Re: [PATCH] objtool: fix bytes check of lea's rex_prefix

On Mon, Dec 05, 2016 at 11:55:51AM +0100, Jiri Slaby wrote: > For the "lea %(rsp), %rbp" case, we check if there is a rex_prefix. But > we check "bytes" which is insn_byte_t[4] in rex_prefix (insn_field > structure). Therefore, the check is always true. > > Instead, check nbytes which is the

Re: [PATCH] objtool: fix bytes check of lea's rex_prefix

On Mon, Dec 05, 2016 at 11:55:51AM +0100, Jiri Slaby wrote: > For the "lea %(rsp), %rbp" case, we check if there is a rex_prefix. But > we check "bytes" which is insn_byte_t[4] in rex_prefix (insn_field > structure). Therefore, the check is always true. > > Instead, check nbytes which is the

[PATCH] objtool: fix bytes check of lea's rex_prefix

For the "lea %(rsp), %rbp" case, we check if there is a rex_prefix. But we check "bytes" which is insn_byte_t[4] in rex_prefix (insn_field structure). Therefore, the check is always true. Instead, check nbytes which is the right one. Signed-off-by: Jiri Slaby Cc: Josh Poimboeuf

[PATCH] objtool: fix bytes check of lea's rex_prefix

For the "lea %(rsp), %rbp" case, we check if there is a rex_prefix. But we check "bytes" which is insn_byte_t[4] in rex_prefix (insn_field structure). Therefore, the check is always true. Instead, check nbytes which is the right one. Signed-off-by: Jiri Slaby Cc: Josh Poimboeuf ---