Opened 8 years ago

Closed 8 years ago

#49801 closed defect (fixed)

__GMP_CC incorrectly has -isysroot ... in __GMP_CC

Reported by: jeremyhu (Jeremy Huddleston Sequoia) Owned by: MarcusCalhoun-Lopez (Marcus Calhoun-Lopez)
Priority: Normal Milestone:
Component: ports Version: 2.3.4
Keywords: Cc:
Port: gmp

Description (last modified by jeremyhu (Jeremy Huddleston Sequoia))

We should not have the -isysroot ... in __GMP_CC. This is fallout from #43262 setting it in CC.

We should solve #43262 another way.

Change History (6)

comment:1 Changed 8 years ago by jeremyhu (Jeremy Huddleston Sequoia)

Description: modified (diff)

comment:2 Changed 8 years ago by jeremyhu (Jeremy Huddleston Sequoia)

This comment in the Portfile suggests that the problem is with the build system itself: 

# Clear all options that affect CFLAGS and CXXFLAGS, since the configure
# script tries to build the fastest library for the build machine if
# CFLAGS and CXXFLAGS are undefined.
#
# Append the -stdlib flags to CXX, since we still want to select the C++ STL.
#
# On PowerPC machines, CFLAGS must be empty to get -force_cpusubtype_ALL.

We should fix that bug properly and upstream the changes, so other packagers benefit as well.

comment:3 Changed 8 years ago by MarcusCalhoun-Lopez (Marcus Calhoun-Lopez)

Status: new → assigned

r142898 has a short term fix.
I attempted to begin a discussion on the ​gmp-discuss mailing list, but my message did not seem to go through.

comment:4 Changed 8 years ago by MarcusCalhoun-Lopez (Marcus Calhoun-Lopez)

I posted a query on an upstream mailing list under the heading "​appending to cflags instead of replacing."

comment:5 Changed 8 years ago by MarcusCalhoun-Lopez (Marcus Calhoun-Lopez)

It has been over two weeks with no upstream response.
Is r142898 good enough to marked this as fixed?

comment:6 Changed 8 years ago by MarcusCalhoun-Lopez (Marcus Calhoun-Lopez)

Resolution: → fixed
Status: assigned → closed
Note: See TracTickets for help on using tickets.