[PATCH] firmware: Simplify FDT header endianness conversion
Anup Patel
anup at brainfault.org
Tue Sep 24 22:18:45 PDT 2024
On Fri, Aug 30, 2024 at 1:45 AM Samuel Holland
<samuel.holland at sifive.com> wrote:
>
> Reduce the code size by using single-byte loads instead of bit
> manipulation. This method also does the right thing on (hypothetical)
> big-endian systems.
>
> Signed-off-by: Samuel Holland <samuel.holland at sifive.com>
LGTM.
Reviewed-by: Anup Patel <anup at brainfault.org>
Applied this patch to the riscv/opensbi repo.
Thanks,
Anup
> ---
>
> firmware/fw_base.S | 26 +++++---------------------
> 1 file changed, 5 insertions(+), 21 deletions(-)
>
> diff --git a/firmware/fw_base.S b/firmware/fw_base.S
> index b950c0b8..0b58cc42 100644
> --- a/firmware/fw_base.S
> +++ b/firmware/fw_base.S
> @@ -242,30 +242,14 @@ _scratch_init:
> beq t1, a1, _fdt_reloc_done
> /* t0 = source FDT start address */
> add t0, a1, zero
> - /* t2 = source FDT size in big-endian */
> -#if __riscv_xlen > 32
> - lwu t2, 4(t0)
> -#else
> - lw t2, 4(t0)
> -#endif
> - /* t3 = bit[15:8] of FDT size */
> - add t3, t2, zero
> - srli t3, t3, 16
> - and t3, t3, a4
> + /* t2 = source FDT size (convert from big-endian) */
> + lbu t2, 7(t0)
> + lbu t3, 6(t0)
> + lbu t4, 5(t0)
> + lbu t5, 4(t0)
> slli t3, t3, 8
> - /* t4 = bit[23:16] of FDT size */
> - add t4, t2, zero
> - srli t4, t4, 8
> - and t4, t4, a4
> slli t4, t4, 16
> - /* t5 = bit[31:24] of FDT size */
> - add t5, t2, zero
> - and t5, t5, a4
> slli t5, t5, 24
> - /* t2 = bit[7:0] of FDT size */
> - srli t2, t2, 24
> - and t2, t2, a4
> - /* t2 = FDT size in little-endian */
> or t2, t2, t3
> or t2, t2, t4
> or t2, t2, t5
> --
> 2.45.1
>
>
> --
> opensbi mailing list
> opensbi at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/opensbi
More information about the opensbi
mailing list