bcm4908: workaround NAND controller #WP issue

There seems to be a problem with setting #WP. On the other hand ignoring
the #WP seems to work. rootfs_data UBI volume seems to persist changes.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>

diff --git a/target/linux/bcm4908/patches-5.4/400-mtd-rawnand-brcmnand-disable-WP-on-BCM4908.patch b/target/linux/bcm4908/patches-5.4/400-mtd-rawnand-brcmnand-disable-WP-on-BCM4908.patch
new file mode 100644 (file)
index 0000000..6506623
--- /dev/null
+++ b/target/linux/bcm4908/patches-5.4/400-mtd-rawnand-brcmnand-disable-WP-on-BCM4908.patch
@@ -0,0 +1,34 @@
+From: =?UTF-8?q?Rafa=C5=82=20Mi=C5=82ecki?= <rafal@milecki.pl>
+Date: Thu, 21 Jan 2021 10:44:53 +0100
+Subject: [PATCH] mtd: rawnand: brcmnand: disable WP on BCM4908
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+BCM4908 contains NAND controller version 0x0701 (v7.1). It means that
+NAND_WP should be available.
+
+For some reason setting #WP on doesn't result in clearing NAND_STATUS_WP
+status bit:
+[    1.077857] bcm63138_nand ff801800.nand: timeout on status poll (expected c0000040 got c00000c0)
+[    1.086832] bcm63138_nand ff801800.nand: nand #WP expected on
+
+For now try working without touching #WP.
+
+Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
+---
+
+--- a/drivers/mtd/nand/raw/brcmnand/brcmnand.c
++++ b/drivers/mtd/nand/raw/brcmnand/brcmnand.c
+@@ -38,7 +38,11 @@
+  * 1: NAND_WP is set by default, cleared for erase/write operations
+  * 2: NAND_WP is always cleared
+  */
++#if IS_ENABLED(CONFIG_ARCH_BCM4908)
++static int wp_on = 0;
++#else
+ static int wp_on = 1;
++#endif
+ module_param(wp_on, int, 0444);
+ 
+ /***********************************************************************