
Hello Tom,
On 07.12.2011 18:14, Tom Rini wrote:
On Wed, Dec 7, 2011 at 2:38 AM, Thomas Weber weber@corscience.de wrote:
Use enable_gpmc_cs_config instead of local writing timing configuration for GPMC.
Signed-off-by: Thomas Weber weber@corscience.de
board/timll/devkit8000/devkit8000.c | 19 ++++++++++++------- 1 files changed, 12 insertions(+), 7 deletions(-)
diff --git a/board/timll/devkit8000/devkit8000.c b/board/timll/devkit8000/devkit8000.c index bcadfc4..dd46c35 100644 --- a/board/timll/devkit8000/devkit8000.c +++ b/board/timll/devkit8000/devkit8000.c @@ -48,6 +48,16 @@
DECLARE_GLOBAL_DATA_PTR;
+static u32 gpmc_net_config[GPMC_MAX_REG] = {
NET_GPMC_CONFIG1,
NET_GPMC_CONFIG2,
NET_GPMC_CONFIG3,
NET_GPMC_CONFIG4,
NET_GPMC_CONFIG5,
NET_GPMC_CONFIG6,
NET_GPMC_CONFIG7,
+};
Value 7 is computed in enable_gpmc_cs_config() so it should be a 0 in the struct (or just a comment for clarity, I've seen both I think).
The calculated value in enable_gpmc_cs_config() is same as NET_GPMC_CONFIG7. The gpmc_net_config[6] is ignored and not used in enable_gpmc_cs_config(). If you want a 0 at this position I will send a second version with a 0 for NET_GPMC_CONFIG7.
Thomas