The patch titled
i2c/tps65010: update for current i2c-omap
has been added to the -mm tree. Its filename is
i2c-tps65010-update-for-current-i2c-omap.patch
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: i2c/tps65010: update for current i2c-omap
From: David Brownell <david-b@xxxxxxxxxxx>
This updates the TPS6501x driver to stop trying to use the faked-out I2C probe
protocol (never guaranteed to work), since the OMAP I2C driver merged to
kernel.org no longer tries to support it.
Until the I2C stack is updated to offer full driver model support from static
board-specific device tables, and is able to pass things like IRQ resources
and chip IDs, the only way to address this type of issue is for drivers to
have board-specific info kick in before registering the driver (for
addressing), and in the probe code (for other resources). So that's what this
patch adds.
Signed-off-by: David Brownell <dbrownell@xxxxxxxxxxxxxxxxxxxxx>
Cc: Jean Delvare <khali@xxxxxxxxxxxx>
Cc: Greg KH <greg@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxx>
---
drivers/i2c/chips/tps65010.c | 38 ++++++++++++++++++++++++++++++---
1 files changed, 35 insertions(+), 3 deletions(-)
diff -puN drivers/i2c/chips/tps65010.c~i2c-tps65010-update-for-current-i2c-omap drivers/i2c/chips/tps65010.c
--- a/drivers/i2c/chips/tps65010.c~i2c-tps65010-update-for-current-i2c-omap
+++ a/drivers/i2c/chips/tps65010.c
@@ -42,15 +42,38 @@
/*-------------------------------------------------------------------------*/
-#define DRIVER_VERSION "2 May 2005"
+#define DRIVER_VERSION "28 Sept 2006"
#define DRIVER_NAME (tps65010_driver.driver.name)
MODULE_DESCRIPTION("TPS6501x Power Management Driver");
MODULE_LICENSE("GPL");
-static unsigned short normal_i2c[] = { 0x48, /* 0x49, */ I2C_CLIENT_END };
-I2C_CLIENT_INSMOD;
+/* The i2c-omap driver doesn't support the i2c request used to fake probes,
+ * so we need board-specific tables saying what devices exist.
+ *
+ * Until the I2C stack doesn't support such tables, we need to fake them
+ * in the module initialization code (and in driver init for IRQs etc).
+ *
+ * Note we expect only ONE tps chip per board; no point in having more.
+ */
+
+static unsigned short force[3] __devinitdata = {
+ 0, 0x48, /* default address */
+ I2C_CLIENT_END,
+};
+static unsigned short *forces[] __devinitdata = { force, NULL, };
+
+static unsigned short ignore[] __devinitdata = {
+ I2C_CLIENT_END, I2C_CLIENT_END,
+};
+
+static struct i2c_client_address_data addr_data __devinitdata = {
+ .forces = forces,
+ .probe = ignore,
+ .normal_i2c = ignore,
+ .ignore = ignore,
+};
static struct i2c_driver tps65010_driver;
@@ -990,6 +1013,15 @@ static int __init tps_init(void)
u32 tries = 3;
int status = -ENODEV;
+#ifdef CONFIG_ARM
+ if (machine_is_omap_osk()
+ || machine_is_omap_h2()
+ || machine_is_omap_h3())
+ /* ok */ ;
+ else
+ return status;
+#endif
+
printk(KERN_INFO "%s: version %s\n", DRIVER_NAME, DRIVER_VERSION);
/* some boards have startup glitches */
_
Patches currently in -mm which might be from david-b@xxxxxxxxxxx are
origin.patch
i2c-tps65010-update-for-current-i2c-omap.patch
git-pcmcia.patch
ide-reprogram-disk-pio-timings-on-resume.patch
rtc-class-kconfig-improvements.patch
rtc-class-uses-subsys_init.patch
rtc-class-error-checks.patch
constify-rtc_class_ops-update-drivers.patch
at91rm9200-rtc-can-issue-system-wakeup-events.patch
omap_cf-works-again-sync-with-linux-omap-tree.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
