2020-08-07 23:23:04

by Crt Mori

Subject: [PATCH v3 1/2] iio:temperature:mlx90632: Reduce number of equal calulcations

TAdut4 was calculated each iteration although it did not change. In light
of near future additions of the Extended range DSP calculations, this
function refactoring will help reduce unrelated changes in that series as
well as reduce the number of new functions needed.

Signed-off-by: Crt Mori <[email protected]>
---
drivers/iio/temperature/mlx90632.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/iio/temperature/mlx90632.c b/drivers/iio/temperature/mlx90632.c
index 51b812bcff2e..763a148a0095 100644
--- a/drivers/iio/temperature/mlx90632.c
+++ b/drivers/iio/temperature/mlx90632.c
@@ -374,11 +374,11 @@ static s32 mlx90632_calc_temp_ambient(s16 ambient_new_raw, s16 ambient_old_raw,
}

static s32 mlx90632_calc_temp_object_iteration(s32 prev_object_temp, s64 object,
- s64 TAdut, s32 Fa, s32 Fb,
+ s64 TAdut, s64 TAdut4, s32 Fa, s32 Fb,
s32 Ga, s16 Ha, s16 Hb,
u16 emissivity)
{
- s64 calcedKsTO, calcedKsTA, ir_Alpha, TAdut4, Alpha_corr;
+ s64 calcedKsTO, calcedKsTA, ir_Alpha, Alpha_corr;
s64 Ha_customer, Hb_customer;

Ha_customer = ((s64)Ha * 1000000LL) >> 14ULL;
@@ -393,10 +393,6 @@ static s32 mlx90632_calc_temp_object_iteration(s32 prev_object_temp, s64 object,
Alpha_corr = emissivity * div64_s64(Alpha_corr, 100000LL);
Alpha_corr = div64_s64(Alpha_corr, 1000LL);
ir_Alpha = div64_s64((s64)object * 10000000LL, Alpha_corr);
- TAdut4 = (div64_s64(TAdut, 10000LL) + 27315) *
- (div64_s64(TAdut, 10000LL) + 27315) *
- (div64_s64(TAdut, 10000LL) + 27315) *
- (div64_s64(TAdut, 10000LL) + 27315);

return (int_sqrt64(int_sqrt64(ir_Alpha * 1000000000000LL + TAdut4))
- 27315 - Hb_customer) * 10;
@@ -406,17 +402,21 @@ static s32 mlx90632_calc_temp_object(s64 object, s64 ambient, s32 Ea, s32 Eb,
s32 Fa, s32 Fb, s32 Ga, s16 Ha, s16 Hb,
u16 tmp_emi)
{
- s64 kTA, kTA0, TAdut;
+ s64 kTA, kTA0, TAdut, TAdut4;
s64 temp = 25000;
s8 i;

kTA = (Ea * 1000LL) >> 16LL;
kTA0 = (Eb * 1000LL) >> 8LL;
TAdut = div64_s64(((ambient - kTA0) * 1000000LL), kTA) + 25 * 1000000LL;
+ TAdut4 = (div64_s64(TAdut, 10000LL) + 27315) *
+ (div64_s64(TAdut, 10000LL) + 27315) *
+ (div64_s64(TAdut, 10000LL) + 27315) *
+ (div64_s64(TAdut, 10000LL) + 27315);

/* Iterations of calculation as described in datasheet */
for (i = 0; i < 5; ++i) {
- temp = mlx90632_calc_temp_object_iteration(temp, object, TAdut,
+ temp = mlx90632_calc_temp_object_iteration(temp, object, TAdut, TAdut4,
Fa, Fb, Ga, Ha, Hb,
tmp_emi);
}
--
2.25.1

2020-08-08 10:33:11

by Andy Shevchenko

Subject: Re: [PATCH v3 1/2] iio:temperature:mlx90632: Reduce number of equal calulcations

On Sat, Aug 8, 2020 at 2:22 AM Crt Mori <[email protected]> wrote:
>
> TAdut4 was calculated each iteration although it did not change. In light
> of near future additions of the Extended range DSP calculations, this
> function refactoring will help reduce unrelated changes in that series as
> well as reduce the number of new functions needed.

I see you provided a helper for this calculus, but I think we may
reduce a lot of churn if you do the following:
- introduce helper and convert one/current user (no functional change!)
- do what you did in this patch, but it will move less LOCs
- do what you have done in the next one using the helper

--
With Best Regards,
Andy Shevchenko