> On Sep 17, 2014, at 7:43 AM, James Greenhalgh <james.greenha...@arm.com> > wrote: > > >> On Wed, Sep 17, 2014 at 09:30:31AM +0100, Richard Earnshaw wrote: >> "=&r" is correct for an early-clobbered scratch. >> >> R. > > In that case... > > How is the attached patch for trunk? I've bootstrapped it on AArch64 > with -fstack-protector-strong and -frename-registers in the BOOT_CFLAGS > without seeing any issues. > > OK? > > Thanks, > James > > --- > gcc/ > > 2014-09-15 James Greenhalgh <james.greenha...@arm.com> > > * config/aarch64/aarch64.md (stack_protect_test_<mode>): Mark > scratch register as an output to placate register renaming. > > gcc/testsuite/ > > 2014-09-15 James Greenhalgh <james.greenha...@arm.com> > > * gcc.target/aarch64/stack_protector_set_1.c: New. > * gcc.target/aarch64/stack_protector_set_2.c: Likewise.
There is nothing aarch64 specific about this testcase so I would place them under gcc.dg and add the extra marker which says this testcase requires stack protector. And maybe even use compile instead of just assemble too. Thanks, Andrew > <0001-Re-PATCH-AArch64-Add-constraint-letter-for-stack_pro.patch>