[PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

Christoph Hellwig and a few others spent a huge effort on removing
set_fs() from most of the important architectures, but about half the
other architectures were never completed even though most of them don't
actually use set_fs() at all.

I did a patch for microblaze at some point, which turned out to be fairly
generic, and now ported it to most other architectures, using new generic
implementations of access_ok() and __{get,put}_kernel_nocheck().

Three architectures (sparc64, ia64, and sh) needed some extra work,
which I also completed.

The final series contains extra cleanup changes that touch all
architectures. Please review and test these, so we can merge them
for v5.18.

The series is available at
https://git.kernel.org/pub/scm/linux/kernel/git/arnd/playground.git/log/?h=set_fs-2
for testing.

Changes in v2:
 - add fixes for more nios2 and microblaze bugs found in the process
 - do final cleanup in a single patch
 - fix sparc64 regression
 - introduce CONFIG_ALTERNATE_USER_ADDRESS_SPACE
 - fix access_ok() in lib/test_lockup.c

Arnd Bergmann (18):
  uaccess: fix integer overflow on access_ok()
  uaccess: fix nios2 and microblaze get_user_8()
  nds32: fix access_ok() checks in get/put_user
  sparc64: add __{get,put}_kernel_nocheck()
  x86: remove __range_not_ok()
  x86: use more conventional access_ok() definition
  nios2: drop access_ok() check from __put_user()
  uaccess: add generic __{get,put}_kernel_nofault
  mips: use simpler access_ok()
  m68k: fix access_ok for coldfire
  arm64: simplify access_ok()
  uaccess: fix type mismatch warnings from access_ok()
  uaccess: generalize access_ok()
  lib/test_lockup: fix kernel pointer check for separate address spaces
  sparc64: remove CONFIG_SET_FS support
  sh: remove CONFIG_SET_FS support
  ia64: remove CONFIG_SET_FS support
  uaccess: drop maining CONFIG_SET_FS users

 arch/Kconfig                              |  10 +-
 arch/alpha/Kconfig                        |   1 -
 arch/alpha/include/asm/processor.h        |   4 -
 arch/alpha/include/asm/thread_info.h      |   2 -
 arch/alpha/include/asm/uaccess.h          |  53 +---------
 arch/arc/Kconfig                          |   1 -
 arch/arc/include/asm/segment.h            |  20 ----
 arch/arc/include/asm/thread_info.h        |   3 -
 arch/arc/include/asm/uaccess.h            |  30 ------
 arch/arc/kernel/process.c                 |   2 +-
 arch/arm/include/asm/uaccess.h            |  22 +---
 arch/arm/kernel/swp_emulate.c             |   2 +-
 arch/arm/kernel/traps.c                   |   2 +-
 arch/arm/lib/uaccess_with_memcpy.c        |  10 --
 arch/arm64/include/asm/uaccess.h          |  29 +-----
 arch/csky/Kconfig                         |   1 -
 arch/csky/include/asm/processor.h         |   2 -
 arch/csky/include/asm/segment.h           |  10 --
 arch/csky/include/asm/thread_info.h       |   2 -
 arch/csky/include/asm/uaccess.h           |  12 ---
 arch/csky/kernel/asm-offsets.c            |   1 -
 arch/csky/kernel/signal.c                 |   2 +-
 arch/h8300/Kconfig                        |   1 -
 arch/h8300/include/asm/processor.h        |   1 -
 arch/h8300/include/asm/segment.h          |  40 --------
 arch/h8300/include/asm/thread_info.h      |   3 -
 arch/h8300/kernel/entry.S                 |   1 -
 arch/h8300/kernel/head_ram.S              |   1 -
 arch/h8300/mm/init.c                      |   6 --
 arch/h8300/mm/memory.c                    |   1 -
 arch/hexagon/Kconfig                      |   1 -
 arch/hexagon/include/asm/thread_info.h    |   6 --
 arch/hexagon/include/asm/uaccess.h        |  25 -----
 arch/hexagon/kernel/process.c             |   1 -
 arch/ia64/Kconfig                         |   1 -
 arch/ia64/include/asm/processor.h         |   4 -
 arch/ia64/include/asm/thread_info.h       |   2 -
 arch/ia64/include/asm/uaccess.h           |  26 ++---
 arch/ia64/kernel/unaligned.c              |  60 +++++++----
 arch/m68k/Kconfig.cpu                     |   1 +
 arch/m68k/include/asm/uaccess.h           |  14 +--
 arch/microblaze/Kconfig                   |   1 -
 arch/microblaze/include/asm/thread_info.h |   6 --
 arch/microblaze/include/asm/uaccess.h     |  61 ++---------
 arch/microblaze/kernel/asm-offsets.c      |   1 -
 arch/microblaze/kernel/process.c          |   1 -
 arch/mips/include/asm/uaccess.h           |  49 +--------
 arch/mips/sibyte/common/sb_tbprof.c       |   6 +-
 arch/nds32/Kconfig                        |   1 -
 arch/nds32/include/asm/thread_info.h      |   4 -
 arch/nds32/include/asm/uaccess.h          |  40 ++++----
 arch/nds32/kernel/process.c               |   5 +-
 arch/nds32/mm/alignment.c                 |   3 -
 arch/nios2/Kconfig                        |   1 -
 arch/nios2/include/asm/thread_info.h      |   9 --
 arch/nios2/include/asm/uaccess.h          | 105 +++++++++----------
 arch/nios2/kernel/signal.c                |  20 ++--
 arch/openrisc/Kconfig                     |   1 -
 arch/openrisc/include/asm/thread_info.h   |   7 --
 arch/openrisc/include/asm/uaccess.h       |  42 +-------
 arch/parisc/Kconfig                       |   1 +
 arch/parisc/include/asm/futex.h           |   6 --
 arch/parisc/include/asm/uaccess.h         |  13 +--
 arch/parisc/lib/memcpy.c                  |   2 +-
 arch/powerpc/include/asm/uaccess.h        |  13 +--
 arch/powerpc/lib/sstep.c                  |   4 +-
 arch/riscv/include/asm/uaccess.h          |  33 +-----
 arch/riscv/kernel/perf_callchain.c        |   2 +-
 arch/s390/Kconfig                         |   1 +
 arch/s390/include/asm/uaccess.h           |  16 +--
 arch/sh/Kconfig                           |   1 -
 arch/sh/include/asm/processor.h           |   1 -
 arch/sh/include/asm/segment.h             |  33 ------
 arch/sh/include/asm/thread_info.h         |   2 -
 arch/sh/include/asm/uaccess.h             |  24 +----
 arch/sh/kernel/io_trapped.c               |   9 +-
 arch/sh/kernel/process_32.c               |   2 -
 arch/sh/kernel/traps_32.c                 |  30 ++++--
 arch/sparc/Kconfig                        |   3 +-
 arch/sparc/include/asm/processor_32.h     |   6 --
 arch/sparc/include/asm/processor_64.h     |   4 -
 arch/sparc/include/asm/switch_to_64.h     |   4 +-
 arch/sparc/include/asm/thread_info_64.h   |   4 +-
 arch/sparc/include/asm/uaccess.h          |   3 -
 arch/sparc/include/asm/uaccess_32.h       |  31 +-----
 arch/sparc/include/asm/uaccess_64.h       | 113 +++++++++++++-------
 arch/sparc/kernel/process_32.c            |   2 -
 arch/sparc/kernel/process_64.c            |  12 ---
 arch/sparc/kernel/signal_32.c             |   2 +-
 arch/sparc/kernel/traps_64.c              |   2 -
 arch/sparc/lib/NGmemcpy.S                 |   3 +-
 arch/sparc/mm/init_64.c                   |   7 +-
 arch/um/include/asm/uaccess.h             |   7 +-
 arch/x86/events/core.c                    |   2 +-
 arch/x86/include/asm/uaccess.h            |  35 +------
 arch/x86/kernel/dumpstack.c               |   2 +-
 arch/x86/kernel/stacktrace.c              |   2 +-
 arch/x86/lib/usercopy.c                   |   2 +-
 arch/xtensa/Kconfig                       |   1 -
 arch/xtensa/include/asm/asm-uaccess.h     |  71 -------------
 arch/xtensa/include/asm/processor.h       |   7 --
 arch/xtensa/include/asm/thread_info.h     |   3 -
 arch/xtensa/include/asm/uaccess.h         |  26 +----
 arch/xtensa/kernel/asm-offsets.c          |   3 -
 drivers/hid/uhid.c                        |   2 +-
 drivers/scsi/sg.c                         |   5 -
 fs/exec.c                                 |   6 --
 include/asm-generic/access_ok.h           |  51 ++++++++++
 include/asm-generic/uaccess.h             |  46 +--------
 include/linux/syscalls.h                  |   4 -
 include/linux/uaccess.h                   |  59 ++++-------
 include/rdma/ib.h                         |   2 +-
 kernel/events/callchain.c                 |   4 -
 kernel/events/core.c                      |   3 -
 kernel/exit.c                             |  14 ---
 kernel/kthread.c                          |   5 -
 kernel/stacktrace.c                       |   3 -
 kernel/trace/bpf_trace.c                  |   4 -
 lib/test_lockup.c                         |  11 +-
 mm/maccess.c                              | 119 ----------------------
 mm/memory.c                               |   8 --
 net/bpfilter/bpfilter_kern.c              |   2 +-
 122 files changed, 387 insertions(+), 1290 deletions(-)
 delete mode 100644 arch/arc/include/asm/segment.h
 delete mode 100644 arch/csky/include/asm/segment.h
 delete mode 100644 arch/h8300/include/asm/segment.h
 delete mode 100644 arch/sh/include/asm/segment.h
 create mode 100644 include/asm-generic/access_ok.h

-- 
2.29.2

Cc: linux@armlinux.org.uk
Cc: will@kernel.org
Cc: guoren@kernel.org
Cc: bcain@codeaurora.org
Cc: geert@linux-m68k.org
Cc: monstr@monstr.eu
Cc: tsbogend@alpha.franken.de
Cc: nickhu@andestech.com
Cc: green.hu@gmail.com
Cc: dinguyen@kernel.org
Cc: shorne@gmail.com
Cc: deller@gmx.de
Cc: mpe@ellerman.id.au
Cc: peterz@infradead.org
Cc: mingo@redhat.com
Cc: mark.rutland@arm.com
Cc: hca@linux.ibm.com
Cc: dalias@libc.org
Cc: davem@davemloft.net
Cc: richard@nod.at
Cc: x86@kernel.org
Cc: jcmvbkbc@gmail.com
Cc: ebiederm@xmission.com
Cc: arnd@arndb.de
Cc: akpm@linux-foundation.org
Cc: ardb@kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-alpha@vger.kernel.org
Cc: linux-snps-arc@lists.infradead.org
Cc: linux-csky@vger.kernel.org
Cc: linux-hexagon@vger.kernel.org
Cc: linux-ia64@vger.kernel.org
Cc: linux-m68k@lists.linux-m68k.org
Cc: linux-mips@vger.kernel.org
Cc: openrisc@lists.librecores.org
Cc: linux-parisc@vger.kernel.org
Cc: linuxppc-dev@lists.ozlabs.org
Cc: linux-riscv@lists.infradead.org
Cc: linux-s390@vger.kernel.org
Cc: linux-sh@vger.kernel.org
Cc: sparclinux@vger.kernel.org
Cc: linux-um@lists.infradead.org
Cc: linux-xtensa@linux-xtensa.org
Cc: linux-arch@vger.kernel.org
Cc: linux-mm@kvack.org

[PATCH v2 01/18] uaccess: fix integer overflow on access_ok()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

Three architectures check the end of a user access against the
address limit without taking a possible overflow into account.
Passing a negative length or another overflow in here returns
success when it should not.

Use the most common correct implementation here, which optimizes
for a constant 'size' argument, and turns the common case into a
single comparison.

Cc: stable@vger.kernel.org
Fixes: da551281947c ("csky: User access")
Fixes: f663b60f5215 ("microblaze: Fix uaccess_ok macro")
Fixes: 7567746e1c0d ("Hexagon: Add user access functions")
Reported-by: David Laight <David.Laight@aculab.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/csky/include/asm/uaccess.h       |  7 +++----
 arch/hexagon/include/asm/uaccess.h    | 18 +++++++++---------
 arch/microblaze/include/asm/uaccess.h | 19 ++++---------------
 3 files changed, 16 insertions(+), 28 deletions(-)

diff --git a/arch/csky/include/asm/uaccess.h b/arch/csky/include/asm/uaccess.h
index c40f06ee8d3e..ac5a54f57d40 100644
--- a/arch/csky/include/asm/uaccess.h
+++ b/arch/csky/include/asm/uaccess.h
@@ -3,14 +3,13 @@
 #ifndef __ASM_CSKY_UACCESS_H
 #define __ASM_CSKY_UACCESS_H
 
-#define user_addr_max() \
-	(uaccess_kernel() ? KERNEL_DS.seg : get_fs().seg)
+#define user_addr_max() (current_thread_info()->addr_limit.seg)
 
 static inline int __access_ok(unsigned long addr, unsigned long size)
 {
-	unsigned long limit = current_thread_info()->addr_limit.seg;
+	unsigned long limit = user_addr_max();
 
-	return ((addr < limit) && ((addr + size) < limit));
+	return (size <= limit) && (addr <= (limit - size));
 }
 #define __access_ok __access_ok
 
diff --git a/arch/hexagon/include/asm/uaccess.h b/arch/hexagon/include/asm/uaccess.h
index ef5bfef8d490..719ba3f3c45c 100644
--- a/arch/hexagon/include/asm/uaccess.h
+++ b/arch/hexagon/include/asm/uaccess.h
@@ -25,17 +25,17 @@
  * Returns true (nonzero) if the memory block *may* be valid, false (zero)
  * if it is definitely invalid.
  *
- * User address space in Hexagon, like x86, goes to 0xbfffffff, so the
- * simple MSB-based tests used by MIPS won't work.  Some further
- * optimization is probably possible here, but for now, keep it
- * reasonably simple and not *too* slow.  After all, we've got the
- * MMU for backup.
  */
+#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
+#define user_addr_max() (uaccess_kernel() ? ~0UL : TASK_SIZE)
 
-#define __access_ok(addr, size) \
-	((get_fs().seg == KERNEL_DS.seg) || \
-	(((unsigned long)addr < get_fs().seg) && \
-	  (unsigned long)size < (get_fs().seg - (unsigned long)addr)))
+static inline int __access_ok(unsigned long addr, unsigned long size)
+{
+	unsigned long limit = TASK_SIZE;
+
+	return (size <= limit) && (addr <= (limit - size));
+}
+#define __access_ok __access_ok
 
 /*
  * When a kernel-mode page fault is taken, the faulting instruction
diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
index d2a8ef9f8978..5b6e0e7788f4 100644
--- a/arch/microblaze/include/asm/uaccess.h
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -39,24 +39,13 @@
 
 # define uaccess_kernel()	(get_fs().seg == KERNEL_DS.seg)
 
-static inline int access_ok(const void __user *addr, unsigned long size)
+static inline int __access_ok(unsigned long addr, unsigned long size)
 {
-	if (!size)
-		goto ok;
+	unsigned long limit = user_addr_max();
 
-	if ((get_fs().seg < ((unsigned long)addr)) ||
-			(get_fs().seg < ((unsigned long)addr + size - 1))) {
-		pr_devel("ACCESS fail at 0x%08x (size 0x%x), seg 0x%08x\n",
-			(__force u32)addr, (u32)size,
-			(u32)get_fs().seg);
-		return 0;
-	}
-ok:
-	pr_devel("ACCESS OK at 0x%08x (size 0x%x), seg 0x%08x\n",
-			(__force u32)addr, (u32)size,
-			(u32)get_fs().seg);
-	return 1;
+	return (size <= limit) && (addr <= (limit - size));
 }
+#define access_ok(addr, size) __access_ok((unsigned long)addr, size)
 
 # define __FIXUP_SECTION	".section .fixup,\"ax\"\n"
 # define __EX_TABLE_SECTION	".section __ex_table,\"a\"\n"
-- 
2.29.2

[PATCH v2 02/18] uaccess: fix nios2 and microblaze get_user_8()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

These two architectures implement 8-byte get_user() through
a memcpy() into a four-byte variable, which won't fit.

Use a temporary 64-bit variable instead here, and use a double
cast the way that risc-v and openrisc do to avoid compile-time
warnings.

Fixes: 6a090e97972d ("arch/microblaze: support get_user() of size 8 bytes")
Fixes: 5ccc6af5e88e ("nios2: Memory management")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/microblaze/include/asm/uaccess.h | 18 +++++++++---------
 arch/nios2/include/asm/uaccess.h      | 26 ++++++++++++++++----------
 2 files changed, 25 insertions(+), 19 deletions(-)

diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
index 5b6e0e7788f4..3fe96979d2c6 100644
--- a/arch/microblaze/include/asm/uaccess.h
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -130,27 +130,27 @@ extern long __user_bad(void);
 
 #define __get_user(x, ptr)						\
 ({									\
-	unsigned long __gu_val = 0;					\
 	long __gu_err;							\
 	switch (sizeof(*(ptr))) {					\
 	case 1:								\
-		__get_user_asm("lbu", (ptr), __gu_val, __gu_err);	\
+		__get_user_asm("lbu", (ptr), x, __gu_err);		\
 		break;							\
 	case 2:								\
-		__get_user_asm("lhu", (ptr), __gu_val, __gu_err);	\
+		__get_user_asm("lhu", (ptr), x, __gu_err);		\
 		break;							\
 	case 4:								\
-		__get_user_asm("lw", (ptr), __gu_val, __gu_err);	\
+		__get_user_asm("lw", (ptr), x, __gu_err);		\
 		break;							\
-	case 8:								\
-		__gu_err = __copy_from_user(&__gu_val, ptr, 8);		\
-		if (__gu_err)						\
-			__gu_err = -EFAULT;				\
+	case 8: {							\
+		__u64 __x = 0;						\
+		__gu_err = raw_copy_from_user(&__x, ptr, 8) ?		\
+							-EFAULT : 0;	\
+		(x) = (typeof(x))(typeof((x) - (x)))__x;		\
 		break;							\
+	}								\
 	default:							\
 		/* __gu_val = 0; __gu_err = -EINVAL;*/ __gu_err = __user_bad();\
 	}								\
-	x = (__force __typeof__(*(ptr))) __gu_val;			\
 	__gu_err;							\
 })
 
diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
index ba9340e96fd4..ca9285a915ef 100644
--- a/arch/nios2/include/asm/uaccess.h
+++ b/arch/nios2/include/asm/uaccess.h
@@ -88,6 +88,7 @@ extern __must_check long strnlen_user(const char __user *s, long n);
 /* Optimized macros */
 #define __get_user_asm(val, insn, addr, err)				\
 {									\
+	unsigned long __gu_val;						\
 	__asm__ __volatile__(						\
 	"       movi    %0, %3\n"					\
 	"1:   " insn " %1, 0(%2)\n"					\
@@ -96,14 +97,20 @@ extern __must_check long strnlen_user(const char __user *s, long n);
 	"       .section __ex_table,\"a\"\n"				\
 	"       .word 1b, 2b\n"						\
 	"       .previous"						\
-	: "=&r" (err), "=r" (val)					\
+	: "=&r" (err), "=r" (__gu_val)					\
 	: "r" (addr), "i" (-EFAULT));					\
+	val = (__force __typeof__(*(addr)))__gu_val;			\
 }
 
-#define __get_user_unknown(val, size, ptr, err) do {			\
+extern void __get_user_unknown(void);
+
+#define __get_user_8(val, ptr, err) do {				\
+	u64 __val = 0;							\
 	err = 0;							\
-	if (__copy_from_user(&(val), ptr, size)) {			\
+	if (raw_copy_from_user(&(__val), ptr, sizeof(val))) {		\
 		err = -EFAULT;						\
+	} else {							\
+		val = (typeof(val))(typeof((val) - (val)))__val;	\
 	}								\
 	} while (0)
 
@@ -119,8 +126,11 @@ do {									\
 	case 4:								\
 		__get_user_asm(val, "ldw", ptr, err);			\
 		break;							\
+	case 8:								\
+		__get_user_8(val, ptr, err);				\
+		break;							\
 	default:							\
-		__get_user_unknown(val, size, ptr, err);		\
+		__get_user_unknown();					\
 		break;							\
 	}								\
 } while (0)
@@ -129,9 +139,7 @@ do {									\
 	({								\
 	long __gu_err = -EFAULT;					\
 	const __typeof__(*(ptr)) __user *__gu_ptr = (ptr);		\
-	unsigned long __gu_val = 0;					\
-	__get_user_common(__gu_val, sizeof(*(ptr)), __gu_ptr, __gu_err);\
-	(x) = (__force __typeof__(x))__gu_val;				\
+	__get_user_common(x, sizeof(*(ptr)), __gu_ptr, __gu_err);	\
 	__gu_err;							\
 	})
 
@@ -139,11 +147,9 @@ do {									\
 ({									\
 	long __gu_err = -EFAULT;					\
 	const __typeof__(*(ptr)) __user *__gu_ptr = (ptr);		\
-	unsigned long __gu_val = 0;					\
 	if (access_ok( __gu_ptr, sizeof(*__gu_ptr)))	\
-		__get_user_common(__gu_val, sizeof(*__gu_ptr),		\
+		__get_user_common(x, sizeof(*__gu_ptr),			\
 			__gu_ptr, __gu_err);				\
-	(x) = (__force __typeof__(x))__gu_val;				\
 	__gu_err;							\
 })
 
-- 
2.29.2

[PATCH v2 03/18] nds32: fix access_ok() checks in get/put_user

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

The get_user()/put_user() functions are meant to check for
access_ok(), while the __get_user()/__put_user() functions
don't.

