chip_ready patch for SST39VF640xB

chip_ready patch for SST39VF640xB

[Niels Gram Jeppesen]

[-- Attachment #1: Type: text/plain, Size: 477 bytes --]


Hi,

In cfi_cmdset_0002.c, chip_ready seems to be unreliable on the SST39VF640xB chip. The attached patch add two extra reads to chip_ready in accordance with the chip data sheet (page 3, "Write Operation Status Detection"). This seems to make the write cycle completion detection reliable. I have not tested the patch on any other chips but I would think that it would be harmless on chips that do not need the extra reads.

Best regards; hope this helps,

Niels



[-- Attachment #2: cfi_cmdset_0002.c.patch --]
[-- Type: application/octet-stream, Size: 529 bytes --]

--- download/linux-2.6.19/drivers/mtd/chips/cfi_cmdset_0002.c	2006-11-29 22:57:37.000000000 +0100
+++ linux-2.6.19-ks8695/drivers/mtd/chips/cfi_cmdset_0002.c	2007-02-02 16:26:55.000000000 +0100
@@ -452,6 +452,14 @@
 	d = map_read(map, addr);
 	t = map_read(map, addr);
 
+	/* Do two extra reads to ensure that status detection works
+	 * on the SST39VF640xB
+	 */
+	if (map_word_equal(map, d, t))
+	  t = map_read(map, addr);
+	if (map_word_equal(map, d, t))
+	  t = map_read(map, addr);
+
 	return map_word_equal(map, d, t);
 }

Re: chip_ready patch for SST39VF640xB

[Artem Bityutskiy]

Hi,

On Fri, 2007-02-02 at 17:03 +0100, Niels Gram Jeppesen wrote:
> In cfi_cmdset_0002.c, chip_ready seems to be unreliable on the SST39VF640xB chip. The attached patch add two extra reads to chip_ready in accordance with the chip data sheet (page 3, "Write Operation Status Detection"). This seems to make the write cycle completion detection reliable. I have not tested the patch on any other chips but I would think that it would be harmless on chips that do not need the extra reads.

1. The patch is not applicable to mtd-2.6.git.
2. Please, add Siged-off-by line.
Look here please: http://www.linux-mtd.infradead.org/source.html

-- 
Best regards,
Artem Bityutskiy (Битюцкий Артём)