On Apr 20, 2008, at 6:28 PM, Wolfgang Denk wrote:
In message Pine.LNX.4.64.0804181156410.4255@blarg.am.freescale.net you wrote:
eg. because of rounding error we can get 799Mhz instead of 800Mhz.
Signed-off-by: Dejan Minic minic@freescale.com Signed-off-by: Srikanth Srinivasan srikanth.srinivasan@freescale.com Signed-off-by: Kumar Gala galak@kernel.crashing.org
cpu/mpc85xx/cpu.c | 18 +++++++++++------- 1 files changed, 11 insertions(+), 7 deletions(-)
diff --git a/cpu/mpc85xx/cpu.c b/cpu/mpc85xx/cpu.c index dcd8817..6972bb1 100644 --- a/cpu/mpc85xx/cpu.c +++ b/cpu/mpc85xx/cpu.c @@ -65,6 +65,11 @@ struct cpu_type cpu_type_list [] = { CPU_TYPE_ENTRY(8572_E), };
+static inline unsigned long integer_round (unsigned long val, unsigned long div) +{
- return ((val + (div/2)) / div);
+}
As already mentioned, "integer_round" is not a good name for this. Also, it seems this should not live in cpu/mpc85xx/cpu.c but in a more central place and be used by lots of other code as well. Actually you might want to make it a macro to allow for different data types as well.
Ok. I've posted v3 of the patch that takes DIV_ROUND_UP from the kernel. See if that works for you.
- k