This broke in 4.19 for nds32, when it gained an extraneous
check in __get_user(), but lost the check it needs in
__put_user().

Fixes: 487913ab18c2 ("nds32: Extract the checking and getting pointer to a macro")
Cc: stable@vger.kernel.org @ v4.19+
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/nds32/include/asm/uaccess.h | 22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/arch/nds32/include/asm/uaccess.h b/arch/nds32/include/asm/uaccess.h
index d4cbf069dc22..37a40981deb3 100644
--- a/arch/nds32/include/asm/uaccess.h
+++ b/arch/nds32/include/asm/uaccess.h
@@ -70,9 +70,7 @@ static inline void set_fs(mm_segment_t fs)
  * versions are void (ie, don't return a value as such).
  */
 
-#define get_user	__get_user					\
-
-#define __get_user(x, ptr)						\
+#define get_user(x, ptr)						\
 ({									\
 	long __gu_err = 0;						\
 	__get_user_check((x), (ptr), __gu_err);				\
@@ -85,6 +83,14 @@ static inline void set_fs(mm_segment_t fs)
 	(void)0;							\
 })
 
+#define __get_user(x, ptr)						\
+({									\
+	long __gu_err = 0;						\
+	const __typeof__(*(ptr)) __user *__p = (ptr);			\
+	__get_user_err((x), __p, (__gu_err));				\
+	__gu_err;							\
+})
+
 #define __get_user_check(x, ptr, err)					\
 ({									\
 	const __typeof__(*(ptr)) __user *__p = (ptr);			\
@@ -165,12 +171,18 @@ do {									\
 		: "r"(addr), "i"(-EFAULT)				\
 		: "cc")
 
-#define put_user	__put_user					\
+#define put_user(x, ptr)						\
+({									\
+	long __pu_err = 0;						\
+	__put_user_check((x), (ptr), __pu_err);				\
+	__pu_err;							\
+})
 
 #define __put_user(x, ptr)						\
 ({									\
 	long __pu_err = 0;						\
-	__put_user_err((x), (ptr), __pu_err);				\
+	__typeof__(*(ptr)) __user *__p = (ptr);				\
+	__put_user_err((x), __p, __pu_err);				\
 	__pu_err;							\
 })
 
-- 
2.29.2

[PATCH v2 04/18] sparc64: add __{get,put}_kernel_nocheck()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

sparc64 is one of the architectures that uses separate address
spaces for kernel and user addresses, so __get_kernel_nofault()
can not just call into the normal __get_user() without the
access_ok() check.

Instead duplicate __get_user() and __put_user() into their
in-kernel versions, with minor changes for the calling conventions
and leaving out the address space modifier on the assembler
instruction.

This could surely be written more elegantly, but duplicating it
gets the job done.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/sparc/include/asm/uaccess_64.h | 78 +++++++++++++++++++++++++++++
 1 file changed, 78 insertions(+)

diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
index 30eb4c6414d1..b283798315b1 100644
--- a/arch/sparc/include/asm/uaccess_64.h
+++ b/arch/sparc/include/asm/uaccess_64.h
@@ -100,6 +100,42 @@ void __retl_efault(void);
 struct __large_struct { unsigned long buf[100]; };
 #define __m(x) ((struct __large_struct *)(x))
 
+#define __put_kernel_nofault(dst, src, type, label)			\
+do {									\
+	type *addr = (type __force *)(dst);				\
+	type data = *(type *)src;					\
+	register int __pu_ret;						\
+	switch (sizeof(type)) {						\
+	case 1: __put_kernel_asm(data, b, addr, __pu_ret); break;	\
+	case 2: __put_kernel_asm(data, h, addr, __pu_ret); break;	\
+	case 4: __put_kernel_asm(data, w, addr, __pu_ret); break;	\
+	case 8: __put_kernel_asm(data, x, addr, __pu_ret); break;	\
+	default: __pu_ret = __put_user_bad(); break;			\
+	}								\
+	if (__pu_ret)							\
+		goto label;						\
+} while (0)
+
+#define __put_kernel_asm(x, size, addr, ret)				\
+__asm__ __volatile__(							\
+		"/* Put kernel asm, inline. */\n"			\
+	"1:\t"	"st"#size " %1, [%2]\n\t"				\
+		"clr	%0\n"						\
+	"2:\n\n\t"							\
+		".section .fixup,#alloc,#execinstr\n\t"			\
+		".align	4\n"						\
+	"3:\n\t"							\
+		"sethi	%%hi(2b), %0\n\t"				\
+		"jmpl	%0 + %%lo(2b), %%g0\n\t"			\
+		" mov	%3, %0\n\n\t"					\
+		".previous\n\t"						\
+		".section __ex_table,\"a\"\n\t"				\
+		".align	4\n\t"						\
+		".word	1b, 3b\n\t"					\
+		".previous\n\n\t"					\
+	       : "=r" (ret) : "r" (x), "r" (__m(addr)),			\
+		 "i" (-EFAULT))
+
 #define __put_user_nocheck(data, addr, size) ({			\
 	register int __pu_ret;					\
 	switch (size) {						\
@@ -134,6 +170,48 @@ __asm__ __volatile__(							\
 
 int __put_user_bad(void);
 
+#define __get_kernel_nofault(dst, src, type, label)			     \
+do {									     \
+	type *addr = (type __force *)(src);		     		     \
+	register int __gu_ret;						     \
+	register unsigned long __gu_val;				     \
+	switch (sizeof(type)) {						     \
+		case 1: __get_kernel_asm(__gu_val, ub, addr, __gu_ret); break; \
+		case 2: __get_kernel_asm(__gu_val, uh, addr, __gu_ret); break; \
+		case 4: __get_kernel_asm(__gu_val, uw, addr, __gu_ret); break; \
+		case 8: __get_kernel_asm(__gu_val, x, addr, __gu_ret); break;  \
+		default:						     \
+			__gu_val = 0;					     \
+			__gu_ret = __get_user_bad();			     \
+			break;						     \
+	} 								     \
+	if (__gu_ret)							     \
+		goto label;						     \
+	*(type *)dst = (__force type) __gu_val;				     \
+} while (0)
+#define __get_kernel_asm(x, size, addr, ret)				\
+__asm__ __volatile__(							\
+		"/* Get kernel asm, inline. */\n"			\
+	"1:\t"	"ld"#size " [%2], %1\n\t"				\
+		"clr	%0\n"						\
+	"2:\n\n\t"							\
+		".section .fixup,#alloc,#execinstr\n\t"			\
+		".align	4\n"						\
+	"3:\n\t"							\
+		"sethi	%%hi(2b), %0\n\t"				\
+		"clr	%1\n\t"						\
+		"jmpl	%0 + %%lo(2b), %%g0\n\t"			\
+		" mov	%3, %0\n\n\t"					\
+		".previous\n\t"						\
+		".section __ex_table,\"a\"\n\t"				\
+		".align	4\n\t"						\
+		".word	1b, 3b\n\n\t"					\
+		".previous\n\t"						\
+	       : "=r" (ret), "=r" (x) : "r" (__m(addr)),		\
+		 "i" (-EFAULT))
+
+#define HAVE_GET_KERNEL_NOFAULT
+
 #define __get_user_nocheck(data, addr, size, type) ({			     \
 	register int __gu_ret;						     \
 	register unsigned long __gu_val;				     \
-- 
2.29.2

[PATCH v2 05/18] x86: remove __range_not_ok()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

The __range_not_ok() helper is an x86 (and sparc64) specific interface
that does roughly the same thing as __access_ok(), but with different
calling conventions.

Change this to use the normal interface in order for consistency as we
clean up all access_ok() implementations.

This changes the limit from TASK_SIZE to TASK_SIZE_MAX, which Al points
out is the right thing do do here anyway.

The callers have to use __access_ok() instead of the normal access_ok()
though, because on x86 that contains a WARN_ON_IN_IRQ() check that cannot
be used inside of NMI context while tracing.

Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Suggested-by: Christoph Hellwig <hch@infradead.org>
Link: https://lore.kernel.org/lkml/YgsUKcXGR7r4nINj@zeniv-ca.linux.org.uk/
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/x86/events/core.c         |  2 +-
 arch/x86/include/asm/uaccess.h | 10 ++++++----
 arch/x86/kernel/dumpstack.c    |  2 +-
 arch/x86/kernel/stacktrace.c   |  2 +-
 arch/x86/lib/usercopy.c        |  2 +-
 5 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index e686c5e0537b..eef816fc216d 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2794,7 +2794,7 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re
 static inline int
 valid_user_frame(const void __user *fp, unsigned long size)
 {
-	return (__range_not_ok(fp, size, TASK_SIZE) == 0);
+	return __access_ok(fp, size);
 }
 
 static unsigned long get_segment_base(unsigned int segment)
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index ac96f9b2d64b..79c4869ccdd6 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -16,8 +16,10 @@
  * Test whether a block of memory is a valid user space address.
  * Returns 0 if the range is valid, nonzero otherwise.
  */
-static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size, unsigned long limit)
+static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size)
 {
+	unsigned long limit = TASK_SIZE_MAX;
+
 	/*
 	 * If we have used "sizeof()" for the size,
 	 * we know it won't overflow the limit (but
@@ -35,10 +37,10 @@ static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size, un
 	return unlikely(addr > limit);
 }
 
-#define __range_not_ok(addr, size, limit)				\
+#define __access_ok(addr, size)						\
 ({									\
 	__chk_user_ptr(addr);						\
-	__chk_range_not_ok((unsigned long __force)(addr), size, limit); \
+	!__chk_range_not_ok((unsigned long __force)(addr), size);	\
 })
 
 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
@@ -69,7 +71,7 @@ static inline bool pagefault_disabled(void);
 #define access_ok(addr, size)					\
 ({									\
 	WARN_ON_IN_IRQ();						\
-	likely(!__range_not_ok(addr, size, TASK_SIZE_MAX));		\
+	likely(__access_ok(addr, size));				\
 })
 
 extern int __get_user_1(void);
diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
index 53de044e5654..da534fb7b5c6 100644
--- a/arch/x86/kernel/dumpstack.c
+++ b/arch/x86/kernel/dumpstack.c
@@ -85,7 +85,7 @@ static int copy_code(struct pt_regs *regs, u8 *buf, unsigned long src,
 	 * Make sure userspace isn't trying to trick us into dumping kernel
 	 * memory by pointing the userspace instruction pointer at it.
 	 */
-	if (__chk_range_not_ok(src, nbytes, TASK_SIZE_MAX))
+	if (!__access_ok((void __user *)src, nbytes))
 		return -EINVAL;
 
 	/*
diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
index 15b058eefc4e..ee117fcf46ed 100644
--- a/arch/x86/kernel/stacktrace.c
+++ b/arch/x86/kernel/stacktrace.c
@@ -90,7 +90,7 @@ copy_stack_frame(const struct stack_frame_user __user *fp,
 {
 	int ret;
 
-	if (__range_not_ok(fp, sizeof(*frame), TASK_SIZE))
+	if (!__access_ok(fp, sizeof(*frame)))
 		return 0;
 
 	ret = 1;
diff --git a/arch/x86/lib/usercopy.c b/arch/x86/lib/usercopy.c
index c3e8a62ca561..ad0139d25401 100644
--- a/arch/x86/lib/usercopy.c
+++ b/arch/x86/lib/usercopy.c
@@ -32,7 +32,7 @@ copy_from_user_nmi(void *to, const void __user *from, unsigned long n)
 {
 	unsigned long ret;
 
-	if (__range_not_ok(from, n, TASK_SIZE))
+	if (!__access_ok(from, n))
 		return n;
 
 	if (!nmi_uaccess_okay())
-- 
2.29.2

[PATCH v2 06/18] x86: use more conventional access_ok() definition

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

The way that access_ok() is defined on x86 is slightly different from
most other architectures, and a bit more complex.

The generic version tends to result in the best output on all
architectures, as it results in single comparison against a constant
limit for calls with a known size.

There are a few callers of __range_not_ok(), all of which use TASK_SIZE
as the limit rather than TASK_SIZE_MAX, but I could not see any reason
for picking this. Changing these to call __access_ok() instead uses the
default limit, but keeps the behavior otherwise.

x86 is the only architecture with a WARN_ON_IN_IRQ() checking
access_ok(), but it's probably best to leave that in place.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/x86/include/asm/uaccess.h | 25 +++----------------------
 1 file changed, 3 insertions(+), 22 deletions(-)

diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index 79c4869ccdd6..a59ba2578e64 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -16,33 +16,14 @@
  * Test whether a block of memory is a valid user space address.
  * Returns 0 if the range is valid, nonzero otherwise.
  */
-static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size)
+static inline bool __access_ok(void __user *ptr, unsigned long size)
 {
 	unsigned long limit = TASK_SIZE_MAX;
+	unsigned long addr = ptr;
 
-	/*
-	 * If we have used "sizeof()" for the size,
-	 * we know it won't overflow the limit (but
-	 * it might overflow the 'addr', so it's
-	 * important to subtract the size from the
-	 * limit, not add it to the address).
-	 */
-	if (__builtin_constant_p(size))
-		return unlikely(addr > limit - size);
-
-	/* Arbitrary sizes? Be careful about overflow */
-	addr += size;
-	if (unlikely(addr < size))
-		return true;
-	return unlikely(addr > limit);
+	return (size <= limit) && (addr <= (limit - size));
 }
 
-#define __access_ok(addr, size)						\
-({									\
-	__chk_user_ptr(addr);						\
-	!__chk_range_not_ok((unsigned long __force)(addr), size);	\
-})
-
 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
 static inline bool pagefault_disabled(void);
 # define WARN_ON_IN_IRQ()	\
-- 
2.29.2

[PATCH v2 07/18] nios2: drop access_ok() check from __put_user()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

Unlike other architectures, the nios2 version of __put_user() has an
extra check for access_ok(), preventing it from being used to implement
__put_kernel_nofault().

Split up put_user() along the same lines as __get_user()/get_user()

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/nios2/include/asm/uaccess.h | 56 +++++++++++++++++++-------------
 1 file changed, 33 insertions(+), 23 deletions(-)

diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
index ca9285a915ef..a5cbe07cf0da 100644
--- a/arch/nios2/include/asm/uaccess.h
+++ b/arch/nios2/include/asm/uaccess.h
@@ -167,34 +167,44 @@ do {									\
 	: "r" (val), "r" (ptr), "i" (-EFAULT));				\
 }
 
-#define put_user(x, ptr)						\
+#define __put_user_common(__pu_val, __pu_ptr)				\
 ({									\
 	long __pu_err = -EFAULT;					\
-	__typeof__(*(ptr)) __user *__pu_ptr = (ptr);			\
-	__typeof__(*(ptr)) __pu_val = (__typeof(*ptr))(x);		\
-	if (access_ok(__pu_ptr, sizeof(*__pu_ptr))) {	\
-		switch (sizeof(*__pu_ptr)) {				\
-		case 1:							\
-			__put_user_asm(__pu_val, "stb", __pu_ptr, __pu_err); \
-			break;						\
-		case 2:							\
-			__put_user_asm(__pu_val, "sth", __pu_ptr, __pu_err); \
-			break;						\
-		case 4:							\
-			__put_user_asm(__pu_val, "stw", __pu_ptr, __pu_err); \
-			break;						\
-		default:						\
-			/* XXX: This looks wrong... */			\
-			__pu_err = 0;					\
-			if (copy_to_user(__pu_ptr, &(__pu_val),		\
-				sizeof(*__pu_ptr)))			\
-				__pu_err = -EFAULT;			\
-			break;						\
-		}							\
+	switch (sizeof(*__pu_ptr)) {					\
+	case 1:								\
+		__put_user_asm(__pu_val, "stb", __pu_ptr, __pu_err);	\
+		break;							\
+	case 2:								\
+		__put_user_asm(__pu_val, "sth", __pu_ptr, __pu_err);	\
+		break;							\
+	case 4:								\
+		__put_user_asm(__pu_val, "stw", __pu_ptr, __pu_err);	\
+		break;							\
+	default:							\
+		/* XXX: This looks wrong... */				\
+		__pu_err = 0;						\
+		if (__copy_to_user(__pu_ptr, &(__pu_val),		\
+			sizeof(*__pu_ptr)))				\
+			__pu_err = -EFAULT;				\
+		break;							\
 	}								\
 	__pu_err;							\
 })
 
-#define __put_user(x, ptr) put_user(x, ptr)
+#define __put_user(x, ptr)						\
+({									\
+	__auto_type __pu_ptr = (ptr);					\
+	typeof(*__pu_ptr) __pu_val = (typeof(*__pu_ptr))(x);		\
+	__put_user_common(__pu_val, __pu_ptr);				\
+})
+
+#define put_user(x, ptr)						\
+({									\
+	__auto_type __pu_ptr = (ptr);					\
+	typeof(*__pu_ptr) __pu_val = (typeof(*__pu_ptr))(x);		\
+	access_ok(__pu_ptr, sizeof(*__pu_ptr)) ?			\
+		__put_user_common(__pu_val, __pu_ptr) :			\
+		-EFAULT;						\
+})
 
 #endif /* _ASM_NIOS2_UACCESS_H */
-- 
2.29.2

[PATCH v2 08/18] uaccess: add generic __{get,put}_kernel_nofault

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

Nine architectures are still missing __{get,put}_kernel_nofault:
alpha, ia64, microblaze, nds32, nios2, openrisc, sh, sparc32, xtensa.

Add a generic version that lets everything use the normal
copy_{from,to}_kernel_nofault() code based on these, removing the last
use of get_fs()/set_fs() from architecture-independent code.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/arm/include/asm/uaccess.h      |   2 -
 arch/arm64/include/asm/uaccess.h    |   2 -
 arch/m68k/include/asm/uaccess.h     |   2 -
 arch/mips/include/asm/uaccess.h     |   2 -
 arch/parisc/include/asm/uaccess.h   |   1 -
 arch/powerpc/include/asm/uaccess.h  |   2 -
 arch/riscv/include/asm/uaccess.h    |   2 -
 arch/s390/include/asm/uaccess.h     |   2 -
 arch/sparc/include/asm/uaccess_64.h |   2 -
 arch/um/include/asm/uaccess.h       |   2 -
 arch/x86/include/asm/uaccess.h      |   2 -
 include/asm-generic/uaccess.h       |   2 -
 include/linux/uaccess.h             |  19 +++++
 mm/maccess.c                        | 108 ----------------------------
 14 files changed, 19 insertions(+), 131 deletions(-)

diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h
index 32dbfd81f42a..d20d78c34b94 100644
--- a/arch/arm/include/asm/uaccess.h
+++ b/arch/arm/include/asm/uaccess.h
@@ -476,8 +476,6 @@ do {									\
 	: "r" (x), "i" (-EFAULT)				\
 	: "cc")
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 do {									\
 	const type *__pk_ptr = (src);					\
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 3a5ff5e20586..2e20879fe3cf 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -26,8 +26,6 @@
 #include <asm/memory.h>
 #include <asm/extable.h>
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 /*
  * Test whether a block of memory is a valid user space address.
  * Returns 1 if the range is valid, 0 otherwise.
diff --git a/arch/m68k/include/asm/uaccess.h b/arch/m68k/include/asm/uaccess.h
index ba670523885c..79617c0b2f91 100644
--- a/arch/m68k/include/asm/uaccess.h
+++ b/arch/m68k/include/asm/uaccess.h
@@ -390,8 +390,6 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 #define INLINE_COPY_FROM_USER
 #define INLINE_COPY_TO_USER
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 do {									\
 	type *__gk_dst = (type *)(dst);					\
diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
index f8f74f9f5883..db9a8e002b62 100644
--- a/arch/mips/include/asm/uaccess.h
+++ b/arch/mips/include/asm/uaccess.h
@@ -296,8 +296,6 @@ struct __large_struct { unsigned long buf[100]; };
 	(val) = __gu_tmp.t;						\
 }
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 do {									\
 	int __gu_err;							\
diff --git a/arch/parisc/include/asm/uaccess.h b/arch/parisc/include/asm/uaccess.h
index ebf8a845b017..0925bbd6db67 100644
--- a/arch/parisc/include/asm/uaccess.h
+++ b/arch/parisc/include/asm/uaccess.h
@@ -95,7 +95,6 @@ struct exception_table_entry {
 	(val) = (__force __typeof__(*(ptr))) __gu_val;	\
 }
 
