601360 - Workaround as much as possible to be able to build with Clang on Mac OS X

Status: RESOLVED FIXED

(Firefox Build System :: General, defect)

Description

[Nomis101]

This Bug is for keeping a patch to workaround as much as possible issues in Mozilla code to be able to build with Clang, until this issues are fixed.
The initial patch is from: http://llvm.org/bugs/show_bug.cgi?id=5511

Comment 1

[Nomis101]

Attachment: Workaround patch v1

This patch works with the current trunk from 2. October. I've got this patch from John Regehr and added a hack to workaround an issue with jsgcinlines.h
This patch also includes the fixes for Bug 579788, Bug 579786 and Bug 583460.

Comment 2

[Kyle Huey (Exited; not receiving bugmail, old account, do not use)]

How is this bug different from bug 574346?

Comment 3

[Nomis101]

(In reply to comment #2)
> How is this bug different from bug 574346?

Bug 574346 is a tracking Bug. This Bug is for hosting a patch that is partly a hack and not a fix, so the issues in /widgets/cocoa can be fixed. See https://bugzilla.mozilla.org/show_bug.cgi?id=583460#c7

Comment 4

[Nomis101]

Attachment: Workaround patch v2

+ 600901 and 600900.

Comment 5

[[e-mail]]

Comment on attachment 480796 [details] [diff] [review]
Workaround patch v2

>   const nsDefaultStringComparator defaultComparator;
>-  const nsASCIICaseInsensitiveStringComparator ciComparator;
>+  const nsASCIICaseInsensitiveStringComparator ciComparator = nsASCIICaseInsensitiveStringComparator();
Out of interest, why doesn't the default comparator need an explicit initialiser?

Comment 6

[Nomis101]

(In reply to comment #5)
> Comment on attachment 480796 [details] [diff] [review]
> Workaround patch v2
> 
> >   const nsDefaultStringComparator defaultComparator;
> >-  const nsASCIICaseInsensitiveStringComparator ciComparator;
> >+  const nsASCIICaseInsensitiveStringComparator ciComparator = nsASCIICaseInsensitiveStringComparator();
> Out of interest, why doesn't the default comparator need an explicit
> initialiser?

I have only copied this from llvm Bug 5511. If you are interested in details for this patch, its better to ask Eli Friedman who has written this lines. I only can say, that Clang doesn't produce an error for this.

Comment 7

[Rafael Ávila de Espíndola (:espindola) (not reading bugmail)]

Can we close this bug? I think 574346 is tracking all that we need and with newer clangs we shouldn't need any workaround.

Comment 8

[Nomis101]

Yes, you are right. We can close this, because its not needed anymore.