[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <149049867976.5373.11775682182530468124.stgit@devbox>
Date: Sun, 26 Mar 2017 12:24:49 +0900
From: Masami Hiramatsu <mhiramat@...nel.org>
To: linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...hat.com>
Cc: Peter Zijlstra <peterz@...radead.org>,
Thomas Gleixner <tglx@...utronix.de>,
"H . Peter Anvin" <hpa@...or.com>,
Ananth N Mavinakayanahalli <ananth@...ux.vnet.ibm.com>,
Anil S Keshavamurthy <anil.s.keshavamurthy@...el.com>,
"David S . Miller" <davem@...emloft.net>,
Andrey Ryabinin <aryabinin@...tuozzo.com>, mhiramat@...nel.org
Subject: [RFC PATCH tip/master 1/8] kprobes/x86: Fix not to boost call far instruction
Fix kprobe-booster not to boost call far instruction,
because call may store the address in singlestep
execution buffer to the stack, which should be modified
after single stepping.
Currently, this instruction will be filtered as not
boostable in resume_execution(), so this is not a
critical issue.
Signed-off-by: Masami Hiramatsu <mhiramat@...nel.org>
---
arch/x86/kernel/kprobes/core.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
index 993fa4f..9eae5a6 100644
--- a/arch/x86/kernel/kprobes/core.c
+++ b/arch/x86/kernel/kprobes/core.c
@@ -200,6 +200,8 @@ int can_boost(kprobe_opcode_t *opcodes, void *addr)
return (opcode != 0x62 && opcode != 0x67);
case 0x70:
return 0; /* can't boost conditional jump */
+ case 0x90:
+ return opcode != 0x9a; /* can't boost call far */
case 0xc0:
/* can't boost software-interruptions */
return (0xc1 < opcode && opcode < 0xcc) || opcode == 0xcf;
Powered by blists - more mailing lists