[PATCH 03/11] riscv: indirect jmp in asm that's static in nature to use sw guarded jump

[Heinrich Schuchardt]

On 25.07.25 01:36, Deepak Gupta wrote:
> Handwritten `__memset` asm routine perform static jumps within
> function and uses `a5` to do that. This would require a landing pad
> instruction at the target. Since its static jump and no memory load is
> involved, use `t2` instead which is exempt from requiring a landing pad.
> 
> Signed-off-by: Deepak Gupta <debug at rivosinc.com>

Acked-by: Heinrich Schuchardt <heinrich.schuchardt at canonical.com>

> ---
>   arch/riscv/lib/memset.S | 6 +++---
>   1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/riscv/lib/memset.S b/arch/riscv/lib/memset.S
> index da23b8347e2d..c4a318d8eef3 100644
> --- a/arch/riscv/lib/memset.S
> +++ b/arch/riscv/lib/memset.S
> @@ -56,12 +56,12 @@ SYM_FUNC_START(__memset)
>   
>   	/* Jump into loop body */
>   	/* Assumes 32-bit instruction lengths */
> -	la a5, 3f
> +	la t2, 3f
>   #ifdef CONFIG_64BIT
>   	srli a4, a4, 1
>   #endif
> -	add a5, a5, a4
> -	jr a5
> +	add t2, t2, a4
> +	jr t2
>   3:
>   	REG_S a1,        0(t0)
>   	REG_S a1,    SZREG(t0)
>