-#define HAVE_GET_KERNEL_NOFAULT
 #define __get_kernel_nofault(dst, src, type, err_label)	\
 {							\
 	type __z;					\
diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h
index 63316100080c..a0032c2e7550 100644
--- a/arch/powerpc/include/asm/uaccess.h
+++ b/arch/powerpc/include/asm/uaccess.h
@@ -467,8 +467,6 @@ do {									\
 		unsafe_put_user(*(u8*)(_src + _i), (u8 __user *)(_dst + _i), e); \
 } while (0)
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 	__get_user_size_goto(*((type *)(dst)),				\
 		(__force type __user *)(src), sizeof(type), err_label)
diff --git a/arch/riscv/include/asm/uaccess.h b/arch/riscv/include/asm/uaccess.h
index c701a5e57a2b..4407b9e48d2c 100644
--- a/arch/riscv/include/asm/uaccess.h
+++ b/arch/riscv/include/asm/uaccess.h
@@ -346,8 +346,6 @@ unsigned long __must_check clear_user(void __user *to, unsigned long n)
 		__clear_user(to, n) : n;
 }
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 do {									\
 	long __kr_err;							\
diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h
index d74e26b48604..29332edf46f0 100644
--- a/arch/s390/include/asm/uaccess.h
+++ b/arch/s390/include/asm/uaccess.h
@@ -282,8 +282,6 @@ static inline unsigned long __must_check clear_user(void __user *to, unsigned lo
 int copy_to_user_real(void __user *dest, void *src, unsigned long count);
 void *s390_kernel_write(void *dst, const void *src, size_t size);
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 int __noreturn __put_kernel_bad(void);
 
 #define __put_kernel_asm(val, to, insn)					\
diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
index b283798315b1..5c12fb46bc61 100644
--- a/arch/sparc/include/asm/uaccess_64.h
+++ b/arch/sparc/include/asm/uaccess_64.h
@@ -210,8 +210,6 @@ __asm__ __volatile__(							\
 	       : "=r" (ret), "=r" (x) : "r" (__m(addr)),		\
 		 "i" (-EFAULT))
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #define __get_user_nocheck(data, addr, size, type) ({			     \
 	register int __gu_ret;						     \
 	register unsigned long __gu_val;				     \
diff --git a/arch/um/include/asm/uaccess.h b/arch/um/include/asm/uaccess.h
index 17d18cfd82a5..1ecfc96bcc50 100644
--- a/arch/um/include/asm/uaccess.h
+++ b/arch/um/include/asm/uaccess.h
@@ -44,8 +44,6 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 }
 
 /* no pagefaults for kernel addresses in um */
-#define HAVE_GET_KERNEL_NOFAULT 1
-
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 do {									\
 	*((type *)dst) = get_unaligned((type *)(src));			\
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index a59ba2578e64..201efcec66b7 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -507,8 +507,6 @@ do {									\
 	unsafe_copy_loop(__ucu_dst, __ucu_src, __ucu_len, u8, label);	\
 } while (0)
 
-#define HAVE_GET_KERNEL_NOFAULT
-
 #ifdef CONFIG_CC_HAS_ASM_GOTO_OUTPUT
 #define __get_kernel_nofault(dst, src, type, err_label)			\
 	__get_user_size(*((type *)(dst)), (__force type __user *)(src),	\
diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
index 10ffa8b5c117..0870fa11a7c5 100644
--- a/include/asm-generic/uaccess.h
+++ b/include/asm-generic/uaccess.h
@@ -77,8 +77,6 @@ do {									\
 		goto err_label;						\
 } while (0)
 
-#define HAVE_GET_KERNEL_NOFAULT 1
-
 static inline __must_check unsigned long
 raw_copy_from_user(void *to, const void __user * from, unsigned long n)
 {
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index ac0394087f7d..67e9bc94dc40 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -368,6 +368,25 @@ long strncpy_from_user_nofault(char *dst, const void __user *unsafe_addr,
 		long count);
 long strnlen_user_nofault(const void __user *unsafe_addr, long count);
 
+#ifndef __get_kernel_nofault
+#define __get_kernel_nofault(dst, src, type, label)	\
+do {							\
+	type __user *p = (type __force __user *)(src);	\
+	type data;					\
+	if (__get_user(data, p))			\
+		goto label;				\
+	*(type *)dst = data;				\
+} while (0)
+
+#define __put_kernel_nofault(dst, src, type, label)	\
+do {							\
+	type __user *p = (type __force __user *)(dst);	\
+	type data = *(type *)src;			\
+	if (__put_user(data, p))			\
+		goto label;				\
+} while (0)
+#endif
+
 /**
  * get_kernel_nofault(): safely attempt to read from a location
  * @val: read into this variable
diff --git a/mm/maccess.c b/mm/maccess.c
index d3f1a1f0b1c1..cbd1b3959af2 100644
--- a/mm/maccess.c
+++ b/mm/maccess.c
@@ -12,8 +12,6 @@ bool __weak copy_from_kernel_nofault_allowed(const void *unsafe_src,
 	return true;
 }
 
-#ifdef HAVE_GET_KERNEL_NOFAULT
-
 #define copy_from_kernel_nofault_loop(dst, src, len, type, err_label)	\
 	while (len >= sizeof(type)) {					\
 		__get_kernel_nofault(dst, src, type, err_label);		\
@@ -102,112 +100,6 @@ long strncpy_from_kernel_nofault(char *dst, const void *unsafe_addr, long count)
 	dst[-1] = '\0';
 	return -EFAULT;
 }
-#else /* HAVE_GET_KERNEL_NOFAULT */
-/**
- * copy_from_kernel_nofault(): safely attempt to read from kernel-space
- * @dst: pointer to the buffer that shall take the data
- * @src: address to read from
- * @size: size of the data chunk
- *
- * Safely read from kernel address @src to the buffer at @dst.  If a kernel
- * fault happens, handle that and return -EFAULT.  If @src is not a valid kernel
- * address, return -ERANGE.
- *
- * We ensure that the copy_from_user is executed in atomic context so that
- * do_page_fault() doesn't attempt to take mmap_lock.  This makes
- * copy_from_kernel_nofault() suitable for use within regions where the caller
- * already holds mmap_lock, or other locks which nest inside mmap_lock.
- */
-long copy_from_kernel_nofault(void *dst, const void *src, size_t size)
-{
-	long ret;
-	mm_segment_t old_fs = get_fs();
-
-	if (!copy_from_kernel_nofault_allowed(src, size))
-		return -ERANGE;
-
-	set_fs(KERNEL_DS);
-	pagefault_disable();
-	ret = __copy_from_user_inatomic(dst, (__force const void __user *)src,
-			size);
-	pagefault_enable();
-	set_fs(old_fs);
-
-	if (ret)
-		return -EFAULT;
-	return 0;
-}
-EXPORT_SYMBOL_GPL(copy_from_kernel_nofault);
-
-/**
- * copy_to_kernel_nofault(): safely attempt to write to a location
- * @dst: address to write to
- * @src: pointer to the data that shall be written
- * @size: size of the data chunk
- *
- * Safely write to address @dst from the buffer at @src.  If a kernel fault
- * happens, handle that and return -EFAULT.
- */
-long copy_to_kernel_nofault(void *dst, const void *src, size_t size)
-{
-	long ret;
-	mm_segment_t old_fs = get_fs();
-
-	set_fs(KERNEL_DS);
-	pagefault_disable();
-	ret = __copy_to_user_inatomic((__force void __user *)dst, src, size);
-	pagefault_enable();
-	set_fs(old_fs);
-
-	if (ret)
-		return -EFAULT;
-	return 0;
-}
-
-/**
- * strncpy_from_kernel_nofault: - Copy a NUL terminated string from unsafe
- *				 address.
- * @dst:   Destination address, in kernel space.  This buffer must be at
- *         least @count bytes long.
- * @unsafe_addr: Unsafe address.
- * @count: Maximum number of bytes to copy, including the trailing NUL.
- *
- * Copies a NUL-terminated string from unsafe address to kernel buffer.
- *
- * On success, returns the length of the string INCLUDING the trailing NUL.
- *
- * If access fails, returns -EFAULT (some data may have been copied and the
- * trailing NUL added).  If @unsafe_addr is not a valid kernel address, return
- * -ERANGE.
- *
- * If @count is smaller than the length of the string, copies @count-1 bytes,
- * sets the last byte of @dst buffer to NUL and returns @count.
- */
-long strncpy_from_kernel_nofault(char *dst, const void *unsafe_addr, long count)
-{
-	mm_segment_t old_fs = get_fs();
-	const void *src = unsafe_addr;
-	long ret;
-
-	if (unlikely(count <= 0))
-		return 0;
-	if (!copy_from_kernel_nofault_allowed(unsafe_addr, count))
-		return -ERANGE;
-
-	set_fs(KERNEL_DS);
-	pagefault_disable();
-
-	do {
-		ret = __get_user(*dst++, (const char __user __force *)src++);
-	} while (dst[-1] && ret == 0 && src - unsafe_addr < count);
-
-	dst[-1] = '\0';
-	pagefault_enable();
-	set_fs(old_fs);
-
-	return ret ? -EFAULT : src - unsafe_addr;
-}
-#endif /* HAVE_GET_KERNEL_NOFAULT */
 
 /**
  * copy_from_user_nofault(): safely attempt to read from a user-space location
-- 
2.29.2

[PATCH v2 09/18] mips: use simpler access_ok()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

Before unifying the mips version of __access_ok() with the generic
code, this converts it to the same algorithm. This is a change in
behavior on mips64, as now address in the user segment, the lower
2^62 bytes, is taken to be valid, relying on a page fault for
addresses that are within that segment but not valid on that CPU.

The new version should be the most effecient way to do this, but
it gets rid of the special handling for size=0 that most other
architectures ignore as well.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/mips/include/asm/uaccess.h | 22 ++++------------------
 1 file changed, 4 insertions(+), 18 deletions(-)

diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
index db9a8e002b62..d7c89dc3426c 100644
--- a/arch/mips/include/asm/uaccess.h
+++ b/arch/mips/include/asm/uaccess.h
@@ -19,6 +19,7 @@
 #ifdef CONFIG_32BIT
 
 #define __UA_LIMIT 0x80000000UL
+#define TASK_SIZE_MAX	__UA_LIMIT
 
 #define __UA_ADDR	".word"
 #define __UA_LA		"la"
@@ -33,6 +34,7 @@
 extern u64 __ua_limit;
 
 #define __UA_LIMIT	__ua_limit
+#define TASK_SIZE_MAX	XKSSEG
 
 #define __UA_ADDR	".dword"
 #define __UA_LA		"dla"
@@ -42,22 +44,6 @@ extern u64 __ua_limit;
 
 #endif /* CONFIG_64BIT */
 
-/*
- * Is a address valid? This does a straightforward calculation rather
- * than tests.
- *
- * Address valid if:
- *  - "addr" doesn't have any high-bits set
- *  - AND "size" doesn't have any high-bits set
- *  - AND "addr+size" doesn't have any high-bits set
- *  - OR we are in kernel mode.
- *
- * __ua_size() is a trick to avoid runtime checking of positive constant
- * sizes; for those we already know at compile time that the size is ok.
- */
-#define __ua_size(size)							\
-	((__builtin_constant_p(size) && (signed long) (size) > 0) ? 0 : (size))
-
 /*
  * access_ok: - Checks if a user space pointer is valid
  * @addr: User space pointer to start of block to check
@@ -79,9 +65,9 @@ extern u64 __ua_limit;
 static inline int __access_ok(const void __user *p, unsigned long size)
 {
 	unsigned long addr = (unsigned long)p;
-	unsigned long end = addr + size - !!size;
+	unsigned long limit = TASK_SIZE_MAX;
 
-	return (__UA_LIMIT & (addr | end | __ua_size(size))) == 0;
+	return (size <= limit) && (addr <= (limit - size));
 }
 
 #define access_ok(addr, size)					\
-- 
2.29.2

[PATCH v2 10/18] m68k: fix access_ok for coldfire

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

While most m68k platforms use separate address spaces for user
and kernel space, at least coldfire does not, and the other
ones have a TASK_SIZE that is less than the entire 4GB address
range.

Using the default implementation of __access_ok() stops coldfire
user space from trivially accessing kernel memory.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/m68k/include/asm/uaccess.h | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/arch/m68k/include/asm/uaccess.h b/arch/m68k/include/asm/uaccess.h
index 79617c0b2f91..8eb625e75452 100644
--- a/arch/m68k/include/asm/uaccess.h
+++ b/arch/m68k/include/asm/uaccess.h
@@ -12,14 +12,21 @@
 #include <asm/extable.h>
 
 /* We let the MMU do all checking */
-static inline int access_ok(const void __user *addr,
+static inline int access_ok(const void __user *ptr,
 			    unsigned long size)
 {
+	unsigned long limit = TASK_SIZE;
+	unsigned long addr = (unsigned long)ptr;
+
 	/*
 	 * XXX: for !CONFIG_CPU_HAS_ADDRESS_SPACES this really needs to check
 	 * for TASK_SIZE!
+	 * Removing this helper is probably sufficient.
 	 */
-	return 1;
+	if (IS_ENABLED(CONFIG_CPU_HAS_ADDRESS_SPACES))
+		return 1;
+
+	return (size <= limit) && (addr <= (limit - size));
 }
 
 /*
-- 
2.29.2

[PATCH v2 11/18] arm64: simplify access_ok()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

arm64 has an inline asm implementation of access_ok() that is derived from
the 32-bit arm version and optimized for the case that both the limit and
the size are variable. With set_fs() gone, the limit is always constant,
and the size usually is as well, so just using the default implementation
reduces the check into a comparison against a constant that can be
scheduled by the compiler.

On a defconfig build, this saves over 28KB of .text.

Acked-by: Robin Murphy <robin.murphy@arm.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/arm64/include/asm/uaccess.h | 34 ++++++++++----------------------
 1 file changed, 10 insertions(+), 24 deletions(-)

diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 2e20879fe3cf..199c553b740a 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -26,6 +26,14 @@
 #include <asm/memory.h>
 #include <asm/extable.h>
 
+static inline int __access_ok(const void __user *ptr, unsigned long size)
+{
+	unsigned long limit = TASK_SIZE_MAX;
+	unsigned long addr = (unsigned long)ptr;
+
+	return (size <= limit) && (addr <= (limit - size));
+}
+
 /*
  * Test whether a block of memory is a valid user space address.
  * Returns 1 if the range is valid, 0 otherwise.
@@ -33,10 +41,8 @@
  * This is equivalent to the following test:
  * (u65)addr + (u65)size <= (u65)TASK_SIZE_MAX
  */
-static inline unsigned long __range_ok(const void __user *addr, unsigned long size)
+static inline int access_ok(const void __user *addr, unsigned long size)
 {
-	unsigned long ret, limit = TASK_SIZE_MAX - 1;
-
 	/*
 	 * Asynchronous I/O running in a kernel thread does not have the
 	 * TIF_TAGGED_ADDR flag of the process owning the mm, so always untag
@@ -46,29 +52,9 @@ static inline unsigned long __range_ok(const void __user *addr, unsigned long si
 	    (current->flags & PF_KTHREAD || test_thread_flag(TIF_TAGGED_ADDR)))
 		addr = untagged_addr(addr);
 
-	__chk_user_ptr(addr);
-	asm volatile(
-	// A + B <= C + 1 for all A,B,C, in four easy steps:
-	// 1: X = A + B; X' = X % 2^64
-	"	adds	%0, %3, %2\n"
-	// 2: Set C = 0 if X > 2^64, to guarantee X' > C in step 4
-	"	csel	%1, xzr, %1, hi\n"
-	// 3: Set X' = ~0 if X >= 2^64. For X == 2^64, this decrements X'
-	//    to compensate for the carry flag being set in step 4. For
-	//    X > 2^64, X' merely has to remain nonzero, which it does.
-	"	csinv	%0, %0, xzr, cc\n"
-	// 4: For X < 2^64, this gives us X' - C - 1 <= 0, where the -1
-	//    comes from the carry in being clear. Otherwise, we are
-	//    testing X' - C == 0, subject to the previous adjustments.
-	"	sbcs	xzr, %0, %1\n"
-	"	cset	%0, ls\n"
-	: "=&r" (ret), "+r" (limit) : "Ir" (size), "0" (addr) : "cc");
-
-	return ret;
+	return likely(__access_ok(addr, size));
 }
 
-#define access_ok(addr, size)	__range_ok(addr, size)
-
 /*
  * User access enabling/disabling.
  */
-- 
2.29.2

[PATCH v2 12/18] uaccess: fix type mismatch warnings from access_ok()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

On some architectures, access_ok() does not do any argument type
checking, so replacing the definition with a generic one causes
a few warnings for harmless issues that were never caught before.

Fix the ones that I found either through my own test builds or
that were reported by the 0-day bot.

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/arc/kernel/process.c           |  2 +-
 arch/arm/kernel/swp_emulate.c       |  2 +-
 arch/arm/kernel/traps.c             |  2 +-
 arch/csky/kernel/signal.c           |  2 +-
 arch/mips/sibyte/common/sb_tbprof.c |  6 +++---
 arch/nios2/kernel/signal.c          | 20 +++++++++++---------
 arch/powerpc/lib/sstep.c            |  4 ++--
 arch/riscv/kernel/perf_callchain.c  |  2 +-
 arch/sparc/kernel/signal_32.c       |  2 +-
 lib/test_lockup.c                   |  4 ++--
 10 files changed, 24 insertions(+), 22 deletions(-)

diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c
index 8e90052f6f05..5f7f5aab361f 100644
--- a/arch/arc/kernel/process.c
+++ b/arch/arc/kernel/process.c
@@ -43,7 +43,7 @@ SYSCALL_DEFINE0(arc_gettls)
 	return task_thread_info(current)->thr_ptr;
 }
 
-SYSCALL_DEFINE3(arc_usr_cmpxchg, int *, uaddr, int, expected, int, new)
+SYSCALL_DEFINE3(arc_usr_cmpxchg, int __user *, uaddr, int, expected, int, new)
 {
 	struct pt_regs *regs = current_pt_regs();
 	u32 uval;
diff --git a/arch/arm/kernel/swp_emulate.c b/arch/arm/kernel/swp_emulate.c
index 6166ba38bf99..b74bfcf94fb1 100644
--- a/arch/arm/kernel/swp_emulate.c
+++ b/arch/arm/kernel/swp_emulate.c
@@ -195,7 +195,7 @@ static int swp_handler(struct pt_regs *regs, unsigned int instr)
 		 destreg, EXTRACT_REG_NUM(instr, RT2_OFFSET), data);
 
 	/* Check access in reasonable access range for both SWP and SWPB */
-	if (!access_ok((address & ~3), 4)) {
+	if (!access_ok((void __user *)(address & ~3), 4)) {
 		pr_debug("SWP{B} emulation: access to %p not allowed!\n",
 			 (void *)address);
 		res = -EFAULT;
diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index da04ed85855a..26c8c8276297 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -576,7 +576,7 @@ do_cache_op(unsigned long start, unsigned long end, int flags)
 	if (end < start || flags)
 		return -EINVAL;
 
-	if (!access_ok(start, end - start))
+	if (!access_ok((void __user *)start, end - start))
 		return -EFAULT;
 
 	return __do_cache_op(start, end);
diff --git a/arch/csky/kernel/signal.c b/arch/csky/kernel/signal.c
index c7b763d2f526..8867ddf3e6c7 100644
--- a/arch/csky/kernel/signal.c
+++ b/arch/csky/kernel/signal.c
@@ -136,7 +136,7 @@ static inline void __user *get_sigframe(struct ksignal *ksig,
 static int
 setup_rt_frame(struct ksignal *ksig, sigset_t *set, struct pt_regs *regs)
 {
-	struct rt_sigframe *frame;
+	struct rt_sigframe __user *frame;
 	int err = 0;
 
 	frame = get_sigframe(ksig, regs, sizeof(*frame));
diff --git a/arch/mips/sibyte/common/sb_tbprof.c b/arch/mips/sibyte/common/sb_tbprof.c
index f80d7a710333..01d00b87d0f5 100644
--- a/arch/mips/sibyte/common/sb_tbprof.c
+++ b/arch/mips/sibyte/common/sb_tbprof.c
@@ -437,13 +437,13 @@ static int sbprof_tb_release(struct inode *inode, struct file *filp)
 	return 0;
 }
 
-static ssize_t sbprof_tb_read(struct file *filp, char *buf,
+static ssize_t sbprof_tb_read(struct file *filp, char __user *buf,
 			      size_t size, loff_t *offp)
 {
 	int cur_sample, sample_off, cur_count, sample_left;
 	char *src;
 	int   count   =	 0;
-	char *dest    =	 buf;
+	char __user *dest = buf;
 	long  cur_off = *offp;
 
 	if (!access_ok(buf, size))
@@ -512,7 +512,7 @@ static long sbprof_tb_ioctl(struct file *filp,
 		if (err)
 			break;
 
-		err = put_user(TB_FULL, (int *) arg);
+		err = put_user(TB_FULL, (int __user *) arg);
 		break;
 	}
 
diff --git a/arch/nios2/kernel/signal.c b/arch/nios2/kernel/signal.c
index 2009ae2d3c3b..386e46443b60 100644
--- a/arch/nios2/kernel/signal.c
+++ b/arch/nios2/kernel/signal.c
@@ -36,10 +36,10 @@ struct rt_sigframe {
 
 static inline int rt_restore_ucontext(struct pt_regs *regs,
 					struct switch_stack *sw,
-					struct ucontext *uc, int *pr2)
+					struct ucontext __user *uc, int *pr2)
 {
 	int temp;
-	unsigned long *gregs = uc->uc_mcontext.gregs;
+	unsigned long __user *gregs = uc->uc_mcontext.gregs;
 	int err;
 
 	/* Always make any pending restarted system calls return -EINTR */
@@ -102,10 +102,11 @@ asmlinkage int do_rt_sigreturn(struct switch_stack *sw)
 {
 	struct pt_regs *regs = (struct pt_regs *)(sw + 1);
 	/* Verify, can we follow the stack back */
-	struct rt_sigframe *frame = (struct rt_sigframe *) regs->sp;
+	struct rt_sigframe __user *frame;
 	sigset_t set;
 	int rval;
 
