Bug 113437

Summary:

fourthTier: JITCode should abstract exactly how the JIT code is structured and where it was allocated

Product:

WebKit

Reporter:

Filip Pizlo <fpizlo>

Component:

JavaScriptCore

Assignee:

Filip Pizlo <fpizlo>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

barraclough, ggaren, mark.lam, mhahnenberg, msaboff, oliver, sam, webkit-ews, webkit.review.bot

Priority:

Version:

528+ (Nightly build)

Hardware:

All

OS:

All

Bug Depends on:

Bug Blocks:

112840

Attachments:

Description	Flags
the patch	webkit-ews: commit-queue-
the patch	mhahnenberg: review+

Filip Pizlo

Reported 2013-03-27 13:24:51 PDT

This will make it easier to bring up the LLVM backend. At first, I'll probably just let LLVM use its own memory allocator. But even if we do switch to having LLVM use our allocator, we will probably want to give LLVM the ability to allocate the code for a function in multiple chunks, if it wanted to. But also, making JITCode be more abstract means that in the future we could even have a DFGJITCode, for example, which contains all of the DFGRareData things in CodeBlock. We could do likewise for the LLInt and Baseline JIT. This will save memory. Anyway, I think this is pure goodness.

Attachments
the patch (55.71 KB, patch) 2013-03-27 14:07 PDT, Filip Pizlo	webkit-ews: commit-queue-	Details Formatted Diff Diff
the patch (55.76 KB, patch) 2013-03-27 14:15 PDT, Filip Pizlo	mhahnenberg: review+	Details Formatted Diff Diff
Show Obsolete (1) View All Add attachment proposed patch, testcase, etc.

Filip Pizlo

Comment 1 2013-03-27 14:07:45 PDT

Created attachment 195391 [details] the patch

WebKit Review Bot

Comment 2 2013-03-27 14:10:22 PDT

Attachment 195391 [details] did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/JavaScriptCore/ChangeLog', u'Source/JavaScriptCore/assembler/RepatchBuffer.h', u'Source/JavaScriptCore/bytecode/CodeBlock.cpp', u'Source/JavaScriptCore/bytecode/CodeBlock.h', u'Source/JavaScriptCore/dfg/DFGDriver.cpp', u'Source/JavaScriptCore/dfg/DFGDriver.h', u'Source/JavaScriptCore/dfg/DFGJITCompiler.cpp', u'Source/JavaScriptCore/dfg/DFGJITCompiler.h', u'Source/JavaScriptCore/dfg/DFGOSREntry.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExit.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExitCompiler32_64.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExitCompiler64.cpp', u'Source/JavaScriptCore/dfg/DFGOperations.cpp', u'Source/JavaScriptCore/interpreter/Interpreter.cpp', u'Source/JavaScriptCore/jit/JIT.cpp', u'Source/JavaScriptCore/jit/JIT.h', u'Source/JavaScriptCore/jit/JITCode.cpp', u'Source/JavaScriptCore/jit/JITCode.h', u'Source/JavaScriptCore/jit/JITDriver.h', u'Source/JavaScriptCore/jit/JITStubs.cpp', u'Source/JavaScriptCore/jit/ThunkGenerators.cpp', u'Source/JavaScriptCore/llint/LLIntEntrypoints.cpp', u'Source/JavaScriptCore/llint/LLIntEntrypoints.h', u'Source/JavaScriptCore/llint/LLIntSlowPaths.cpp', u'Source/JavaScriptCore/runtime/Executable.cpp', u'Source/JavaScriptCore/runtime/Executable.h', u'Source/JavaScriptCore/runtime/ExecutionHarness.h']" exit_code: 1 Source/JavaScriptCore/jit/JITCode.h:50: enum members should use InterCaps with an initial capital letter. [readability/enum_casing] [4] Source/JavaScriptCore/jit/JITCode.h:79: The parameter name "jitType" adds no information, so it should be removed. [readability/parameter_name] [5] Source/JavaScriptCore/jit/JITCode.h:119: The parameter name "ref" adds no information, so it should be removed. [readability/parameter_name] [5] Source/JavaScriptCore/jit/JITCode.h:119: The parameter name "jitType" adds no information, so it should be removed. [readability/parameter_name] [5] Total errors found: 4 in 27 files If any of these errors are false positives, please file a bug against check-webkit-style.

