On 01/28/2015 02:04 PM, Matt Turner wrote:
> On Wed, Jan 28, 2015 at 1:53 PM, Ian Romanick wrote:
>> On 01/28/2015 11:51 AM, Matt Turner wrote:
>>> What I'm asking is whether we suspect that they specifically want
>>> half-up behavior (speculation, so not likely insightful), or if
>>> there's a wa
On Wed, Jan 28, 2015 at 1:53 PM, Ian Romanick wrote:
> On 01/28/2015 11:51 AM, Matt Turner wrote:
>> What I'm asking is whether we suspect that they specifically want
>> half-up behavior (speculation, so not likely insightful), or if
>> there's a way we can emulate round-half-up behavior using rou
On 01/28/2015 11:51 AM, Matt Turner wrote:
> On Wed, Jan 28, 2015 at 11:20 AM, Ian Romanick wrote:
>> On 01/28/2015 10:31 AM, Matt Turner wrote:
>>> Note: this will round differently for x.5 where x is even.
>>>
>>> total instructions in shared programs: 5953897 -> 5948654 (-0.09%)
>>> instruction
On Wed, Jan 28, 2015 at 11:20 AM, Ian Romanick wrote:
> On 01/28/2015 10:31 AM, Matt Turner wrote:
>> Note: this will round differently for x.5 where x is even.
>>
>> total instructions in shared programs: 5953897 -> 5948654 (-0.09%)
>> instructions in affected programs: 88619 -> 83376 (-5.92%
On 01/28/2015 10:31 AM, Matt Turner wrote:
> Note: this will round differently for x.5 where x is even.
>
> total instructions in shared programs: 5953897 -> 5948654 (-0.09%)
> instructions in affected programs: 88619 -> 83376 (-5.92%)
> helped:696
> ---
> If we
Note: this will round differently for x.5 where x is even.
total instructions in shared programs: 5953897 -> 5948654 (-0.09%)
instructions in affected programs: 88619 -> 83376 (-5.92%)
helped:696
---
If we implemented round() differently from roundEven(), we sho
