* Roger Quadros [off-list ref] [170508 03:44]:

Hi Tony,

On 06/05/17 01:37, Tony Lindgren wrote:

quoted

The width needs to be configured in bytes with 1 meaning 8-bit
access and 2 meaning 16-bit access.

As per the TRM, for GPMC_CONFIG1.DEVICESIZE
0 means 8-bit
1 means 16-bit

quoted

Cc: Peter Ujfalusi <redacted>
Cc: Roger Quadros <redacted>
Signed-off-by: Tony Lindgren <tony@atomide.com>
---
 drivers/memory/omap-gpmc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/memory/omap-gpmc.c b/drivers/memory/omap-gpmc.c
--- a/drivers/memory/omap-gpmc.c
+++ b/drivers/memory/omap-gpmc.c

@@ -513,7 +513,7 @@ static void gpmc_cs_show_timings(int cs, const char *desc)
 	pr_info("gpmc cs%i access configuration:\n", cs);
 	GPMC_GET_RAW_BOOL(GPMC_CS_CONFIG1,  4,  4, "time-para-granularity");
 	GPMC_GET_RAW(GPMC_CS_CONFIG1,  8,  9, "mux-add-data");
-	GPMC_GET_RAW_MAX(GPMC_CS_CONFIG1, 12, 13,
+	GPMC_GET_RAW_SHIFT_MAX(GPMC_CS_CONFIG1, 12, 13, 1,
 			 GPMC_CONFIG1_DEVICESIZE_MAX, "device-width");

setting shift 1 means the value will be multiplied by 2
so this will read 0 for 8-bit and 2 for 16-bit.

Is this what we want?

Yeah we already have debug output that is supposed to be usable
for configuring a new device based on debug output of the bootloader
timings. And the configured parameters use bytes for device-width.

I'd still prefer the old way of showing RAW values than interpreted ones.

We have that too in the debug output as the raw values of
all the GPMC_CS_CONFIG registers.

Regards,

Tony