+	frame = (struct rt_sigframe __user *) regs->sp;
 	if (!access_ok(frame, sizeof(*frame)))
 		goto badframe;
 
@@ -124,10 +125,10 @@ asmlinkage int do_rt_sigreturn(struct switch_stack *sw)
 	return 0;
 }
 
-static inline int rt_setup_ucontext(struct ucontext *uc, struct pt_regs *regs)
+static inline int rt_setup_ucontext(struct ucontext __user *uc, struct pt_regs *regs)
 {
 	struct switch_stack *sw = (struct switch_stack *)regs - 1;
-	unsigned long *gregs = uc->uc_mcontext.gregs;
+	unsigned long __user *gregs = uc->uc_mcontext.gregs;
 	int err = 0;
 
 	err |= __put_user(MCONTEXT_VERSION, &uc->uc_mcontext.version);
@@ -162,8 +163,9 @@ static inline int rt_setup_ucontext(struct ucontext *uc, struct pt_regs *regs)
 	return err;
 }
 
-static inline void *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
-				 size_t frame_size)
+static inline void __user *get_sigframe(struct ksignal *ksig,
+					struct pt_regs *regs,
+					size_t frame_size)
 {
 	unsigned long usp;
 
@@ -174,13 +176,13 @@ static inline void *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
 	usp = sigsp(usp, ksig);
 
 	/* Verify, is it 32 or 64 bit aligned */
-	return (void *)((usp - frame_size) & -8UL);
+	return (void __user *)((usp - frame_size) & -8UL);
 }
 
 static int setup_rt_frame(struct ksignal *ksig, sigset_t *set,
 			  struct pt_regs *regs)
 {
-	struct rt_sigframe *frame;
+	struct rt_sigframe __user *frame;
 	int err = 0;
 
 	frame = get_sigframe(ksig, regs, sizeof(*frame));
diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
index a94b0cd0bdc5..022d23ae300b 100644
--- a/arch/powerpc/lib/sstep.c
+++ b/arch/powerpc/lib/sstep.c
@@ -112,9 +112,9 @@ static nokprobe_inline long address_ok(struct pt_regs *regs,
 {
 	if (!user_mode(regs))
 		return 1;
-	if (__access_ok(ea, nb))
+	if (access_ok((void __user *)ea, nb))
 		return 1;
-	if (__access_ok(ea, 1))
+	if (access_ok((void __user *)ea, 1))
 		/* Access overlaps the end of the user region */
 		regs->dar = TASK_SIZE_MAX - 1;
 	else
diff --git a/arch/riscv/kernel/perf_callchain.c b/arch/riscv/kernel/perf_callchain.c
index 1fc075b8f764..f0c7bb98119a 100644
--- a/arch/riscv/kernel/perf_callchain.c
+++ b/arch/riscv/kernel/perf_callchain.c
@@ -15,7 +15,7 @@ static unsigned long user_backtrace(struct perf_callchain_entry_ctx *entry,
 {
 	struct stackframe buftail;
 	unsigned long ra = 0;
-	unsigned long *user_frame_tail =
+	unsigned long __user *user_frame_tail =
 			(unsigned long *)(fp - sizeof(struct stackframe));
 
 	/* Check accessibility of one struct frame_tail beyond */
diff --git a/arch/sparc/kernel/signal_32.c b/arch/sparc/kernel/signal_32.c
index ffab16369bea..74f80443b195 100644
--- a/arch/sparc/kernel/signal_32.c
+++ b/arch/sparc/kernel/signal_32.c
@@ -65,7 +65,7 @@ struct rt_signal_frame {
  */
 static inline bool invalid_frame_pointer(void __user *fp, int fplen)
 {
-	if ((((unsigned long) fp) & 15) || !__access_ok((unsigned long)fp, fplen))
+	if ((((unsigned long) fp) & 15) || !access_ok(fp, fplen))
 		return true;
 
 	return false;
diff --git a/lib/test_lockup.c b/lib/test_lockup.c
index 906b598740a7..6a0f329a794a 100644
--- a/lib/test_lockup.c
+++ b/lib/test_lockup.c
@@ -417,8 +417,8 @@ static bool test_kernel_ptr(unsigned long addr, int size)
 		return false;
 
 	/* should be at least readable kernel address */
-	if (access_ok(ptr, 1) ||
-	    access_ok(ptr + size - 1, 1) ||
+	if (access_ok((void __user *)ptr, 1) ||
+	    access_ok((void __user *)ptr + size - 1, 1) ||
 	    get_kernel_nofault(buf, ptr) ||
 	    get_kernel_nofault(buf, ptr + size - 1)) {
 		pr_err("invalid kernel ptr: %#lx\n", addr);
-- 
2.29.2

[PATCH v2 13/18] uaccess: generalize access_ok()

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

There are many different ways that access_ok() is defined across
architectures, but in the end, they all just compare against the
user_addr_max() value or they accept anything.

Provide one definition that works for most architectures, checking
against TASK_SIZE_MAX for user processes or skipping the check inside
of uaccess_kernel() sections.

For architectures without CONFIG_SET_FS(), this should be the fastest
check, as it comes down to a single comparison of a pointer against a
compile-time constant, while the architecture specific versions tend to
do something more complex for historic reasons or get something wrong.

Type checking for __user annotations is handled inconsistently across
architectures, but this is easily simplified as well by using an inline
function that takes a 'const void __user *' argument. A handful of
callers need an extra __user annotation for this.

Some architectures had trick to use 33-bit or 65-bit arithmetic on the
addresses to calculate the overflow, however this simpler version uses
fewer registers, which means it can produce better object code in the
end despite needing a second (statically predicted) branch.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Mark Rutland <mark.rutland@arm.com> [arm64, asm-generic]
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/Kconfig                          |  7 ++++
 arch/alpha/include/asm/uaccess.h      | 34 +++------------
 arch/arc/include/asm/uaccess.h        | 29 -------------
 arch/arm/include/asm/uaccess.h        | 20 +--------
 arch/arm64/include/asm/uaccess.h      | 11 ++---
 arch/csky/include/asm/uaccess.h       |  8 ----
 arch/hexagon/include/asm/uaccess.h    | 25 ------------
 arch/ia64/include/asm/uaccess.h       |  5 +--
 arch/m68k/Kconfig.cpu                 |  1 +
 arch/m68k/include/asm/uaccess.h       | 19 +--------
 arch/microblaze/include/asm/uaccess.h |  8 +---
 arch/mips/include/asm/uaccess.h       | 29 +------------
 arch/nds32/include/asm/uaccess.h      |  7 +---
 arch/nios2/include/asm/uaccess.h      | 11 +----
 arch/openrisc/include/asm/uaccess.h   | 19 +--------
 arch/parisc/Kconfig                   |  1 +
 arch/parisc/include/asm/uaccess.h     | 12 ++----
 arch/powerpc/include/asm/uaccess.h    | 11 +----
 arch/riscv/include/asm/uaccess.h      | 31 +-------------
 arch/s390/Kconfig                     |  1 +
 arch/s390/include/asm/uaccess.h       | 14 +------
 arch/sh/include/asm/uaccess.h         | 22 +---------
 arch/sparc/Kconfig                    |  1 +
 arch/sparc/include/asm/uaccess.h      |  3 --
 arch/sparc/include/asm/uaccess_32.h   | 18 ++------
 arch/sparc/include/asm/uaccess_64.h   | 12 +-----
 arch/um/include/asm/uaccess.h         |  5 ++-
 arch/x86/include/asm/uaccess.h        | 14 +------
 arch/xtensa/include/asm/uaccess.h     | 10 +----
 include/asm-generic/access_ok.h       | 59 +++++++++++++++++++++++++++
 include/asm-generic/uaccess.h         | 21 +---------
 include/linux/uaccess.h               |  7 ----
 32 files changed, 109 insertions(+), 366 deletions(-)
 create mode 100644 include/asm-generic/access_ok.h

diff --git a/arch/Kconfig b/arch/Kconfig
index 678a80713b21..fa5db36bda67 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -898,6 +898,13 @@ config HAVE_SOFTIRQ_ON_OWN_STACK
 	  Architecture provides a function to run __do_softirq() on a
 	  separate stack.
 
+config ALTERNATE_USER_ADDRESS_SPACE
+	bool
+	help
+	  Architectures set this when the CPU uses separate address
+	  spaces for kernel and user space pointers. In this case, the
+	  access_ok() check on a __user pointer is skipped.
+
 config PGTABLE_LEVELS
 	int
 	default 2
diff --git a/arch/alpha/include/asm/uaccess.h b/arch/alpha/include/asm/uaccess.h
index 1b6f25efa247..82c5743fc9cd 100644
--- a/arch/alpha/include/asm/uaccess.h
+++ b/arch/alpha/include/asm/uaccess.h
@@ -20,28 +20,7 @@
 #define get_fs()  (current_thread_info()->addr_limit)
 #define set_fs(x) (current_thread_info()->addr_limit = (x))
 
-#define uaccess_kernel()	(get_fs().seg == KERNEL_DS.seg)
-
-/*
- * Is a address valid? This does a straightforward calculation rather
- * than tests.
- *
- * Address valid if:
- *  - "addr" doesn't have any high-bits set
- *  - AND "size" doesn't have any high-bits set
- *  - AND "addr+size-(size != 0)" doesn't have any high-bits set
- *  - OR we are in kernel mode.
- */
-#define __access_ok(addr, size) ({				\
-	unsigned long __ao_a = (addr), __ao_b = (size);		\
-	unsigned long __ao_end = __ao_a + __ao_b - !!__ao_b;	\
-	(get_fs().seg & (__ao_a | __ao_b | __ao_end)) == 0; })
-
-#define access_ok(addr, size)				\
-({							\
-	__chk_user_ptr(addr);				\
-	__access_ok(((unsigned long)(addr)), (size));	\
-})
+#include <asm-generic/access_ok.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
@@ -105,7 +84,7 @@ extern void __get_user_unknown(void);
 	long __gu_err = -EFAULT;				\
 	unsigned long __gu_val = 0;				\
 	const __typeof__(*(ptr)) __user *__gu_addr = (ptr);	\
-	if (__access_ok((unsigned long)__gu_addr, size)) {	\
+	if (__access_ok(__gu_addr, size)) {			\
 		__gu_err = 0;					\
 		switch (size) {					\
 		  case 1: __get_user_8(__gu_addr); break;	\
@@ -200,7 +179,7 @@ extern void __put_user_unknown(void);
 ({								\
 	long __pu_err = -EFAULT;				\
 	__typeof__(*(ptr)) __user *__pu_addr = (ptr);		\
-	if (__access_ok((unsigned long)__pu_addr, size)) {	\
+	if (__access_ok(__pu_addr, size)) {			\
 		__pu_err = 0;					\
 		switch (size) {					\
 		  case 1: __put_user_8(x, __pu_addr); break;	\
@@ -316,17 +295,14 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long len)
 
 extern long __clear_user(void __user *to, long len);
 
-extern inline long
+static inline long
 clear_user(void __user *to, long len)
 {
-	if (__access_ok((unsigned long)to, len))
+	if (__access_ok(to, len))
 		len = __clear_user(to, len);
 	return len;
 }
 
-#define user_addr_max() \
-        (uaccess_kernel() ? ~0UL : TASK_SIZE)
-
 extern long strncpy_from_user(char *dest, const char __user *src, long count);
 extern __must_check long strnlen_user(const char __user *str, long n);
 
diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h
index 783bfdb3bfa3..30f80b4be2ab 100644
--- a/arch/arc/include/asm/uaccess.h
+++ b/arch/arc/include/asm/uaccess.h
@@ -23,35 +23,6 @@
 
 #include <linux/string.h>	/* for generic string functions */
 
-
-#define __kernel_ok		(uaccess_kernel())
-
-/*
- * Algorithmically, for __user_ok() we want do:
- * 	(start < TASK_SIZE) && (start+len < TASK_SIZE)
- * where TASK_SIZE could either be retrieved from thread_info->addr_limit or
- * emitted directly in code.
- *
- * This can however be rewritten as follows:
- *	(len <= TASK_SIZE) && (start+len < TASK_SIZE)
- *
- * Because it essentially checks if buffer end is within limit and @len is
- * non-ngeative, which implies that buffer start will be within limit too.
- *
- * The reason for rewriting being, for majority of cases, @len is generally
- * compile time constant, causing first sub-expression to be compile time
- * subsumed.
- *
- * The second part would generate weird large LIMMs e.g. (0x6000_0000 - 0x10),
- * so we check for TASK_SIZE using get_fs() since the addr_limit load from mem
- * would already have been done at this call site for __kernel_ok()
- *
- */
-#define __user_ok(addr, sz)	(((sz) <= TASK_SIZE) && \
-				 ((addr) <= (get_fs() - (sz))))
-#define __access_ok(addr, sz)	(unlikely(__kernel_ok) || \
-				 likely(__user_ok((addr), (sz))))
-
 /*********** Single byte/hword/word copies ******************/
 
 #define __get_user_fn(sz, u, k)					\
diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h
index d20d78c34b94..2fcbec9c306c 100644
--- a/arch/arm/include/asm/uaccess.h
+++ b/arch/arm/include/asm/uaccess.h
@@ -55,21 +55,6 @@ extern int __put_user_bad(void);
 
 #ifdef CONFIG_MMU
 
-/*
- * We use 33-bit arithmetic here.  Success returns zero, failure returns
- * addr_limit.  We take advantage that addr_limit will be zero for KERNEL_DS,
- * so this will always return success in that case.
- */
-#define __range_ok(addr, size) ({ \
-	unsigned long flag, roksum; \
-	__chk_user_ptr(addr);	\
-	__asm__(".syntax unified\n" \
-		"adds %1, %2, %3; sbcscc %1, %1, %0; movcc %0, #0" \
-		: "=&r" (flag), "=&r" (roksum) \
-		: "r" (addr), "Ir" (size), "0" (TASK_SIZE) \
-		: "cc"); \
-	flag; })
-
 /*
  * This is a type: either unsigned long, if the argument fits into
  * that type, or otherwise unsigned long long.
@@ -241,15 +226,12 @@ extern int __put_user_8(void *, unsigned long long);
 
 #else /* CONFIG_MMU */
 
-#define __addr_ok(addr)		((void)(addr), 1)
-#define __range_ok(addr, size)	((void)(addr), 0)
-
 #define get_user(x, p)	__get_user(x, p)
 #define __put_user_check __put_user_nocheck
 
 #endif /* CONFIG_MMU */
 
-#define access_ok(addr, size)	(__range_ok(addr, size) == 0)
+#include <asm-generic/access_ok.h>
 
 #ifdef CONFIG_CPU_SPECTRE
 /*
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 199c553b740a..e8dce0cc5eaa 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -26,13 +26,7 @@
 #include <asm/memory.h>
 #include <asm/extable.h>
 
-static inline int __access_ok(const void __user *ptr, unsigned long size)
-{
-	unsigned long limit = TASK_SIZE_MAX;
-	unsigned long addr = (unsigned long)ptr;
-
-	return (size <= limit) && (addr <= (limit - size));
-}
+static inline int __access_ok(const void __user *ptr, unsigned long size);
 
 /*
  * Test whether a block of memory is a valid user space address.
@@ -54,6 +48,9 @@ static inline int access_ok(const void __user *addr, unsigned long size)
 
 	return likely(__access_ok(addr, size));
 }
+#define access_ok access_ok
+
+#include <asm-generic/access_ok.h>
 
 /*
  * User access enabling/disabling.
diff --git a/arch/csky/include/asm/uaccess.h b/arch/csky/include/asm/uaccess.h
index ac5a54f57d40..fec8f77ffc99 100644
--- a/arch/csky/include/asm/uaccess.h
+++ b/arch/csky/include/asm/uaccess.h
@@ -5,14 +5,6 @@
 
 #define user_addr_max() (current_thread_info()->addr_limit.seg)
 
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	unsigned long limit = user_addr_max();
-
-	return (size <= limit) && (addr <= (limit - size));
-}
-#define __access_ok __access_ok
-
 /*
  * __put_user_fn
  */
diff --git a/arch/hexagon/include/asm/uaccess.h b/arch/hexagon/include/asm/uaccess.h
index 719ba3f3c45c..bff77efc0d9a 100644
--- a/arch/hexagon/include/asm/uaccess.h
+++ b/arch/hexagon/include/asm/uaccess.h
@@ -12,31 +12,6 @@
  */
 #include <asm/sections.h>
 
-/*
- * access_ok: - Checks if a user space pointer is valid
- * @addr: User space pointer to start of block to check
- * @size: Size of block to check
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Checks if a pointer to a block of memory in user space is valid.
- *
- * Returns true (nonzero) if the memory block *may* be valid, false (zero)
- * if it is definitely invalid.
- *
- */
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
-#define user_addr_max() (uaccess_kernel() ? ~0UL : TASK_SIZE)
-
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	unsigned long limit = TASK_SIZE;
-
-	return (size <= limit) && (addr <= (limit - size));
-}
-#define __access_ok __access_ok
-
 /*
  * When a kernel-mode page fault is taken, the faulting instruction
  * address is checked against a table of exception_table_entries.
diff --git a/arch/ia64/include/asm/uaccess.h b/arch/ia64/include/asm/uaccess.h
index e19d2dcc0ced..e242a3cc1330 100644
--- a/arch/ia64/include/asm/uaccess.h
+++ b/arch/ia64/include/asm/uaccess.h
@@ -50,8 +50,6 @@
 #define get_fs()  (current_thread_info()->addr_limit)
 #define set_fs(x) (current_thread_info()->addr_limit = (x))
 
-#define uaccess_kernel()	(get_fs().seg == KERNEL_DS.seg)
-
 /*
  * When accessing user memory, we need to make sure the entire area really is in
  * user-level space.  In order to do this efficiently, we make sure that the page at
@@ -65,7 +63,8 @@ static inline int __access_ok(const void __user *p, unsigned long size)
 	return likely(addr <= seg) &&
 	 (seg == KERNEL_DS.seg || likely(REGION_OFFSET(addr) < RGN_MAP_LIMIT));
 }
-#define access_ok(addr, size)	__access_ok((addr), (size))
+#define __access_ok __access_ok
+#include <asm-generic/access_ok.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
diff --git a/arch/m68k/Kconfig.cpu b/arch/m68k/Kconfig.cpu
index 0d00ef5117dc..16ea9a67723c 100644
--- a/arch/m68k/Kconfig.cpu
+++ b/arch/m68k/Kconfig.cpu
@@ -453,6 +453,7 @@ config CPU_HAS_NO_UNALIGNED
 
 config CPU_HAS_ADDRESS_SPACES
 	bool
+	select ALTERNATE_USER_ADDRESS_SPACE
 
 config FPU
 	bool
diff --git a/arch/m68k/include/asm/uaccess.h b/arch/m68k/include/asm/uaccess.h
index 8eb625e75452..64914872a5c9 100644
--- a/arch/m68k/include/asm/uaccess.h
+++ b/arch/m68k/include/asm/uaccess.h
@@ -10,24 +10,7 @@
 #include <linux/compiler.h>
 #include <linux/types.h>
 #include <asm/extable.h>
-
-/* We let the MMU do all checking */
-static inline int access_ok(const void __user *ptr,
-			    unsigned long size)
-{
-	unsigned long limit = TASK_SIZE;
-	unsigned long addr = (unsigned long)ptr;
-
-	/*
-	 * XXX: for !CONFIG_CPU_HAS_ADDRESS_SPACES this really needs to check
-	 * for TASK_SIZE!
-	 * Removing this helper is probably sufficient.
-	 */
-	if (IS_ENABLED(CONFIG_CPU_HAS_ADDRESS_SPACES))
-		return 1;
-
-	return (size <= limit) && (addr <= (limit - size));
-}
+#include <asm-generic/access_ok.h>
 
 /*
  * Not all varients of the 68k family support the notion of address spaces.
diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
index 3fe96979d2c6..bf9b7657a65a 100644
--- a/arch/microblaze/include/asm/uaccess.h
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -39,13 +39,7 @@
 
 # define uaccess_kernel()	(get_fs().seg == KERNEL_DS.seg)
 
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	unsigned long limit = user_addr_max();
-
-	return (size <= limit) && (addr <= (limit - size));
-}
-#define access_ok(addr, size) __access_ok((unsigned long)addr, size)
+#include <asm-generic/access_ok.h>
 
 # define __FIXUP_SECTION	".section .fixup,\"ax\"\n"
 # define __EX_TABLE_SECTION	".section __ex_table,\"a\"\n"
diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
index d7c89dc3426c..436248652b28 100644
--- a/arch/mips/include/asm/uaccess.h
+++ b/arch/mips/include/asm/uaccess.h
@@ -44,34 +44,7 @@ extern u64 __ua_limit;
 
 #endif /* CONFIG_64BIT */
 
-/*
- * access_ok: - Checks if a user space pointer is valid
- * @addr: User space pointer to start of block to check
- * @size: Size of block to check
- *
- * Context: User context only. This function may sleep if pagefaults are
- *          enabled.
- *
- * Checks if a pointer to a block of memory in user space is valid.
- *
- * Returns true (nonzero) if the memory block may be valid, false (zero)
- * if it is definitely invalid.
- *
- * 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.
- */
-
-static inline int __access_ok(const void __user *p, unsigned long size)
-{
-	unsigned long addr = (unsigned long)p;
-	unsigned long limit = TASK_SIZE_MAX;
-
-	return (size <= limit) && (addr <= (limit - size));
-}
-
-#define access_ok(addr, size)					\
-	likely(__access_ok((addr), (size)))
+#include <asm-generic/access_ok.h>
 
