On Tuesday, May 01, 2018 01:02:39 PM Daniel Lezcano wrote: > On Thu, Apr 26, 2018 at 01:51:31PM +0200, Bartlomiej Zolnierkiewicz wrote: > > Cleanup code for enabling threshold interrupts in ->tmu_control > > method implementations. > > > > There should be no functional changes caused by this patch. > > > > Signed-off-by: Bartlomiej Zolnierkiewicz <[email protected]> > > --- > > drivers/thermal/samsung/exynos_tmu.c | 101 > > ++++++++++++----------------------- > > 1 file changed, 34 insertions(+), 67 deletions(-) > > > > diff --git a/drivers/thermal/samsung/exynos_tmu.c > > b/drivers/thermal/samsung/exynos_tmu.c > > index abe0737..9639acf 100644 > > --- a/drivers/thermal/samsung/exynos_tmu.c > > +++ b/drivers/thermal/samsung/exynos_tmu.c > > @@ -76,9 +76,6 @@ > > #define EXYNOS_TMU_THERM_TRIP_EN_SHIFT 12 > > > > #define EXYNOS_TMU_INTEN_RISE0_SHIFT 0 > > -#define EXYNOS_TMU_INTEN_RISE1_SHIFT 4 > > -#define EXYNOS_TMU_INTEN_RISE2_SHIFT 8 > > -#define EXYNOS_TMU_INTEN_RISE3_SHIFT 12 > > #define EXYNOS_TMU_INTEN_FALL0_SHIFT 16 > > > > #define EXYNOS_EMUL_TIME 0x57F0 > > @@ -136,13 +133,6 @@ > > #define EXYNOS7_TMU_TEMP_MASK 0x1ff > > #define EXYNOS7_PD_DET_EN_SHIFT 23 > > #define EXYNOS7_TMU_INTEN_RISE0_SHIFT 0 > > -#define EXYNOS7_TMU_INTEN_RISE1_SHIFT 1 > > -#define EXYNOS7_TMU_INTEN_RISE2_SHIFT 2 > > -#define EXYNOS7_TMU_INTEN_RISE3_SHIFT 3 > > -#define EXYNOS7_TMU_INTEN_RISE4_SHIFT 4 > > -#define EXYNOS7_TMU_INTEN_RISE5_SHIFT 5 > > -#define EXYNOS7_TMU_INTEN_RISE6_SHIFT 6 > > -#define EXYNOS7_TMU_INTEN_RISE7_SHIFT 7 > > #define EXYNOS7_EMUL_DATA_SHIFT 7 > > #define EXYNOS7_EMUL_DATA_MASK 0x1ff > > > > @@ -615,29 +605,27 @@ static void exynos4210_tmu_control(struct > > platform_device *pdev, bool on) > > { > > struct exynos_tmu_data *data = platform_get_drvdata(pdev); > > struct thermal_zone_device *tz = data->tzd; > > - unsigned int con, interrupt_en; > > + unsigned int con, interrupt_en = 0, i; > > > > con = get_con_reg(data, readl(data->base + EXYNOS_TMU_REG_CONTROL)); > > > > if (on) { > > - con |= (1 << EXYNOS_TMU_CORE_EN_SHIFT); > > - interrupt_en = > > - (of_thermal_is_trip_valid(tz, 3) > > - << EXYNOS_TMU_INTEN_RISE3_SHIFT) | > > - (of_thermal_is_trip_valid(tz, 2) > > - << EXYNOS_TMU_INTEN_RISE2_SHIFT) | > > - (of_thermal_is_trip_valid(tz, 1) > > - << EXYNOS_TMU_INTEN_RISE1_SHIFT) | > > - (of_thermal_is_trip_valid(tz, 0) > > - << EXYNOS_TMU_INTEN_RISE0_SHIFT); > > + for (i = 0; i < data->ntrip; i++) { > > + if (!of_thermal_is_trip_valid(tz, i)) > > + continue; > > + > > + interrupt_en |= > > + (1 << (EXYNOS_TMU_INTEN_RISE0_SHIFT + i * 4)); > > + } > > As EXYNOS_TMU_INTEN_RISE0_SHIFT is equal to zero, may be you can replace this > by BITS(i * 4) ? > > Same comments for exynos5433 and exynos7 below.
Good point, I will replace it by using BIT() macro. > I don't know which one was intended : The one that doesn't change the driver behavior as stated in the patch description. > ((EXYNOS_TMU_INTEN_RISE0_SHIFT + i) * 4) or > (EXYNOS_TMU_INTEN_RISE0_SHIFT + (i * 4)) > > but if it is the former it is lucky it works because the macro is zero. > > Is it possible to have the registers layout, that would facilitate the review. I'm sorry but Exynos TMU documentation is not publicly available. Best regards, -- Bartlomiej Zolnierkiewicz Samsung R&D Institute Poland Samsung Electronics
