1050127 - Mismatched #define and #undef in js/src/jit/Recover.cpp

Status: RESOLVED FIXED

(Core :: JavaScript Engine: JIT, defect)

Description

[Aetf]

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:34.0) Gecko/20100101 Firefox/34.0 (Beta/Release)
Build ID: 20140723030202

Steps to reproduce:

open js/src/jit/Recover.cpp in editor.


Actual results:

in RInstruction::readRecoverData, a typo in the undef of MATCH_OPCODES_

::: js/src/jit/Recover.cpp
@@ 40,9 @@
> #   define MATCH_OPCODES_(op)                                           \
>       case Recover_##op:                                                \
>         static_assert(sizeof(R##op) <= sizeof(RInstructionStorage),     \
>                       "Storage space is too small to decode R" #op " instructions."); \
>         new (raw->addr()) R##op(reader);                                \
>         break;
> 
>         RECOVER_OPCODE_LIST(MATCH_OPCODES_)
> #   undef DEFINE_OPCODES_


Expected results:

It should be this

::: js/src/jit/Recover.cpp
@@ 40,9 @@
> #   define MATCH_OPCODES_(op)                                           \
>       case Recover_##op:                                                \
>         static_assert(sizeof(R##op) <= sizeof(RInstructionStorage),     \
>                       "Storage space is too small to decode R" #op " instructions."); \
>         new (raw->addr()) R##op(reader);                                \
>         break;
> 
>         RECOVER_OPCODE_LIST(MATCH_OPCODES_)
> #   undef MATCH_OPCODES_

Comment 1

[Aetf]

Attachment: Change DEFINE_OPCODES_ to MATCH_OPCODES_

A patch that fixes this

Comment 2

[Aetf]

Hi Nicolas. I thought maybe you have too many review request so I picked up the first one who is suggested by the page ;P

So how to land this change into the code base? Will this patch be sent to the try-server?

Comment 3

[Nicolas B. Pierron [:nbp] {away}]

(In reply to Aetf from comment #2)
> Hi Nicolas. I thought maybe you have too many review request so I picked up
> the first one who is suggested by the page ;P

Thanks for your consideration :)
Still, this is my typo …

> So how to land this change into the code base? Will this patch be sent to
> the try-server?

This patch is trivial and is unlikely to cause regressions.  So I will just land it with the other patches from Bug 1038592.

Comment 4

[Nicolas B. Pierron [:nbp] {away}]

https://hg.mozilla.org/integration/mozilla-inbound/rev/40e8e63d4f75

Comment 5

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/40e8e63d4f75