/*
************************************************************************************************************************
* NAND FLASH LOGIC MANAGE LAYER READ-RECLAIM
*
*Description: Repair the logic block whose data has reach the limit of the ability of
* the HW ECC module correct.
*
*Arguments : nPage the page address where need be repaired.
*
*Return : read-reclaim result;
* = 0 do read-reclaim successful;
* = -1 do read-reclaim failed.
*
*Notes : if read a physical page millions of times, there may be some bit error in
* the page, and the error bit number will increase along with the read times,
* so, if the number of the error bit reachs the ECC limit, the data should be
* read out and write to another physical blcok.
************************************************************************************************************************
*/
__s32 LML_ReadReclaim(__u32 nPage)
{
__s32 result;
#if CFG_SUPPORT_READ_RECLAIM
LOGICCTL_ERR("[LOGICCTL_ERR] read reclaim go\n");
//flush the page cache to nand flash first, because need use the buffer
result = LML_FlushPageCache();
if(result < 0)
{
LOGICCTL_ERR("[LOGICCTL_ERR] Flush page cache failed when do read reclaim! Error:0x%x\n", result);
return -1;
}
//read the full page data to buffer
result = LML_PageRead(nPage, FULL_BITMAP_OF_LOGIC_PAGE, LML_WRITE_PAGE_CACHE);
if(result < 0)
{
return -1;
}
//the data in the page cache is full, write it to nand flash
result = LML_PageWrite(nPage, FULL_BITMAP_OF_LOGIC_PAGE, LML_WRITE_PAGE_CACHE);
if(result < 0)
{
return -1;
}
#endif
return 0;
}
