Skip to contentSkip to conversation

pata_sc1200: Fix crash on boot

PATCH Applied Alan Cox 6 participants 9 messages Oct 6, 2009 – Oct 16, 2009 libata

lore
Share & tools
lore

Alan Cox

author
#0 lore raw

PATCH 1/1 pata_sc1200: Fix crash on boot

The SC1200 needs a NULL terminator or it may cause a crash on boot.

Bug #14227

Also correct a bogus comment as the driver had serializing added so can run dual port.

---

 drivers/ata/pata_sc1200.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)
drivers/ata/pata_sc1200.c +1 -2
index f49814d..3bbed83 100644
@@ -235,8 +235,7 @@ static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA2,
 		.port_ops = &sc1200_port_ops
 	};
-	/* Can't enable port 2 yet, see top comments */
-	const struct ata_port_info *ppi[] = { &info, };
+	const struct ata_port_info *ppi[] = { &info, NULL };
 
 	return ata_pci_sff_init_one(dev, ppi, &sc1200_sht, NULL);
 }
#1 lore raw
Quoted 19 lines · On Tue, 06 Oct 2009 16:07:51 +0100 Alan Cox <[email protected]> wrote:

The SC1200 needs a NULL terminator or it may cause a crash on boot.

Bug #14227

Also correct a bogus comment as the driver had serializing added so can run dual port.

---

 drivers/ata/pata_sc1200.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)
drivers/ata/pata_sc1200.c +0 -0
index f49814d..3bbed83 100644

drivers/ata/pata_sc1200.c static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id) · line 241

Quoted 9 lines
drivers/ata/pata_sc1200.c +1 -2
@@ -235,8 +235,7 @@ static int sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA2,
 		.port_ops = &sc1200_port_ops
 	};
-	/* Can't enable port 2 yet, see top comments */
-	const struct ata_port_info *ppi[] = { &info, };
+	const struct ata_port_info *ppi[] = { &info, NULL };
 
 	return ata_pci_sff_init_one(dev, ppi, &sc1200_sht, NULL);
 }

This patch applies to 2.6.31 and 2.6.30 (at least). How far back should it be ported?

There was no Cc:stable in the changelog.

There was no Cc:stable in the mail headers. Nor cc:linux-ide.

Adding a Reported-by: is nice.

Please use the form

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=14227

when mentioning bugzilla reports within changelogs. To provide a nice greppable pattern for those people who might close off old reports for us.

#2 lore raw
Quoted 1 line · Andrew Morton

On Tue, 06 Oct 2009 16:07:51 +0100 Alan Cox <[email protected]>

Quoted 39 lines · wrote:
Quoted 30 lines

The SC1200 needs a NULL terminator or it may cause a crash on boot.

Bug #14227

Also correct a bogus comment as the driver had serializing added so can run
dual port.

---

 drivers/ata/pata_sc1200.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)
drivers/ata/pata_sc1200.c +1 -2
index f49814d..3bbed83 100644
@@ -235,8 +235,7 @@ static int sc1200_init_one(struct pci_dev *dev,
 const struct pci_device_id *id)
 		.udma_mask = ATA_UDMA2,
 		.port_ops = &sc1200_port_ops
 	};
-	/* Can't enable port 2 yet, see top comments */
-	const struct ata_port_info *ppi[] = { &info, };
+	const struct ata_port_info *ppi[] = { &info, NULL };
 
 	return ata_pci_sff_init_one(dev, ppi, &sc1200_sht, NULL);
Quoted 1 line

}

This patch applies to 2.6.31 and 2.6.30 (at least). How far back should it be ported?

There was no Cc:stable in the changelog.

There was no Cc:stable in the mail headers. Nor cc:linux-ide.

Quoted 8 lines

Adding a Reported-by: is nice.

Please use the form

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=14227

when mentioning bugzilla reports within changelogs. To provide a nice greppable pattern for those people who might close off old reports for

us.

According to the git-tree this feature was introduced on linux kernel 2.6.30:
http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.30.y.git;a=commitdiff;h=c0f2ee34a5a0b79fd98d965ad8ae765d4639bfa5

2.6.29 and older kernels do not have this change/feature

#3 lore raw
In reply to Andrew Morton
Quoted 2 lines

This patch applies to 2.6.31 and 2.6.30 (at least). How far back should it be ported?

