linux-kernel - [PATCH 1/4] x86: fix jprobe_saved

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-Id: <1197926820.23402.28.camel@brick>
Date:	Mon, 17 Dec 2007 13:26:55 -0800
From:	Harvey Harrison <harvey.harrison@...il.com>
To:	Ingo Molnar <mingo@...e.hu>
Cc:	Ananth N Mavinakayanahalli <ananth@...ibm.com>,
	Jim Keniston <jkenisto@...ibm.com>,
	Roland McGrath <roland@...hat.com>,
	Arjan van de Ven <arjan@...radead.org>,
	prasanna@...ibm.com, anil.s.keshavamurthy@...el.com,
	davem@...emloft.net, systemtap-ml <systemtap@...rces.redhat.com>,
	LKML <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: [PATCH 1/4] x86: fix jprobe_saved_sp handling

This patch fixes a bug of jprobe and cleans up. jprobe for x86-64
can cause kernel page fault when the jprobe_return() is called from
incorrect function. Anyway, that path finally invokes BUG() macro,
so this is not so serious.

Based on patch from Masami Hiramatsu <mhiramat@...hat.com>

- Use jprobe_saved_regs instead getting it from stack.
  (Especially on x86-64, it may get incorrect data, because
   pt_regs can not be get by using container_of(rsp))
- Change the type of stack pointer to unsigned long *.

Signed-off-by: Harvey Harrison <harvey.harrison@...il.com>
---
 arch/x86/kernel/kprobes.c |    8 ++------
 include/asm-x86/kprobes.h |    2 +-
 2 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index e5b7d07..9130c01 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -1022,16 +1022,12 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	if ((addr > (u8 *) jprobe_return) && (addr < (u8 *) jprobe_return_end)) {
 #ifdef CONFIG_X86_32
 		if (&regs->sp != kcb->jprobe_saved_sp) {
-			struct pt_regs *saved_regs =
-			    container_of(kcb->jprobe_saved_sp,
-					    struct pt_regs, sp);
+			struct pt_regs *saved_regs = &kcb->jprobe_saved_sp;
 			printk("current sp %p does not match saved sp %p\n",
 			       &regs->sp, kcb->jprobe_saved_sp);
 #else
 		if ((long *)regs->sp != kcb->jprobe_saved_sp) {
-			struct pt_regs *saved_regs =
-			    container_of(kcb->jprobe_saved_sp,
-					    struct pt_regs, sp);
+			struct pt_regs *saved_regs = &kcb->jprobe_saved_sp;
 			printk("current sp %p does not match saved sp %p\n",
 			       (long *)regs->sp, kcb->jprobe_saved_sp);
 #endif
diff --git a/include/asm-x86/kprobes.h b/include/asm-x86/kprobes.h
index e348ed6..7319c62 100644
--- a/include/asm-x86/kprobes.h
+++ b/include/asm-x86/kprobes.h
@@ -79,7 +79,7 @@ struct kprobe_ctlblk {
 	unsigned long kprobe_status;
 	unsigned long kprobe_old_flags;
 	unsigned long kprobe_saved_flags;
-	long *jprobe_saved_sp;
+	unsigned long *jprobe_saved_sp;
 	struct pt_regs jprobe_saved_regs;
 	kprobe_opcode_t jprobes_stack[MAX_STACK_SIZE];
 	struct prev_kprobe prev_kprobe;
-- 
1.5.4.rc0.1083.gf568


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/