gcc++ 5.1

Dan Horák dan at danny.cz
Wed Jun 3 20:11:14 UTC 2015


On Wed, 3 Jun 2015 19:54:46 +0000
Neale Ferguson <neale at sinenomine.net> wrote:

> I think there may be a problem with the compiler. Building v8z and

v8z like v8 ported to s390x? Is it an open source port?

> get:
> 
> ../src/hydrogen-check-elimination.cc: In member function ‘void
> v8::internal::HFlowEngine<State,
> Effects>::AnalyzeDominatedBlocks(v8::internal::HBasicBlock*, State*)
> Effects>[with
> State = v8::internal::HCheckTable; Effects =
> v8::internal::HCheckMapsEffects]’:
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
>        HCheckTableEntry* entry = &entries_[i];
>                                             ^
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
>        HCheckTableEntry* entry = &entries_[i];
>                                             ^
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
> ../src/hydrogen-check-elimination.cc:668:44: error: array subscript is
> above array bounds [-Werror=array-bounds]
> 
> 
> i goes from 0 to size and the array size is non-zero, so I do not
> think this is a valid error. It compiles cleanly under 4.4.7.

sounds similar to https://bugzilla.redhat.com/show_bug.cgi?id=663080
which should be fixed. Please report it as a bug in
https://bugzilla.redhat.com. We are discovering compiler bugs in every
new gcc release, but the toolchain is usually very responsive to our
findings.


		Dan


More information about the s390x mailing list