Reviewers: Jakob,
Message:
Please take a look. I was trying to use counters to debug a performance
issue
and discovered that some counters do not work as callback is installed too
late
when stubs are already generated.
Description:
Make counter and histogram related callbacks part of the
https://codereview.chromium.org/961283002/diff/1/src/hydrogen-instructions.h
File src/hydrogen-instructions.h (right):
https://codereview.chromium.org/961283002/diff/1/src/hydrogen-instructions.h#newcode1194
src/hydrogen-instructions.h:1194: void SetOperandPositions(Zone* zone,
SourcePosition le
lgtm
https://codereview.chromium.org/962593005/
--
--
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
https://codereview.chromium.org/935393002/diff/40001/src/hydrogen.cc
File src/hydrogen.cc (right):
https://codereview.chromium.org/935393002/diff/40001/src/hydrogen.cc#newcode10497
src/hydrogen.cc:10497: left->IsConstant() &&
HConstant::cast(right)->HasStringValue()
one left is still here :)
ht
I think when I was copy pasting code for my comment I did it super wrong :)
https://codereview.chromium.org/935393002/diff/20001/src/hydrogen.cc
File src/hydrogen.cc (right):
https://codereview.chromium.org/935393002/diff/20001/src/hydrogen.cc#newcode10494
src/hydrogen.cc:10494: ? HConstant::ca
LGTM
if it compiles and runs :)
https://codereview.chromium.org/935393002/diff/1/src/hydrogen.cc
File src/hydrogen.cc (right):
https://codereview.chromium.org/935393002/diff/1/src/hydrogen.cc#newcode10502
src/hydrogen.cc:10502: if (left->IsConstant() &&
HConstant::cast(left)->HasStringValue()
https://codereview.chromium.org/931233002/diff/60001/tools/gcmole/gcmole.lua
File tools/gcmole/gcmole.lua (right):
https://codereview.chromium.org/931233002/diff/60001/tools/gcmole/gcmole.lua#newcode409
tools/gcmole/gcmole.lua:409: local function SearchForErrors(filename,
lines)
On 2015/02/18 11
https://codereview.chromium.org/931233002/diff/60001/tools/gcmole/gcmole.lua
File tools/gcmole/gcmole.lua (right):
https://codereview.chromium.org/931233002/diff/60001/tools/gcmole/gcmole.lua#newcode117
tools/gcmole/gcmole.lua:117: function IterTable(t)
my comment should obviously read:
n
lua code lgtm
https://codereview.chromium.org/931233002/diff/60001/tools/gcmole/gcmole.lua
File tools/gcmole/gcmole.lua (right):
https://codereview.chromium.org/931233002/diff/60001/tools/gcmole/gcmole.lua#newcode117
tools/gcmole/gcmole.lua:117: function IterTable(t)
I'd write it
local functio
Reviewers: Sven Panne,
Message:
PTAL
On Windows line endings normalization makes original line endings
unrecoverable
and as a result messes up source position matching in IRHydra.
Description:
CodeTracer should open file in binary mode to avoid line endings
normalization.
Please review t
LGTM (though it's kinda useless, I am not in OWNERs)
it would be good if you could run the test suite in the debug mode with
--hydrogen-track-position and see what/if something fails.
https://codereview.chromium.org/914413007/diff/20001/src/compiler.h
File src/compiler.h (right):
https://coder
Some comments from me.
I would like to also note that this tracking was designed with two
requirements
in mind:
- to work using whatever source is dumped, without access to full *live*
source;
- to be path sensitive, e.g. if you have function
function foo() {
bar();
bar();
}
and you
DBC (there's a bug, see below)
(nice sNaN idea!)
https://codereview.chromium.org/863633002/diff/21/src/hydrogen-instructions.cc
File src/hydrogen-instructions.cc (right):
https://codereview.chromium.org/863633002/diff/21/src/hydrogen-instructions.cc#newcode4056
src/hydrogen-instruction
fyi: this regressed EarleyBoyer by 40%
https://codereview.chromium.org/767253002/
--
--
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
LGTM,
I don't think this is used in IRHydra
[thanks for CCing me!]
https://codereview.chromium.org/716793003/
--
--
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.
-me +jkummerow
https://codereview.chromium.org/688533002/
--
--
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 emai
DBC
https://codereview.chromium.org/637313002/diff/150001/src/compiler/graph-visualizer.cc
File src/compiler/graph-visualizer.cc (right):
https://codereview.chromium.org/637313002/diff/150001/src/compiler/graph-visualizer.cc#newcode638
src/compiler/graph-visualizer.cc:638: os_ << " <|@\n";
coul
Committed patchset #1 (id:1) manually as 23814 (presubmit successful).
https://codereview.chromium.org/556453008/
--
--
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.
