Reviewers: Mads Ager, William Hesse,
Description:
Daylight savings intervals caching. There're two caches: small one for 6
intervals, and bigger one which is allocated only
when needed.
This is the new version of http://codereview.chromium.org/1758013/show
Sorry for not making changes there. I c
Of the other benchmarks, closure-datetime spends about half of its
time in DST calculations. And judging from the results of JSC, they do
cache DST intensively.
I'll upload a new version of my change later today. It will cache less
aggressively, then it does now.
On Mon, May 17, 2010 at 11:29,
Sorry for not answering this right away. I was on vacations.
I investigated some date tests and found that in some of them
(http://compute1.aar:9013/golem/date-parse.html) more then 80% of time
is taken by determining the DST offset.
I'm working right now to eliminate system call to check DST off
Reviewers: Mads Ager,
Description:
Rewrite the functions determining DST in C++ and also cache all the DST
intervals.
Please review this at http://codereview.chromium.org/1758013/show
Affected files:
M src/date.js
M src/platform-posix.cc
M src/runtime.h
M src/runtime.cc
--
v8-dev mail
I've changed the return value to newly allocated HeapNumber.
http://codereview.chromium.org/1694004/show
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.com/group/v8-dev
The current code used NewNumberFromDouble only in a couple of places and
NumberFromDouble in 20 or so places.
http://codereview.chromium.org/1694004/diff/10001/11001
File src/heap.cc (right):
http://codereview.chromium.org/1694004/diff/10001/11001#newcode1832
src/heap.cc:1832: if (rep.bits == p
Reviewers: Mads Ager,
Description:
Simplify double to number convertion.
Please review this at http://codereview.chromium.org/1694004/show
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/heap.h
M src/heap.cc
M src/runtime.cc
--
v8-dev m
LGTM
http://codereview.chromium.org/1563002
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.com/group/v8-dev
To unsubscribe from this group, send email to v8-dev+unsubscribegooglegroups.com or reply
to this email with the words "REMOVE ME" as the subject.
Reviewers: Mads Ager,
Description:
Fix bug http://code.google.com/p/v8/issues/detail?id=659. Move the limits
check
for date before the time zone offset is applied.
Please review this at http://codereview.chromium.org/1075016
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Af
Reviewers: Mads Ager,
Description:
Modify FastD2I to use static_cast instead of lrint(). Benchmarks show that
it's
much faster this way.
Please review this at http://codereview.chromium.org/1128010
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/c
Definitely LGTM then.
On Fri, Mar 19, 2010 at 16:01, Mads Sig Ager wrote:
> This one was not purely about readability. GCC gave warnings on this
> one because there were combinations of && and || with no parenthesis
> to clearify precedence.
>
> -- Mads
>
> On Fri,
To me extra parentheses make the code less readable, but obviously you
know the project style guidelines better than myself.
On Fri, Mar 19, 2010 at 10:52, wrote:
> LMRTM (Looks More Readable To Me)
>
> http://codereview.chromium.org/1132002
>
--
v8-dev mailing list
v8-dev@googlegroups.com
htt
On 2010/03/18 08:21:22, Erik Corry wrote:
I think Rico's objection could be solved with a rename eg to
Runtime_RoundNumber() so that it is clear that the input must be a number.
Done.
http://codereview.chromium.org/1008004
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.co
http://codereview.chromium.org/1008004/diff/5001/6002
File src/runtime.cc (right):
http://codereview.chromium.org/1008004/diff/5001/6002#newcode5330
src/runtime.cc:5330: // Must be SMI. Return the argument unchanged for
all the other types
On 2010/03/17 23:29:19, fschneider wrote:
SMI -> smi
Reviewers: Mads Ager, fschneider, Vladislav Kaznacheev,
Description:
Improve Math.round(). Fix the bug in r4146. Further improve performance by
checking the exponent instead of comparing doubles. Add several tests for
numbers near the limits of SMI and several tests from WebKit.
Please review t
On Tue, Mar 16, 2010 at 16:58, Oleg Eterevsky wrote:
>>> > Drive-by comment: Is it worth doing some of the work in math.js like we
>>> > do for floor?
>>> You mean, check for SMI in math.js rather then in runtime.cc? What
>>> difference does it make?
>> > Drive-by comment: Is it worth doing some of the work in math.js like we
>> > do for floor?
>>
>> You mean, check for SMI in math.js rather then in runtime.cc? What
>> difference does it make? We'll still have to call C++ function, so
>> there'll be an overhead of CEntryStub.
>
> No, you can c
Ok, I'll do it like that. I was not sure about reference counting and
such things, so I decided to play safe. Sorry, I still do not
understand V8 internals well enough.
