netdev - [PATCH 06/18] x86, barrier: stop speculation for failed access

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

Date:   Fri, 05 Jan 2018 17:10:26 -0800
From:   Dan Williams <dan.j.williams@...el.com>
To:     linux-kernel@...r.kernel.org
Cc:     linux-arch@...r.kernel.org, Andi Kleen <ak@...ux.intel.com>,
        Arnd Bergmann <arnd@...db.de>, gregkh@...uxfoundation.org,
        peterz@...radead.org, netdev@...r.kernel.org, x86@...nel.org,
        Ingo Molnar <mingo@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>, tglx@...utronix.de,
        torvalds@...ux-foundation.org, alan@...ux.intel.com
Subject: [PATCH 06/18] x86, barrier: stop speculation for failed access_ok

From: Andi Kleen <ak@...ux.intel.com>

When access_ok fails we should always stop speculating.
Add the required barriers to the x86 access_ok macro.

Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Ingo Molnar <mingo@...hat.com>
Cc: "H. Peter Anvin" <hpa@...or.com>
Cc: Arnd Bergmann <arnd@...db.de>
Cc: x86@...nel.org
Signed-off-by: Andi Kleen <ak@...ux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@...el.com>
---
 arch/x86/include/asm/uaccess.h |   17 +++++++++++++----
 include/asm-generic/barrier.h  |    6 +++---
 2 files changed, 16 insertions(+), 7 deletions(-)

diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index 574dff4d2913..9b6f20cfaeb9 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -43,6 +43,8 @@ static inline void set_fs(mm_segment_t fs)
 /*
  * Test whether a block of memory is a valid user space address.
  * Returns 0 if the range is valid, nonzero otherwise.
+ *
+ * We also disable speculation when a check fails.
  */
 static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size, unsigned long limit)
 {
@@ -53,14 +55,19 @@ static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size, un
 	 * important to subtract the size from the
 	 * limit, not add it to the address).
 	 */
-	if (__builtin_constant_p(size))
-		return unlikely(addr > limit - size);
+	if (__builtin_constant_p(size)) {
+		if (unlikely(addr > limit - size))
+			return true;
+		nospec_barrier();
+		return false;
+	}
 
 	/* Arbitrary sizes? Be careful about overflow */
 	addr += size;
-	if (unlikely(addr < size))
+	if (unlikely(addr < size || addr > limit))
 		return true;
-	return unlikely(addr > limit);
+	nospec_barrier();
+	return false;
 }
 
 #define __range_not_ok(addr, size, limit)				\
@@ -94,6 +101,8 @@ static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size, un
  * Note that, depending on architecture, this function probably just
  * checks that the pointer is in the user space range - after calling
  * this function, memory access functions may still return -EFAULT.
+ *
+ * Stops speculation automatically
  */
 #define access_ok(type, addr, size)					\
 ({									\
diff --git a/include/asm-generic/barrier.h b/include/asm-generic/barrier.h
index 91c3071f49e5..a11765eba860 100644
--- a/include/asm-generic/barrier.h
+++ b/include/asm-generic/barrier.h
@@ -59,7 +59,9 @@
  *
  * Architectures with a suitable memory barrier should provide an
  * implementation. This is non-portable, and generic code should use
- * nospec_ptr().
+ * nospec_{array_ptr,ptr}. Arch-specific code should define and use
+ * nospec_barrier() for usages where nospec_{array_ptr,ptr} is
+ * unsuitable.
  */
 #ifndef __nospec_barrier
 #define __nospec_barrier()		do { } while (0)
@@ -120,8 +122,6 @@
 	nospec_ptr(__arr + __idx, __arr, __arr + __sz);			\
 })
 
-#undef __nospec_barrier
-
 #ifndef __smp_mb
 #define __smp_mb()	mb()
 #endif