Skip to contentSkip to conversation

Use PCI_DEVICE_ID_88ALP01 for CAFÉ chip, rather than PCI_DEVICE_ID_CAFE.

PATCH In review David Woodhouse 3 participants 5 messages Sep 3, 2008 1 message not in the archive

Subsystems (6)
lore
Share & tools
lore
#1 lore raw

PATCH 1/3 Use PCI_DEVICE_ID_88ALP01 for CAFÉ chip, rather than PCI_DEVICE_ID_CAFE.

In reply to a message not in the archive

Probably better to use the official designation.

---
(Carried in git://git.infradead.org/~dwmw2/random-2.6.git)

 drivers/mmc/host/sdhci-pci.c |    2 +-
 include/linux/pci_ids.h      |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/mmc/host/sdhci-pci.c +1 -1
index fcb14c2..0341cfb 100644
@@ -326,7 +326,7 @@ static const struct pci_device_id pci_ids[] __devinitdata = {
 
 	{
 		.vendor         = PCI_VENDOR_ID_MARVELL,
-		.device         = PCI_DEVICE_ID_MARVELL_CAFE_SD,
+		.device         = PCI_DEVICE_ID_MARVELL_88ALP01_SD,
 		.subvendor      = PCI_ANY_ID,
 		.subdevice      = PCI_ANY_ID,
 		.driver_data    = (kernel_ulong_t)&sdhci_cafe,
include/linux/pci_ids.h +1 -1
index f1624b3..ef6ef64 100644
@@ -1521,7 +1521,7 @@
 #define PCI_DEVICE_ID_MARVELL_GT64260	0x6430
 #define PCI_DEVICE_ID_MARVELL_MV64360	0x6460
 #define PCI_DEVICE_ID_MARVELL_MV64460	0x6480
-#define PCI_DEVICE_ID_MARVELL_CAFE_SD	0x4101
+#define PCI_DEVICE_ID_MARVELL_88ALP01_SD	0x4101
 
 #define PCI_VENDOR_ID_V3		0x11b0
 #define PCI_DEVICE_ID_V3_V960		0x0001
Signature · 4 lines
1.5.5.1
-- 
David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation
#3 lore raw

PATCH 2/3 Define and use PCI_DEVICE_ID_MARVELL_88ALP01_NAND for CAFÉ

In reply to a message not in the archive

---
(Carried in git://git.infradead.org/~dwmw2/random-2.6.git)

 drivers/mtd/nand/cafe_nand.c |    6 +++++-
 include/linux/pci_ids.h      |    1 +
 2 files changed, 6 insertions(+), 1 deletions(-)
drivers/mtd/nand/cafe_nand.c +5 -1
index 95345d0..b8064bf 100644
@@ -1,6 +1,9 @@
 /*
  * Driver for One Laptop Per Child ‘CAFÉ’ controller, aka Marvell 88ALP01
  *
+ * The data sheet for this device can be found at:
+ *    http://www.marvell.com/products/pcconn/88ALP01.jsp
+ *
  * Copyright © 2006 Red Hat, Inc.
  * Copyright © 2006 David Woodhouse <[email protected]>
  */
@@ -842,7 +845,8 @@ static void __devexit cafe_nand_remove(struct pci_dev *pdev)
 }
 
 static struct pci_device_id cafe_nand_tbl[] = {
-	{ 0x11ab, 0x4100, PCI_ANY_ID, PCI_ANY_ID },
+	{ PCI_VENDOR_ID_MARVELL, PCI_DEVICE_ID_MARVELL_88ALP01_NAND,
+	  PCI_ANY_ID, PCI_ANY_ID },
 	{ }
 };
 
include/linux/pci_ids.h +1 -0
index ef6ef64..e6240b7 100644
@@ -1521,6 +1521,7 @@
 #define PCI_DEVICE_ID_MARVELL_GT64260	0x6430
 #define PCI_DEVICE_ID_MARVELL_MV64360	0x6460
 #define PCI_DEVICE_ID_MARVELL_MV64460	0x6480
+#define PCI_DEVICE_ID_MARVELL_88ALP01_NAND	0x4100
 #define PCI_DEVICE_ID_MARVELL_88ALP01_SD	0x4101
 
