[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161011015307.su5oe4ki4otddrb6@treble>
Date: Mon, 10 Oct 2016 20:53:07 -0500
From: Josh Poimboeuf <jpoimboe@...hat.com>
To: Arnd Bergmann <arnd@...db.de>
Cc: Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>, x86@...nel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH] objtool: support '-mtune=atom' stack frame setup instruction
>From 60c982d4d04014adb3bde1ebee6ca95320ffb213 Mon Sep 17 00:00:00 2001
Message-Id: <60c982d4d04014adb3bde1ebee6ca95320ffb213.1476150736.git.jpoimboe@...hat.com>
From: Josh Poimboeuf <jpoimboe@...hat.com>
Date: Mon, 10 Oct 2016 15:24:01 -0500
Subject: [PATCH] objtool: support '-mtune=atom' stack frame setup instruction
Arnd reported that enabling CONFIG_MATOM results in a bunch of objtool
false positive frame pointer warnings:
arch/x86/events/intel/ds.o: warning: objtool: intel_pmu_pebs_del()+0x43: call without frame pointer save/setup
security/keys/keyring.o: warning: objtool: keyring_read()+0x59: call without frame pointer save/setup
kernel/signal.o: warning: objtool: __dequeue_signal()+0xd8: call without frame pointer save/setup
...
objtool gets confused by the fact that the '-mtune=atom' gcc option
sometimes uses 'lea (%rsp),%rbp' instead of 'mov %rsp,%rbp'. The
instructions are effectively the same, but objtool doesn't know about
the 'lea' variant.
Fix the false warnings by adding support for 'lea (%rsp),%rbp' in the
objtool decoder.
Reported-by: Arnd Bergmann <arnd@...db.de>
Signed-off-by: Josh Poimboeuf <jpoimboe@...hat.com>
---
tools/objtool/arch/x86/decode.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/tools/objtool/arch/x86/decode.c b/tools/objtool/arch/x86/decode.c
index c0c0b26..b63a31b 100644
--- a/tools/objtool/arch/x86/decode.c
+++ b/tools/objtool/arch/x86/decode.c
@@ -98,6 +98,15 @@ int arch_decode_instruction(struct elf *elf, struct section *sec,
*type = INSN_FP_SETUP;
break;
+ case 0x8d:
+ if (insn.rex_prefix.bytes &&
+ insn.rex_prefix.bytes[0] == 0x48 &&
+ insn.modrm.nbytes && insn.modrm.bytes[0] == 0x2c &&
+ insn.sib.nbytes && insn.sib.bytes[0] == 0x24)
+ /* lea %(rsp), %rbp */
+ *type = INSN_FP_SETUP;
+ break;
+
case 0x90:
*type = INSN_NOP;
break;
--
2.7.4
Powered by blists - more mailing lists