s390x: explicitly mark stack as not executable
PATCH 1/1 s390x: explicitly mark stack as not executable
With somewhat recent GCC versions, we get this warning on s390x:
/usr/bin/ld: warning: s390x/cpu.o: missing .note.GNU-stack section implies executable stack /usr/bin/ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linker
We don't really care whether stack is executable or not since we set it up ourselves and we're running DAT off mostly anyways.
Silence the warning by explicitly marking the stack as not executable.
--- s390x/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
s390x/Makefile +1 -1
@@ -79,7 +79,7 @@ CFLAGS += -O2 CFLAGS += -march=zEC12 CFLAGS += -mbackchain CFLAGS += -fno-delete-null-pointer-checks -LDFLAGS += -nostdlib -Wl,--build-id=none +LDFLAGS += -nostdlib -Wl,--build-id=none -z noexecstack # We want to keep intermediate files .PRECIOUS: %.o %.lds
Signature · 1 line
2.41.0
Quoted 11 lines · On Wed, 9 Aug 2023 11:17:08 +0200 Nico Boehr <[email protected]> wrote:
With somewhat recent GCC versions, we get this warning on s390x:
/usr/bin/ld: warning:s390x/cpu.o: missing .note.GNU-stack section implies executable stack/usr/bin/ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linkerWe don't really care whether stack is executable or not since we set it up ourselves and we're running DAT off mostly anyways.
Silence the warning by explicitly marking the stack as not executable.
Signed-off-by Nico Boehr <[email protected]>
Quoted 17 lines
--- s390x/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)s390x/Makefile +1
-1@@ -79,7 +79,7 @@ CFLAGS += -O2 CFLAGS += -march=zEC12 CFLAGS += -mbackchain CFLAGS += -fno-delete-null-pointer-checks -LDFLAGS += -nostdlib -Wl,--build-id=none +LDFLAGS += -nostdlib -Wl,--build-id=none -z noexecstack # We want to keep intermediate files .PRECIOUS: %.o %.lds
Quoted 20 lines · Nico Boehr
With somewhat recent GCC versions, we get this warning on s390x:
/usr/bin/ld: warning:s390x/cpu.o: missing .note.GNU-stack section implies executable stack/usr/bin/ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linkerWe don't really care whether stack is executable or not since we set it up ourselves and we're running DAT off mostly anyways.
Silence the warning by explicitly marking the stack as not executable.
Signed-off-by Nico Boehr <[email protected]>--- s390x/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)s390x/Makefile +0
-0
s390x/Makefile CFLAGS += -O2 · line 82
Quoted 5 lines
s390x/Makefile +1
-1@@ -79,7 +79,7 @@ CFLAGS += -O2 CFLAGS += -march=zEC12 CFLAGS += -mbackchain CFLAGS += -fno-delete-null-pointer-checks -LDFLAGS += -nostdlib -Wl,--build-id=none +LDFLAGS += -nostdlib -Wl,--build-id=none -z noexecstack
I already did a similar patch some weeks ago:
https://lore.kernel.org/kvm/[email protected]/
... we need it for x86, too, so I guess I should go ahead and commit it - and ask Sean to respin his conflicting series.
Thomas
Quoting Thomas Huth (2023-08-13 11:50:00)
Quoted 29 lines · Nico Boehr
Quoted 20 lines
With somewhat recent GCC versions, we get this warning on s390x:
/usr/bin/ld: warning:s390x/cpu.o: missing .note.GNU-stack section implies executable stack/usr/bin/ld: NOTE: This behaviour is deprecated and will be removed in a future version of the linkerWe don't really care whether stack is executable or not since we set it up ourselves and we're running DAT off mostly anyways.
Silence the warning by explicitly marking the stack as not executable.
Signed-off-by Nico Boehr <[email protected]>--- s390x/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)s390x/Makefile +0
-0s390x/Makefile CFLAGS += -O2 · line 82
Quoted 5 lines
s390x/Makefile +1
-1@@ -79,7 +79,7 @@ CFLAGS += -O2 CFLAGS += -march=zEC12 CFLAGS += -mbackchain CFLAGS += -fno-delete-null-pointer-checks -LDFLAGS += -nostdlib -Wl,--build-id=none +LDFLAGS += -nostdlib -Wl,--build-id=none -z noexecstackI already did a similar patch some weeks ago:
Quoted 4 lines
https://lore.kernel.org/kvm/[email protected]/
... we need it for x86, too, so I guess I should go ahead and commit it - and ask Sean to respin his conflicting series.
Thanks!