Thanks, I've incorporated the changes into the v2 patch series.
-- Rebecca On 1/3/2024 3:56 PM, Ard Biesheuvel wrote:
Hi Rebecca, On Wed, 3 Jan 2024 at 21:44, Rebecca Cran <rebe...@os.amperecomputing.com> wrote:Fix the calculation of the timer period in GenericWatchdogDxe: we need to multiply before dividing to keep the values as integers. Signed-off-by: Rebecca Cran <rebe...@os.amperecomputing.com> --- ArmPkg/Drivers/GenericWatchdogDxe/GenericWatchdogDxe.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ArmPkg/Drivers/GenericWatchdogDxe/GenericWatchdogDxe.c b/ArmPkg/Drivers/GenericWatchdogDxe/GenericWatchdogDxe.c index 05df101d5f4b..8f02f38c64e3 100644 --- a/ArmPkg/Drivers/GenericWatchdogDxe/GenericWatchdogDxe.c +++ b/ArmPkg/Drivers/GenericWatchdogDxe/GenericWatchdogDxe.c @@ -119,7 +119,7 @@ WatchdogInterruptHandler ( // the timer period plus 1. // if (mWatchdogNotify != NULL) { - TimerPeriod = ((TIME_UNITS_PER_SECOND / mTimerFrequencyHz) * mNumTimerTicks); + TimerPeriod = ((TIME_UNITS_PER_SECOND * mNumTimerTicks) / mTimerFrequencyHz);Could we just store the timer period in a global mTimerPeriod, and get rid of mNumTimerTicks entirely? AFAICT, that would get rid of these calculations as well.mWatchdogNotify (TimerPeriod + 1); } @@ -260,7 +260,7 @@ WatchdogGetTimerPeriod ( return EFI_INVALID_PARAMETER; } - *TimerPeriod = ((TIME_UNITS_PER_SECOND / mTimerFrequencyHz) * mNumTimerTicks); + *TimerPeriod = ((TIME_UNITS_PER_SECOND * mNumTimerTicks) / mTimerFrequencyHz); return EFI_SUCCESS; } -- 2.34.1
-=-=-=-=-=-=-=-=-=-=-=- Groups.io Links: You receive all messages sent to this group. View/Reply Online (#113214): https://edk2.groups.io/g/devel/message/113214 Mute This Topic: https://groups.io/mt/103510102/21656 Group Owner: devel+ow...@edk2.groups.io Unsubscribe: https://edk2.groups.io/g/devel/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
