objtool: Allow no-op CFI ops in alternatives

Randy reported a false-positive:

  arch/x86/hyperv/hv_apic.o: warning: objtool: hv_apic_write()+0x25: alternative modifies stack

What happens is that:

	alternative_io("movl %0, %P1", "xchgl %0, %P1", X86_BUG_11AP,
 13d:   89 9d 00 d0 7f ff       mov    %ebx,-0x803000(%rbp)

decodes to an instruction with CFI-ops because it modifies RBP.
However, due to this being a !frame-pointer build, that should not in
fact change the CFI state.

So instead of dis-allowing any CFI-op, verify the op would've actually
changed the CFI state.

Fixes: 7117f16bf460 ("objtool: Fix ORC vs alternatives")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Tested-by: Randy Dunlap <rdunlap@infradead.org>

diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 32dea5f..196a551 100644 (file)
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -2060,17 +2060,18 @@ static int handle_insn_ops(struct instruction *insn, struct insn_state *state)
        struct stack_op *op;
 
        list_for_each_entry(op, &insn->stack_ops, list) {
+               struct cfi_state old_cfi = state->cfi;
                int res;
 
-               if (insn->alt_group) {
-                       WARN_FUNC("alternative modifies stack", insn->sec, insn->offset);
-                       return -1;
-               }
-
                res = update_cfi_state(insn, &state->cfi, op);
                if (res)
                        return res;
 
+               if (insn->alt_group && memcmp(&state->cfi, &old_cfi, sizeof(struct cfi_state))) {
+                       WARN_FUNC("alternative modifies stack", insn->sec, insn->offset);
+                       return -1;
+               }
+
                if (op->dest.type == OP_DEST_PUSHF) {
                        if (!state->uaccess_stack) {
                                state->uaccess_stack = 1;