Bug 112665 – Implement and32 on MIPS platform

• Status: RESOLVED
• Resolution: FIXED
• Component: JavaScriptCore

[Csaba Osztrogonác]

+++ This bug was initially created as a clone of Bug #112663 +++

http://trac.webkit.org/changeset/146164 broke the MIPS
build because of missing add32 implementation:

/data/buildbot/mips-1/qt-linux-mipsel-mips32r2-release/build/Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp: In member function 'void JSC::DFG::SpeculativeJIT::compileAdd(JSC::DFG::Node*)':
/data/buildbot/mips-1/qt-linux-mipsel-mips32r2-release/build/Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp:3136: error: no matching function for call to 'JSC::DFG::JITCompiler::and32(JSC::AbstractMacroAssembler<JSC::MIPSAssembler>::Address, JSC::DFG::GPRReg&)'
/data/buildbot/mips-1/qt-linux-mipsel-mips32r2-release/build/Source/JavaScriptCore/assembler/MacroAssemblerMIPS.h:270: note: candidates are: void JSC::MacroAssemblerMIPS::and32(JSC::MIPSRegisters::RegisterID, JSC::MIPSRegisters::RegisterID)
/data/buildbot/mips-1/qt-linux-mipsel-mips32r2-release/build/Source/JavaScriptCore/assembler/MacroAssemblerMIPS.h:275: note:                 void JSC::MacroAssemblerMIPS::and32(JSC::MIPSRegisters::RegisterID, JSC::MIPSRegisters::RegisterID, JSC::MIPSRegisters::RegisterID)
/data/buildbot/mips-1/qt-linux-mipsel-mips32r2-release/build/Source/JavaScriptCore/assembler/MacroAssemblerMIPS.h:280: note:                 void JSC::MacroAssemblerMIPS::and32(JSC::AbstractMacroAssembler<JSC::MIPSAssembler>::TrustedImm32, JSC::MIPSRegisters::RegisterID)
/data/buildbot/mips-1/qt-linux-mipsel-mips32r2-release/build/Source/JavaScriptCore/assembler/MacroAssemblerMIPS.h:296: note:                 void JSC::MacroAssemblerMIPS::and32(JSC::AbstractMacroAssembler<JSC::MIPSAssembler>::TrustedImm32, JSC::MIPSRegisters::RegisterID, JSC::MIPSRegisters::RegisterID)
make[3]: *** [.obj/release-shared/dfg/DFGSpeculativeJIT.o] Error 1

[Peter Gal]

Please rename the bug. The 'and32' is missing not the 'add32'.

[Peter Gal]

Created attachment 193755 [details]
speculative fix

A MIPS guru should check this patch just to be on safe side.

[Gergely Kis]

Thanks for the patch, Balázs will check it first thing.

[Balazs Kilvady]

(In reply to comment #2)
> Created an attachment (id=193755) [details]
> speculative fix

The patch seems to be OK. But I must work on an other issue around operationToString (likely new bug report/patch) to be able to run all the tests.

[Peter Gal]

(In reply to comment #4)
> (In reply to comment #2)
> > Created an attachment (id=193755) [details] [details]
> > speculative fix
> 
> The patch seems to be OK. But I must work on an other issue around operationToString (likely new bug report/patch) to be able to run all the tests.

Ok, in the meantime I think we should commit this in, just to get the bot a little greener.

[Balazs Kilvady]

(In reply to comment #5)
> (In reply to comment #4)
> > (In reply to comment #2)
> > > Created an attachment (id=193755) [details] [details] [details]
> > > speculative fix
> > 
> > The patch seems to be OK. But I must work on an other issue around operationToString (likely new bug report/patch) to be able to run all the tests.
> 
> Ok, in the meantime I think we should commit this in, just to get the bot a little greener.

Agree.

[Zoltan Herczeg]

Comment on attachment 193755 [details]
speculative fix

róme

[WebKit Review Bot]

Comment on attachment 193755 [details]
speculative fix

Clearing flags on attachment: 193755

Committed r146339: <http://trac.webkit.org/changeset/146339>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.