Skip to contentSkip to conversation

OneNAND: Introduce chip_probe function

PATCH In review Kyungmin Park 2 participants 7 messages May 28, 2010 – Aug 5, 2010

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 OneNAND: Introduce chip_probe function

Samsung SoCs use the own OneNAND controler and detect OneNAND chip at power on. To use this feature, introduce the chip_probe function.

Also remove workaround for Samsung SoCs.

---

drivers/mtd/onenand/onenand_base.c +30 -12
index 26caf25..0b12102 100644
@@ -3730,17 +3730,16 @@ out:
 }
 
 /**
- * onenand_probe - [OneNAND Interface] Probe the OneNAND device
+ * onenand_chip_probe - [OneNAND Interface] The generic chip probe
  * @param mtd		MTD device structure
  *
  * OneNAND detection method:
  *   Compare the values from command with ones from register
  */
-static int onenand_probe(struct mtd_info *mtd)
+static int onenand_chip_probe(struct mtd_info *mtd)
 {
 	struct onenand_chip *this = mtd->priv;
-	int bram_maf_id, bram_dev_id, maf_id, dev_id, ver_id;
-	int density;
+	int bram_maf_id, bram_dev_id, maf_id, dev_id;
 	int syscfg;
 
 	/* Save system configuration 1 */
@@ -3763,12 +3762,6 @@ static int onenand_probe(struct mtd_info *mtd)
 	/* Restore system configuration 1 */
 	this->write_word(syscfg, this->base + ONENAND_REG_SYS_CFG1);
 
-	/* Workaround */
-	if (syscfg & ONENAND_SYS_CFG1_SYNC_WRITE) {
-		bram_maf_id = this->read_word(this->base + ONENAND_REG_MANUFACTURER_ID);
-		bram_dev_id = this->read_word(this->base + ONENAND_REG_DEVICE_ID);
-	}
-
 	/* Check manufacturer ID */
 	if (onenand_check_maf(bram_maf_id))
 		return -ENXIO;
@@ -3776,13 +3769,35 @@ static int onenand_probe(struct mtd_info *mtd)
 	/* Read manufacturer and device IDs from Register */
 	maf_id = this->read_word(this->base + ONENAND_REG_MANUFACTURER_ID);
 	dev_id = this->read_word(this->base + ONENAND_REG_DEVICE_ID);
-	ver_id = this->read_word(this->base + ONENAND_REG_VERSION_ID);
-	this->technology = this->read_word(this->base + ONENAND_REG_TECHNOLOGY);
 
 	/* Check OneNAND device */
 	if (maf_id != bram_maf_id || dev_id != bram_dev_id)
 		return -ENXIO;
 
+	return 0;
+}
+
+/**
+ * onenand_probe - [OneNAND Interface] Probe the OneNAND device
+ * @param mtd		MTD device structure
+ */
+static int onenand_probe(struct mtd_info *mtd)
+{
+	struct onenand_chip *this = mtd->priv;
+	int maf_id, dev_id, ver_id;
+	int density;
+	int ret;
+
+	ret = this->chip_probe(mtd);
+	if (ret)
+		return ret;
+
+	/* Read manufacturer and device IDs from Register */
+	maf_id = this->read_word(this->base + ONENAND_REG_MANUFACTURER_ID);
+	dev_id = this->read_word(this->base + ONENAND_REG_DEVICE_ID);
+	ver_id = this->read_word(this->base + ONENAND_REG_VERSION_ID);
+	this->technology = this->read_word(this->base + ONENAND_REG_TECHNOLOGY);
+
 	/* Flash device information */
 	onenand_print_device_info(dev_id, ver_id);
 	this->device_id = dev_id;
@@ -3909,6 +3924,9 @@ int onenand_scan(struct mtd_info *mtd, int maxchips)
 	if (!this->unlock_all)
 		this->unlock_all = onenand_unlock_all;
 
