Message ID | CAOMZO5B5q6d-JXSTvAAsQ_ngx2Vgxm0eQ9V+zwT_EjuU-Opvtg@mail.gmail.com |
---|---|
State | Changes Requested |
Delegated to: | Stefano Babic |
Headers | show |
Dirk, On Sun, Apr 29, 2012 at 1:26 PM, Fabio Estevam <festevam@gmail.com> wrote: > This should be fixed, but it seems it does not explain the problem you > are seeing. We need to check whether get_ahb_clk() is using the weak > definition of get_periph_clk() instead the one from clock.c. Yes, I confirmed that that get_ahb_clk was using the weak definition of get_periph_clk(). I removed the weak definition of get_periph_clk() and made it non-static and now it is exported via sys_proto.h. I sent v2 which should fix the problem you reported in v1. Thanks, Fabio Estevam
Hi Dirk,
On Sun, Apr 29, 2012 at 3:15 PM, Fabio Estevam <festevam@gmail.com> wrote:
> I sent v2 which should fix the problem you reported in v1.
When you have a chance, could you please try this version and let us
know if it fixes the issues you reported previously?
Thanks,
Fabio Estevam
--- a/arch/arm/cpu/armv7/imx-common/cpu.c +++ b/arch/arm/cpu/armv7/imx-common/cpu.c @@ -147,6 +147,4 @@ u32 get_ahb_clk(void) ahb_podf = reg >> MXC_CCM_CBCDR_AHB_PODF_OFFSET; return get_periph_clk() / (ahb_podf + 1); - - return 0; }
Dirk, On Sun, Apr 29, 2012 at 11:54 AM, Fabio Estevam <festevam@gmail.com> wrote: > Thanks for testing it. I found one mistake and will submit a fix shortly. Here is one issue I found: This should be fixed, but it seems it does not explain the problem you are seeing. We need to check whether get_ahb_clk() is using the weak definition of get_periph_clk() instead the one from clock.c. Regards, Fabio Estevam