svn commit: r439595 - in head/devel: aarch64-gcc aarch64-none-elf-gcc amd64-gcc arm-none-eabi-gcc arm-none-eabi-gcc492 mips-gcc mips64-gcc powerpc64-gcc riscv64-gcc sparc64-gcc

[Mark Millard]

On 2017-Apr-28, at 8:37 PM, Mark Linimon <linimon at lonesome.com> wrote:

> On Fri, Apr 28, 2017 at 07:41:33PM -0700, Mark Millard wrote:
>> I do have access to powerpc64 and tested it in my context.
>> It is powerpc that I'm far from an appropriate configuration
>> of for now.
> 
> OK, I misread that.

Your powerpc64 testing might be of gcc 4.2.1 and
old libstdc++ as a context. That would be different
than my powerpc64 tests.


As for armv7 building/installing
devel/arm-none-eabi-gcc492 . . .

I grep'd the armv7 logs and fancy_abort does occur
in mine twice as parts of a warning, not an error:

/usr/obj/portswork/usr/ports/devel/arm-none-eabi-gcc492/work/gcc-4.9.2/gcc/ipa-devirt.c:1557:7: warning: variable 'can_refer' is used uninitialized whenever '?
:' condition is false
      [-Wsometimes-uninitialized]
      gcc_assert (odr_violation_reported);
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/obj/portswork/usr/ports/devel/arm-none-eabi-gcc492/work/gcc-4.9.2/gcc/system.h:692:12: note: expanded from macro 'gcc_assert'
   ((void)(!(EXPR) ? fancy_abort (__FILE__, __LINE__, __FUNCTION__), 0 : 0))
           ^~~~~~~

/usr/obj/portswork/usr/ports/devel/arm-none-eabi-gcc492/work/gcc-4.9.2/gcc/ipa-devirt.c:1557:7: note: remove the '?:' if its condition is always true
      gcc_assert (odr_violation_reported);
      ^
/usr/obj/portswork/usr/ports/devel/arm-none-eabi-gcc492/work/gcc-4.9.2/gcc/system.h:692:12: note: expanded from macro 'gcc_assert'
   ((void)(!(EXPR) ? fancy_abort (__FILE__, __LINE__, __FUNCTION__), 0 : 0))
           ^

So both instances are tied to devel/arm-none-eabi-gcc492
but are not from build errors in my context.

In my context devel/arm-none-eabi-gcc492 had the
problem of conflicting with devel/arm-none-eabi-gcc
file(s) when I'd built and installed
devel/arm-none-eabi-gcc first. (In the other order
it would be devel/arm-none-eabi-gcc that showed the
conflict.)

Otherwise devel/arm-none-eabi-gcc and
devel/arm-none-eabi-gcc492 seemed to build fine
on armv7 and the other hosts.

===
Mark Millard
markmi at dsl-only.net