Hi Gerhard,
On 2014-03-08 18:38, Gerhard Sittig wrote:
On Fri, Mar 07, 2014 at 18:56 +0100, Hannes Petermaier wrote:
- fix: return-value of 'i2c_set_bus_speed' was interpreted wrong
Signed-off-by: Hannes Petermaier <oe5...@oevsv.at>
---
board/BuR/kwb/board.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/board/BuR/kwb/board.c b/board/BuR/kwb/board.c
index 8aa16bc..8fb5e68 100644
--- a/board/BuR/kwb/board.c
+++ b/board/BuR/kwb/board.c
@@ -120,7 +120,7 @@ void am33xx_spl_board_init(void)
/* power-ON 3V3 via Resetcontroller */
oldspeed = i2c_get_bus_speed();
- if (0 != i2c_set_bus_speed(CONFIG_SYS_OMAP24_I2C_SPEED_PSOC)) {
+ if (0 <= i2c_set_bus_speed(CONFIG_SYS_OMAP24_I2C_SPEED_PSOC)) {
buf = RSTCTRL_FORCE_PWR_NEN;
i2c_write(RSTCTRL_ADDR, RSTCTRL_CTRLREG, 1,
(uint8_t *)&buf, sizeof(buf));
While you are at it, can you fixup this Yoda programming style
and use the regular idiom instead? It hurts the brain to have to
stop and read code "backwards" before seeing what's going on.
OK. I'll send v2. Maybe i can change my personal idiom.
I've learned to do this so to avoid a unintended assignment to a variable.
For exmaple:
if (var = 10)
compiler will do assignment to variable.
if (10 = var)
compiler will generate error.
But in these days "checkpatch.pl" will warn in such case.
virtually yours
Gerhard Sittig
best regards,
Hannes
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot
