| Message ID | a5aae850-5768-481a-9a89-380e65bd661e@AM1EHSMHS015.ehs.local |
|---|---|
| State | Accepted |
| Delegated to: | Michal Simek |
| Headers | show |
On 02/17/2014 07:12 AM, Siva Durga Prasad Paladugu wrote: > Typecast the argument with unsigned long > long for proper calculation of lldiv > > Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com> > --- > -Fixes the issue of rounding of the multiplication result to > 32-bit value We found this in connection to eeprom write test. Signed-off-by: Michal Simek <michal.simek@xilinx.com> Albert: This is the bug fixed I was talking you about on IRC. Will add it to my zynq pull request if this is fine for others. Thanks, Michal
Hi Michal, On Mon, 17 Feb 2014 07:50:26 +0100, Michal Simek <monstr@monstr.eu> wrote: > On 02/17/2014 07:12 AM, Siva Durga Prasad Paladugu wrote: > > Typecast the argument with unsigned long > > long for proper calculation of lldiv > > > > Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com> > > --- > > -Fixes the issue of rounding of the multiplication result to > > 32-bit value > > We found this in connection to eeprom write test. > > Signed-off-by: Michal Simek <michal.simek@xilinx.com> > > Albert: This is the bug fixed I was talking you about on IRC. > Will add it to my zynq pull request if this is fine for others. Fine for me at least. As soon as my current PR to u-boot/master is applied (cc: Tom), I'll resume applying patches, starting with your Zynq PR, so if it was updated meanwhile, I'll apply the updated one. > Thanks, > Michal Amicalement,
On 02/17/2014 10:07 AM, Albert ARIBAUD wrote: > Hi Michal, > > On Mon, 17 Feb 2014 07:50:26 +0100, Michal Simek <monstr@monstr.eu> > wrote: > >> On 02/17/2014 07:12 AM, Siva Durga Prasad Paladugu wrote: >>> Typecast the argument with unsigned long >>> long for proper calculation of lldiv >>> >>> Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com> >>> --- >>> -Fixes the issue of rounding of the multiplication result to >>> 32-bit value >> >> We found this in connection to eeprom write test. >> >> Signed-off-by: Michal Simek <michal.simek@xilinx.com> >> >> Albert: This is the bug fixed I was talking you about on IRC. >> Will add it to my zynq pull request if this is fine for others. > > Fine for me at least. As soon as my current PR to u-boot/master is > applied (cc: Tom), I'll resume applying patches, starting with your Zynq > PR, so if it was updated meanwhile, I'll apply the updated one. I am keeping my eyes on it too. Thanks, Michal
diff --git a/arch/arm/cpu/armv7/zynq/timer.c b/arch/arm/cpu/armv7/zynq/timer.c index 2be253c..09a5547 100644 --- a/arch/arm/cpu/armv7/zynq/timer.c +++ b/arch/arm/cpu/armv7/zynq/timer.c @@ -107,7 +107,8 @@ void __udelay(unsigned long usec) if (usec == 0) return; - countticks = lldiv(TIMER_TICK_HZ * usec, 1000000); + countticks = lldiv(((unsigned long long)TIMER_TICK_HZ * usec), + 1000000); /* decrementing timer */ timeend = readl(&timer_base->counter) - countticks;
Typecast the argument with unsigned long long for proper calculation of lldiv Signed-off-by: Siva Durga Prasad Paladugu <sivadur@xilinx.com> --- -Fixes the issue of rounding of the multiplication result to 32-bit value --- arch/arm/cpu/armv7/zynq/timer.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-)