns on "pci_use_crs" by
> default on a board that needs it.

Hi Jesse,

This patch is against upstream (891003abb0d), and I think it also
applies cleanly on your linux-next branch.  It fixes a regression and
shouldn't affect other machines, so you might consider it even for
3.3.  Otherwise it can go in the merge window.

Bjorn

Re: [PATCH] x86/PCI: use host bridge _CRS info on MSI MS-7253

[Jesse Barnes]

[-- Attachment #1: Type: text/plain, Size: 1102 bytes --]

On Tue, 28 Feb 2012 11:51:10 -0700
Bjorn Helgaas <bhelgaas@google.com> wrote:

> From: Jonathan Nieder <jrnieder@gmail.com>
> 
> In the spirit of commit 29cf7a30f8a0 ("x86/PCI: use host bridge _CRS
> info on ASUS M2V-MX SE"), this DMI quirk turns on "pci_use_crs" by
> default on a board that needs it.
> 
> This fixes boot failures and oopses introduced in 3e3da00c01d0
> ("x86/pci: AMD one chain system to use pci read out res").  The quirk
> is quite targetted (to a specific board and BIOS version) for two
> reasons:
> 
>  (1) to emphasize that this method of tackling the problem one quirk
>      at a time is a little insane
> 
>  (2) to give BIOS vendors an opportunity to use simpler tables and
>      allow us to return to generic behavior (whatever that happens to
>      be) with a later BIOS update
> 
> In other words, I am not at all happy with having quirks like this.
> But it is even worse for the kernel not to work out of the box on
> these machines, so...

Applied to my for-linus branch, thanks.

-- 
Jesse Barnes, Intel Open Source Technology Center

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Jonathan Nieder]

Carlos was getting

	WARNING: at drivers/pci/pci.c:118 pci_ioremap_bar+0x24/0x52()

when probing his sound card, and sound did not work.  After adding
pci=use_crs to the kernel command line, no more trouble.

Ok, we can add a quirk.  dmidecode output reveals that this is an MSI
MS-7253, for which we already have a quirk, but the short-sighted
author tied the quirk to a single BIOS version, making it not kick in
on Carlos's machine with BIOS V1.2.  If a later BIOS update makes it
no longer necessary to look at the _CRS info it will still be
harmless, so let's stop trying to guess which versions have and don't
have accurate _CRS tables.

Addresses https://bugtrack.alsa-project.org/alsa-bug/view.php?id=5533
Also see <https://bugzilla.kernel.org/show_bug.cgi?id=42619>.

Reported-by: Carlos Luna <caralu74@gmail.com>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
---
Hi,

Could you squash this into the patch "x86/PCI: use host bridge _CRS
info on MSI MS-7253", if it's not too late?  Sorry for the thinko.

Thanks,
Jonathan

 arch/x86/pci/acpi.c |    1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 2c5cbdb5d602..d066e11159f3 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -68,7 +68,6 @@ static const struct dmi_system_id pci_use_crs_table[] __initconst = {
 			DMI_MATCH(DMI_BOARD_VENDOR, "MICRO-STAR INTERNATIONAL CO., LTD"),
 			DMI_MATCH(DMI_BOARD_NAME, "MS-7253"),
 			DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"),
-			DMI_MATCH(DMI_BIOS_VERSION, "V1.6"),
 		},
 	},
 
-- 
1.7.9.2

Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Bjorn Helgaas]

On Tue, Feb 28, 2012 at 2:31 PM, Jonathan Nieder <jrnieder@gmail.com> wrote:
> Carlos was getting
>
>        WARNING: at drivers/pci/pci.c:118 pci_ioremap_bar+0x24/0x52()
>
> when probing his sound card, and sound did not work.  After adding
> pci=use_crs to the kernel command line, no more trouble.
>
> Ok, we can add a quirk.  dmidecode output reveals that this is an MSI
> MS-7253, for which we already have a quirk, but the short-sighted
> author tied the quirk to a single BIOS version, making it not kick in
> on Carlos's machine with BIOS V1.2.  If a later BIOS update makes it
> no longer necessary to look at the _CRS info it will still be
> harmless, so let's stop trying to guess which versions have and don't
> have accurate _CRS tables.
>
> Addresses https://bugtrack.alsa-project.org/alsa-bug/view.php?id=5533
> Also see <https://bugzilla.kernel.org/show_bug.cgi?id=42619>.

Thanks, Jonathan.  I was going to do this, but you saved me the trouble :)

Reviewed-by: Bjorn Helgaas <bhelgaas@google.com>

> Reported-by: Carlos Luna <caralu74@gmail.com>
> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
> ---
> Hi,
>
> Could you squash this into the patch "x86/PCI: use host bridge _CRS
> info on MSI MS-7253", if it's not too late?  Sorry for the thinko.
>
> Thanks,
> Jonathan
>
>  arch/x86/pci/acpi.c |    1 -
>  1 file changed, 1 deletion(-)
>
> diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
> index 2c5cbdb5d602..d066e11159f3 100644
> --- a/arch/x86/pci/acpi.c
> +++ b/arch/x86/pci/acpi.c
> @@ -68,7 +68,6 @@ static const struct dmi_system_id pci_use_crs_table[] __initconst = {
>                        DMI_MATCH(DMI_BOARD_VENDOR, "MICRO-STAR INTERNATIONAL CO., LTD"),
>                        DMI_MATCH(DMI_BOARD_NAME, "MS-7253"),
>                        DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"),
> -                       DMI_MATCH(DMI_BIOS_VERSION, "V1.6"),
>                },
>        },
>
> --
> 1.7.9.2
>

Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Jesse Barnes]

[-- Attachment #1: Type: text/plain, Size: 1126 bytes --]

On Tue, 28 Feb 2012 15:31:35 -0600
Jonathan Nieder <jrnieder@gmail.com> wrote:

> Carlos was getting
> 
> 	WARNING: at drivers/pci/pci.c:118 pci_ioremap_bar+0x24/0x52()
> 
> when probing his sound card, and sound did not work.  After adding
> pci=use_crs to the kernel command line, no more trouble.
> 
> Ok, we can add a quirk.  dmidecode output reveals that this is an MSI
> MS-7253, for which we already have a quirk, but the short-sighted
> author tied the quirk to a single BIOS version, making it not kick in
> on Carlos's machine with BIOS V1.2.  If a later BIOS update makes it
> no longer necessary to look at the _CRS info it will still be
> harmless, so let's stop trying to guess which versions have and don't
> have accurate _CRS tables.
> 
> Addresses https://bugtrack.alsa-project.org/alsa-bug/view.php?id=5533
> Also see <https://bugzilla.kernel.org/show_bug.cgi?id=42619>.
> 
> Reported-by: Carlos Luna <caralu74@gmail.com>
> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
> ---

Applied to my for-linus branch, thanks.

-- 
Jesse Barnes, Intel Open Source Technology Center

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Jonathan Nieder]

Hi Bjorn and Jesse,

Jesse Barnes wrote:

> Applied to my for-linus branch, thanks.

Thanks for taking care of it.  I'll look into submitting these to
-stable once they hit mainline if I notice.

[3.0.y, 3.2.y, 3.3.y] Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Jonathan Nieder]

Hi Greg,

Jonathan Nieder wrote:
> Jesse Barnes wrote:

>> Applied to my for-linus branch, thanks.
>
> Thanks for taking care of it.  I'll look into submitting these to
> -stable once they hit mainline if I notice.

Please queue

 84113717 (x86/PCI: use host bridge _CRS info on MSI MS-7253)
 a97f4f5e (x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS
           version)

for maintained stable trees 2.6.34.y and newer.

Those two patches address a regression that bisects to
v2.6.34-rc1~218^2~26 (x86/pci: AMD one chain system to use pci read
out res, 2010-02-10) and on the affected machines (Microstar MSI
MS-7253) results in misconfiguration and instability --- some example
symptoms were audio being silenced and crashes at boot time.  They
work by automatically enabling the pci=use_crs parameter so the kernel
can pay attention to information in ACPI tables that can save it from
trouble.

The patch was applied to mainline during the 3.3 merge window.

https://bugzilla.kernel.org/show_bug.cgi?id=42619 has details.
Thanks to Svante Signell and Carlos Luna for reporting it and Bjorn
Helgas and Paul Menzel for tracking it down.

Jonathan

Re: [3.0.y, 3.2.y, 3.3.y] Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Jonathan Nieder]

Jonathan Nieder wrote:

> [Subject: [3.0.y, 3.2.y, 3.3.y] Re: [PATCH] x86/PCI: do not tie MSI
> MS-7253 use_crs quirk to BIOS version]

Since v3.3 already contains the patches, 3.3.y does not need this.
3.0.y and 3.2.y do, though.  Sorry for the nonsense.

Re: [3.0.y, 3.2.y, 3.3.y] Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Greg Kroah-Hartman]

On Mon, Apr 09, 2012 at 06:47:55PM -0500, Jonathan Nieder wrote:
> Hi Greg,
> 
> Jonathan Nieder wrote:
> > Jesse Barnes wrote:
> 
> >> Applied to my for-linus branch, thanks.
> >
> > Thanks for taking care of it.  I'll look into submitting these to
> > -stable once they hit mainline if I notice.
> 
> Please queue
> 
>  84113717 (x86/PCI: use host bridge _CRS info on MSI MS-7253)
>  a97f4f5e (x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS
>            version)
> 
> for maintained stable trees 2.6.34.y and newer.

Applied to 3.0-stable and newer.

greg k-h

Re: [3.0.y, 3.2.y, 3.3.y] Re: [PATCH] x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS version

[Jonathan Nieder]

Greg Kroah-Hartman wrote:
> On Mon, Apr 09, 2012 at 06:47:55PM -0500, Jonathan Nieder wrote:

>> Please queue
>>
>>  84113717 (x86/PCI: use host bridge _CRS info on MSI MS-7253)
>>  a97f4f5e (x86/PCI: do not tie MSI MS-7253 use_crs quirk to BIOS
>>            version)
>>
>> for maintained stable trees 2.6.34.y and newer.
>
> Applied to 3.0-stable and newer.

Thanks much.

Paul Gortmaker: FYI.  [1] has context.

Ciao,
Jonathan

[1] http://thread.gmane.org/gmane.linux.kernel.pci/13825/focus=14618