Reviewers: jarin,
Description:
Fix perf jitdump integration.
When emitting JIT_CODE_LOAD for Crankshafted code exclude Safepoint Table
from
the size of the code.
This cleans up perf annotate output - because otherwise it confusingly
tries to
disassemble safepoint table.
R=ja...@chromium.
Ok, here is another attempt.
Digging back in history it seems that current AsUC16 behaves pretty much
like %k
used to behave. %k was almost exclusively used for by jsregexp printing code
(probably to make output compatible with dot file format and avoid 0 bytes
in
char* strings).
I think %
Hmm. Can't land this. There is a RegExp test that actually expects this
kind of
escaping.
I don't understand how this works without escaping \ though. I will dig and
update this CL.
https://codereview.chromium.org/458533002/
--
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.goog
Thanks for the review, Sven!
https://codereview.chromium.org/458533002/diff/1/src/ostreams.cc
File src/ostreams.cc (right):
https://codereview.chromium.org/458533002/diff/1/src/ostreams.cc#newcode169
src/ostreams.cc:169: (c.value == '\r') ||
On 2014/08/25 13:07:20, Sven Panne wrote:
This inden
Committed patchset #2 manually as 23366 (presubmit successful).
https://codereview.chromium.org/488993004/
--
--
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 un
Committed patchset #2 manually as 23366 (presubmit successful).
https://codereview.chromium.org/488993004/
--
--
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 un
+svenpanne, +verwaest as reviewers
https://codereview.chromium.org/488993004/
--
--
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
Reviewers: Yang,
https://codereview.chromium.org/488993004/diff/1/src/objects.h
File src/objects.h (right):
https://codereview.chromium.org/488993004/diff/1/src/objects.h#newcode5962
src/objects.h:5962: enum { kGroupCount =
kAllocationSiteTransitionChangedGroup + 1 };
On 2014/08/25 12:12:43, Ya
How about this one instead
https://codereview.chromium.org/488993004
?
https://codereview.chromium.org/467183002/
--
--
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" grou
https://codereview.chromium.org/467183002/diff/20001/src/objects.cc
File src/objects.cc (right):
https://codereview.chromium.org/467183002/diff/20001/src/objects.cc#newcode11878
src/objects.cc:11878: default:
On 2014/08/19 11:08:04, Toon Verwaest wrote:
Any reason why you added a default case?
Committed patchset #2 manually as 23191 (presubmit successful).
https://codereview.chromium.org/467183002/
--
--
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 un
Thanks!
I will sit on this until Chrome is branched, as the tree is throttled right
now.
https://codereview.chromium.org/467183002/
--
--
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
Reviewers: Yang,
Message:
Please take a look.
This is a usability improvement for --trace-deopt.
Description:
When marking dependent code for deoptimization print the group that is being
deoptimized.
Otherwise it is impossible to figure out from the --trace-deoptimization
output
what is goi
There might be some reason for escaping those non-printable characters,
however
this makes impossible reconstruction of source from the output dumps as
escaping
is not reversible unless it will also escape '\'. Which is another possible
solution.
https://codereview.chromium.org/458533002/
--
Reviewers: Sven Panne,
Message:
Sven, could please take a look?
Description:
Fix disassembly redirection from stdout into a file.
Pass \n, \r and \t through OStream without escaping.
BUG=
Please review this at https://codereview.chromium.org/458533002/
SVN Base: https://v8.googlecode.com/svn
DBC on the benchmarking side
One important difference is that %StringBuilderConcat builds Sequential
string
while optimization builds cons string.
Hence benchmarks that don't account for this difference are not completely
representative: you would pay the cost of flattening at use sites, inst
DBC:
I have seen IC based heuristics misfire for code that is high on direct
function
calls but low on ICs.
e.g. see http://jsperf.com/standalone-vs-method/2 where OSR never hits the
loop
because we use amount of initialized ICs to drive optimization decisions.
https://codereview.chromium
Committed patchset #1 manually as r21376 (presubmit successful).
https://codereview.chromium.org/291123002/
--
--
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 u
Reviewers: Yang,
Message:
Please take a look.
Description:
Ensure that interruptor callback registered through API is called outside of
ExecutionAccess lock.
Such a coarse locking can cause a dead-lock when another thread is
attempting to
clear an interrupt while we are waiting in the interr
37 matches
Mail list logo
