Estimate quotient digit in Bigint division using an integer division instead of

Estimate quotient digit in Bigint division using an integer division instead of
a floating point division.

R=srdjan@google.com

Committed: https://code.google.com/p/dart/source/detail?r=40365

[regis, 2014-09-17 00:17:39]

regis@google.com changed reviewers:
+ srdjan@google.com

[srdjan, 2014-09-17 00:31:09]

lgtm

[regis, 2014-09-17 00:36:48]

On 2014/09/17 00:31:09, srdjan wrote:
> lgtm

Although all the tests pass, I need to double check what happens when y0 is
exactly 1, i.e. yt becomes 0. I probably have a bug. I wish we could represent
all uint64_t as mint :-)

[regis, 2014-09-17 01:44:20]

On 2014/09/17 00:36:48, regis wrote:
> On 2014/09/17 00:31:09, srdjan wrote:
> > lgtm
> 
> Although all the tests pass, I need to double check what happens when y0 is
> exactly 1, i.e. yt becomes 0. I probably have a bug. I wish we could represent
> all uint64_t as mint :-)

Never mind. y is normalized, so yt cannot become 0. Added comment.

[regis, 2014-09-17 01:44:38]

Committed patchset #2 (id:20001) manually as r40365 (presubmit successful).