diff mbox series

[5.11,443/775] powerpc/uaccess: Avoid might_fault() when user access is enabled

Message ID 20210301161223.444826869@linuxfoundation.org
State Superseded
Headers show
Series None | expand

Commit Message

Greg Kroah-Hartman March 1, 2021, 4:10 p.m. UTC
From: Alexey Kardashevskiy <aik@ozlabs.ru>

[ Upstream commit 7d506ca97b665b95e698a53697dad99fae813c1a ]

The amount of code executed with enabled user space access (unlocked
KUAP) should be minimal. However with CONFIG_PROVE_LOCKING or
CONFIG_DEBUG_ATOMIC_SLEEP enabled, might_fault() calls into various
parts of the kernel, and may even end up replaying interrupts which in
turn may access user space and forget to restore the KUAP state.

The problem places are:
  1. strncpy_from_user (and similar) which unlock KUAP and call
     unsafe_get_user -> __get_user_allowed -> __get_user_nocheck()
     with do_allow=false to skip KUAP as the caller took care of it.
  2. __unsafe_put_user_goto() which is called with unlocked KUAP.

eg:
  WARNING: CPU: 30 PID: 1 at arch/powerpc/include/asm/book3s/64/kup.h:324 arch_local_irq_restore+0x160/0x190
  NIP arch_local_irq_restore+0x160/0x190
  LR  lock_is_held_type+0x140/0x200
  Call Trace:
    0xc00000007f392ff8 (unreliable)
    ___might_sleep+0x180/0x320
    __might_fault+0x50/0xe0
    filldir64+0x2d0/0x5d0
    call_filldir+0xc8/0x180
    ext4_readdir+0x948/0xb40
    iterate_dir+0x1ec/0x240
    sys_getdents64+0x80/0x290
    system_call_exception+0x160/0x280
    system_call_common+0xf0/0x27c

Change __get_user_nocheck() to look at `do_allow` to decide whether to
skip might_fault(). Since strncpy_from_user/etc call might_fault()
anyway before unlocking KUAP, there should be no visible change.

Drop might_fault() in __unsafe_put_user_goto() as it is only called
from unsafe_put_user(), which already has KUAP unlocked.

Since keeping might_fault() is still desirable for debugging, add
calls to it in user_[read|write]_access_begin(). That also allows us
to drop the is_kernel_addr() test, because there should be no code
using user_[read|write]_access_begin() in order to access a kernel
address.

Fixes: de78a9c42a79 ("powerpc: Add a framework for Kernel Userspace Access Protection")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[mpe: Combine with related patch from myself, merge change logs]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210204121612.32721-1-aik@ozlabs.ru
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/powerpc/include/asm/uaccess.h | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
diff mbox series

Patch

diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h
index 501c9a79038c0..f53bfefb4a577 100644
--- a/arch/powerpc/include/asm/uaccess.h
+++ b/arch/powerpc/include/asm/uaccess.h
@@ -216,8 +216,6 @@  do {								\
 #define __put_user_nocheck_goto(x, ptr, size, label)		\
 do {								\
 	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
-	if (!is_kernel_addr((unsigned long)__pu_addr))		\
-		might_fault();					\
 	__chk_user_ptr(ptr);					\
 	__put_user_size_goto((x), __pu_addr, (size), label);	\
 } while (0)
@@ -313,7 +311,7 @@  do {								\
 	__typeof__(size) __gu_size = (size);			\
 								\
 	__chk_user_ptr(__gu_addr);				\
-	if (!is_kernel_addr((unsigned long)__gu_addr))		\
+	if (do_allow && !is_kernel_addr((unsigned long)__gu_addr)) \
 		might_fault();					\
 	barrier_nospec();					\
 	if (do_allow)								\
@@ -508,6 +506,9 @@  static __must_check inline bool user_access_begin(const void __user *ptr, size_t
 {
 	if (unlikely(!access_ok(ptr, len)))
 		return false;
+
+	might_fault();
+
 	allow_read_write_user((void __user *)ptr, ptr, len);
 	return true;
 }
@@ -521,6 +522,9 @@  user_read_access_begin(const void __user *ptr, size_t len)
 {
 	if (unlikely(!access_ok(ptr, len)))
 		return false;
+
+	might_fault();
+
 	allow_read_from_user(ptr, len);
 	return true;
 }
@@ -532,6 +536,9 @@  user_write_access_begin(const void __user *ptr, size_t len)
 {
 	if (unlikely(!access_ok(ptr, len)))
 		return false;
+
+	might_fault();
+
 	allow_write_to_user((void __user *)ptr, len);
 	return true;
 }