GCC bug 121567 is actually a duplicate of GCC bug 81426, which already
existed upstream, and for which we already had
BR2_TOOLCHAIN_HAS_GCC_BUG_81246.
In addition, BR2_TOOLCHAIN_HAS_GCC_BUG_81246 is correctly defined as
affecting SuperH regardless of the optimization level. Indeed, when
BR2_TOOLCHAIN_HAS_GCC_BUG_121567 was introduced, it was thought that
only -O2 and -O3 levels are affected, but -Og, -O1 and -Ofast are also
affected, causing the python3 build to fail.
Fixes:
https://autobuild.buildroot.net/results/
aca1812415ee84eef223b0fdff88f31a39775294/ (-O1 build)
https://autobuild.buildroot.net/results/
3fd91ce3ff8727d2f9c12f6721df922c60282d70/ (-Ofast build)
https://autobuild.buildroot.net/results/
e5536dcf598a51f5006343513a76f4e223e80a55/ (-Og build)
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
PYTHON3_CFLAGS = $(TARGET_CFLAGS)
-ifeq ($(BR2_TOOLCHAIN_HAS_GCC_BUG_121567),y)
+ifeq ($(BR2_TOOLCHAIN_HAS_GCC_BUG_81426),y)
PYTHON3_CFLAGS += -O0
endif
default y if BR2_microblaze
# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81426
+# error: unable to find a register to spill in class ‘R0_REGS’
+# The bug exists at least since gcc 6.x, so we are assuming it affects
+# all GCC versions.
config BR2_TOOLCHAIN_HAS_GCC_BUG_81426
bool
default y if BR2_sh
BR2_OPTIMIZE_2 || \
BR2_OPTIMIZE_3
-# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121567
-# error: unable to find a register to spill in class ‘R0_REGS’
-# The bug exists at least since gcc 6.x, so we are assuming it affects
-# all GCC versions.
-config BR2_TOOLCHAIN_HAS_GCC_BUG_121567
- bool
- default y if BR2_sh
- depends on BR2_OPTIMIZE_2 || \
- BR2_OPTIMIZE_3
-
config BR2_TOOLCHAIN_HAS_NATIVE_RPC
bool