+	if (!this->chip_probe)
+		this->chip_probe = onenand_chip_probe;
+
 	if (!this->read_bufferram)
 		this->read_bufferram = onenand_read_bufferram;
 	if (!this->write_bufferram)
include/linux/mtd/onenand.h +2 -0
index c26ff86..0c8815b 100644
@@ -68,6 +68,7 @@ struct onenand_bufferram {
  * @write_word:		[REPLACEABLE] hardware specific function for write
  *			register of OneNAND
  * @mmcontrol:		sync burst read function
+ * @chip_probe:		[REPLACEABLE] hardware specific function for chip probe
  * @block_markbad:	function to mark a block as bad
  * @scan_bbt:		[REPLACEALBE] hardware specific function for scanning
  *			Bad block Table
@@ -114,6 +115,7 @@ struct onenand_chip {
 	unsigned short (*read_word)(void __iomem *addr);
 	void (*write_word)(unsigned short value, void __iomem *addr);
 	void (*mmcontrol)(struct mtd_info *mtd, int sync_read);
+	int (*chip_probe)(struct mtd_info *mtd);
 	int (*block_markbad)(struct mtd_info *mtd, loff_t ofs);
 	int (*scan_bbt)(struct mtd_info *mtd);
 
#1 lore raw
Quoted 6 lines · Kyungmin Park

Samsung SoCs use the own OneNAND controler and detect OneNAND chip at power on. To use this feature, introduce the chip_probe function.

Also remove workaround for Samsung SoCs.

Pushed both to my l2-mtd-2.6.git / dunno.

Signature · 2 lines
Best Regards,
Artem Bityutskiy (Артём Битюцкий)
#2 lore raw

Hi Artem,

Can merge it next merge window. It's clean method for samsung socs and no functional changes for others.

Thank you,
Kyungmin Park

Quoted 15 lines · Artem Bityutskiy
Quoted 6 lines · Kyungmin Park

Samsung SoCs use the own OneNAND controler and detect OneNAND chip at power on. To use this feature, introduce the chip_probe function.

Also remove workaround for Samsung SoCs.

Pushed both to my l2-mtd-2.6.git / dunno.

--
Best Regards,
Artem Bityutskiy (Артём Битюцкий)

#3 lore raw
Quoted 4 lines · Kyungmin Park

Hi Artem,

Can merge it next merge window. It's clean method for samsung socs and no functional changes for others.

Hi,

I do not merge things, I'm just a "secretary" who picks up patches and passes them to dwmw2. He merges those he approves. So, closer to the merge window he should look at my l2 tree (he usually does) and pick up your stuff.

Signature · 2 lines
Best Regards,
Artem Bityutskiy (Артём Битюцкий)
#4 lore raw
Quoted 6 lines · Artem Bityutskiy
Quoted 4 lines · Kyungmin Park

Hi Artem,

Can merge it next merge window. It's clean method for samsung socs and no functional changes for others.

Quoted 6 lines

Hi,

I do not merge things, I'm just a "secretary" who picks up patches and passes them to dwmw2. He merges those he approves. So, closer to the merge window he should look at my l2 tree (he usually does) and pick up your stuff.

I see
At that time, I will notice again if it's missing.

Thank you,
Kyungmin Park

#5 lore raw

Hi,

Can you check it again?
It seems not merged at linux-next tree.

Thank you,
Kyungmin Park

Quoted 20 lines · Kyungmin Park
Quoted 6 lines · Artem Bityutskiy
Quoted 4 lines · Kyungmin Park

Hi Artem,

Can merge it next merge window. It's clean method for samsung socs and no functional changes for others.

Quoted 6 lines

Hi,

I do not merge things, I'm just a "secretary" who picks up patches and passes them to dwmw2. He merges those he approves. So, closer to the merge window he should look at my l2 tree (he usually does) and pick up your stuff.

I see
At that time, I will notice again if it's missing.

Thank you,
Kyungmin Park

#6 lore raw
Quoted 4 lines · Kyungmin Park

Hi,

Can you check it again?
It seems not merged at linux-next tree.

It is now in the mtd tree.

Signature · 2 lines
Best Regards,
Artem Bityutskiy (Артём Битюцкий)