All done.
Showcase counters:
[java]Compile of permutations succeeded
[java] - Perf Counters -
[java] DataflowOptimizer.exec:
15/24.70754483/1.647169655333/2.51847443/15/1.0
(count/total_time/avg_time/max_time/slow_count/slow_ratio)
Resending counters with rich format:
[java]Compile of permutations succeeded
[java] - Perf Counters -
[java] DataflowOptimizer.exec:
15/24.70754483/1.647169655333/2.51847443/15/1.0
(count/total_time/avg_time/max_time/slow_count/slow_
LGTM
http://gwt-code-reviews.appspot.com/295802/show
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
To unsubscribe, reply using "remove me" as the subject.
Reviewers: Lex,
Description:
Fixing gflow nested switch CFG generation
Added couple of tostrings to aid in debugging.
Please review this at http://gwt-code-reviews.appspot.com/180801
Affected files:
dev/core/src/com/google/gwt/dev/jjs/impl/gflow/AnalysisSolver.java
dev/core/src/com/google/g
Reviewers: Lex,
Message:
Lex,
This adds some debugging facilities and fixes recursion problems in copy
propagation analysis. Please take a look.
Please review this at http://gwt-code-reviews.appspot.com/154811
Affected files:
M dev/core/src/com/google/gwt/dev/jjs/impl/gflow/AnalysisSol
http://gwt-code-reviews.appspot.com/153807/diff/1002/82
File dev/core/src/com/google/gwt/dev/jjs/impl/gflow/cfg/CfgBuilder.java
(right):
http://gwt-code-reviews.appspot.com/153807/diff/1002/82#newcode736
Line 736: } else if (typeOracle.canTriviallyCast(catchType,
exceptionType)) {
According to t
Lex,
All done. All tests pass. PTAL.
http://gwt-code-reviews.appspot.com/153807/diff/1/9
File dev/core/src/com/google/gwt/dev/jjs/ast/JReferenceType.java
(right):
http://gwt-code-reviews.appspot.com/153807/diff/1/9#newcode28
Line 28: public static JType dereference(JType type) {
On 2010/02/26
Reviewers: Lex,
Message:
Lex,
these changes make gflow framework to pass all smoke tests. Please take
a look.
Please review this at http://gwt-code-reviews.appspot.com/153807
Affected files:
M dev/core/src/com/google/gwt/dev/jjs/ast/JIfStatement.java
M dev/core/src/com/google/gwt
Reviewers: Lex,
Message:
Lex, I'm adressing all comments from previous review here, since we
checked in gflow.
Feel free to comment on original changelists:
http://gwt-code-reviews.appspot.com/130812/show
http://gwt-code-reviews.appspot.com/130813/show
http://gwt-code-reviews.appspot.com/130814
All done in:
http://gwt-code-reviews.appspot.com/143812/show
http://gwt-code-reviews.appspot.com/130812/diff/1/2
File
dev/core/src/com/google/gwt/dev/jjs/impl/gflow/constants/AssumptionDeducer.java
(right):
http://gwt-code-reviews.appspot.com/130812/diff/1/2#newcode135
Line 135: // Only last e
Reviewers: Lex,
Message:
Lex,
Please review.
Description:
SameParameterValueOptimizer should ignore methods called from native
code.
Please review this at http://gwt-code-reviews.appspot.com/141812
Affected files:
M
dev/core/src/com/google/gwt/dev/jjs/impl/SameParameterValueOptimizer.java
Reviewers: Lex,
Message:
Requesting review.
Description:
Analysis framework update driven by compile time optimization.
Please review this at http://gwt-code-reviews.appspot.com/132813
Affected files:
M dev/core/src/com/google/gwt/dev/jjs/impl/gflow/Analysis.java
M dev/core/src/com/google/
Reviewers: Lex,
Message:
Requesting review.
Description:
Copy propagation analysis/optimization.
Detects when variable is the copy of another and uses older var instead.
Please review this at http://gwt-code-reviews.appspot.com/132814
Affected files:
A dev/core/src/com/google/gwt/dev/jjs/im
Reviewers: Lex,
Message:
Requesting review
Description:
Unreachable analysis/optimization.
Determines CFG nodes which are not reachable, and deletes them from
source
tree if possible.
Please review this at http://gwt-code-reviews.appspot.com/130814
Affected files:
A
dev/core/src/com/googl
Reviewers: Lex,
Message:
Requesting review.
Description:
Liveness analysis/optimization.
Detects when variable is not used after the assignment,
and eliminates assignment.
Please review this at http://gwt-code-reviews.appspot.com/130813
Affected files:
A
dev/core/src/com/google/gwt/dev/jj
Reviewers: Lex,
Message:
Requesting review.
Description:
Constant analysis.
Detects situation when variable value is constant and replaces variable
access with constant value.
Please review this at http://gwt-code-reviews.appspot.com/130812
Affected files:
A
dev/core/src/com/google/gwt/de
Merged into http://gwt-code-reviews.appspot.com/126803
http://gwt-code-reviews.appspot.com/117805
--
http://groups.google.com/group/Google-Web-Toolkit-Contributors
Lex,
Next patch will contains 3 optimizations. Promise! :)
http://gwt-code-reviews.appspot.com/117805/diff/2047/1046
File dev/core/src/com/google/gwt/dev/util/Either.java (right):
http://gwt-code-reviews.appspot.com/117805/diff/2047/1046#newcode76
Line 76: return true;
On 2009/12/15 23:00:37, L
Reviewers: Lex,
Message:
Lex,
This is the second installment. Take a look.
