On Sun, Dec 10, 2017 at 07:35:00PM +0100, Lukas Bulwahn wrote:
> Just use the right value and avoid an implicit conversion between the two
> enumeration types that just happened to work in this case. It seems that
> this must have just been overlooked in the new implementation of objtool
> in commit baa41469a7b9 ("objtool: Implement stack validation 2.0").
> 
> I found this when compiling the kernel with clang-5.0, i.e., executing
> 
>   make HOSTCC=clang-5.0 CC=clang-5.0 defconfig && \
>   make HOSTCC=clang-5.0 CC=clang-5.0
> 
> fails with:
> 
> ```
> arch/x86/decode.c:141:20: error: implicit conversion from enumeration type 
> 'enum op_src_type' to different enumeration type 'enum op_dest_type' 
> [-Werror,-Wenum-conversion]
>                         op->dest.type = OP_SRC_REG;
>                                       ~ ^~~~~~~~~~
> 1 error generated.
> ```

given the declaration in tools/objtool/arch.h
enum op_dest_type {
        OP_DEST_REG,
        ...

enum op_src_type {
        OP_SRC_REG,
        ...

the change seems correct and it is clear why it worked with the wrong type

> 
> Fixes: baa41469a7b9 ("objtool: Implement stack validation 2.0")
> Signed-off-by: Lukas Bulwahn <lukas.bulw...@gmail.com>
Reviewed-by: Nicholas Mc Guire <der.h...@hofr.at>

> ---
>  tools/objtool/arch/x86/decode.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c
> index 8acfc47..540a209 100644
> --- a/tools/objtool/arch/x86/decode.c
> +++ b/tools/objtool/arch/x86/decode.c
> @@ -138,7 +138,7 @@ int arch_decode_instruction(struct elf *elf, struct 
> section *sec,
>                       *type = INSN_STACK;
>                       op->src.type = OP_SRC_ADD;
>                       op->src.reg = op_to_cfi_reg[modrm_reg][rex_r];
> -                     op->dest.type = OP_SRC_REG;
> +                     op->dest.type = OP_DEST_REG;
>                       op->dest.reg = CFI_SP;
>               }
>               break;
> -- 
> 2.7.4
> 
> _______________________________________________
> SIL2review mailing list
> sil2rev...@lists.osadl.org
> https://lists.osadl.org/mailman/listinfo/sil2review

Reply via email to