Filip Pizlo

Comment 3 2013-03-27 14:12:27 PDT

(In reply to comment #2) > Attachment 195391 [details] did not pass style-queue: > > Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/JavaScriptCore/ChangeLog', u'Source/JavaScriptCore/assembler/RepatchBuffer.h', u'Source/JavaScriptCore/bytecode/CodeBlock.cpp', u'Source/JavaScriptCore/bytecode/CodeBlock.h', u'Source/JavaScriptCore/dfg/DFGDriver.cpp', u'Source/JavaScriptCore/dfg/DFGDriver.h', u'Source/JavaScriptCore/dfg/DFGJITCompiler.cpp', u'Source/JavaScriptCore/dfg/DFGJITCompiler.h', u'Source/JavaScriptCore/dfg/DFGOSREntry.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExit.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExitCompiler32_64.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExitCompiler64.cpp', u'Source/JavaScriptCore/dfg/DFGOperations.cpp', u'Source/JavaScriptCore/interpreter/Interpreter.cpp', u'Source/JavaScriptCore/jit/JIT.cpp', u'Source/JavaScriptCore/jit/JIT.h', u'Source/JavaScriptCore/jit/JITCode.cpp', u'Source/JavaScriptCore/jit/JITCode.h', u'Source/JavaScriptCore/jit/JITDriver.h', u'Source/JavaScriptCore/jit/JITStubs.cpp', u'Source/JavaScriptCore/jit/ThunkGenerators.cpp', u'Source/JavaScriptCore/llint/LLIntEntrypoints.cpp', u'Source/JavaScriptCore/llint/LLIntEntrypoints.h', u'Source/JavaScriptCore/llint/LLIntSlowPaths.cpp', u'Source/JavaScriptCore/runtime/Executable.cpp', u'Source/JavaScriptCore/runtime/Executable.h', u'Source/JavaScriptCore/runtime/ExecutionHarness.h']" exit_code: 1 > Source/JavaScriptCore/jit/JITCode.h:50: enum members should use InterCaps with an initial capital letter. [readability/enum_casing] [4] Nope. > Source/JavaScriptCore/jit/JITCode.h:79: The parameter name "jitType" adds no information, so it should be removed. [readability/parameter_name] [5] > Source/JavaScriptCore/jit/JITCode.h:119: The parameter name "ref" adds no information, so it should be removed. [readability/parameter_name] [5] > Source/JavaScriptCore/jit/JITCode.h:119: The parameter name "jitType" adds no information, so it should be removed. [readability/parameter_name] [5] I fixed them. > Total errors found: 4 in 27 files > > > If any of these errors are false positives, please file a bug against check-webkit-style.

Early Warning System Bot

Comment 4 2013-03-27 14:14:13 PDT

Comment on attachment 195391 [details] the patch Attachment 195391 [details] did not pass qt-wk2-ews (qt): Output: http://webkit-commit-queue.appspot.com/results/17304362

Filip Pizlo

Comment 5 2013-03-27 14:15:48 PDT

Created attachment 195393 [details] the patch Fixing unresolved symbols.

WebKit Review Bot

Comment 6 2013-03-27 14:18:37 PDT

Attachment 195393 [details] did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/JavaScriptCore/ChangeLog', u'Source/JavaScriptCore/assembler/RepatchBuffer.h', u'Source/JavaScriptCore/bytecode/CodeBlock.cpp', u'Source/JavaScriptCore/bytecode/CodeBlock.h', u'Source/JavaScriptCore/dfg/DFGDriver.cpp', u'Source/JavaScriptCore/dfg/DFGDriver.h', u'Source/JavaScriptCore/dfg/DFGJITCompiler.cpp', u'Source/JavaScriptCore/dfg/DFGJITCompiler.h', u'Source/JavaScriptCore/dfg/DFGOSREntry.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExit.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExitCompiler32_64.cpp', u'Source/JavaScriptCore/dfg/DFGOSRExitCompiler64.cpp', u'Source/JavaScriptCore/dfg/DFGOperations.cpp', u'Source/JavaScriptCore/interpreter/Interpreter.cpp', u'Source/JavaScriptCore/jit/JIT.cpp', u'Source/JavaScriptCore/jit/JIT.h', u'Source/JavaScriptCore/jit/JITCode.cpp', u'Source/JavaScriptCore/jit/JITCode.h', u'Source/JavaScriptCore/jit/JITDriver.h', u'Source/JavaScriptCore/jit/JITStubs.cpp', u'Source/JavaScriptCore/jit/ThunkGenerators.cpp', u'Source/JavaScriptCore/llint/LLIntEntrypoints.cpp', u'Source/JavaScriptCore/llint/LLIntEntrypoints.h', u'Source/JavaScriptCore/llint/LLIntSlowPaths.cpp', u'Source/JavaScriptCore/runtime/Executable.cpp', u'Source/JavaScriptCore/runtime/Executable.h', u'Source/JavaScriptCore/runtime/ExecutionHarness.h']" exit_code: 1 Source/JavaScriptCore/jit/JITCode.h:50: enum members should use InterCaps with an initial capital letter. [readability/enum_casing] [4] Total errors found: 1 in 27 files If any of these errors are false positives, please file a bug against check-webkit-style.

Mark Hahnenberg

Comment 7 2013-03-27 14:22:35 PDT

Comment on attachment 195393 [details] the patch r=me

Geoffrey Garen

Comment 8 2013-03-27 15:40:14 PDT

> But also, making JITCode be more abstract means that in the future we could even have a DFGJITCode, for example, which contains all of the DFGRareData things in CodeBlock. We could do likewise for the LLInt and Baseline JIT. This will save memory. For a given piece of data, how would you describe the guideline for whether it belongs in JITCode or CodeBlock?

Filip Pizlo

Comment 9 2013-03-27 15:49:44 PDT

(In reply to comment #8) > > But also, making JITCode be more abstract means that in the future we could even have a DFGJITCode, for example, which contains all of the DFGRareData things in CodeBlock. We could do likewise for the LLInt and Baseline JIT. This will save memory. > > For a given piece of data, how would you describe the guideline for whether it belongs in JITCode or CodeBlock? Specific to a particular execution engine? -> put it in JITCode not CodeBlock.

Filip Pizlo

Comment 10 2013-03-27 15:56:55 PDT

Landed in http://trac.webkit.org/changeset/147014

Note You need to log in before you can comment on or make changes to this bug.