Bug 141362 - Use a PairHash for DFGByteCodeParser's ConstantBufferKey
Summary: Use a PairHash for DFGByteCodeParser's ConstantBufferKey
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Benjamin Poulain
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-02-07 14:35 PST by Benjamin Poulain
Modified: 2015-02-07 15:49 PST (History)
0 users

See Also:

Attachments
Patch (2.62 KB, patch)
2015-02-07 14:37 PST, Benjamin Poulain 		darin: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Benjamin Poulain 2015-02-07 14:35:52 PST 
Use a PairHash for DFGByteCodeParser's ConstantBufferKey
Comment 1 Benjamin Poulain 2015-02-07 14:37:22 PST 
Created attachment 246218 [details]
Patch
Comment 2 Darin Adler 2015-02-07 15:49:14 PST 
Comment on attachment 246218 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=246218&action=review

> Source/JavaScriptCore/ChangeLog:10
> +        Use nullptr for pointers when possible.

Seems like we’d also want to use member initialization rather than initializing in each constructor. Makes it harder to accidentally leave m_codeBlock uninitialized.

> Source/JavaScriptCore/dfg/DFGByteCodeParser.cpp:84
> +        return WTF::pairIntHash(PtrHash<CodeBlock*>::hash(m_codeBlock), IntHash<unsigned>::hash(m_index));

Doesn’t pairIntHash already do enough hashing that we don’t need the IntHash<unsigned>::hash here? Also, that could just be WTF::intHash(m_index). I’m also not sure why HashFunctions.h doesn’t do using WTF::intHash and WTF::pairIntHash. No reason for us to be inconsistent about things like that.