[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <YCOb2byLJhLOjhrL@hirez.programming.kicks-ass.net>
Date: Wed, 10 Feb 2021 09:39:53 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Josh Poimboeuf <jpoimboe@...hat.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Dave Hansen <dave.hansen@...el.com>, x86@...nel.org,
linux-kernel@...r.kernel.org
Cc: Steven Rostedt <rostedt@...dmis.org>, yu-cheng.yu@...el.com
Subject: [RFC][PATCH] objtool: WARN about ENDBR instructions
Given all the ENDBR fun we recently had, do we want the below? Until
someone comes and fixes up kprobes/ftrace/livepatch etc.. having them is
a giant pain and we'd better warn about it.
---
diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c
index 931cef78b857..e708731b10cd 100644
--- a/tools/objtool/arch/x86/decode.c
+++ b/tools/objtool/arch/x86/decode.c
@@ -91,7 +91,7 @@ int arch_decode_instruction(const struct elf *elf, const struct section *sec,
{
struct insn insn;
int x86_64, sign;
- unsigned char op1, op2, rex = 0, rex_b = 0, rex_r = 0, rex_w = 0,
+ unsigned char op1, op2, pfx = 0, rex = 0, rex_b = 0, rex_r = 0, rex_w = 0,
rex_x = 0, modrm = 0, modrm_mod = 0, modrm_rm = 0,
modrm_reg = 0, sib = 0;
struct stack_op *op = NULL;
@@ -118,6 +118,9 @@ int arch_decode_instruction(const struct elf *elf, const struct section *sec,
op1 = insn.opcode.bytes[0];
op2 = insn.opcode.bytes[1];
+ if (insn.prefixes.nbytes)
+ pfx = insn.prefixes.bytes[0];
+
if (insn.rex_prefix.nbytes) {
rex = insn.rex_prefix.bytes[0];
rex_w = X86_REX_W(rex) >> 3;
@@ -444,6 +447,11 @@ int arch_decode_instruction(const struct elf *elf, const struct section *sec,
/* nopl/nopw */
*type = INSN_NOP;
+ } else if (op2 == 0x1e && pfx == 0xf3 && (modrm == 0xfa || modrm == 0xfb)) {
+
+ /* endbr32/endbr64 */
+ WARN("endbr32/64 instruction at %s:0x%lx", sec->name, offset);
+
} else if (op2 == 0xa0 || op2 == 0xa8) {
/* push fs/gs */
Powered by blists - more mailing lists