Add a build test to ensure that LD can properly align sections. --- Makefile | 4 ++-- tools/test-gcc.sh | 35 ++++++++++++++++++++++++++++++++++- 2 files changed, 36 insertions(+), 3 deletions(-) diff --git a/Makefile b/Makefile index 9affb39..d17f85a 100644 --- a/Makefile +++ b/Makefile @@ -70,9 +70,9 @@ vpath %.S src vgasrc ################ Build rules # Verify the gcc configuration and test if -fwhole-program works. -TESTGCC:=$(shell CC="$(CC)" tools/test-gcc.sh) +TESTGCC:=$(shell CC="$(CC)" LD="$(LD)" tools/test-gcc.sh) ifeq "$(TESTGCC)" "-1" -$(error "Please upgrade GCC") +$(error "Please upgrade GCC and/or binutils") endif ifndef COMPSTRAT diff --git a/tools/test-gcc.sh b/tools/test-gcc.sh index ce3497b..935f211 100755 --- a/tools/test-gcc.sh +++ b/tools/test-gcc.sh @@ -4,10 +4,43 @@ mkdir -p out TMPFILE1=out/tmp_testcompile1.c TMPFILE1o=out/tmp_testcompile1.o +TMPFILE1_ld=out/tmp_testcompile1.lds TMPFILE2=out/tmp_testcompile2.c TMPFILE2o=out/tmp_testcompile2.o TMPFILE3o=out/tmp_testcompile3.o +# Test if ld's alignment handling is correct. This is a known problem +# with the linker that ships with Ubuntu 11.04. +cat - > $TMPFILE1 <<EOF +const char v1[] __attribute__((section(".text.v1"))) = "0123456789"; +const char v2[] __attribute__((section(".text.v2"))) = "0123456789"; +EOF +cat - > $TMPFILE1_ld <<EOF +SECTIONS +{ + .mysection 0x88f0 : { +. = 0x10 ; +*(.text.v1) +. = 0x20 ; +*(.text.v2) +. = 0x30 ; + } +} +EOF +$CC -O -g -c $TMPFILE1 -o $TMPFILE1o > /dev/null 2>&1 +$LD -T $TMPFILE1_ld $TMPFILE1o -o $TMPFILE2o > /dev/null 2>&1 +if [ $? -ne 0 ]; then + echo "The version of LD on this system does not properly handle" > /dev/fd/2 + echo "alignments. As a result, this project can not be built." > /dev/fd/2 + echo "" > /dev/fd/2 + echo "The problem may be the result of this LD bug report:" > /dev/fd/2 + echo " http://sourceware.org/bugzilla/show_bug.cgi?id=12726" > /dev/fd/2 + echo "" > /dev/fd/2 + echo "Please update to a working version of binutils and retry." > /dev/fd/2 + echo -1 + exit 0 +fi + # Test for "-fwhole-program". Older versions of gcc (pre v4.1) don't # support the whole-program optimization - detect that. $CC -fwhole-program -S -o /dev/null -xc /dev/null > /dev/null 2>&1 @@ -85,4 +118,4 @@ fi # "ebp" register is clobberred in an "asm" statement. The code has # been modified to not clobber "ebp" - no test is available yet. -rm -f $TMPFILE1 $TMPFILE1o $TMPFILE2 $TMPFILE2o $TMPFILE3o +rm -f $TMPFILE1 $TMPFILE1o $TMPFILE1_ld $TMPFILE2 $TMPFILE2o $TMPFILE3o -- 1.7.4.4