No idea. I was just getting it fixed for the maintainer while waiting for something else to build.

#4 lore raw
Quoted 5 lines · On Wednesday 14 October 2009 14:00:46 Alan Cox wrote:
Quoted 2 lines

This patch applies to 2.6.31 and 2.6.30 (at least). How far back should it be ported?

No idea. I was just getting it fixed for the maintainer while waiting for something else to build.

BTW do you happen to know who is the current maintainer of this driver (ditto for other PATA drivers)?

MAINTAINERS lists only:

SERIAL ATA (SATA) SUBSYSTEM
P: Jeff Garzik
M: [email protected]
L: [email protected]
T: git git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
S: Supported
F: drivers/ata/
F: include/linux/ata.h
F: include/linux/libata.h

but this driver surely ain't SATA.

#5 lore raw
Quoted 15 lines

BTW do you happen to know who is the current maintainer of this driver (ditto for other PATA drivers)?

MAINTAINERS lists only:

SERIAL ATA (SATA) SUBSYSTEM
P: Jeff Garzik
M: [email protected]
L: [email protected]
T: git git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
S: Supported
F: drivers/ata/
F: include/linux/ata.h
F: include/linux/libata.h

Quoted 1 line

but this driver surely ain't SATA.

The subsystem wants renaming. Patches go to Jeff who though he rarely appears on listy and stuff keeps the tree and is doing all the submission stuff.

Alan

#7 lore raw
In reply to Andrew Morton
Quoted 2 lines

This patch applies to 2.6.31 and 2.6.30 (at least). How far back should it be ported?

Having had a dig the bug was introduced in
c0f2ee34a5a0b79fd98d965ad8ae765d4639bfa5

I've not however tested the patch with any older releases.

Alan

#6 lore raw
In reply to Bartlomiej Zolnierkiewicz
Quoted 22 lines · Bartlomiej Zolnierkiewicz
Quoted 5 lines · On Wednesday 14 October 2009 14:00:46 Alan Cox wrote:
Quoted 2 lines

This patch applies to 2.6.31 and 2.6.30 (at least). How far back should it be ported?

No idea. I was just getting it fixed for the maintainer while waiting for something else to build.

BTW do you happen to know who is the current maintainer of this driver (ditto for other PATA drivers)?

MAINTAINERS lists only:

SERIAL ATA (SATA) SUBSYSTEM
P: Jeff Garzik
M: [email protected]
L: [email protected]
T: git git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
S: Supported
F: drivers/ata/
F: include/linux/ata.h
F: include/linux/libata.h

Quoted 1 line

but this driver surely ain't SATA.

FWIW, I've been treating Alan as "PATA sub-maintainer", and tend to wait for an Alan review of a PATA driver patch before applying it [unless it's blindingly obvious or Bart chimes etc.].

But yes, overall, libata is certainly not SATA-only anymore, and MAINTAINERS should reflect that.

Maybe a patch like the attached? (DaveM CC'd, since I modified the old-IDE entry as well)

Jeff

MAINTAINERS +2 -2
index 69e31aa..1a756bd 100644
@@ -2581,7 +2581,7 @@ W:	http://www.developer.ibm.com/welcome/netfinity/serveraid.html
 S:	Supported
 F:	drivers/scsi/ips.*
 
-IDE SUBSYSTEM
+Old IDE SUBSYSTEM
 M:	"David S. Miller" <[email protected]>
 L:	[email protected]
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/davem/ide-2.6.git
@@ -4644,7 +4644,7 @@ S:	Maintained
 F:	drivers/misc/phantom.c
 F:	include/linux/phantom.h
 
-SERIAL ATA (SATA) SUBSYSTEM
+ATA SUBSYSTEM
 M:	Jeff Garzik <[email protected]>
 L:	[email protected]
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev.git
Applied
#8 lore raw
In reply to Alan Cox
Quoted 4 lines · Alan Cox

The SC1200 needs a NULL terminator or it may cause a crash on boot.

Bug #14227

Quoted 8 lines

Also correct a bogus comment as the driver had serializing added so can run dual port.

---

  drivers/ata/pata_sc1200.c |    3 +--
  1 files changed, 1 insertions(+), 2 deletions(-)

applied