 /*
  * put_user: - Write a simple value into user space.
diff --git a/arch/nds32/include/asm/uaccess.h b/arch/nds32/include/asm/uaccess.h
index 37a40981deb3..832d642a4068 100644
--- a/arch/nds32/include/asm/uaccess.h
+++ b/arch/nds32/include/asm/uaccess.h
@@ -38,18 +38,15 @@ extern int fixup_exception(struct pt_regs *regs);
 
 #define get_fs()	(current_thread_info()->addr_limit)
 #define user_addr_max	get_fs
+#define uaccess_kernel() (get_fs() == KERNEL_DS)
 
 static inline void set_fs(mm_segment_t fs)
 {
 	current_thread_info()->addr_limit = fs;
 }
 
-#define uaccess_kernel()	(get_fs() == KERNEL_DS)
+#include <asm-generic/access_ok.h>
 
-#define __range_ok(addr, size) (size <= get_fs() && addr <= (get_fs() -size))
-
-#define access_ok(addr, size)	\
-	__range_ok((unsigned long)addr, (unsigned long)size)
 /*
  * Single-value transfer routines.  They automatically use the right
  * size if we just have the right pointer type.  Note that the functions
diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
index a5cbe07cf0da..6664ddc0e8e5 100644
--- a/arch/nios2/include/asm/uaccess.h
+++ b/arch/nios2/include/asm/uaccess.h
@@ -30,19 +30,10 @@
 #define get_fs()		(current_thread_info()->addr_limit)
 #define set_fs(seg)		(current_thread_info()->addr_limit = (seg))
 
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
-
-#define __access_ok(addr, len)			\
-	(((signed long)(((long)get_fs().seg) &	\
-		((long)(addr) | (((long)(addr)) + (len)) | (len)))) == 0)
-
-#define access_ok(addr, len)		\
-	likely(__access_ok((unsigned long)(addr), (unsigned long)(len)))
+#include <asm-generic/access_ok.h>
 
 # define __EX_TABLE_SECTION	".section __ex_table,\"a\"\n"
 
-#define user_addr_max() (uaccess_kernel() ? ~0UL : TASK_SIZE)
-
 /*
  * Zero Userspace
  */
diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h
index 120f5005461b..8f049ec99b3e 100644
--- a/arch/openrisc/include/asm/uaccess.h
+++ b/arch/openrisc/include/asm/uaccess.h
@@ -45,21 +45,7 @@
 
 #define uaccess_kernel()	(get_fs() == KERNEL_DS)
 
-/* Ensure that the range from addr to addr+size is all within the process'
- * address space
- */
-static inline int __range_ok(unsigned long addr, unsigned long size)
-{
-	const mm_segment_t fs = get_fs();
-
-	return size <= fs && addr <= (fs - size);
-}
-
-#define access_ok(addr, size)						\
-({ 									\
-	__chk_user_ptr(addr);						\
-	__range_ok((unsigned long)(addr), (size));			\
-})
+#include <asm-generic/access_ok.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
@@ -268,9 +254,6 @@ clear_user(void __user *addr, unsigned long size)
 	return size;
 }
 
-#define user_addr_max() \
-	(uaccess_kernel() ? ~0UL : TASK_SIZE)
-
 extern long strncpy_from_user(char *dest, const char __user *src, long count);
 
 extern __must_check long strnlen_user(const char __user *str, long n);
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 43c1c880def6..15039fdd5413 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -1,6 +1,7 @@
 # SPDX-License-Identifier: GPL-2.0
 config PARISC
 	def_bool y
+	select ALTERNATE_USER_ADDRESS_SPACE
 	select ARCH_32BIT_OFF_T if !64BIT
 	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select HAVE_FUNCTION_TRACER
diff --git a/arch/parisc/include/asm/uaccess.h b/arch/parisc/include/asm/uaccess.h
index 0925bbd6db67..187f4bdff13e 100644
--- a/arch/parisc/include/asm/uaccess.h
+++ b/arch/parisc/include/asm/uaccess.h
@@ -11,15 +11,9 @@
 #include <linux/bug.h>
 #include <linux/string.h>
 
-/*
- * Note that since kernel addresses are in a separate address space on
- * parisc, we don't need to do anything for access_ok().
- * We just let the page fault handler do the right thing. This also means
- * that put_user is the same as __put_user, etc.
- */
-
-#define access_ok(uaddr, size)	\
-	( (uaddr) == (uaddr) )
+#define TASK_SIZE_MAX DEFAULT_TASK_SIZE
+#include <asm/pgtable.h>
+#include <asm-generic/access_ok.h>
 
 #define put_user __put_user
 #define get_user __get_user
diff --git a/arch/powerpc/include/asm/uaccess.h b/arch/powerpc/include/asm/uaccess.h
index a0032c2e7550..2e83217f52de 100644
--- a/arch/powerpc/include/asm/uaccess.h
+++ b/arch/powerpc/include/asm/uaccess.h
@@ -11,18 +11,9 @@
 #ifdef __powerpc64__
 /* We use TASK_SIZE_USER64 as TASK_SIZE is not constant */
 #define TASK_SIZE_MAX		TASK_SIZE_USER64
-#else
-#define TASK_SIZE_MAX		TASK_SIZE
 #endif
 
-static inline bool __access_ok(unsigned long addr, unsigned long size)
-{
-	return addr < TASK_SIZE_MAX && size <= TASK_SIZE_MAX - addr;
-}
-
-#define access_ok(addr, size)		\
-	(__chk_user_ptr(addr),		\
-	 __access_ok((unsigned long)(addr), (size)))
+#include <asm-generic/access_ok.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
diff --git a/arch/riscv/include/asm/uaccess.h b/arch/riscv/include/asm/uaccess.h
index 4407b9e48d2c..855450bed9f5 100644
--- a/arch/riscv/include/asm/uaccess.h
+++ b/arch/riscv/include/asm/uaccess.h
@@ -21,42 +21,13 @@
 #include <asm/byteorder.h>
 #include <asm/extable.h>
 #include <asm/asm.h>
+#include <asm-generic/access_ok.h>
 
 #define __enable_user_access()							\
 	__asm__ __volatile__ ("csrs sstatus, %0" : : "r" (SR_SUM) : "memory")
 #define __disable_user_access()							\
 	__asm__ __volatile__ ("csrc sstatus, %0" : : "r" (SR_SUM) : "memory")
 
-/**
- * access_ok: - Checks if a user space pointer is valid
- * @addr: User space pointer to start of block to check
- * @size: Size of block to check
- *
- * Context: User context only.  This function may sleep.
- *
- * Checks if a pointer to a block of memory in user space is valid.
- *
- * Returns true (nonzero) if the memory block may be valid, false (zero)
- * if it is definitely invalid.
- *
- * 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.
- */
-#define access_ok(addr, size) ({					\
-	__chk_user_ptr(addr);						\
-	likely(__access_ok((unsigned long __force)(addr), (size)));	\
-})
-
-/*
- * Ensure that the range [addr, addr+size) is within the process's
- * address space
- */
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	return size <= TASK_SIZE && addr <= TASK_SIZE - size;
-}
-
 /*
  * The exception table consists of pairs of addresses: the first is the
  * address of an instruction that is allowed to fault, and the second is
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index be9f39fd06df..fb48a62aa985 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -55,6 +55,7 @@ config S390
 	# Note: keep this list sorted alphabetically
 	#
 	imply IMA_SECURE_AND_OR_TRUSTED_BOOT
+	select ALTERNATE_USER_ADDRESS_SPACE
 	select ARCH_32BIT_USTAT_F_TINODE
 	select ARCH_BINFMT_ELF_STATE
 	select ARCH_ENABLE_MEMORY_HOTPLUG if SPARSEMEM
diff --git a/arch/s390/include/asm/uaccess.h b/arch/s390/include/asm/uaccess.h
index 29332edf46f0..5cb258cd9d29 100644
--- a/arch/s390/include/asm/uaccess.h
+++ b/arch/s390/include/asm/uaccess.h
@@ -17,22 +17,10 @@
 #include <asm/ctl_reg.h>
 #include <asm/extable.h>
 #include <asm/facility.h>
+#include <asm-generic/access_ok.h>
 
 void debug_user_asce(int exit);
 
-static inline int __range_ok(unsigned long addr, unsigned long size)
-{
-	return 1;
-}
-
-#define __access_ok(addr, size)				\
-({							\
-	__chk_user_ptr(addr);				\
-	__range_ok((unsigned long)(addr), (size));	\
-})
-
-#define access_ok(addr, size) __access_ok(addr, size)
-
 unsigned long __must_check
 raw_copy_from_user(void *to, const void __user *from, unsigned long n);
 
diff --git a/arch/sh/include/asm/uaccess.h b/arch/sh/include/asm/uaccess.h
index 8867bb04b00e..ccd219d74851 100644
--- a/arch/sh/include/asm/uaccess.h
+++ b/arch/sh/include/asm/uaccess.h
@@ -5,28 +5,10 @@
 #include <asm/segment.h>
 #include <asm/extable.h>
 
-#define __addr_ok(addr) \
-	((unsigned long __force)(addr) < current_thread_info()->addr_limit.seg)
-
-/*
- * __access_ok: Check if address with size is OK or not.
- *
- * Uhhuh, this needs 33-bit arithmetic. We have a carry..
- *
- * sum := addr + size;  carry? --> flag = true;
- * if (sum >= addr_limit) flag = true;
- */
-#define __access_ok(addr, size)	({				\
-	unsigned long __ao_a = (addr), __ao_b = (size);		\
-	unsigned long __ao_end = __ao_a + __ao_b - !!__ao_b;	\
-	__ao_end >= __ao_a && __addr_ok(__ao_end); })
-
-#define access_ok(addr, size)	\
-	(__chk_user_ptr(addr),		\
-	 __access_ok((unsigned long __force)(addr), (size)))
-
 #define user_addr_max()	(current_thread_info()->addr_limit.seg)
 
+#include <asm-generic/access_ok.h>
+
 /*
  * Uh, these should become the main single-value transfer routines ...
  * They automatically use the right size if we just have the right
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 1cab1b284f1a..9f6f9bce5292 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -62,6 +62,7 @@ config SPARC32
 
 config SPARC64
 	def_bool 64BIT
+	select ALTERNATE_USER_ADDRESS_SPACE
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_KRETPROBES
diff --git a/arch/sparc/include/asm/uaccess.h b/arch/sparc/include/asm/uaccess.h
index 390094200fc4..ee75f69e3fcd 100644
--- a/arch/sparc/include/asm/uaccess.h
+++ b/arch/sparc/include/asm/uaccess.h
@@ -10,9 +10,6 @@
 #include <asm/uaccess_32.h>
 #endif
 
-#define user_addr_max() \
-	(uaccess_kernel() ? ~0UL : TASK_SIZE)
-
 long strncpy_from_user(char *dest, const char __user *src, long count);
 
 #endif
diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h
index 4a12346bb69c..367747116260 100644
--- a/arch/sparc/include/asm/uaccess_32.h
+++ b/arch/sparc/include/asm/uaccess_32.h
@@ -25,17 +25,7 @@
 #define get_fs()	(current->thread.current_ds)
 #define set_fs(val)	((current->thread.current_ds) = (val))
 
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
-
-/* We have there a nice not-mapped page at PAGE_OFFSET - PAGE_SIZE, so that this test
- * can be fairly lightweight.
- * No one can read/write anything from userland in the kernel space by setting
- * large size and address near to PAGE_OFFSET - a fault will break his intentions.
- */
-#define __user_ok(addr, size) ({ (void)(size); (addr) < STACK_TOP; })
-#define __kernel_ok (uaccess_kernel())
-#define __access_ok(addr, size) (__user_ok((addr) & get_fs().seg, (size)))
-#define access_ok(addr, size) __access_ok((unsigned long)(addr), size)
+#include <asm-generic/access_ok.h>
 
 /* Uh, these should become the main single-value transfer routines..
  * They automatically use the right size if we just have the right
@@ -47,13 +37,13 @@
  * and hide all the ugliness from the user.
  */
 #define put_user(x, ptr) ({ \
-	unsigned long __pu_addr = (unsigned long)(ptr); \
+	void __user *__pu_addr = (ptr); \
 	__chk_user_ptr(ptr); \
 	__put_user_check((__typeof__(*(ptr)))(x), __pu_addr, sizeof(*(ptr))); \
 })
 
 #define get_user(x, ptr) ({ \
-	unsigned long __gu_addr = (unsigned long)(ptr); \
+	const void __user *__gu_addr = (ptr); \
 	__chk_user_ptr(ptr); \
 	__get_user_check((x), __gu_addr, sizeof(*(ptr)), __typeof__(*(ptr))); \
 })
@@ -232,7 +222,7 @@ static inline unsigned long __clear_user(void __user *addr, unsigned long size)
 
 static inline unsigned long clear_user(void __user *addr, unsigned long n)
 {
-	if (n && __access_ok((unsigned long) addr, n))
+	if (n && __access_ok(addr, n))
 		return __clear_user(addr, n);
 	else
 		return n;
diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
index 5c12fb46bc61..59b9a545df23 100644
--- a/arch/sparc/include/asm/uaccess_64.h
+++ b/arch/sparc/include/asm/uaccess_64.h
@@ -31,7 +31,7 @@
 
 #define get_fs() ((mm_segment_t){(current_thread_info()->current_ds)})
 
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
+#include <asm-generic/access_ok.h>
 
 #define set_fs(val)								\
 do {										\
@@ -61,16 +61,6 @@ static inline bool __chk_range_not_ok(unsigned long addr, unsigned long size, un
 	__chk_range_not_ok((unsigned long __force)(addr), size, limit); \
 })
 
-static inline int __access_ok(const void __user * addr, unsigned long size)
-{
-	return 1;
-}
-
-static inline int access_ok(const void __user * addr, unsigned long size)
-{
-	return 1;
-}
-
 void __retl_efault(void);
 
 /* Uh, these should become the main single-value transfer routines..
diff --git a/arch/um/include/asm/uaccess.h b/arch/um/include/asm/uaccess.h
index 1ecfc96bcc50..7d9d60e41e4e 100644
--- a/arch/um/include/asm/uaccess.h
+++ b/arch/um/include/asm/uaccess.h
@@ -25,7 +25,7 @@
 extern unsigned long raw_copy_from_user(void *to, const void __user *from, unsigned long n);
 extern unsigned long raw_copy_to_user(void __user *to, const void *from, unsigned long n);
 extern unsigned long __clear_user(void __user *mem, unsigned long len);
-static inline int __access_ok(unsigned long addr, unsigned long size);
+static inline int __access_ok(const void __user *ptr, unsigned long size);
 
 /* Teach asm-generic/uaccess.h that we have C functions for these. */
 #define __access_ok __access_ok
@@ -36,8 +36,9 @@ static inline int __access_ok(unsigned long addr, unsigned long size);
 
 #include <asm-generic/uaccess.h>
 
-static inline int __access_ok(unsigned long addr, unsigned long size)
+static inline int __access_ok(const void __user *ptr, unsigned long size)
 {
+	unsigned long addr = (unsigned long)ptr;
 	return __addr_range_nowrap(addr, size) &&
 		(__under_task_size(addr, size) ||
 		 __access_ok_vsyscall(addr, size));
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index 201efcec66b7..f78e2b3501a1 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -12,18 +12,6 @@
 #include <asm/smap.h>
 #include <asm/extable.h>
 
-/*
- * Test whether a block of memory is a valid user space address.
- * Returns 0 if the range is valid, nonzero otherwise.
- */
-static inline bool __access_ok(void __user *ptr, unsigned long size)
-{
-	unsigned long limit = TASK_SIZE_MAX;
-	unsigned long addr = ptr;
-
-	return (size <= limit) && (addr <= (limit - size));
-}
-
 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
 static inline bool pagefault_disabled(void);
 # define WARN_ON_IN_IRQ()	\
@@ -55,6 +43,8 @@ static inline bool pagefault_disabled(void);
 	likely(__access_ok(addr, size));				\
 })
 
+#include <asm-generic/access_ok.h>
+
 extern int __get_user_1(void);
 extern int __get_user_2(void);
 extern int __get_user_4(void);
diff --git a/arch/xtensa/include/asm/uaccess.h b/arch/xtensa/include/asm/uaccess.h
index 75bd8fbf52ba..0edd9e4b23d0 100644
--- a/arch/xtensa/include/asm/uaccess.h
+++ b/arch/xtensa/include/asm/uaccess.h
@@ -35,15 +35,7 @@
 #define get_fs()	(current->thread.current_ds)
 #define set_fs(val)	(current->thread.current_ds = (val))
 
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
-
-#define __kernel_ok (uaccess_kernel())
-#define __user_ok(addr, size) \
-		(((size) <= TASK_SIZE)&&((addr) <= TASK_SIZE-(size)))
-#define __access_ok(addr, size) (__kernel_ok || __user_ok((addr), (size)))
-#define access_ok(addr, size) __access_ok((unsigned long)(addr), (size))
-
-#define user_addr_max() (uaccess_kernel() ? ~0UL : TASK_SIZE)
+#include <asm-generic/access_ok.h>
 
 /*
  * These are the main single-value transfer routines.  They
diff --git a/include/asm-generic/access_ok.h b/include/asm-generic/access_ok.h
new file mode 100644
index 000000000000..1aad8964d2ed
--- /dev/null
+++ b/include/asm-generic/access_ok.h
@@ -0,0 +1,59 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_GENERIC_ACCESS_OK_H__
+#define __ASM_GENERIC_ACCESS_OK_H__
+
+/*
+ * Checking whether a pointer is valid for user space access.
+ * These definitions work on most architectures, but overrides can
+ * be used where necessary.
+ */
+
+/*
+ * architectures with compat tasks have a variable TASK_SIZE and should
+ * override this to a constant.
+ */
+#ifndef TASK_SIZE_MAX
+#define TASK_SIZE_MAX			TASK_SIZE
+#endif
+
+#ifndef uaccess_kernel
+#ifdef CONFIG_SET_FS
+#define uaccess_kernel()		(get_fs().seg == KERNEL_DS.seg)
+#else
+#define uaccess_kernel()		(0)
+#endif
+#endif
+
+#ifndef user_addr_max
+#define user_addr_max()			(uaccess_kernel() ? ~0UL : TASK_SIZE_MAX)
+#endif
+
+#ifndef __access_ok
+/*
+ * 'size' is a compile-time constant for most callers, so optimize for
+ * this case to turn the check into a single comparison against a constant
+ * limit and catch all possible overflows.
+ * On architectures with separate user address space (m68k, s390, parisc,
+ * sparc64) or those without an MMU, this should always return true.
+ *
+ * This version was originally contributed by Jonas Bonn for the
+ * OpenRISC architecture, and was found to be the most efficient
+ * for constant 'size' and 'limit' values.
+ */
+static inline int __access_ok(const void __user *ptr, unsigned long size)
+{
+	unsigned long limit = user_addr_max();
+	unsigned long addr = (unsigned long)ptr;
+
+	if (IS_ENABLED(CONFIG_ALTERNATE_USER_ADDRESS_SPACE))
+		return true;
+
+	return (size <= limit) && (addr <= (limit - size));
+}
+#endif
+
+#ifndef access_ok
+#define access_ok(addr, size) likely(__access_ok(addr, size))
+#endif
+
+#endif
diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
index 0870fa11a7c5..ebc685dc8d74 100644
--- a/include/asm-generic/uaccess.h
+++ b/include/asm-generic/uaccess.h
@@ -114,28 +114,9 @@ static inline void set_fs(mm_segment_t fs)
 }
 #endif
 
-#ifndef uaccess_kernel
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
-#endif
-
-#ifndef user_addr_max
-#define user_addr_max() (uaccess_kernel() ? ~0UL : TASK_SIZE)
-#endif
-
 #endif /* CONFIG_SET_FS */
 
-#define access_ok(addr, size) __access_ok((unsigned long)(addr),(size))
-
-/*
- * The architecture should really override this if possible, at least
- * doing a check on the get_fs()
- */
-#ifndef __access_ok
-static inline int __access_ok(unsigned long addr, unsigned long size)
-{
-	return 1;
-}
-#endif
+#include <asm-generic/access_ok.h>
 
 /*
  * These are the main single-value transfer routines.  They automatically
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 67e9bc94dc40..2c31667e62e0 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -33,13 +33,6 @@ typedef struct {
 	/* empty dummy */
 } mm_segment_t;
 
-#ifndef TASK_SIZE_MAX
-#define TASK_SIZE_MAX			TASK_SIZE
-#endif
-
-#define uaccess_kernel()		(false)
-#define user_addr_max()			(TASK_SIZE_MAX)
-
 static inline mm_segment_t force_uaccess_begin(void)
 {
 	return (mm_segment_t) { };
-- 
2.29.2

[PATCH v2 14/18] lib/test_lockup: fix kernel pointer check for separate address spaces

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

test_kernel_ptr() uses access_ok() to figure out if a given address
points to user space instead of kernel space. However on architectures
that set CONFIG_ALTERNATE_USER_ADDRESS_SPACE, a pointer can be valid
for both, and the check always fails because access_ok() returns true.

