Compilation fails on OS X 10.5 PPC

[interfect]

Building (in the v8ppc dependency) fails due to the absence of an "elf.h". It appears that this file is expected to exist on all non-AIX systems, but OS X 10.5 does not seem to have it by default.

Is there something I have forgotten to install?

Here is the tail end of my build:

  touch /Users/anovak/build/node/out/Release/obj.target/deps/v8ppc/tools/gyp/js2c.stamp
  g++ '-D_DARWIN_USE_64_BIT_INODE=1' '-DENABLE_DEBUGGER_SUPPORT' '-DENABLE_EXTRA_CHECKS' '-DV8_TARGET_ARCH_PPC' -I../deps/v8ppc/src  -O3 -gdwarf-2 -fstrict-aliasing -mmacosx-version-min=10.5 -arch i386 -Wall -Wendif-labels -W -Wno-unused-parameter -fno-rtti -fno-exceptions -fno-threadsafe-statics -fno-strict-aliasing -MMD -MF /Users/anovak/build/node/out/Release/.deps//Users/anovak/build/node/out/Release/obj.target/v8_base/deps/v8ppc/src/accessors.o.d.raw  -c -o /Users/anovak/build/node/out/Release/obj.target/v8_base/deps/v8ppc/src/accessors.o ../deps/v8ppc/src/accessors.cc
In file included from ../deps/v8ppc/src/macro-assembler.h:64,
                 from ../deps/v8ppc/src/deoptimizer.h:34,
                 from ../deps/v8ppc/src/accessors.cc:32:
../deps/v8ppc/src/ppc/assembler-ppc.h:48:17: error: elf.h: No such file or directory
../deps/v8ppc/src/objects.h:1052: warning: integer overflow in expression
make[1]: *** [/Users/anovak/build/node/out/Release/obj.target/v8_base/deps/v8ppc/src/accessors.o] Error 1
make: *** [node] Error 2
bash-3.2$ 

[andrewlow]

This is effectively a duplicate of #14

If you're able to read/modify C++ code - it should be easy enough to #ifdef out the code that is related to the elf.h module. This code shouldn't be strictly required to have node run.

[interfect]

I never did manage to make this work; I got it to compile, but it was
writing out elf binaries on OS X due to something wrong with the build
system, and I didn't get a chance to fix it before losing access to my test
machine.

If you wait long enough nobody will want to use OS X 10.5 anymore anyway,
so I suppose that's one way to solve the problem.

On Tue, Sep 9, 2014 at 11:44 AM, Andrew Low [email protected]
wrote:

Closed #36 #36.

—
Reply to this email directly or view it on GitHub
#36 (comment).

[andrewlow]

I closed this because it was basically a duplicate and there hadn't been any updates. Issue #14 will live on until we get around to fixing it.

I like the idea of having this code work on lots of hardware - but it's really a matter of priority and interest. In any case, there is an open issue to track OSX PPC - just not this one.