summaryrefslogtreecommitdiffstats
path: root/meta/recipes-devtools/gcc/gcc-8.3/0019-export-CPP.patch
diff options
context:
space:
mode:
Diffstat (limited to 'meta/recipes-devtools/gcc/gcc-8.3/0019-export-CPP.patch')
-rw-r--r--meta/recipes-devtools/gcc/gcc-8.3/0019-export-CPP.patch53
1 files changed, 0 insertions, 53 deletions
diff --git a/meta/recipes-devtools/gcc/gcc-8.3/0019-export-CPP.patch b/meta/recipes-devtools/gcc/gcc-8.3/0019-export-CPP.patch
deleted file mode 100644
index 140c67a486..0000000000
--- a/meta/recipes-devtools/gcc/gcc-8.3/0019-export-CPP.patch
+++ /dev/null
@@ -1,53 +0,0 @@
1From a2936382da81aefa9b69c1fc625f6c706b7ea1d8 Mon Sep 17 00:00:00 2001
2From: Khem Raj <raj.khem@gmail.com>
3Date: Fri, 20 Feb 2015 09:40:59 +0000
4Subject: [PATCH 19/40] export CPP
5
6The OE environment sets and exports CPP as being the target gcc. When
7building gcc-cross-canadian for a mingw targetted sdk, the following can be found
8in build.x86_64-pokysdk-mingw32.i586-poky-linux/build-x86_64-linux/libiberty/config.log:
9
10configure:3641: checking for _FILE_OFFSET_BITS value needed for large files
11configure:3666: gcc -c -isystem/media/build1/poky/build/tmp/sysroots/x86_64-linux/usr/include -O2 -pipe conftest.c >&5
12configure:3666: $? = 0
13configure:3698: result: no
14configure:3786: checking how to run the C preprocessor
15configure:3856: result: x86_64-pokysdk-mingw32-gcc -E --sysroot=/media/build1/poky/build/tmp/sysroots/x86_64-nativesdk-mingw32-pokysdk-mingw32
16configure:3876: x86_64-pokysdk-mingw32-gcc -E --sysroot=/media/build1/poky/build/tmp/sysroots/x86_64-nativesdk-mingw32-pokysdk-mingw32 conftest.c
17configure:3876: $? = 0
18
19Note this is a *build* target (in build-x86_64-linux) so it should be
20using the host "gcc", not x86_64-pokysdk-mingw32-gcc. Since the mingw32
21headers are very different, using the wrong cpp is a real problem. It is leaking
22into configure through the CPP variable. Ultimately this leads to build
23failures related to not being able to include a process.h file for pem-unix.c.
24
25The fix is to ensure we export a sane CPP value into the build
26environment when using build targets. We could define a CPP_FOR_BUILD value which may be
27the version which needs to be upstreamed but for now, this fix is good enough to
28avoid the problem.
29
30RP 22/08/2013
31
32Upstream-Status: Pending
33
34Signed-off-by: Khem Raj <raj.khem@gmail.com>
35---
36 Makefile.in | 1 +
37 1 file changed, 1 insertion(+)
38
39diff --git a/Makefile.in b/Makefile.in
40index 38774f542a6..b426d4f9350 100644
41--- a/Makefile.in
42+++ b/Makefile.in
43@@ -149,6 +149,7 @@ BUILD_EXPORTS = \
44 AR="$(AR_FOR_BUILD)"; export AR; \
45 AS="$(AS_FOR_BUILD)"; export AS; \
46 CC="$(CC_FOR_BUILD)"; export CC; \
47+ CPP="$(CC_FOR_BUILD) -E"; export CPP; \
48 CFLAGS="$(CFLAGS_FOR_BUILD)"; export CFLAGS; \
49 CONFIG_SHELL="$(SHELL)"; export CONFIG_SHELL; \
50 CXX="$(CXX_FOR_BUILD)"; export CXX; \
51--
522.21.0
53