fsl_elbc_nand: workaround for hangs during nand write
authorAnton Vorontsov <avorontsov@ru.mvista.com>
Fri, 28 Mar 2008 19:10:54 +0000 (22:10 +0300)
committerScott Wood <scottwood@freescale.com>
Tue, 12 Aug 2008 16:31:26 +0000 (11:31 -0500)
Using current driver elbc sometimes hangs during nand write. Reading back
last byte helps though (thanks to Scott Wood for the idea).

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Scott Wood <scottwood@freescale.com>
drivers/mtd/nand/fsl_elbc_nand.c

index c1644c0..ab3e0fd 100644 (file)
@@ -489,7 +489,7 @@ static void fsl_elbc_write_buf(struct mtd_info *mtd, const u8 *buf, int len)
        struct fsl_elbc_ctrl *ctrl = priv->ctrl;
        unsigned int bufsize = mtd->writesize + mtd->oobsize;
 
-       if (len < 0) {
+       if (len <= 0) {
                printf("write_buf of %d bytes", len);
                ctrl->status = 0;
                return;
@@ -503,6 +503,15 @@ static void fsl_elbc_write_buf(struct mtd_info *mtd, const u8 *buf, int len)
        }
 
        memcpy_toio(&ctrl->addr[ctrl->index], buf, len);
+       /*
+        * This is workaround for the weird elbc hangs during nand write,
+        * Scott Wood says: "...perhaps difference in how long it takes a
+        * write to make it through the localbus compared to a write to IMMR
+        * is causing problems, and sync isn't helping for some reason."
+        * Reading back the last byte helps though.
+        */
+       in_8(&ctrl->addr[ctrl->index] + len - 1);
+
        ctrl->index += len;
 }