From: mhench
The eLBC NAND driver currently follows up each program/write operation with a
read-back of the page, in order to [ostensibly] fill in ECC data for the
caller. However, the page address used for this read is always -1, so the read
will never work correctly. Remove this useless (and p
The eLBC NAND driver currently follows up each program/write operation
with a
read-back of the page, in order to [ostensibly] fill in ECC data for the
caller. However, the page address used for this read is always -1, so
the read
will never work correctly. Remove this useless (and potentially
prob
Ok, so to port Liu Shuo's huge page nand to fake 2k page patch to uboot
We either need to introduce a callback after mtd initialization
(Which would have to be outside the driver) or lie about the ID info
coming from the driver.
I choose the second.
Does this look reasonable to you guys?
The IPL bo
nand_wait();
+
+ buf_ofs = (page_in_block & 1) * 4096;
+
+ for (j = 0; j < 2048; j++)
+ dst[pos + j] = buf[buf_ofs + j];
+
+ page_addr2k++;
+ pos += 2048;
+ }
+}
+#else
static void nand_load(unsigned int offs, i
4 matches
Mail list logo
