Monday, October 24, 2022

Re: aarch64 bulk build report

> http://build-failures.rhaalovely.net/aarch64/2022-10-18/x11/gnustep/base.log

././GSEasyHandle.h:48:40: error: blocks support disabled - compile with -fblocks or pick a deployment target that supports them
result: (void (^)(GSEasyHandleWriteBufferResult result, NSInteger length, NSData *data))result;
^
In file included from GSEasyHandle.m:3:
In file included from ././GSTimeoutSource.h:5:
././GSDispatch.h:128:9: warning: 'DISPATCH_QUEUE_SERIAL' macro redefined [-Wmacro-redefined]
#define DISPATCH_QUEUE_SERIAL 0

This happens if devel/libdispatch is present during build. I suggest we
neuter this, see diff below.


I also noticed that bfd.h usability in cofigure output depends on
devel/gdb being present or not. I haven't looked into this, but that
should probably also be neutered one way or the other.


I have no idea what causes the latest build failure starting with:

checking whether objc really works... no

http://build-failures.rhaalovely.net/aarch64/2022-10-22/x11/gnustep/base.log


Index: Makefile
===================================================================
RCS file: /cvs/ports/x11/gnustep/base/Makefile,v
retrieving revision 1.84
diff -u -p -r1.84 Makefile
--- Makefile 11 Mar 2022 20:15:52 -0000 1.84
+++ Makefile 24 Oct 2022 21:07:41 -0000
@@ -1,7 +1,7 @@
COMMENT= GNUstep base library

DISTNAME= gnustep-base-1.28.0
-REVISION= 3
+REVISION= 4

SHARED_LIBS= gnustep-base 10.5
CATEGORIES= devel
@@ -37,7 +37,8 @@ LIB_DEPENDS += converters/libiconv \

CONFIGURE_STYLE=gnu
CONFIGURE_ENV= ac_cv_header_execinfo_h=false
-CONFIGURE_ARGS += --disable-procfs \
+CONFIGURE_ARGS += --disable-libdispatch \
+ --disable-procfs \
--disable-procfs-psinfo
MAKE_FLAGS += libgnustep-base_INTERFACE_VERSION=${LIBgnustep-base_VERSION}

No comments:

Post a Comment