Make the check for user space pointers conditional on the type of
address space layout.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 lib/test_lockup.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/lib/test_lockup.c b/lib/test_lockup.c
index 6a0f329a794a..c3fd87d6c2dd 100644
--- a/lib/test_lockup.c
+++ b/lib/test_lockup.c
@@ -417,9 +417,14 @@ static bool test_kernel_ptr(unsigned long addr, int size)
 		return false;
 
 	/* should be at least readable kernel address */
-	if (access_ok((void __user *)ptr, 1) ||
-	    access_ok((void __user *)ptr + size - 1, 1) ||
-	    get_kernel_nofault(buf, ptr) ||
+	if (!IS_ENABLED(CONFIG_ALTERNATE_USER_ADDRESS_SPACE) &&
+	    (access_ok((void __user *)ptr, 1) ||
+	     access_ok((void __user *)ptr + size - 1, 1))) {
+		pr_err("user space ptr invalid in kernel: %#lx\n", addr);
+		return true;
+	}
+
+	if (get_kernel_nofault(buf, ptr) ||
 	    get_kernel_nofault(buf, ptr + size - 1)) {
 		pr_err("invalid kernel ptr: %#lx\n", addr);
 		return true;
-- 
2.29.2

[PATCH v2 15/18] sparc64: remove CONFIG_SET_FS support

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

sparc64 uses address space identifiers to differentiate between kernel
and user space, using ASI_P for kernel threads but ASI_AIUS for normal
user space, with the option of changing between them.

As nothing really changes the ASI any more, just hardcode ASI_AIUS
everywhere. Kernel threads are not allowed to access __user pointers
anyway.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/sparc/include/asm/processor_64.h   |  4 ----
 arch/sparc/include/asm/switch_to_64.h   |  4 +---
 arch/sparc/include/asm/thread_info_64.h |  4 +---
 arch/sparc/include/asm/uaccess_64.h     | 20 +-------------------
 arch/sparc/kernel/process_64.c          | 12 ------------
 arch/sparc/kernel/traps_64.c            |  2 --
 arch/sparc/lib/NGmemcpy.S               |  3 +--
 arch/sparc/mm/init_64.c                 |  7 ++++---
 8 files changed, 8 insertions(+), 48 deletions(-)

diff --git a/arch/sparc/include/asm/processor_64.h b/arch/sparc/include/asm/processor_64.h
index ae851e8fce4c..89850dff6b03 100644
--- a/arch/sparc/include/asm/processor_64.h
+++ b/arch/sparc/include/asm/processor_64.h
@@ -47,10 +47,6 @@
 
 #ifndef __ASSEMBLY__
 
-typedef struct {
-	unsigned char seg;
-} mm_segment_t;
-
 /* The Sparc processor specific thread struct. */
 /* XXX This should die, everything can go into thread_info now. */
 struct thread_struct {
diff --git a/arch/sparc/include/asm/switch_to_64.h b/arch/sparc/include/asm/switch_to_64.h
index b1d4e2e3210f..14f3c49bfdbc 100644
--- a/arch/sparc/include/asm/switch_to_64.h
+++ b/arch/sparc/include/asm/switch_to_64.h
@@ -20,10 +20,8 @@ do {						\
 	 */
 #define switch_to(prev, next, last)					\
 do {	save_and_clear_fpu();						\
-	/* If you are tempted to conditionalize the following */	\
-	/* so that ASI is only written if it changes, think again. */	\
 	__asm__ __volatile__("wr %%g0, %0, %%asi"			\
-	: : "r" (task_thread_info(next)->current_ds));\
+	: : "r" (ASI_AIUS));						\
 	trap_block[current_thread_info()->cpu].thread =			\
 		task_thread_info(next);					\
 	__asm__ __volatile__(						\
diff --git a/arch/sparc/include/asm/thread_info_64.h b/arch/sparc/include/asm/thread_info_64.h
index 8047a9caab2f..1a44372e2bc0 100644
--- a/arch/sparc/include/asm/thread_info_64.h
+++ b/arch/sparc/include/asm/thread_info_64.h
@@ -46,7 +46,7 @@ struct thread_info {
 	struct pt_regs		*kregs;
 	int			preempt_count;	/* 0 => preemptable, <0 => BUG */
 	__u8			new_child;
-	__u8			current_ds;
+	__u8			__pad;
 	__u16			cpu;
 
 	unsigned long		*utraps;
@@ -81,7 +81,6 @@ struct thread_info {
 #define TI_KREGS	0x00000028
 #define TI_PRE_COUNT	0x00000030
 #define TI_NEW_CHILD	0x00000034
-#define TI_CURRENT_DS	0x00000035
 #define TI_CPU		0x00000036
 #define TI_UTRAPS	0x00000038
 #define TI_REG_WINDOW	0x00000040
@@ -116,7 +115,6 @@ struct thread_info {
 #define INIT_THREAD_INFO(tsk)				\
 {							\
 	.task		=	&tsk,			\
-	.current_ds	=	ASI_P,			\
 	.preempt_count	=	INIT_PREEMPT_COUNT,	\
 	.kregs		=	(struct pt_regs *)(init_stack+THREAD_SIZE)-1 \
 }
diff --git a/arch/sparc/include/asm/uaccess_64.h b/arch/sparc/include/asm/uaccess_64.h
index 59b9a545df23..94266a5c5b04 100644
--- a/arch/sparc/include/asm/uaccess_64.h
+++ b/arch/sparc/include/asm/uaccess_64.h
@@ -12,33 +12,15 @@
 #include <asm/spitfire.h>
 
 #include <asm/processor.h>
+#include <asm-generic/access_ok.h>
 
 /*
  * Sparc64 is segmented, though more like the M68K than the I386.
  * We use the secondary ASI to address user memory, which references a
  * completely different VM map, thus there is zero chance of the user
  * doing something queer and tricking us into poking kernel memory.
- *
- * What is left here is basically what is needed for the other parts of
- * the kernel that expect to be able to manipulate, erum, "segments".
- * Or perhaps more properly, permissions.
- *
- * "For historical reasons, these macros are grossly misnamed." -Linus
  */
 
-#define KERNEL_DS   ((mm_segment_t) { ASI_P })
-#define USER_DS     ((mm_segment_t) { ASI_AIUS })	/* har har har */
-
-#define get_fs() ((mm_segment_t){(current_thread_info()->current_ds)})
-
-#include <asm-generic/access_ok.h>
-
-#define set_fs(val)								\
-do {										\
-	current_thread_info()->current_ds = (val).seg;				\
-	__asm__ __volatile__ ("wr %%g0, %0, %%asi" : : "r" ((val).seg));	\
-} while(0)
-
 /*
  * Test whether a block of memory is a valid user space address.
  * Returns 0 if the range is valid, nonzero otherwise.
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index f5b2cac8669f..9a2ceb080ac9 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -106,18 +106,13 @@ static void show_regwindow32(struct pt_regs *regs)
 {
 	struct reg_window32 __user *rw;
 	struct reg_window32 r_w;
-	mm_segment_t old_fs;
 	
 	__asm__ __volatile__ ("flushw");
 	rw = compat_ptr((unsigned int)regs->u_regs[14]);
-	old_fs = get_fs();
-	set_fs (USER_DS);
 	if (copy_from_user (&r_w, rw, sizeof(r_w))) {
-		set_fs (old_fs);
 		return;
 	}
 
-	set_fs (old_fs);			
 	printk("l0: %08x l1: %08x l2: %08x l3: %08x "
 	       "l4: %08x l5: %08x l6: %08x l7: %08x\n",
 	       r_w.locals[0], r_w.locals[1], r_w.locals[2], r_w.locals[3],
@@ -136,7 +131,6 @@ static void show_regwindow(struct pt_regs *regs)
 	struct reg_window __user *rw;
 	struct reg_window *rwk;
 	struct reg_window r_w;
-	mm_segment_t old_fs;
 
 	if ((regs->tstate & TSTATE_PRIV) || !(test_thread_flag(TIF_32BIT))) {
 		__asm__ __volatile__ ("flushw");
@@ -145,14 +139,10 @@ static void show_regwindow(struct pt_regs *regs)
 		rwk = (struct reg_window *)
 			(regs->u_regs[14] + STACK_BIAS);
 		if (!(regs->tstate & TSTATE_PRIV)) {
-			old_fs = get_fs();
-			set_fs (USER_DS);
 			if (copy_from_user (&r_w, rw, sizeof(r_w))) {
-				set_fs (old_fs);
 				return;
 			}
 			rwk = &r_w;
-			set_fs (old_fs);			
 		}
 	} else {
 		show_regwindow32(regs);
@@ -598,7 +588,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg,
 		memset(child_trap_frame, 0, child_stack_sz);
 		__thread_flag_byte_ptr(t)[TI_FLAG_BYTE_CWP] = 
 			(current_pt_regs()->tstate + 1) & TSTATE_CWP;
-		t->current_ds = ASI_P;
 		t->kregs->u_regs[UREG_G1] = sp; /* function */
 		t->kregs->u_regs[UREG_G2] = arg;
 		return 0;
@@ -613,7 +602,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg,
 	t->kregs->u_regs[UREG_FP] = sp;
 	__thread_flag_byte_ptr(t)[TI_FLAG_BYTE_CWP] = 
 		(regs->tstate + 1) & TSTATE_CWP;
-	t->current_ds = ASI_AIUS;
 	if (sp != regs->u_regs[UREG_FP]) {
 		unsigned long csp;
 
diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c
index 21077821f427..5b4de4a89dec 100644
--- a/arch/sparc/kernel/traps_64.c
+++ b/arch/sparc/kernel/traps_64.c
@@ -2857,8 +2857,6 @@ void __init trap_init(void)
 		     TI_PRE_COUNT != offsetof(struct thread_info,
 					      preempt_count) ||
 		     TI_NEW_CHILD != offsetof(struct thread_info, new_child) ||
-		     TI_CURRENT_DS != offsetof(struct thread_info,
-						current_ds) ||
 		     TI_KUNA_REGS != offsetof(struct thread_info,
 					      kern_una_regs) ||
 		     TI_KUNA_INSN != offsetof(struct thread_info,
diff --git a/arch/sparc/lib/NGmemcpy.S b/arch/sparc/lib/NGmemcpy.S
index 8e4d22a6ba0b..ee51c1230689 100644
--- a/arch/sparc/lib/NGmemcpy.S
+++ b/arch/sparc/lib/NGmemcpy.S
@@ -10,8 +10,7 @@
 #include <asm/thread_info.h>
 #define GLOBAL_SPARE	%g7
 #define RESTORE_ASI(TMP)	\
-	ldub	[%g6 + TI_CURRENT_DS], TMP;  \
-	wr	TMP, 0x0, %asi;
+	wr	%g0, ASI_AIUS, %asi
 #else
 #define GLOBAL_SPARE	%g5
 #define RESTORE_ASI(TMP)	\
diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c
index 1b23639e2fcd..8b1911591581 100644
--- a/arch/sparc/mm/init_64.c
+++ b/arch/sparc/mm/init_64.c
@@ -709,9 +709,10 @@ static void __init inherit_prom_mappings(void)
 
 void prom_world(int enter)
 {
-	if (!enter)
-		set_fs(get_fs());
-
+	/*
+	 * No need to change the address space any more, just flush
+	 * the register windows
+	 */
 	__asm__ __volatile__("flushw");
 }
 
-- 
2.29.2

[PATCH v2 16/18] sh: remove CONFIG_SET_FS support

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

sh uses set_fs/get_fs only in one file, to handle address
errors in both user and kernel memory.

It already has an abstraction to differentiate between I/O
and memory, so adding a third class for kernel memory fits
into the same scheme and lets us kill off CONFIG_SET_FS.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/sh/Kconfig                   |  1 -
 arch/sh/include/asm/processor.h   |  1 -
 arch/sh/include/asm/segment.h     | 33 -------------------------------
 arch/sh/include/asm/thread_info.h |  2 --
 arch/sh/include/asm/uaccess.h     |  4 ----
 arch/sh/kernel/io_trapped.c       |  9 ++-------
 arch/sh/kernel/process_32.c       |  2 --
 arch/sh/kernel/traps_32.c         | 30 +++++++++++++++++-----------
 8 files changed, 21 insertions(+), 61 deletions(-)
 delete mode 100644 arch/sh/include/asm/segment.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 2474a04ceac4..f676e92b7d5b 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -65,7 +65,6 @@ config SUPERH
 	select PERF_EVENTS
 	select PERF_USE_VMALLOC
 	select RTC_LIB
-	select SET_FS
 	select SPARSE_IRQ
 	select TRACE_IRQFLAGS_SUPPORT
 	help
diff --git a/arch/sh/include/asm/processor.h b/arch/sh/include/asm/processor.h
index 3820d698846e..85a6c1c3c16e 100644
--- a/arch/sh/include/asm/processor.h
+++ b/arch/sh/include/asm/processor.h
@@ -3,7 +3,6 @@
 #define __ASM_SH_PROCESSOR_H
 
 #include <asm/cpu-features.h>
-#include <asm/segment.h>
 #include <asm/cache.h>
 
 #ifndef __ASSEMBLY__
diff --git a/arch/sh/include/asm/segment.h b/arch/sh/include/asm/segment.h
deleted file mode 100644
index 02e54a3335d6..000000000000
--- a/arch/sh/include/asm/segment.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef __ASM_SH_SEGMENT_H
-#define __ASM_SH_SEGMENT_H
-
-#ifndef __ASSEMBLY__
-
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-
-/*
- * The fs value determines whether argument validity checking should be
- * performed or not.  If get_fs() == USER_DS, checking is performed, with
- * get_fs() == KERNEL_DS, checking is bypassed.
- *
- * For historical reasons, these macros are grossly misnamed.
- */
-#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFFUL)
-#ifdef CONFIG_MMU
-#define USER_DS		MAKE_MM_SEG(PAGE_OFFSET)
-#else
-#define USER_DS		KERNEL_DS
-#endif
-
-#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
-
-#define get_fs()	(current_thread_info()->addr_limit)
-#define set_fs(x)	(current_thread_info()->addr_limit = (x))
-
-#endif /* __ASSEMBLY__ */
-#endif /* __ASM_SH_SEGMENT_H */
diff --git a/arch/sh/include/asm/thread_info.h b/arch/sh/include/asm/thread_info.h
index 598d0184ffea..b119b859a0a3 100644
--- a/arch/sh/include/asm/thread_info.h
+++ b/arch/sh/include/asm/thread_info.h
@@ -30,7 +30,6 @@ struct thread_info {
 	__u32			status;		/* thread synchronous flags */
 	__u32			cpu;
 	int			preempt_count; /* 0 => preemptable, <0 => BUG */
-	mm_segment_t		addr_limit;	/* thread address space */
 	unsigned long		previous_sp;	/* sp of previous stack in case
 						   of nested IRQ stacks */
 	__u8			supervisor_stack[0];
@@ -58,7 +57,6 @@ struct thread_info {
 	.status		= 0,			\
 	.cpu		= 0,			\
 	.preempt_count	= INIT_PREEMPT_COUNT,	\
-	.addr_limit	= KERNEL_DS,		\
 }
 
 /* how to get the current stack pointer from C */
diff --git a/arch/sh/include/asm/uaccess.h b/arch/sh/include/asm/uaccess.h
index ccd219d74851..a79609eb14be 100644
--- a/arch/sh/include/asm/uaccess.h
+++ b/arch/sh/include/asm/uaccess.h
@@ -2,11 +2,7 @@
 #ifndef __ASM_SH_UACCESS_H
 #define __ASM_SH_UACCESS_H
 
-#include <asm/segment.h>
 #include <asm/extable.h>
-
-#define user_addr_max()	(current_thread_info()->addr_limit.seg)
-
 #include <asm-generic/access_ok.h>
 
 /*
diff --git a/arch/sh/kernel/io_trapped.c b/arch/sh/kernel/io_trapped.c
index 004ad0130b10..e803b14ef12e 100644
--- a/arch/sh/kernel/io_trapped.c
+++ b/arch/sh/kernel/io_trapped.c
@@ -270,7 +270,6 @@ static struct mem_access trapped_io_access = {
 
 int handle_trapped_io(struct pt_regs *regs, unsigned long address)
 {
-	mm_segment_t oldfs;
 	insn_size_t instruction;
 	int tmp;
 
@@ -281,16 +280,12 @@ int handle_trapped_io(struct pt_regs *regs, unsigned long address)
 
 	WARN_ON(user_mode(regs));
 
-	oldfs = get_fs();
-	set_fs(KERNEL_DS);
-	if (copy_from_user(&instruction, (void *)(regs->pc),
-			   sizeof(instruction))) {
-		set_fs(oldfs);
+	if (copy_from_kernel_nofault(&instruction, (void *)(regs->pc),
+				     sizeof(instruction))) {
 		return 0;
 	}
 
 	tmp = handle_unaligned_access(instruction, regs,
 				      &trapped_io_access, 1, address);
-	set_fs(oldfs);
 	return tmp == 0;
 }
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index 1c28e3cddb60..ca01286a0610 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -123,7 +123,6 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg,
 #if defined(CONFIG_SH_FPU)
 		childregs->sr |= SR_FD;
 #endif
-		ti->addr_limit = KERNEL_DS;
 		ti->status &= ~TS_USEDFPU;
 		p->thread.fpu_counter = 0;
 		return 0;
@@ -132,7 +131,6 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg,
 
 	if (usp)
 		childregs->regs[15] = usp;
-	ti->addr_limit = USER_DS;
 
 	if (clone_flags & CLONE_SETTLS)
 		childregs->gbr = tls;
diff --git a/arch/sh/kernel/traps_32.c b/arch/sh/kernel/traps_32.c
index b3c715bc254b..6cdda3a621a1 100644
--- a/arch/sh/kernel/traps_32.c
+++ b/arch/sh/kernel/traps_32.c
@@ -75,6 +75,23 @@ static struct mem_access user_mem_access = {
 	copy_to_user,
 };
 
+static unsigned long copy_from_kernel_wrapper(void *dst, const void __user *src,
+					      unsigned long cnt)
+{
+	return copy_from_kernel_nofault(dst, (const void __force *)src, cnt);
+}
+
+static unsigned long copy_to_kernel_wrapper(void __user *dst, const void *src,
+					    unsigned long cnt)
+{
+	return copy_to_kernel_nofault((void __force *)dst, src, cnt);
+}
+
+static struct mem_access kernel_mem_access = {
+	copy_from_kernel_wrapper,
+	copy_to_kernel_wrapper,
+};
+
 /*
  * handle an instruction that does an unaligned memory access by emulating the
  * desired behaviour
@@ -473,7 +490,6 @@ asmlinkage void do_address_error(struct pt_regs *regs,
 				 unsigned long address)
 {
 	unsigned long error_code = 0;
-	mm_segment_t oldfs;
 	insn_size_t instruction;
 	int tmp;
 
@@ -489,13 +505,10 @@ asmlinkage void do_address_error(struct pt_regs *regs,
 		local_irq_enable();
 		inc_unaligned_user_access();
 
-		oldfs = force_uaccess_begin();
 		if (copy_from_user(&instruction, (insn_size_t __user *)(regs->pc & ~1),
 				   sizeof(instruction))) {
-			force_uaccess_end(oldfs);
 			goto uspace_segv;
 		}
-		force_uaccess_end(oldfs);
 
 		/* shout about userspace fixups */
 		unaligned_fixups_notify(current, instruction, regs);
@@ -518,11 +531,9 @@ asmlinkage void do_address_error(struct pt_regs *regs,
 			goto uspace_segv;
 		}
 
-		oldfs = force_uaccess_begin();
 		tmp = handle_unaligned_access(instruction, regs,
 					      &user_mem_access, 0,
 					      address);
-		force_uaccess_end(oldfs);
 
 		if (tmp == 0)
 			return; /* sorted */
@@ -538,21 +549,18 @@ asmlinkage void do_address_error(struct pt_regs *regs,
 		if (regs->pc & 1)
 			die("unaligned program counter", regs, error_code);
 
-		set_fs(KERNEL_DS);
-		if (copy_from_user(&instruction, (void __user *)(regs->pc),
+		if (copy_from_kernel_nofault(&instruction, (void *)(regs->pc),
 				   sizeof(instruction))) {
 			/* Argh. Fault on the instruction itself.
 			   This should never happen non-SMP
 			*/
-			set_fs(oldfs);
 			die("insn faulting in do_address_error", regs, 0);
 		}
 
 		unaligned_fixups_notify(current, instruction, regs);
 
-		handle_unaligned_access(instruction, regs, &user_mem_access,
+		handle_unaligned_access(instruction, regs, &kernel_mem_access,
 					0, address);
-		set_fs(oldfs);
 	}
 }
 
-- 
2.29.2

[PATCH v2 17/18] ia64: remove CONFIG_SET_FS support

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

ia64 only uses set_fs() in one file to handle unaligned access for
both user space and kernel instructions. Rewrite this to explicitly
pass around a flag about which one it is and drop the feature from
the architecture.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/ia64/Kconfig                   |  1 -
 arch/ia64/include/asm/processor.h   |  4 --
 arch/ia64/include/asm/thread_info.h |  2 -
 arch/ia64/include/asm/uaccess.h     | 21 +++-------
 arch/ia64/kernel/unaligned.c        | 60 +++++++++++++++++++----------
 5 files changed, 45 insertions(+), 43 deletions(-)

