[PATCH 1/4] time: Avoid 64-bit by 64-bit division in time64_to_tm()

[Thomas Weißschuh (Schneider Electric)]

As the divisor is static and known to fit into 32 bits, a 64/32 division
is enough. The 64/64 is unnecessary and very expensive on architectures
without native 64/64 division instructions.

This brings the time64_to_tm_test_date_range() runtime from 45 seconds
down to 30 seconds in an emulated 32-bit ARM machine.

Signed-off-by: Thomas Weißschuh (Schneider Electric) <thomas.weissschuh@xxxxxxxxxxxxx>
---
 kernel/time/timeconv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/time/timeconv.c b/kernel/time/timeconv.c
index 59b922c826e7..779b91898276 100644
--- a/kernel/time/timeconv.c
+++ b/kernel/time/timeconv.c
@@ -105,8 +105,8 @@ void time64_to_tm(time64_t totalsecs, int offset, struct tm *result)
 	udays	= ((u64) days) + 2305843009213814918ULL;
 
 	u64tmp		= 4 * udays + 3;
-	century		= div64_u64_rem(u64tmp, 146097, &u64tmp);
-	day_of_century	= (u32) (u64tmp / 4);
+	century		= div_u64_rem(u64tmp, 146097, &u32tmp);
+	day_of_century	= u32tmp / 4;
 
 	u32tmp		= 4 * day_of_century + 3;
 	u64tmp		= 2939745ULL * u32tmp;

-- 
2.55.0