 #define PCI_VENDOR_ID_V3		0x11b0
Signature · 4 lines
1.5.5.1
-- 
David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation
#4 lore raw

PATCH 3/3 Define and use PCI_DEVICE_ID_MARVELL_88ALP01_CCIC for CAFÉ camera driver

In reply to a message not in the archive

Also, stop looking at the NAND controller (0x4100) and checking the device class. For a while during development, all three functions on the chip had the same ID. We made them fix that fairly promptly, and we can forget about it now.

---
(Carried in git://git.infradead.org/~dwmw2/random-2.6.git)

 drivers/media/video/cafe_ccic.c |   13 +++----------
 include/linux/pci_ids.h         |    1 +
 2 files changed, 4 insertions(+), 10 deletions(-)
drivers/media/video/cafe_ccic.c +3 -10
index c149b7d..ea0db81 100644
@@ -2091,15 +2091,8 @@ static int cafe_pci_probe(struct pci_dev *pdev,
 		const struct pci_device_id *id)
 {
 	int ret;
-	u16 classword;
 	struct cafe_camera *cam;
-	/*
-	 * Make sure we have a camera here - we'll get calls for
-	 * the other cafe devices as well.
-	 */
-	pci_read_config_word(pdev, PCI_CLASS_DEVICE, &classword);
-	if (classword != PCI_CLASS_MULTIMEDIA_VIDEO)
-		return -ENODEV;
+
 	/*
 	 * Start putting together one of our big camera structures.
 	 */
@@ -2287,8 +2280,8 @@ static int cafe_pci_resume(struct pci_dev *pdev)
 
 
 static struct pci_device_id cafe_ids[] = {
-	{ PCI_DEVICE(0x11ab, 0x4100) }, /* Eventual real ID */
-	{ PCI_DEVICE(0x11ab, 0x4102) }, /* Really eventual real ID */
+	{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL,
+		     PCI_DEVICE_ID_MARVELL_88ALP01_CCIC) },
 	{ 0, }
 };
 
include/linux/pci_ids.h +1 -0
index e6240b7..6cf53f4 100644
@@ -1523,6 +1523,7 @@
 #define PCI_DEVICE_ID_MARVELL_MV64460	0x6480
 #define PCI_DEVICE_ID_MARVELL_88ALP01_NAND	0x4100
 #define PCI_DEVICE_ID_MARVELL_88ALP01_SD	0x4101
+#define PCI_DEVICE_ID_MARVELL_88ALP01_CCIC	0x4102
 
 #define PCI_VENDOR_ID_V3		0x11b0
 #define PCI_DEVICE_ID_V3_V960		0x0001
Signature · 4 lines
1.5.5.1
-- 
David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation
Acked-by on v1 1/3
#2 lore raw
In reply to David Woodhouse
Quoted 4 lines · On Wed, 03 Sep 2008 11:48:17 +0100 David Woodhouse <[email protected]> wrote:

Probably better to use the official designation.

---

No objection here.

Signature · 6 lines
     -- Pierre Ossman

  WARNING: This correspondence is being monitored by the
  Swedish government. Make sure your server uses encryption
  for SMTP traffic and consider using PGP for end-to-end
  encryption.
Acked-by on v1 3/3
#5 lore raw

Re: [PATCH 3/3] Define and use PCI_DEVICE_ID_MARVELL_88ALP01_CCIC for CAFÉ camera driver

In reply to David Woodhouse
Quoted 4 lines · On Wed, 03 Sep 2008 11:48:22 +0100 David Woodhouse <[email protected]> wrote:

Also, stop looking at the NAND controller (0x4100) and checking the device class. For a while during development, all three functions on the chip had the same ID. We made them fix that fairly promptly, and we can forget about it now.

I see no harm in having the driver check the device to ensure it's what it thinks it is - we can probably afford the runtime overhead. But whatever, I'm not that concerned; toss in an

if you wish.

jon