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index a7e01573abd8..6b6a35b3d959 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -61,7 +61,6 @@ config IA64
 	select NEED_SG_DMA_LENGTH
 	select NUMA if !FLATMEM
 	select PCI_MSI_ARCH_FALLBACKS if PCI_MSI
-	select SET_FS
 	select ZONE_DMA32
 	default y
 	help
diff --git a/arch/ia64/include/asm/processor.h b/arch/ia64/include/asm/processor.h
index 45365c2ef598..7cbce290f4e5 100644
--- a/arch/ia64/include/asm/processor.h
+++ b/arch/ia64/include/asm/processor.h
@@ -243,10 +243,6 @@ DECLARE_PER_CPU(struct cpuinfo_ia64, ia64_cpu_info);
 
 extern void print_cpu_info (struct cpuinfo_ia64 *);
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
 #define SET_UNALIGN_CTL(task,value)								\
 ({												\
 	(task)->thread.flags = (((task)->thread.flags & ~IA64_THREAD_UAC_MASK)			\
diff --git a/arch/ia64/include/asm/thread_info.h b/arch/ia64/include/asm/thread_info.h
index 51d20cb37706..ef83493e6778 100644
--- a/arch/ia64/include/asm/thread_info.h
+++ b/arch/ia64/include/asm/thread_info.h
@@ -27,7 +27,6 @@ struct thread_info {
 	__u32 cpu;			/* current CPU */
 	__u32 last_cpu;			/* Last CPU thread ran on */
 	__u32 status;			/* Thread synchronous flags */
-	mm_segment_t addr_limit;	/* user-level address space limit */
 	int preempt_count;		/* 0=premptable, <0=BUG; will also serve as bh-counter */
 #ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
 	__u64 utime;
@@ -48,7 +47,6 @@ struct thread_info {
 	.task		= &tsk,			\
 	.flags		= 0,			\
 	.cpu		= 0,			\
-	.addr_limit	= KERNEL_DS,		\
 	.preempt_count	= INIT_PREEMPT_COUNT,	\
 }
 
diff --git a/arch/ia64/include/asm/uaccess.h b/arch/ia64/include/asm/uaccess.h
index e242a3cc1330..60adadeb3e9e 100644
--- a/arch/ia64/include/asm/uaccess.h
+++ b/arch/ia64/include/asm/uaccess.h
@@ -42,26 +42,17 @@
 #include <asm/extable.h>
 
 /*
- * For historical reasons, the following macros are grossly misnamed:
- */
-#define KERNEL_DS	((mm_segment_t) { ~0UL })		/* cf. access_ok() */
-#define USER_DS		((mm_segment_t) { TASK_SIZE-1 })	/* cf. access_ok() */
-
-#define get_fs()  (current_thread_info()->addr_limit)
-#define set_fs(x) (current_thread_info()->addr_limit = (x))
-
-/*
- * When accessing user memory, we need to make sure the entire area really is in
- * user-level space.  In order to do this efficiently, we make sure that the page at
- * address TASK_SIZE is never valid.  We also need to make sure that the address doesn't
+ * When accessing user memory, we need to make sure the entire area really is
+ * in user-level space.  We also need to make sure that the address doesn't
  * point inside the virtually mapped linear page table.
  */
 static inline int __access_ok(const void __user *p, unsigned long size)
 {
+	unsigned long limit = TASK_SIZE;
 	unsigned long addr = (unsigned long)p;
-	unsigned long seg = get_fs().seg;
-	return likely(addr <= seg) &&
-	 (seg == KERNEL_DS.seg || likely(REGION_OFFSET(addr) < RGN_MAP_LIMIT));
+
+	return likely((size <= limit) && (addr <= (limit - size)) &&
+		 likely(REGION_OFFSET(addr) < RGN_MAP_LIMIT));
 }
 #define __access_ok __access_ok
 #include <asm-generic/access_ok.h>
diff --git a/arch/ia64/kernel/unaligned.c b/arch/ia64/kernel/unaligned.c
index 6c1a8951dfbb..0acb5a0cd7ab 100644
--- a/arch/ia64/kernel/unaligned.c
+++ b/arch/ia64/kernel/unaligned.c
@@ -749,9 +749,25 @@ emulate_load_updates (update_t type, load_store_t ld, struct pt_regs *regs, unsi
 	}
 }
 
+static int emulate_store(unsigned long ifa, void *val, int len, bool kernel_mode)
+{
+	if (kernel_mode)
+		return copy_to_kernel_nofault((void *)ifa, val, len);
+
+	return copy_to_user((void __user *)ifa, val, len);
+}
+
+static int emulate_load(void *val, unsigned long ifa, int len, bool kernel_mode)
+{
+	if (kernel_mode)
+	       return copy_from_kernel_nofault(val, (void *)ifa, len);
+
+	return copy_from_user(val, (void __user *)ifa, len);
+}
 
 static int
-emulate_load_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
+emulate_load_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs,
+		  bool kernel_mode)
 {
 	unsigned int len = 1 << ld.x6_sz;
 	unsigned long val = 0;
@@ -774,7 +790,7 @@ emulate_load_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
 		return -1;
 	}
 	/* this assumes little-endian byte-order: */
-	if (copy_from_user(&val, (void __user *) ifa, len))
+	if (emulate_load(&val, ifa, len, kernel_mode))
 		return -1;
 	setreg(ld.r1, val, 0, regs);
 
@@ -872,7 +888,8 @@ emulate_load_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
 }
 
 static int
-emulate_store_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
+emulate_store_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs,
+		   bool kernel_mode)
 {
 	unsigned long r2;
 	unsigned int len = 1 << ld.x6_sz;
@@ -901,7 +918,7 @@ emulate_store_int (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
 	}
 
 	/* this assumes little-endian byte-order: */
-	if (copy_to_user((void __user *) ifa, &r2, len))
+	if (emulate_store(ifa, &r2, len, kernel_mode))
 		return -1;
 
 	/*
@@ -1021,7 +1038,7 @@ float2mem_double (struct ia64_fpreg *init, struct ia64_fpreg *final)
 }
 
 static int
-emulate_load_floatpair (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
+emulate_load_floatpair (unsigned long ifa, load_store_t ld, struct pt_regs *regs, bool kernel_mode)
 {
 	struct ia64_fpreg fpr_init[2];
 	struct ia64_fpreg fpr_final[2];
@@ -1050,8 +1067,8 @@ emulate_load_floatpair (unsigned long ifa, load_store_t ld, struct pt_regs *regs
 		 * This assumes little-endian byte-order.  Note that there is no "ldfpe"
 		 * instruction:
 		 */
-		if (copy_from_user(&fpr_init[0], (void __user *) ifa, len)
-		    || copy_from_user(&fpr_init[1], (void __user *) (ifa + len), len))
+		if (emulate_load(&fpr_init[0], ifa, len, kernel_mode)
+		    || emulate_load(&fpr_init[1], (ifa + len), len, kernel_mode))
 			return -1;
 
 		DPRINT("ld.r1=%d ld.imm=%d x6_sz=%d\n", ld.r1, ld.imm, ld.x6_sz);
@@ -1126,7 +1143,8 @@ emulate_load_floatpair (unsigned long ifa, load_store_t ld, struct pt_regs *regs
 
 
 static int
-emulate_load_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
+emulate_load_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs,
+	            bool kernel_mode)
 {
 	struct ia64_fpreg fpr_init;
 	struct ia64_fpreg fpr_final;
@@ -1152,7 +1170,7 @@ emulate_load_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
 	 * See comments in ldX for descriptions on how the various loads are handled.
 	 */
 	if (ld.x6_op != 0x2) {
-		if (copy_from_user(&fpr_init, (void __user *) ifa, len))
+		if (emulate_load(&fpr_init, ifa, len, kernel_mode))
 			return -1;
 
 		DPRINT("ld.r1=%d x6_sz=%d\n", ld.r1, ld.x6_sz);
@@ -1202,7 +1220,8 @@ emulate_load_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
 
 
 static int
-emulate_store_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
+emulate_store_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs,
+		     bool kernel_mode)
 {
 	struct ia64_fpreg fpr_init;
 	struct ia64_fpreg fpr_final;
@@ -1244,7 +1263,7 @@ emulate_store_float (unsigned long ifa, load_store_t ld, struct pt_regs *regs)
 	DDUMP("fpr_init =", &fpr_init, len);
 	DDUMP("fpr_final =", &fpr_final, len);
 
-	if (copy_to_user((void __user *) ifa, &fpr_final, len))
+	if (emulate_store(ifa, &fpr_final, len, kernel_mode))
 		return -1;
 
 	/*
@@ -1295,7 +1314,6 @@ void
 ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 {
 	struct ia64_psr *ipsr = ia64_psr(regs);
-	mm_segment_t old_fs = get_fs();
 	unsigned long bundle[2];
 	unsigned long opcode;
 	const struct exception_table_entry *eh = NULL;
@@ -1304,6 +1322,7 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 		load_store_t insn;
 	} u;
 	int ret = -1;
+	bool kernel_mode = false;
 
 	if (ia64_psr(regs)->be) {
 		/* we don't support big-endian accesses */
@@ -1367,13 +1386,13 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 			if (unaligned_dump_stack)
 				dump_stack();
 		}
-		set_fs(KERNEL_DS);
+		kernel_mode = true;
 	}
 
 	DPRINT("iip=%lx ifa=%lx isr=%lx (ei=%d, sp=%d)\n",
 	       regs->cr_iip, ifa, regs->cr_ipsr, ipsr->ri, ipsr->it);
 
-	if (__copy_from_user(bundle, (void __user *) regs->cr_iip, 16))
+	if (emulate_load(bundle, regs->cr_iip, 16, kernel_mode))
 		goto failure;
 
 	/*
@@ -1467,7 +1486,7 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 	      case LDCCLR_IMM_OP:
 	      case LDCNC_IMM_OP:
 	      case LDCCLRACQ_IMM_OP:
-		ret = emulate_load_int(ifa, u.insn, regs);
+		ret = emulate_load_int(ifa, u.insn, regs, kernel_mode);
 		break;
 
 	      case ST_OP:
@@ -1478,7 +1497,7 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 		fallthrough;
 	      case ST_IMM_OP:
 	      case STREL_IMM_OP:
-		ret = emulate_store_int(ifa, u.insn, regs);
+		ret = emulate_store_int(ifa, u.insn, regs, kernel_mode);
 		break;
 
 	      case LDF_OP:
@@ -1486,21 +1505,21 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 	      case LDFCCLR_OP:
 	      case LDFCNC_OP:
 		if (u.insn.x)
-			ret = emulate_load_floatpair(ifa, u.insn, regs);
+			ret = emulate_load_floatpair(ifa, u.insn, regs, kernel_mode);
 		else
-			ret = emulate_load_float(ifa, u.insn, regs);
+			ret = emulate_load_float(ifa, u.insn, regs, kernel_mode);
 		break;
 
 	      case LDF_IMM_OP:
 	      case LDFA_IMM_OP:
 	      case LDFCCLR_IMM_OP:
 	      case LDFCNC_IMM_OP:
-		ret = emulate_load_float(ifa, u.insn, regs);
+		ret = emulate_load_float(ifa, u.insn, regs, kernel_mode);
 		break;
 
 	      case STF_OP:
 	      case STF_IMM_OP:
-		ret = emulate_store_float(ifa, u.insn, regs);
+		ret = emulate_store_float(ifa, u.insn, regs, kernel_mode);
 		break;
 
 	      default:
@@ -1521,7 +1540,6 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
 
 	DPRINT("ipsr->ri=%d iip=%lx\n", ipsr->ri, regs->cr_iip);
   done:
-	set_fs(old_fs);		/* restore original address limit */
 	return;
 
   failure:
-- 
2.29.2

[PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

There are no remaining callers of set_fs(), so CONFIG_SET_FS
can be removed globally, along with the thread_info field and
any references to it.

This turns access_ok() into a cheaper check against TASK_SIZE_MAX.

With CONFIG_SET_FS gone, so drop all remaining references to
set_fs()/get_fs(), mm_segment_t and uaccess_kernel().

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 arch/Kconfig                              |  3 -
 arch/alpha/Kconfig                        |  1 -
 arch/alpha/include/asm/processor.h        |  4 --
 arch/alpha/include/asm/thread_info.h      |  2 -
 arch/alpha/include/asm/uaccess.h          | 19 ------
 arch/arc/Kconfig                          |  1 -
 arch/arc/include/asm/segment.h            | 20 -------
 arch/arc/include/asm/thread_info.h        |  3 -
 arch/arc/include/asm/uaccess.h            |  1 -
 arch/arm/lib/uaccess_with_memcpy.c        | 10 ----
 arch/csky/Kconfig                         |  1 -
 arch/csky/include/asm/processor.h         |  2 -
 arch/csky/include/asm/segment.h           | 10 ----
 arch/csky/include/asm/thread_info.h       |  2 -
 arch/csky/include/asm/uaccess.h           |  3 -
 arch/csky/kernel/asm-offsets.c            |  1 -
 arch/h8300/Kconfig                        |  1 -
 arch/h8300/include/asm/processor.h        |  1 -
 arch/h8300/include/asm/segment.h          | 40 -------------
 arch/h8300/include/asm/thread_info.h      |  3 -
 arch/h8300/kernel/entry.S                 |  1 -
 arch/h8300/kernel/head_ram.S              |  1 -
 arch/h8300/mm/init.c                      |  6 --
 arch/h8300/mm/memory.c                    |  1 -
 arch/hexagon/Kconfig                      |  1 -
 arch/hexagon/include/asm/thread_info.h    |  6 --
 arch/hexagon/kernel/process.c             |  1 -
 arch/microblaze/Kconfig                   |  1 -
 arch/microblaze/include/asm/thread_info.h |  6 --
 arch/microblaze/include/asm/uaccess.h     | 24 --------
 arch/microblaze/kernel/asm-offsets.c      |  1 -
 arch/microblaze/kernel/process.c          |  1 -
 arch/nds32/Kconfig                        |  1 -
 arch/nds32/include/asm/thread_info.h      |  4 --
 arch/nds32/include/asm/uaccess.h          | 15 +----
 arch/nds32/kernel/process.c               |  5 +-
 arch/nds32/mm/alignment.c                 |  3 -
 arch/nios2/Kconfig                        |  1 -
 arch/nios2/include/asm/thread_info.h      |  9 ---
 arch/nios2/include/asm/uaccess.h          | 12 ----
 arch/openrisc/Kconfig                     |  1 -
 arch/openrisc/include/asm/thread_info.h   |  7 ---
 arch/openrisc/include/asm/uaccess.h       | 23 --------
 arch/parisc/include/asm/futex.h           |  6 --
 arch/parisc/lib/memcpy.c                  |  2 +-
 arch/sparc/Kconfig                        |  2 +-
 arch/sparc/include/asm/processor_32.h     |  6 --
 arch/sparc/include/asm/uaccess_32.h       | 13 -----
 arch/sparc/kernel/process_32.c            |  2 -
 arch/xtensa/Kconfig                       |  1 -
 arch/xtensa/include/asm/asm-uaccess.h     | 71 -----------------------
 arch/xtensa/include/asm/processor.h       |  7 ---
 arch/xtensa/include/asm/thread_info.h     |  3 -
 arch/xtensa/include/asm/uaccess.h         | 16 -----
 arch/xtensa/kernel/asm-offsets.c          |  3 -
 drivers/hid/uhid.c                        |  2 +-
 drivers/scsi/sg.c                         |  5 --
 fs/exec.c                                 |  6 --
 include/asm-generic/access_ok.h           | 10 +---
 include/asm-generic/uaccess.h             | 25 +-------
 include/linux/syscalls.h                  |  4 --
 include/linux/uaccess.h                   | 33 -----------
 include/rdma/ib.h                         |  2 +-
 kernel/events/callchain.c                 |  4 --
 kernel/events/core.c                      |  3 -
 kernel/exit.c                             | 14 -----
 kernel/kthread.c                          |  5 --
 kernel/stacktrace.c                       |  3 -
 kernel/trace/bpf_trace.c                  |  4 --
 mm/maccess.c                              | 11 ----
 mm/memory.c                               |  8 ---
 net/bpfilter/bpfilter_kern.c              |  2 +-
 72 files changed, 10 insertions(+), 522 deletions(-)
 delete mode 100644 arch/arc/include/asm/segment.h
 delete mode 100644 arch/csky/include/asm/segment.h
 delete mode 100644 arch/h8300/include/asm/segment.h

diff --git a/arch/Kconfig b/arch/Kconfig
index fa5db36bda67..99349547afed 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -24,9 +24,6 @@ config KEXEC_ELF
 config HAVE_IMA_KEXEC
 	bool
 
-config SET_FS
-	bool
-
 config HOTPLUG_SMT
 	bool
 
diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 4e87783c90ad..eee8b5b0a58b 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -35,7 +35,6 @@ config ALPHA
 	select OLD_SIGSUSPEND
 	select CPU_NO_EFFICIENT_FFS if !ALPHA_EV67
 	select MMU_GATHER_NO_RANGE
-	select SET_FS
 	select SPARSEMEM_EXTREME if SPARSEMEM
 	select ZONE_DMA
 	help
diff --git a/arch/alpha/include/asm/processor.h b/arch/alpha/include/asm/processor.h
index 090499c99c1c..43e234c518b1 100644
--- a/arch/alpha/include/asm/processor.h
+++ b/arch/alpha/include/asm/processor.h
@@ -26,10 +26,6 @@
 #define TASK_UNMAPPED_BASE \
   ((current->personality & ADDR_LIMIT_32BIT) ? 0x40000000 : TASK_SIZE / 2)
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
 /* This is dead.  Everything has been moved to thread_info.  */
 struct thread_struct { };
 #define INIT_THREAD  { }
diff --git a/arch/alpha/include/asm/thread_info.h b/arch/alpha/include/asm/thread_info.h
index 2592356e3215..fdc485d7787a 100644
--- a/arch/alpha/include/asm/thread_info.h
+++ b/arch/alpha/include/asm/thread_info.h
@@ -19,7 +19,6 @@ struct thread_info {
 	unsigned int		flags;		/* low level flags */
 	unsigned int		ieee_state;	/* see fpu.h */
 
-	mm_segment_t		addr_limit;	/* thread address space */
 	unsigned		cpu;		/* current CPU */
 	int			preempt_count; /* 0 => preemptable, <0 => BUG */
 	unsigned int		status;		/* thread-synchronous flags */
@@ -35,7 +34,6 @@ struct thread_info {
 #define INIT_THREAD_INFO(tsk)			\
 {						\
 	.task		= &tsk,			\
-	.addr_limit	= KERNEL_DS,		\
 	.preempt_count	= INIT_PREEMPT_COUNT,	\
 }
 
diff --git a/arch/alpha/include/asm/uaccess.h b/arch/alpha/include/asm/uaccess.h
index 82c5743fc9cd..c32c2584c0b7 100644
--- a/arch/alpha/include/asm/uaccess.h
+++ b/arch/alpha/include/asm/uaccess.h
@@ -2,26 +2,7 @@
 #ifndef __ALPHA_UACCESS_H
 #define __ALPHA_UACCESS_H
 
-/*
- * The fs value determines whether argument validity checking should be
- * performed or not.  If get_fs() == USER_DS, checking is performed, with
- * get_fs() == KERNEL_DS, checking is bypassed.
- *
- * Or at least it did once upon a time.  Nowadays it is a mask that
- * defines which bits of the address space are off limits.  This is a
- * wee bit faster than the above.
- *
- * For historical reasons, these macros are grossly misnamed.
- */
-
-#define KERNEL_DS	((mm_segment_t) { 0UL })
-#define USER_DS		((mm_segment_t) { -0x40000000000UL })
-
-#define get_fs()  (current_thread_info()->addr_limit)
-#define set_fs(x) (current_thread_info()->addr_limit = (x))
-
 #include <asm-generic/access_ok.h>
-
 /*
  * These are the main single-value transfer routines.  They automatically
  * use the right size if we just have the right pointer type.
diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 3c2a4753d09b..e0a60a27e14d 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -45,7 +45,6 @@ config ARC
 	select PCI_SYSCALL if PCI
 	select PERF_USE_VMALLOC if ARC_CACHE_VIPT_ALIASING
 	select HAVE_ARCH_JUMP_LABEL if ISA_ARCV2 && !CPU_ENDIAN_BE32
-	select SET_FS
 	select TRACE_IRQFLAGS_SUPPORT
 
 config LOCKDEP_SUPPORT
diff --git a/arch/arc/include/asm/segment.h b/arch/arc/include/asm/segment.h
deleted file mode 100644
index 871f8ab11bfd..000000000000
--- a/arch/arc/include/asm/segment.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com)
- */
-
-#ifndef __ASMARC_SEGMENT_H
-#define __ASMARC_SEGMENT_H
-
-#ifndef __ASSEMBLY__
-
-typedef unsigned long mm_segment_t;
-
-#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-
-#define KERNEL_DS		MAKE_MM_SEG(0)
-#define USER_DS			MAKE_MM_SEG(TASK_SIZE)
-#define uaccess_kernel()	(get_fs() == KERNEL_DS)
-
-#endif /* __ASSEMBLY__ */
-#endif /* __ASMARC_SEGMENT_H */
diff --git a/arch/arc/include/asm/thread_info.h b/arch/arc/include/asm/thread_info.h
index d36863e34bfc..1e0b2e3914d5 100644
--- a/arch/arc/include/asm/thread_info.h
+++ b/arch/arc/include/asm/thread_info.h
@@ -27,7 +27,6 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/thread_info.h>
-#include <asm/segment.h>
 
