2019-06-25 19:16:53

by Dianzhang Chen

Subject: [PATCH v2] x86/ptrace: Fix possible spectre-v1 in ptrace_get_debugreg()

The index to access the threads ptrace_bps is controlled by userspace
via syscall: sys_ptrace(), hence, hence leading to a potential
exploitation of the Spectre variant 1 vulnerability.
The n can be controlled from:
ptrace -> arch_ptrace -> ptrace_get_debugreg.

Fix this by sanitizing n before using it to index thread->ptrace_bps.

Signed-off-by: Dianzhang Chen <[email protected]>
---
arch/x86/kernel/ptrace.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index a166c96..cbac646 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -25,6 +25,7 @@
#include <linux/rcupdate.h>
#include <linux/export.h>
#include <linux/context_tracking.h>
+#include <linux/nospec.h>

#include <linux/uaccess.h>
#include <asm/pgtable.h>
@@ -643,9 +644,11 @@ static unsigned long ptrace_get_debugreg(struct task_struct *tsk, int n)
{
struct thread_struct *thread = &tsk->thread;
unsigned long val = 0;
+ int index = n;

if (n < HBP_NUM) {
- struct perf_event *bp = thread->ptrace_bps[n];
+ index = array_index_nospec(index, HBP_NUM);
+ struct perf_event *bp = thread->ptrace_bps[index];

if (bp)
val = bp->hw.info.address;
--
2.7.4

2019-06-27 21:52:29

by tip-bot for Vasyl Gomonovych

Subject: [tip:x86/pti] x86/ptrace: Fix possible spectre-v1 in ptrace_get_debugreg()

Commit-ID: 31a2fbb390fee4231281b939e1979e810f945415
Gitweb: https://git.kernel.org/tip/31a2fbb390fee4231281b939e1979e810f945415
Author: Dianzhang Chen <[email protected]>
AuthorDate: Tue, 25 Jun 2019 23:30:17 +0800
Committer: Thomas Gleixner <[email protected]>
CommitDate: Thu, 27 Jun 2019 23:48:04 +0200

x86/ptrace: Fix possible spectre-v1 in ptrace_get_debugreg()

The index to access the threads ptrace_bps is controlled by userspace via
syscall: sys_ptrace(), hence leading to a potential exploitation of the
Spectre variant 1 vulnerability.

The index can be controlled from:
ptrace -> arch_ptrace -> ptrace_get_debugreg.

Fix this by sanitizing the user supplied index before using it access
thread->ptrace_bps.

Signed-off-by: Dianzhang Chen <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: https://lkml.kernel.org/r/[email protected]

---
arch/x86/kernel/ptrace.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index a166c960bc9e..cbac64659dc4 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -25,6 +25,7 @@
#include <linux/rcupdate.h>
#include <linux/export.h>
#include <linux/context_tracking.h>
+#include <linux/nospec.h>

#include <linux/uaccess.h>
#include <asm/pgtable.h>
@@ -643,9 +644,11 @@ static unsigned long ptrace_get_debugreg(struct task_struct *tsk, int n)
{
struct thread_struct *thread = &tsk->thread;
unsigned long val = 0;
+ int index = n;

if (n < HBP_NUM) {
- struct perf_event *bp = thread->ptrace_bps[n];
+ index = array_index_nospec(index, HBP_NUM);
+ struct perf_event *bp = thread->ptrace_bps[index];

if (bp)
val = bp->hw.info.address;