On Mon, Mar 15, 2010 at 23:11, wrote:
>
> http://codereview.chromium.org/981002/diff/1/2
> File src/runtime.cc (right):
>
> htt
You mean, check for SMI in math.js rather then in runtime.cc? What
difference does it make? We'll still have to call C++ function, so
there'll be an overhead of CEntryStub.
On Mon, Mar 15, 2010 at 23:07, wrote:
> Drive-by comment: Is it worth doing some of the work in math.js like we do
> for
>
Reviewers: Mads Ager, Vladislav Kaznacheev,
Description:
Add a special case in Math.round for a SMI result. Also change the
implementation for non-SMI case.
http://compute3.aar:9013/golem/r4138-v8-olege-round-vs-4138-v8.html
Expect gain on http://compute3.aar:9013/golem/pinto-random.html
Pleas
http://codereview.chromium.org/968001
--
v8-dev mailing list
v8-dev@googlegroups.com
http://groups.google.com/group/v8-dev
Reviewers: Mads Ager,
Please review this at http://codereview.chromium.org/968001
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/date-delay.js
Index: src/date-delay.js
===
--- src/dat
http://codereview.chromium.org/811006/diff/8001/9003
File src/runtime.cc (right):
http://codereview.chromium.org/811006/diff/8001/9003#newcode5150
src/runtime.cc:5150: class YearMonthDate {
On 2010/03/11 13:48:53, Mads Ager wrote:
Instead of having a class that does all the work in it's constru
Reviewers: Mads Ager, Vladislav Kaznacheev,
Message:
I'm not sure about the style details: whether it's ok to create a class in
runtime.cc to return the triplet (year, month, day), is it ok to insert the
tables of dates directly into C++ file, or should I put it in some separate
place...
Descrip
http://codereview.chromium.org/661466/diff/5/6
File src/runtime.cc (right):
http://codereview.chromium.org/661466/diff/5/6#newcode4981
src/runtime.cc:4981: static const int base_day = 365*(1970 + year_delta)
+ (1970 + year_delta)/4 -
On 2010/03/03 14:09:37, Mads Ager wrote:
Please add space aro
Reviewers: Mads Ager, antonm,
Description:
Fix broken test in WebKit test suite and add the test in question to V8
tests.
Please review this at http://codereview.chromium.org/661466
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/runtime.cc
M
Reviewers: Vitaly, Mads Ager,
Description:
Prevent "control reaches end of non-void function" warning from gcc on Mac.
Please review this at http://codereview.chromium.org/660379
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/runtime.cc
Index: src
Reviewers: William Hesse,
Description:
Fix minor syntax error which broke the compilation.
tbr=whe...@chromium.org
Please review this at http://codereview.chromium.org/660377
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affected files:
M src/rewriter.cc
Index: src/r
http://codereview.chromium.org/660375/diff/7/1002
File src/runtime.cc (right):
http://codereview.chromium.org/660375/diff/7/1002#newcode4947
src/runtime.cc:4947: ASSERT(args[0]->IsSmi());
On 2010/03/02 14:07:53, antonm wrote:
I think we should either have ASSERTs and use Smi::cast or
CONVERT_S
Reviewers: Mads Ager, antonm,
Description:
Revert change to EquivalentTime in date.js. The checks that I've removed in
the
last changed were essential after all.
Please review this at http://codereview.chromium.org/660375
SVN Base: http://v8.googlecode.com/svn/branches/bleeding_edge/
Affec
http://codereview.chromium.org/661366/diff/1/3
File src/date-delay.js (right):
http://codereview.chromium.org/661366/diff/1/3#newcode368
src/date-delay.js:368: // Now we presume that year, month and date are
actually SMIs.
On 2010/03/02 13:04:08, Mads Ager wrote:
Since the runtime function prob
http://codereview.chromium.org/660084/diff/1/2
File src/runtime.cc (right):
http://codereview.chromium.org/660084/diff/1/2#newcode4157
src/runtime.cc:4157: // We do not include 0 so that we didn't have to
treat +0 / -0 cases.
On 2010/02/25 15:11:03, Mads Ager wrote:
didn't -> don't
Done.
htt
Reviewers: Mads Ager, Vladislav Kaznacheev, Vitaly,
Description:
Optimize three Number2Integer functions in runtime.cc: remove the check
that was
already done in JavaScript, add simpler code in case we are getting a SMI
result.
Please review this at http://codereview.chromium.org/660084
SVN
Reviewers: Mads Ager, Vitaly,
Description:
Change the implementation of lastIndexOf method of String. Convert the
strings
in question to Vector or Vector and execute simple search
algorithm
on them.
The difference in speed on 500k character string is about 10x.
Please review this at http:/
34 matches
Mail list logo
