On 02/01/2018 03:53 PM, Alexey Brodkin wrote:
Hi Marek,
On Wed, 2018-01-31 at 16:13 +0100, Marek Vasut wrote:
On 01/31/2018 03:56 PM, Alexey Brodkin wrote:
For some reason from day one we used to have both CONFIG_DWC2_UTMI_WIDTH mentioned in dwc2.h and in scripts/config_whitelist.txt but never really used and CONFIG_DWC2_UTMI_PHY_WIDTH used in real code in dwc2.c (but never defined).
Moreover even though CONFIG_DWC2_UTMI_WIDTH might be either 8 or 16 depending on hardware (and the same is said in a comment for it in dwc2.h) but then 8 is hardcoded in the header leaving no ability to override this value in board's configuration.
Signed-off-by: Alexey Brodkin abrodkin@synopsys.com Cc: Marek Vasut marex@denx.de
Applied, thanks.
btw How the heck did that code ever even compile ?
Well my change doesn't really fix anything in existing code except makes once check meaningful compared to dummy always negative as it was before :)
Ha, now I see it.
I'd say it means the driver was not very widely used on different hardwares. Still it does work [at least] to some extent which is really nice.
Probably. All this info should be pulled from DT anyway.