https://codereview.chromium.org/1157663007/diff/20001/src/compiler/register-allocator.h
File src/compiler/register-allocator.h (right):
https://codereview.chromium.org/1157663007/diff/20001/src/compiler/register-allocator.h#newcode485
src/compiler/register-allocator.h:485: float weight_;
On 2015/06/03 15:06:16, jarin wrote:
Could you please split the fields into groups by their use in the
allocator
(three groups: (1) used by linear scan only, (2) used by greedy only,
(3) used
by both). I am worried that too much bloat might hurt performance of
linear
scan, the split could help with experiments measuring impact of the
bloat. It
would also be good for documentation purposes.
Actually, the greedy introduces 3 fields only, the rest end up being
used by both. So there's no category 1, at least just yet.
https://codereview.chromium.org/1157663007/diff/20001/src/compiler/register-allocator.h#newcode940
src/compiler/register-allocator.h:940: LiveRange* operator*() { return
pos_->range; }
On 2015/06/03 15:06:16, jarin wrote:
Should not the star operator return a reference rather than a pointer?
It returns the item "pointed at" by the iterator, so LiveRange*.
https://codereview.chromium.org/1157663007/diff/20001/src/compiler/register-allocator.h#newcode950
src/compiler/register-allocator.h:950: bool operator!=(const
conflict_iterator& other) {
On 2015/06/03 15:06:16, jarin wrote:
Why is not this is simply negation of the equality?
Even if there is reason to do it this way, please, do not use bitwise
xor for
bools. You can use != instead.
Done.
https://codereview.chromium.org/1157663007/diff/20001/src/disassembler.cc
File src/disassembler.cc (right):
https://codereview.chromium.org/1157663007/diff/20001/src/disassembler.cc#newcode149
src/disassembler.cc:149: out.AddFormatted("%p %4X ", prev_pc, prev_pc
- begin);
On 2015/06/03 15:06:16, jarin wrote:
Why did you change this?
I am worried that tools relying on formatting here might be unhappy.
"perf" (the tool) outputs offsets as hex, so this makes perf analysis
much easier. I share your worry though. I'll change to having a flag
perhaps?
https://codereview.chromium.org/1157663007/
--
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.com/group/v8-dev
---
You received this message because you are subscribed to the Google Groups "v8-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to v8-dev+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
