[ANN] OPAL Digest

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[ANN] OPAL Digest

Stéphane Ducasse
hi guys

Since Opalers are working hard in the shadow, I thought that I can do a little digest for them.
Each time I will delete ten entries I will do a small digest.
Thanks marcus, Jorge and JB for their time and energy.

Stef

PS: OPAL is the new compiler framework we will have in 1.3.
Really cool. OPAL comes from the Opal coast near to lille with cool dunes and birds.

==================== Summary ====================

Name: OpalCompiler-Core-MarcusDenker.33
Author: MarcusDenker
Time: 9 September 2010, 8:12:20 pm
UUID: beb4fce5-38ef-4ff1-ba41-924245ff6456
Ancestors: OpalCompiler-Core-MarcusDenker.32

- cleanups (unsent methods)
- make sure to hand over array of copiedTemos to IRBuilder,
not Set.

final one for today... things start to work. Failing tests down to 24 (from 40). TestCoverage 81%

==================== Summary ====================

Name: OpalCompiler-Tests-MarcusDenker.29
Author: MarcusDenker
Time: 9 September 2010, 7:47:08 pm
UUID: aa19bd88-0895-4ab8-b55c-3f62f1698e33
Ancestors: OpalCompiler-Tests-MarcusDenker.28

- ASTChecker tests does not use verifyin, but just the ASTChecker
- misc changes

==================== Summary ====================

Name: OpalCompiler-Core-MarcusDenker.32
Author: MarcusDenker
Time: 9 September 2010, 7:46:21 pm
UUID: 845092d2-e58f-441a-9304-43106311e5a1
Ancestors: OpalCompiler-Core-MarcusDenker.31

-> ASTTraslator now requests a tempVector from IRBuilder if needed
-> simplified ASTClosureAnalyzer --> moved code to block/method scope
-> tempVector now is not a temp anymore, just a copying one if needed
-> verifyIn: now calls both visitors.

==================== Summary ====================

Name: OpalCompiler-Core-MarcusDenker.31
Author: MarcusDenker
Time: 9 September 2010, 5:49:20 pm
UUID: ab2dc77f-71b4-4f82-9cb4-4a2c1261e3fe
Ancestors: OpalCompiler-Core-MarcusDenker.30

- simplify a bit re-lookup of variables in OCASTClosureAnalyzer
- scopes now have a unique name after ASTChecker pass
 (this is then used as the name of the vector)


==================== Summary ====================

Name: OpalCompiler-Tests-MarcusDenker.28
Author: MarcusDenker
Time: 9 September 2010, 5:49:34 pm
UUID: 4821aa22-5ca5-4973-9c18-df0b1bb0297f
Ancestors: OpalCompiler-Tests-MarcusDenker.27

more tests...

==================== Summary ====================

Name: OpalCompiler-Tests-MarcusDenker.27
Author: MarcusDenker
Time: 9 September 2010, 5:09:43 pm
UUID: 49dfe8ab-d8af-4b61-9bcc-066fcf7e5aa0
Ancestors: OpalCompiler-Tests-JorgeRessia.26

fix tests...

==================== Summary ====================

Name: OpalCompiler-Core-MarcusDenker.30
Author: MarcusDenker
Time: 9 September 2010, 5:09:27 pm
UUID: 2485ef0c-25fe-4863-bd49-fad1aaa0f8bc
Ancestors: OpalCompiler-Core-jb.29, OpalCompiler-Core-JorgeRessia.29

more work for scope analysis...
-> after variables are re-looked up, add copying temps to outer scop
-> Fixes

==================== Summary ====================

Name: OpalCompiler-Tests-JorgeRessia.26
Author: JorgeRessia
Time: 7 September 2010, 9:29:08 pm
UUID: 4def3fb1-2057-45fe-b7f7-62f057c999fb
Ancestors: OpalCompiler-Tests-MarcusDenker.25

- Introduction of optimized block scopes.
- the closure analysis is been moved out of the astchecker

==================== Summary ====================

Name: OpalCompiler-Core-JorgeRessia.29
Author: JorgeRessia
Time: 7 September 2010, 9:28:46 pm
UUID: 02e0a230-43fb-4c08-97f2-a767968faaf8
Ancestors: OpalCompiler-Core-MarcusDenker.28

- Introduction of optimized block scopes.
- the closure analysis is been moved out of the astchecker

==================== Summary ====================

Name: OpalCompiler-Core-jb.29
Author: jb
Time: 7 September 2010, 11:31:01 am
UUID: 06331cc1-732f-4e49-8327-1d6eef92bf26
Ancestors: OpalCompiler-Core-MarcusDenker.28

Fix some bug about Jump in decompiler and clean a bit the code.



_______________________________________________
Pharo-project mailing list
[hidden email]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project