【现象】
实现了Linux kernel中的nand flash driver后，用mtd oob test去测试oob数据读写，出现错误：
# insmod mtd_oobtest.ko dev=3
=================================================
mtd_oobtest: MTD device: 3
mtd_oobtest: MTD device size 67108864, eraseblock size 524288, page size 4096, count of eraseblocks 128, pages per eraseblock 128, OOB size 128
mtd_oobtest: scanning for bad eraseblocks
mtd_oobtest: scanned 128 eraseblocks, 0 are bad
mtd_oobtest: test 1 of 5
mtd_oobtest: erasing whole device
mtd_oobtest: erased 128 eraseblocks
mtd_oobtest: writing OOBs of whole device
mtd_oobtest: written up to eraseblock 0
mtd_oobtest: written 128 eraseblocks
mtd_oobtest: verifying all eraseblocks
mtd_oobtest: error: verify failed at 0x0
mtd_oobtest: error: verify failed at 0x1000
mtd_oobtest: error: verify failed at 0x2000
mtd_oobtest: error: verify failed at 0x3000
mtd_oobtest: error: verify failed at 0x4000
mtd_oobtest: error: verify failed at 0x5000
mtd_oobtest: error: verify failed at 0x6000
mtd_oobtest: error: verify failed at 0x7000
。。。。。
mtd_oobtest: error: verify failed at 0x7e000
mtd_oobtest: error: verify failed at 0x7f000
mtd_oobtest: verified up to eraseblock 0
mtd_oobtest: error: verify failed at 0x80000
mtd_oobtest: error: verify failed at 0x81000
。。。
mtd_oobtest: error: verify failed at 0x3e7000
mtd_oobtest: error: verify failed at 0x3e8000
mtd_oobtest: error: too many errors
mtd_oobtest: error -1 occurred
【解决过程】
1.开始怀疑是数据没有写入进去，后来在mtd_oobtest.c中加入代码调试，发现数据写入是正确的。
2.后来，终于搞懂了，原来是nand_base.c中的，oob 的layout设置有误，
原先是：
static struct nand_ecclayout nand_oob_64 = {
.eccbytes = 32,
.eccpos = {
32, 33, 34, 35, 36, 37, 38, 39,
40, 41, 42, 43, 44, 45, 46, 47,
48, 49, 50, 51, 52, 53, 54, 55,
56, 57, 58, 59, 60, 61, 62, 63},
.oobfree = {
{.offset = 2,
.length = 30}}
};
static struct nand_ecclayout nand_oob_128 = {
.eccbytes = 64,
.eccpos = {
64, 65, 66, 67, 68, 69, 70, 71,
72, 73, 74, 75, 76, 77, 78, 79,
80, 81, 82, 83, 84, 85, 86, 87,
88, 89, 90, 91, 92, 93, 94, 95,
96, 97, 98, 99, 100, 101, 102, 103,
104, 105, 106, 107, 108, 109, 110, 111,
112, 113, 114, 115, 116, 117, 118, 119,
120, 121, 122, 123, 124, 125, 126, 127},
.oobfree = {
{.offset = 2,
.length = 30}}
};
应该改成：
static struct nand_ecclayout nand_oob_64 = {
.eccbytes = 32,
.eccpos = {
32, 33, 34, 35, 36, 37, 38, 39,
40, 41, 42, 43, 44, 45, 46, 47,
48, 49, 50, 51, 52, 53, 54, 55,
56, 57, 58, 59, 60, 61, 62, 63},
.oobfree = {
{.offset = 2,
.length = 22/* from 24-63 store HW BCH4 ECC if HW BCH4 ECC is enabled */}}
};
static struct nand_ecclayout nand_oob_128 = {
.eccbytes = 64,
.eccpos = {
64, 65, 66, 67, 68, 69, 70, 71,
72, 73, 74, 75, 76, 77, 78, 79,
80, 81, 82, 83, 84, 85, 86, 87,
88, 89, 90, 91, 92, 93, 94, 95,
96, 97, 98, 99, 100, 101, 102, 103,
104, 105, 106, 107, 108, 109, 110, 111,
112, 113, 114, 115, 116, 117, 118, 119,
120, 121, 122, 123, 124, 125, 126, 127},
.oobfree = {
{.offset = 2,
.length = 54/* from 56-127 store HW BCH4 ECC if HW BCH4 ECC is enabled */}}
};
因为我这里的特殊的HW 的BCH4 的ECC，oob区域：
对于2k pagesize的nand，占用了24到63；
对于4k pagesize的nand，占用了54到127；
所以，要改成对应的值，否则，还是像上面那样写成30和62的话，那oob空闲区域（oob中，除了开始的2个字节的用于判定是否坏块和ecc的大小之外）的最后8个字节，实际就是被HW BCH4的ecc占用了，
所以，准备好buffer后，每次去写入oob，再读出来的值，后面那8个字节都是不匹配的，所以oob test一直报错。像上面那样修改，留给足够的ECC的空间用于存放ecc的值之后，就解决问题了。
其他人如果遇到此类问题，也是根据你的具体情况去修改oob layout，也就可以了。
转载请注明：在路上 » 【已解决】MTD oob test出错：mtd_oobtest: error: verify failed at XXXX