[PATCH 0/2] bus: sunxi-rsb: fix primary PMIC and peripheral IC address mapping

[PATCH 0/2] bus: sunxi-rsb: fix primary PMIC and peripheral IC address mapping

[Chen-Yu Tsai]

Hi Arnd, Olof,

Here are 2 fixes for the sunxi-rsb driver. There were some typos in the
hardware address mapping table, but as long as the assigned address
matched what the bootloader used, it still worked fine. The only
indication was "failed to set runtime address" error messages which I
mistook for some hardware quirk.

I looked deeper when the AC100 on my A80 boards refused to respond.

Patch 1 fixes the hardware address for primary PMICs (AXP223, AXP809...).
This is the part that still worked fine despite the error because the
runtime address matches the one set by the bootloader.

Patch 2 fixes the assigned runtime address for the peripheral IC
(AC100/AC200). This part does not work due to the runtime address being
invalid.

Maxime asked me to send them directly to you. Please pick them up for 4.4.


Regards
ChenYu

P.S. my primary email is down this week, but should be fixed by the end
of the week.

Chen-Yu Tsai (2):
  bus: sunxi-rsb: Fix primary PMIC mapping hardware address
  bus: sunxi-rsb: Fix peripheral IC mapping runtime address

 drivers/bus/sunxi-rsb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

-- 
2.6.4

[PATCH 1/2] bus: sunxi-rsb: Fix primary PMIC mapping hardware address

[Chen-Yu Tsai]

The primary PMICs use 0x3a3 as their hardware address, not 0x3e3.

Signed-off-by: Chen-Yu Tsai <wens@csie.org>
---
 drivers/bus/sunxi-rsb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bus/sunxi-rsb.c b/drivers/bus/sunxi-rsb.c
index 0cfcb39c53f4..f951b3555175 100644
--- a/drivers/bus/sunxi-rsb.c
+++ b/drivers/bus/sunxi-rsb.c
@@ -527,7 +527,7 @@ static int sunxi_rsb_init_device_mode(struct sunxi_rsb *rsb)
  */
 
 static const struct sunxi_rsb_addr_map sunxi_rsb_addr_maps[] = {
-	{ 0x3e3, 0x2d }, /* Primary PMIC: AXP223, AXP809, AXP81X, ... */
+	{ 0x3a3, 0x2d }, /* Primary PMIC: AXP223, AXP809, AXP81X, ... */
 	{ 0x745, 0x3a }, /* Secondary PMIC: AXP806, ... */
 	{ 0xe89, 0x45 }, /* Peripheral IC: AC100, ... */
 };
-- 
2.6.4

[PATCH 2/2] bus: sunxi-rsb: Fix peripheral IC mapping runtime address

[Chen-Yu Tsai]

0x4e is the runtime address normally associated with perihperal ICs.
0x45 is not a valid runtime address.

Signed-off-by: Chen-Yu Tsai <wens@csie.org>
---
 drivers/bus/sunxi-rsb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/bus/sunxi-rsb.c b/drivers/bus/sunxi-rsb.c
index f951b3555175..25996e256110 100644
--- a/drivers/bus/sunxi-rsb.c
+++ b/drivers/bus/sunxi-rsb.c
@@ -529,7 +529,7 @@ static int sunxi_rsb_init_device_mode(struct sunxi_rsb *rsb)
 static const struct sunxi_rsb_addr_map sunxi_rsb_addr_maps[] = {
 	{ 0x3a3, 0x2d }, /* Primary PMIC: AXP223, AXP809, AXP81X, ... */
 	{ 0x745, 0x3a }, /* Secondary PMIC: AXP806, ... */
-	{ 0xe89, 0x45 }, /* Peripheral IC: AC100, ... */
+	{ 0xe89, 0x4e }, /* Peripheral IC: AC100, ... */
 };
 
 static u8 sunxi_rsb_get_rtaddr(u16 hwaddr)
-- 
2.6.4

[PATCH 0/2] bus: sunxi-rsb: fix primary PMIC and peripheral IC address mapping

[Olof Johansson]

On Wed, Dec 16, 2015 at 05:14:44PM +0800, Chen-Yu Tsai wrote:
> Hi Arnd, Olof,
> 
> Here are 2 fixes for the sunxi-rsb driver. There were some typos in the
> hardware address mapping table, but as long as the assigned address
> matched what the bootloader used, it still worked fine. The only
> indication was "failed to set runtime address" error messages which I
> mistook for some hardware quirk.
> 
> I looked deeper when the AC100 on my A80 boards refused to respond.
> 
> Patch 1 fixes the hardware address for primary PMICs (AXP223, AXP809...).
> This is the part that still worked fine despite the error because the
> runtime address matches the one set by the bootloader.
> 
> Patch 2 fixes the assigned runtime address for the peripheral IC
> (AC100/AC200). This part does not work due to the runtime address being
> invalid.
> 
> Maxime asked me to send them directly to you. Please pick them up for 4.4.

Applied, thanks.

In the future, for fixes late in the cycle like this, please describe in a few
words in the commit message just what the error is for someone trying to use
kernel without the fix, so someone just reading the changelog can determine if
it's a high-priority fix or not.


Thanks!


-Olof