On 08/29/2013 12:07 PM, Joe Perches wrote:
>> On the topic of multiply vs. divide (which probably is not relevant to topic
>> at
>> hand though), since ARCompact doesn't have native divide, we end up
>> emulating it
>> using libgcc routines. That makes it slightly non-deterministic (not a big
>>
On Thu, 2013-08-29 at 06:24 +, Vineet Gupta wrote:
> On 08/29/2013 11:31 AM, Joe Perches wrote:
> > I think the whole thing is odd and it should simply be
> >
> > loops = loops_per_jiffy * usecs_to_jiffies(usecs)
[]
> On the topic of multiply vs. divide (which probably is not relevant to to
On 08/29/2013 11:31 AM, Joe Perches wrote:
> On Thu, 2013-08-29 at 05:55 +, Vineet Gupta wrote:
>
>> The intent of writing orig code was to generate only 1 MPYHU insn (32*32 =
>> high-part-64) for the whole math, at any optimization level whatsoever. If
>> the
>> first MPY is overflowing, u r
On Thu, 2013-08-29 at 05:55 +, Vineet Gupta wrote:
> The intent of writing orig code was to generate only 1 MPYHU insn (32*32 =
> high-part-64) for the whole math, at any optimization level whatsoever. If the
> first MPY is overflowing, u r likely spinning for > 10,000 usec (10ms) which
> is
On 08/29/2013 12:00 AM, Mischa Jonker wrote:
> Make sure that usecs is casted to long long, to ensure that the
> (usecs * 4295 * HZ) multiplication is 64 bit.
>
> Initially, the (usecs * 4295 * HZ) part was done as a 32 bit
> multiplication, with the result casted to 64 bit. This led to some bits
>
On Wed, 2013-08-28 at 19:12 +, Mischa Jonker wrote:
> Hello Joe,
>
> > I don't see the loops_per_jiffy initial shift << 32.
>
> loops_per_jiffy * HZ = loops_per_second
> loops_per_jiffy * HZ = 1,000,000 * loops_per_us
> loops_per_jiffy * HZ * 4295 = 4,295,000 * loops_per_us
>
> loops_per_jif
Hello Joe,
> I don't see the loops_per_jiffy initial shift << 32.
loops_per_jiffy * HZ = loops_per_second
loops_per_jiffy * HZ = 1,000,000 * loops_per_us
loops_per_jiffy * HZ * 4295 = 4,295,000 * loops_per_us
loops_per_jiffy * HZ * 4294.967296 = 2^32 * loops_per_us
> > > > - loops = ((lon
On Wed, 2013-08-28 at 18:53 +, Mischa Jonker wrote:
> > > Make sure that usecs is casted to long long, to ensure that the (usecs
> > > * 4295 * HZ) multiplication is 64 bit.
> > >
> > > Initially, the (usecs * 4295 * HZ) part was done as a 32 bit
> > > multiplication, with the result casted to
> > Make sure that usecs is casted to long long, to ensure that the (usecs
> > * 4295 * HZ) multiplication is 64 bit.
> >
> > Initially, the (usecs * 4295 * HZ) part was done as a 32 bit
> > multiplication, with the result casted to 64 bit. This led to some
> > bits falling off.
> >
> > Signed-off-
On Wed, 2013-08-28 at 20:29 +0200, Mischa Jonker wrote:
> Make sure that usecs is casted to long long, to ensure that the
> (usecs * 4295 * HZ) multiplication is 64 bit.
>
> Initially, the (usecs * 4295 * HZ) part was done as a 32 bit
> multiplication, with the result casted to 64 bit. This led to
Make sure that usecs is casted to long long, to ensure that the
(usecs * 4295 * HZ) multiplication is 64 bit.
Initially, the (usecs * 4295 * HZ) part was done as a 32 bit
multiplication, with the result casted to 64 bit. This led to some bits
falling off.
Signed-off-by: Mischa Jonker
---
arch/a
11 matches
Mail list logo
