We can reduce the number of instructions needed to use available_harts_lock by using the aq and rl suffixes for AMOs.
Signed-off-by: Sean Anderson seanga2@gmail.com ---
Changes in v2: - Remove fences after amoswaps - Reword commit message
arch/riscv/cpu/start.S | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-)
diff --git a/arch/riscv/cpu/start.S b/arch/riscv/cpu/start.S index e3222b1ea7..66ca1c7020 100644 --- a/arch/riscv/cpu/start.S +++ b/arch/riscv/cpu/start.S @@ -125,14 +125,12 @@ call_board_init_f_0:
#ifndef CONFIG_XIP la t0, available_harts_lock - fence rw, w - amoswap.w zero, zero, 0(t0) + amoswap.w.rl zero, zero, 0(t0)
wait_for_gd_init: la t0, available_harts_lock li t1, 1 -1: amoswap.w t1, t1, 0(t0) - fence r, rw +1: amoswap.w.aq t1, t1, 0(t0) bnez t1, 1b
/* register available harts in the available_harts mask */ @@ -142,8 +140,7 @@ wait_for_gd_init: or t2, t2, t1 SREG t2, GD_AVAILABLE_HARTS(gp)
- fence rw, w - amoswap.w zero, zero, 0(t0) + amoswap.w.rl zero, zero, 0(t0)
/* * Continue on hart lottery winner, others branch to