Description:
Control flow graph abstraction & builder for CFG optimizations.
Doesn't include actual optimizations.
Please review this at http://gwt-code-reviews.appspot.com/117805
Affected files:
M dev/core/src/com/g
> Sounds good. Git is optional; please do if it's easy, but don't worry
> if it's not. I can patch in multiple patches just fine.
I use git for development anyway.
On Tue, Dec 1, 2009 at 3:32 PM, Lex Spoon wrote:
> On Tue, Dec 1, 2009 at 5:59 PM, Mike Aizatsky wrot
> Well, it would be sitting in svn and not being tested in any way
> except that it compiles. If you want to commit the pieces somewhere,
> why don't we make an svn branch?
I don't think svn branch will help much. It will only add headaches.
What do you say if I would merge all LGTM'ed changes to
Forwarding since I wasn't member of the group.
-- Forwarded message --
From:
Date: Tue, Dec 1, 2009 at 1:52 PM
Subject: Re: Flow analysis framework definition and solver.
To: sp...@google.com,
google-web-toolkit-contributors@googlegroups.com,
cromwell...@google.com
Reviewers:
ctually rather skip optimizations during initial construction in order to
surgically call specific optimizers. Thoughts?
>
>
> On Tue, Dec 16, 2008 at 9:58 PM, Mike Aizatsky mike.aizat...@gmail.com>
wrote:
>
> Scott,
>
>
>
> I'm sorry I've disappeared, but I h
Scott,
I'm sorry I've disappeared, but I had really busy couple of months. I
have resurrected the change (and all forthcoming).
Please take another look at the test mini-framework:
http://codereview.appspot.com/11060
On Tue, Oct 14, 2008 at 5:23 AM, Scott Blum wrote:
> On Mon, Oct 6, 2008 at
BTW, something was changed recently to use context class loader. This
is corrected patch.
On Mon, Oct 6, 2008 at 10:36 AM, Mike Aizatsky <[EMAIL PROTECTED]> wrote:
> Sorry, forgot to attach new patch. Here it is.
>
> On Mon, Oct 6, 2008 at 10:32 AM, Mike Aizatsky <[EMAIL
> Gotcha. But if you don't compile multiple times, how do you prevent
> optimizations that occur in one test method from impacting ones that occur
> in another test method?
I can't. I do this by limiting the scope optimization should work on
(see MethodInlineTest)...
> By the way, I've already
Scott,
>> > - Could briefly summarize the motivation for the
>> > suite()/JJSOptimizerTestDecator design?
>>
>> This way the module will be compiled only once. Unfortunately
>> JavaToJavaScriptCompiler doesn't have any easily extract compiling
>> pass and I didn't want to seriously refactor it.
>
ck
> for workspace configuration there, and it makes the build file itself
> cleaner.
> General concept is fine, of course, but can we make those changes?
>
>
> On Fri, Oct 3, 2008 at 8:28 PM, Scott Blum <[EMAIL PROTECTED]> wrote:
>>
>> Freeland, it's all you.
Sorry, forgot to attach new patch. Here it is.
On Mon, Oct 6, 2008 at 10:32 AM, Mike Aizatsky <[EMAIL PROTECTED]> wrote:
> Scott,
>
>> - Could briefly summarize the motivation for the
>> suite()/JJSOptimizerTestDecator design?
>
> This way the module will be com
Scott,
> - Could briefly summarize the motivation for the
> suite()/JJSOptimizerTestDecator design?
This way the module will be compiled only once. Unfortunately
JavaToJavaScriptCompiler doesn't have any easily extract compiling
pass and I didn't want to seriously refactor it.
> - JJSOptimizerT
Here's svn diff patch.
On Wed, Oct 1, 2008 at 2:32 PM, Mike Aizatsky <[EMAIL PROTECTED]> wrote:
> Hi!
>
> I created a small test framework for jjs optimizers. I've also
> attached a small test for MethodInliner to show how it looks like.
>
> Patch stats:
&g
Nope. I just didn't see it in gwtc. Thought it was lost somehow.
On Fri, Oct 3, 2008 at 5:28 PM, Scott Blum <[EMAIL PROTECTED]> wrote:
> Wait, I got the one you sent yesterday; is this one different?
>
> On Fri, Oct 3, 2008 at 8:25 PM, Mike Aizatsky <[EMAIL PROTECTED]>
Hi!
I've fixed build file to support working in non-svn workdir (e.g. git
:). Can you please get this in?
--
Regards,
Mike
--~--~-~--~~~---~--~~
http://groups.google.com/group/Google-Web-Toolkit-Contributors
-~--~~~~--~~--~--~---
Scott,
This indeed fixes the problem. I'll send a patch with test soon.
On Wed, Oct 1, 2008 at 1:36 PM, Scott Blum <[EMAIL PROTECTED]> wrote:
> I suspect the block at 596 should be run only when x.canBePolyMorphic().
>
> On Wed, Oct 1, 2008 at 4:12 PM, Mike Aizatsky <[EM
Hi!
I created a small test framework for jjs optimizers. I've also
attached a small test for MethodInliner to show how it looks like.
Patch stats:
.../com/google/gwt/dev/jjs/impl/MethodInliner.java | 11 +-
.../gwt/dev/jjs/impl/JJSOptimizerTestCase.java | 195
.../gw
Hi!
I'm currently working on a new optimization for GWT. And
apparently it triggered a bug in TypeTightener. I'm not sure why you
can't see it without my additional pass (and, actually, reordered some
others), but here's what is going on:
- There's a FlexTable.insertRow method, which just calls
36 matches
Mail list logo
