Agree @Jeff and updated the testcase . Thank you ~Y
On Tue, Jun 17, 2025 at 8:46 AM Jeff Law <jeffreya...@gmail.com> wrote: > > > On 6/11/25 2:47 AM, Umesh Kalappa wrote: > > gcc/ChangeLog: > > > > * config/riscv/sync.md (lrsc_atomic_exchange<mode>): > > Use the scratch reg for rd. > > > > gcc/testsuite/ChangeLog: > > > > * gcc.target/riscv/zalrsc.c: New test. > I was about to push this to the trunk when I realized the test isn't > really all that useful. > > The net of this change is we get to drop the unnecessary mv. But the > test just verifies that we have a lr/sc pair, which was true both before > and after this change. > > Is it possible to improve the test such that it fails without this > change and passes with this change? > > Thanks, > jeff > >
