One example I observed sometime ago is in Crankshaft: If a phi determines 
that its type to be a heap number, and one of its incoming edges is a 
constant smi, we use a heap number to contain that smi.

On Monday, December 8, 2014 at 10:33:14 PM UTC+1, Adam Klein wrote:
>
> Makes sense, thanks.
>
> On Mon, Dec 8, 2014 at 1:12 PM, 'Andreas Rossberg' via v8-users <
> [email protected] <javascript:>> wrote:
>
>> On 8 December 2014 at 20:53, Adam Klein <[email protected] <javascript:>> 
>> wrote:
>> > Subject says it all. I was surprised not to find a DCHECK in
>> > Heap::AllocateHeapNumber asserting this is the case, and wondered why.
>>
>> You mean, do we normalise all numbers within smi range to smis? No,
>> since that would require potentially expensive extra logic for all
>> operators and intrinsic functions that could possibly return a number
>> in smi range for some arguments.
>>
>> /Andreas
>>
>> --
>> --
>> v8-users mailing list
>> [email protected] <javascript:>
>> http://groups.google.com/group/v8-users
>> ---
>> You received this message because you are subscribed to the Google Groups 
>> "v8-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 
-- 
v8-users mailing list
[email protected]
http://groups.google.com/group/v8-users
--- 
You received this message because you are subscribed to the Google Groups 
"v8-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to