On 04/19/2018 03:58 AM, Alex Bennée wrote: > -cc_i386=i386-pc-linux-gnu-gcc > libs_qga="" > debug_info="yes" > stack_protector="" > @@ -457,6 +456,8 @@ docker="no" > cross_cc_aarch64="aarch64-linux-gnu-gcc" > cross_cc_arm="arm-linux-gnueabihf-gcc" > cross_cc_powerpc="powerpc-linux-gnu-gcc" > +cross_cc_i386="i386-pc-linux-gnu-gcc" > +cross_cc_cflags_i386="" ... > > enabled_cross_compilers="" > > @@ -687,12 +688,10 @@ case "$cpu" in > i386|i486|i586|i686|i86pc|BePC) > cpu="i386" > supported_cpu="yes" > - cross_cc_i386=gcc > ;; > x86_64|amd64) > cpu="x86_64" > supported_cpu="yes" > - cross_cc_x86_64=gcc > ;; > armv*b|armv*l|arm) > cpu="arm" > @@ -1435,7 +1434,6 @@ case "$cpu" in > i386) > CPU_CFLAGS="-m32" > LDFLAGS="-m32 $LDFLAGS" > - cc_i386='$(CC) -m32'
Why is the i386 case not handled like... > ;; > x86_64) > # ??? Only extremely old AMD cpus do not have cmpxchg16b. > @@ -1443,12 +1441,14 @@ case "$cpu" in > # runtime and generate the fallback to serial emulation. > CPU_CFLAGS="-m64 -mcx16" > LDFLAGS="-m64 $LDFLAGS" > - cc_i386='$(CC) -m32' > + cross_cc_x86_64=$cc > + cross_cc_cflags_x86_64=$CPU_CFLAGS ... the x86_64 case? Also, does cross_cc_cflags_foo really have value over including the flags directly in the cross_cc_foo variable? r~