* [PATCH][kprobe] disallow kprobes on emulate_step and branch_taken
@ 2006-11-01 2:50 Gui,Jian
0 siblings, 0 replies; only message in thread
From: Gui,Jian @ 2006-11-01 2:50 UTC (permalink / raw)
To: linuxppc-dev
Probing on emulate_step function will crash 2.6.18.1 when it is
triggered. This problem is powerpc dependent.
When kprobe is triggered, emulate_step() is on its kernel path and
will cause recursive kprobe fault. And branch_taken() is called
in emulate_step(). This patch can disallow kprobes on both of them.
diff -Nur orig/arch/powerpc/lib/sstep.c linux-2.6.18.1/arch/powerpc/lib/sstep.c
--- orig/arch/powerpc/lib/sstep.c 2006-11-01 08:45:06.000000000 -0500
+++ linux-2.6.18.1/arch/powerpc/lib/sstep.c 2006-11-01 08:50:19.000000000 -0500
@@ -9,6 +9,7 @@
* 2 of the License, or (at your option) any later version.
*/
#include <linux/kernel.h>
+#include <linux/kprobes.h>
#include <linux/ptrace.h>
#include <asm/sstep.h>
#include <asm/processor.h>
@@ -25,7 +26,7 @@
/*
* Determine whether a conditional branch instruction would branch.
*/
-static int branch_taken(unsigned int instr, struct pt_regs *regs)
+static int __kprobes branch_taken(unsigned int instr, struct pt_regs *regs)
{
unsigned int bo = (instr >> 21) & 0x1f;
unsigned int bi;
@@ -51,7 +52,7 @@
* or -1 if the instruction is one that should not be stepped,
* such as an rfid, or a mtmsrd that would clear MSR_RI.
*/
-int emulate_step(struct pt_regs *regs, unsigned int instr)
+int __kprobes emulate_step(struct pt_regs *regs, unsigned int instr)
{
unsigned int opcode, rd;
unsigned long int imm;
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2006-11-01 2:51 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-11-01 2:50 [PATCH][kprobe] disallow kprobes on emulate_step and branch_taken Gui,Jian
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).