Thanks. I meant whether Smi should be untagged to the real int value
(without tag) when assigning to Operand immediate field.
On Saturday, April 21, 2018 at 1:27:44 AM UTC-7, Ben Noordhuis wrote:
>
> On Sat, Apr 21, 2018 at 2:58 AM, Thomson Tan <lil...@gmail.com
> <javascript:>> wrote:
> > The ARM64 backend could convert Smi to Operand by assigning tagged Smi
> to
> > immediate filed in Operand class, as below code snippet. Then the
> immediate
> > field is tagged Smi value, would it cause problem when encoding this
> > operand?
> >
> > // src\arm64\assember-arm64-inl.h
> > template<>
> > struct ImmediateInitializer<Smi*> {
> > static const bool kIsIntType = false;
> > static inline RelocInfo::Mode rmode_for(Smi* t) { return
> RelocInfo::NONE;
> > }
> > static inline int64_t immediate_for(Smi* t) {;
> > return reinterpret_cast<int64_t>(t);
> > }
> > };
>
> No, If I understand your question correctly.
>
> Smis are not heap objects and don't need additional bookkeeping.
> That's unlike embedded object pointers, those need to be made visible
> to the garbage collector.
>
--
--
v8-users mailing list
v8-users@googlegroups.com
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 v8-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