 /*
  * low level task data that entry.S needs immediate access to
@@ -40,7 +39,6 @@ struct thread_info {
 	unsigned long flags;		/* low level flags */
 	int preempt_count;		/* 0 => preemptable, <0 => BUG */
 	struct task_struct *task;	/* main task structure */
-	mm_segment_t addr_limit;	/* thread address space */
 	__u32 cpu;			/* current CPU */
 	unsigned long thr_ptr;		/* TLS ptr */
 };
@@ -56,7 +54,6 @@ struct thread_info {
 	.flags      = 0,			\
 	.cpu        = 0,			\
 	.preempt_count  = INIT_PREEMPT_COUNT,	\
-	.addr_limit = KERNEL_DS,		\
 }
 
 static inline __attribute_const__ struct thread_info *current_thread_info(void)
diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h
index 30f80b4be2ab..99712471c96a 100644
--- a/arch/arc/include/asm/uaccess.h
+++ b/arch/arc/include/asm/uaccess.h
@@ -638,7 +638,6 @@ extern unsigned long arc_clear_user_noinline(void __user *to,
 #define __clear_user(d, n)		arc_clear_user_noinline(d, n)
 #endif
 
-#include <asm/segment.h>
 #include <asm-generic/uaccess.h>
 
 #endif
diff --git a/arch/arm/lib/uaccess_with_memcpy.c b/arch/arm/lib/uaccess_with_memcpy.c
index 106f83a5ea6d..c30b689bec2e 100644
--- a/arch/arm/lib/uaccess_with_memcpy.c
+++ b/arch/arm/lib/uaccess_with_memcpy.c
@@ -92,11 +92,6 @@ __copy_to_user_memcpy(void __user *to, const void *from, unsigned long n)
 	unsigned long ua_flags;
 	int atomic;
 
-	if (uaccess_kernel()) {
-		memcpy((void *)to, from, n);
-		return 0;
-	}
-
 	/* the mmap semaphore is taken only if not in an atomic context */
 	atomic = faulthandler_disabled();
 
@@ -165,11 +160,6 @@ __clear_user_memset(void __user *addr, unsigned long n)
 {
 	unsigned long ua_flags;
 
-	if (uaccess_kernel()) {
-		memset((void *)addr, 0, n);
-		return 0;
-	}
-
 	mmap_read_lock(current->mm);
 	while (n) {
 		pte_t *pte;
diff --git a/arch/csky/Kconfig b/arch/csky/Kconfig
index 132f43f12dd8..75ef86605d69 100644
--- a/arch/csky/Kconfig
+++ b/arch/csky/Kconfig
@@ -79,7 +79,6 @@ config CSKY
 	select PCI_DOMAINS_GENERIC if PCI
 	select PCI_SYSCALL if PCI
 	select PCI_MSI if PCI
-	select SET_FS
 	select TRACE_IRQFLAGS_SUPPORT
 
 config LOCKDEP_SUPPORT
diff --git a/arch/csky/include/asm/processor.h b/arch/csky/include/asm/processor.h
index 817dd60ff152..688c7548b559 100644
--- a/arch/csky/include/asm/processor.h
+++ b/arch/csky/include/asm/processor.h
@@ -4,7 +4,6 @@
 #define __ASM_CSKY_PROCESSOR_H
 
 #include <linux/bitops.h>
-#include <asm/segment.h>
 #include <asm/ptrace.h>
 #include <asm/current.h>
 #include <asm/cache.h>
@@ -59,7 +58,6 @@ struct thread_struct {
  */
 #define start_thread(_regs, _pc, _usp)					\
 do {									\
-	set_fs(USER_DS); /* reads from user space */			\
 	(_regs)->pc = (_pc);						\
 	(_regs)->regs[1] = 0; /* ABIV1 is R7, uClibc_main rtdl arg */	\
 	(_regs)->regs[2] = 0;						\
diff --git a/arch/csky/include/asm/segment.h b/arch/csky/include/asm/segment.h
deleted file mode 100644
index 5bc1cc62b87f..000000000000
--- a/arch/csky/include/asm/segment.h
+++ /dev/null
@@ -1,10 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-
-#ifndef __ASM_CSKY_SEGMENT_H
-#define __ASM_CSKY_SEGMENT_H
-
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-#endif /* __ASM_CSKY_SEGMENT_H */
diff --git a/arch/csky/include/asm/thread_info.h b/arch/csky/include/asm/thread_info.h
index 8c349a8f904d..b5ed788f0c68 100644
--- a/arch/csky/include/asm/thread_info.h
+++ b/arch/csky/include/asm/thread_info.h
@@ -16,7 +16,6 @@ struct thread_info {
 	unsigned long		flags;
 	int			preempt_count;
 	unsigned long		tp_value;
-	mm_segment_t		addr_limit;
 	struct restart_block	restart_block;
 	struct pt_regs		*regs;
 	unsigned int		cpu;
@@ -26,7 +25,6 @@ struct thread_info {
 {						\
 	.task		= &tsk,			\
 	.preempt_count  = INIT_PREEMPT_COUNT,	\
-	.addr_limit     = KERNEL_DS,		\
 	.cpu		= 0,			\
 	.restart_block = {			\
 		.fn = do_no_restart_syscall,	\
diff --git a/arch/csky/include/asm/uaccess.h b/arch/csky/include/asm/uaccess.h
index fec8f77ffc99..2e927c21d8a1 100644
--- a/arch/csky/include/asm/uaccess.h
+++ b/arch/csky/include/asm/uaccess.h
@@ -3,8 +3,6 @@
 #ifndef __ASM_CSKY_UACCESS_H
 #define __ASM_CSKY_UACCESS_H
 
-#define user_addr_max() (current_thread_info()->addr_limit.seg)
-
 /*
  * __put_user_fn
  */
@@ -200,7 +198,6 @@ unsigned long raw_copy_to_user(void *to, const void *from, unsigned long n);
 unsigned long __clear_user(void __user *to, unsigned long n);
 #define __clear_user __clear_user
 
-#include <asm/segment.h>
 #include <asm-generic/uaccess.h>
 
 #endif /* __ASM_CSKY_UACCESS_H */
diff --git a/arch/csky/kernel/asm-offsets.c b/arch/csky/kernel/asm-offsets.c
index 1cbcba4b0dd1..d1e903579473 100644
--- a/arch/csky/kernel/asm-offsets.c
+++ b/arch/csky/kernel/asm-offsets.c
@@ -25,7 +25,6 @@ int main(void)
 	/* offsets into the thread_info struct */
 	DEFINE(TINFO_FLAGS,       offsetof(struct thread_info, flags));
 	DEFINE(TINFO_PREEMPT,     offsetof(struct thread_info, preempt_count));
-	DEFINE(TINFO_ADDR_LIMIT,  offsetof(struct thread_info, addr_limit));
 	DEFINE(TINFO_TP_VALUE,   offsetof(struct thread_info, tp_value));
 	DEFINE(TINFO_TASK,        offsetof(struct thread_info, task));
 
diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index 3e3e0f16f7e0..fe48c4f26cc8 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -24,7 +24,6 @@ config H8300
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_HASH
 	select CPU_NO_EFFICIENT_FFS
-	select SET_FS
 	select UACCESS_MEMCPY
 
 config CPU_BIG_ENDIAN
diff --git a/arch/h8300/include/asm/processor.h b/arch/h8300/include/asm/processor.h
index 141a23eb62b7..ba171aa4dacb 100644
--- a/arch/h8300/include/asm/processor.h
+++ b/arch/h8300/include/asm/processor.h
@@ -13,7 +13,6 @@
 #define __ASM_H8300_PROCESSOR_H
 
 #include <linux/compiler.h>
-#include <asm/segment.h>
 #include <asm/ptrace.h>
 #include <asm/current.h>
 
diff --git a/arch/h8300/include/asm/segment.h b/arch/h8300/include/asm/segment.h
deleted file mode 100644
index 37950725d9b9..000000000000
--- a/arch/h8300/include/asm/segment.h
+++ /dev/null
@@ -1,40 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _H8300_SEGMENT_H
-#define _H8300_SEGMENT_H
-
-/* define constants */
-#define USER_DATA     (1)
-#ifndef __USER_DS
-#define __USER_DS     (USER_DATA)
-#endif
-#define USER_PROGRAM  (2)
-#define SUPER_DATA    (3)
-#ifndef __KERNEL_DS
-#define __KERNEL_DS   (SUPER_DATA)
-#endif
-#define SUPER_PROGRAM (4)
-
-#ifndef __ASSEMBLY__
-
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
-#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-#define USER_DS		MAKE_MM_SEG(__USER_DS)
-#define KERNEL_DS	MAKE_MM_SEG(__KERNEL_DS)
-
-/*
- * Get/set the SFC/DFC registers for MOVES instructions
- */
-
-static inline mm_segment_t get_fs(void)
-{
-	return USER_DS;
-}
-
-#define uaccess_kernel()	(get_fs().seg == KERNEL_DS.seg)
-
-#endif /* __ASSEMBLY__ */
-
-#endif /* _H8300_SEGMENT_H */
diff --git a/arch/h8300/include/asm/thread_info.h b/arch/h8300/include/asm/thread_info.h
index a518214d4ddd..ff2d873749a4 100644
--- a/arch/h8300/include/asm/thread_info.h
+++ b/arch/h8300/include/asm/thread_info.h
@@ -10,7 +10,6 @@
 #define _ASM_THREAD_INFO_H
 
 #include <asm/page.h>
-#include <asm/segment.h>
 
 #ifdef __KERNEL__
 
@@ -31,7 +30,6 @@ struct thread_info {
 	unsigned long	   flags;		/* low level flags */
 	int		   cpu;			/* cpu we're on */
 	int		   preempt_count;	/* 0 => preemptable, <0 => BUG */
-	mm_segment_t		addr_limit;
 };
 
 /*
@@ -43,7 +41,6 @@ struct thread_info {
 	.flags =	0,			\
 	.cpu =		0,			\
 	.preempt_count = INIT_PREEMPT_COUNT,	\
-	.addr_limit	= KERNEL_DS,		\
 }
 
 /* how to get the thread information struct from C */
diff --git a/arch/h8300/kernel/entry.S b/arch/h8300/kernel/entry.S
index c6e289b5f1f2..42db87c17917 100644
--- a/arch/h8300/kernel/entry.S
+++ b/arch/h8300/kernel/entry.S
@@ -17,7 +17,6 @@
 #include <linux/sys.h>
 #include <asm/unistd.h>
 #include <asm/setup.h>
-#include <asm/segment.h>
 #include <asm/linkage.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
diff --git a/arch/h8300/kernel/head_ram.S b/arch/h8300/kernel/head_ram.S
index dbf8429f5fab..489462f0ee57 100644
--- a/arch/h8300/kernel/head_ram.S
+++ b/arch/h8300/kernel/head_ram.S
@@ -4,7 +4,6 @@
 #include <linux/init.h>
 #include <asm/unistd.h>
 #include <asm/setup.h>
-#include <asm/segment.h>
 #include <asm/linkage.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c
index f7bf4693e3b2..9fa13312720a 100644
--- a/arch/h8300/mm/init.c
+++ b/arch/h8300/mm/init.c
@@ -34,7 +34,6 @@
 #include <linux/gfp.h>
 
 #include <asm/setup.h>
-#include <asm/segment.h>
 #include <asm/page.h>
 #include <asm/sections.h>
 
@@ -71,11 +70,6 @@ void __init paging_init(void)
 		panic("%s: Failed to allocate %lu bytes align=0x%lx\n",
 		      __func__, PAGE_SIZE, PAGE_SIZE);
 
-	/*
-	 * Set up SFC/DFC registers (user data space).
-	 */
-	set_fs(USER_DS);
-
 	pr_debug("before free_area_init\n");
 
 	pr_debug("free_area_init -> start_mem is %#lx\nvirtual_end is %#lx\n",
diff --git a/arch/h8300/mm/memory.c b/arch/h8300/mm/memory.c
index 4a60e2b5eb96..c950571064d2 100644
--- a/arch/h8300/mm/memory.c
+++ b/arch/h8300/mm/memory.c
@@ -24,7 +24,6 @@
 #include <linux/types.h>
 
 #include <asm/setup.h>
-#include <asm/segment.h>
 #include <asm/page.h>
 #include <asm/traps.h>
 #include <asm/io.h>
diff --git a/arch/hexagon/Kconfig b/arch/hexagon/Kconfig
index 15dd8f38b698..54eadf265178 100644
--- a/arch/hexagon/Kconfig
+++ b/arch/hexagon/Kconfig
@@ -30,7 +30,6 @@ config HEXAGON
 	select GENERIC_CLOCKEVENTS_BROADCAST
 	select MODULES_USE_ELF_RELA
 	select GENERIC_CPU_DEVICES
-	select SET_FS
 	select ARCH_WANT_LD_ORPHAN_WARN
 	select TRACE_IRQFLAGS_SUPPORT
 	help
diff --git a/arch/hexagon/include/asm/thread_info.h b/arch/hexagon/include/asm/thread_info.h
index 535976665bf0..e90f280b9ce3 100644
--- a/arch/hexagon/include/asm/thread_info.h
+++ b/arch/hexagon/include/asm/thread_info.h
@@ -22,10 +22,6 @@
 
 #ifndef __ASSEMBLY__
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
 /*
  * This is union'd with the "bottom" of the kernel stack.
  * It keeps track of thread info which is handy for routines
@@ -37,7 +33,6 @@ struct thread_info {
 	unsigned long		flags;          /* low level flags */
 	__u32                   cpu;            /* current cpu */
 	int                     preempt_count;  /* 0=>preemptible,<0=>BUG */
-	mm_segment_t            addr_limit;     /* segmentation sux */
 	/*
 	 * used for syscalls somehow;
 	 * seems to have a function pointer and four arguments
@@ -66,7 +61,6 @@ struct thread_info {
 	.flags          = 0,                    \
 	.cpu            = 0,                    \
 	.preempt_count  = 1,                    \
-	.addr_limit     = KERNEL_DS,            \
 	.sp = 0,				\
 	.regs = NULL,			\
 }
diff --git a/arch/hexagon/kernel/process.c b/arch/hexagon/kernel/process.c
index 232dfd8956aa..dfa6b2757c05 100644
--- a/arch/hexagon/kernel/process.c
+++ b/arch/hexagon/kernel/process.c
@@ -105,7 +105,6 @@ int copy_thread(unsigned long clone_flags, unsigned long usp, unsigned long arg,
 	/*
 	 * Parent sees new pid -- not necessary, not even possible at
 	 * this point in the fork process
-	 * Might also want to set things like ti->addr_limit
 	 */
 
 	return 0;
diff --git a/arch/microblaze/Kconfig b/arch/microblaze/Kconfig
index 59798e43cdb0..1fb1cec087b7 100644
--- a/arch/microblaze/Kconfig
+++ b/arch/microblaze/Kconfig
@@ -42,7 +42,6 @@ config MICROBLAZE
 	select CPU_NO_EFFICIENT_FFS
 	select MMU_GATHER_NO_RANGE
 	select SPARSE_IRQ
-	select SET_FS
 	select ZONE_DMA
 	select TRACE_IRQFLAGS_SUPPORT
 
diff --git a/arch/microblaze/include/asm/thread_info.h b/arch/microblaze/include/asm/thread_info.h
index 44f5ca331862..a0ddd2a36fb9 100644
--- a/arch/microblaze/include/asm/thread_info.h
+++ b/arch/microblaze/include/asm/thread_info.h
@@ -56,17 +56,12 @@ struct cpu_context {
 	__u32	fsr;
 };
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
 struct thread_info {
 	struct task_struct	*task; /* main task structure */
 	unsigned long		flags; /* low level flags */
 	unsigned long		status; /* thread-synchronous flags */
 	__u32			cpu; /* current CPU */
 	__s32			preempt_count; /* 0 => preemptable,< 0 => BUG*/
-	mm_segment_t		addr_limit; /* thread address space */
 
 	struct cpu_context	cpu_context;
 };
@@ -80,7 +75,6 @@ struct thread_info {
 	.flags		= 0,			\
 	.cpu		= 0,			\
 	.preempt_count	= INIT_PREEMPT_COUNT,	\
-	.addr_limit	= KERNEL_DS,		\
 }
 
 /* how to get the thread information struct from C */
diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
index bf9b7657a65a..3aab2f17e046 100644
--- a/arch/microblaze/include/asm/uaccess.h
+++ b/arch/microblaze/include/asm/uaccess.h
@@ -15,30 +15,6 @@
 #include <linux/pgtable.h>
 #include <asm/extable.h>
 #include <linux/string.h>
-
-/*
- * On Microblaze the fs value is actually the top of the corresponding
- * address space.
- *
- * The fs value determines whether argument validity checking should be
- * performed or not. If get_fs() == USER_DS, checking is performed, with
- * get_fs() == KERNEL_DS, checking is bypassed.
- *
- * For historical reasons, these macros are grossly misnamed.
- *
- * For non-MMU arch like Microblaze, KERNEL_DS and USER_DS is equal.
- */
-# define MAKE_MM_SEG(s)       ((mm_segment_t) { (s) })
-
-#  define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFF)
-#  define USER_DS	MAKE_MM_SEG(TASK_SIZE - 1)
-
-# define get_fs()	(current_thread_info()->addr_limit)
-# define set_fs(val)	(current_thread_info()->addr_limit = (val))
-# define user_addr_max() get_fs().seg
-
-# define uaccess_kernel()	(get_fs().seg == KERNEL_DS.seg)
-
 #include <asm-generic/access_ok.h>
 
 # define __FIXUP_SECTION	".section .fixup,\"ax\"\n"
diff --git a/arch/microblaze/kernel/asm-offsets.c b/arch/microblaze/kernel/asm-offsets.c
index b77dd188dec4..47ee409508b1 100644
--- a/arch/microblaze/kernel/asm-offsets.c
+++ b/arch/microblaze/kernel/asm-offsets.c
@@ -86,7 +86,6 @@ int main(int argc, char *argv[])
 	/* struct thread_info */
 	DEFINE(TI_TASK, offsetof(struct thread_info, task));
 	DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
-	DEFINE(TI_ADDR_LIMIT, offsetof(struct thread_info, addr_limit));
 	DEFINE(TI_CPU_CONTEXT, offsetof(struct thread_info, cpu_context));
 	DEFINE(TI_PREEMPT_COUNT, offsetof(struct thread_info, preempt_count));
 	BLANK();
diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c
index 5e2b91c1e8ce..1b944d319d73 100644
--- a/arch/microblaze/kernel/process.c
+++ b/arch/microblaze/kernel/process.c
@@ -18,7 +18,6 @@
 #include <linux/tick.h>
 #include <linux/bitops.h>
 #include <linux/ptrace.h>
-#include <linux/uaccess.h> /* for USER_DS macros */
 #include <asm/cacheflush.h>
 
 void show_regs(struct pt_regs *regs)
diff --git a/arch/nds32/Kconfig b/arch/nds32/Kconfig
index 4d1421b18734..013249430fa3 100644
--- a/arch/nds32/Kconfig
+++ b/arch/nds32/Kconfig
@@ -44,7 +44,6 @@ config NDS32
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_FTRACE_MCOUNT_RECORD
 	select HAVE_DYNAMIC_FTRACE
-	select SET_FS
 	select TRACE_IRQFLAGS_SUPPORT
 	help
 	  Andes(nds32) Linux support.
diff --git a/arch/nds32/include/asm/thread_info.h b/arch/nds32/include/asm/thread_info.h
index d3967ad184f0..bd8f81cf2ce5 100644
--- a/arch/nds32/include/asm/thread_info.h
+++ b/arch/nds32/include/asm/thread_info.h
@@ -16,8 +16,6 @@ struct task_struct;
 #include <asm/ptrace.h>
 #include <asm/types.h>
 
-typedef unsigned long mm_segment_t;
-
 /*
  * low level task data that entry.S needs immediate access to.
  * __switch_to() assumes cpu_context follows immediately after cpu_domain.
@@ -25,12 +23,10 @@ typedef unsigned long mm_segment_t;
 struct thread_info {
 	unsigned long flags;	/* low level flags */
 	__s32 preempt_count;	/* 0 => preemptable, <0 => bug */
-	mm_segment_t addr_limit;	/* address limit */
 };
 #define INIT_THREAD_INFO(tsk)						\
 {									\
 	.preempt_count	= INIT_PREEMPT_COUNT,				\
-	.addr_limit	= KERNEL_DS,					\
 }
 #define thread_saved_pc(tsk) ((unsigned long)(tsk->thread.cpu_context.pc))
 #define thread_saved_fp(tsk) ((unsigned long)(tsk->thread.cpu_context.fp))
diff --git a/arch/nds32/include/asm/uaccess.h b/arch/nds32/include/asm/uaccess.h
index 832d642a4068..377548d4451a 100644
--- a/arch/nds32/include/asm/uaccess.h
+++ b/arch/nds32/include/asm/uaccess.h
@@ -11,6 +11,7 @@
 #include <asm/errno.h>
 #include <asm/memory.h>
 #include <asm/types.h>
+#include <asm-generic/access_ok.h>
 
 #define __asmeq(x, y)  ".ifnc " x "," y " ; .err ; .endif\n\t"
 
@@ -33,20 +34,6 @@ struct exception_table_entry {
 
 extern int fixup_exception(struct pt_regs *regs);
 
-#define KERNEL_DS 	((mm_segment_t) { ~0UL })
-#define USER_DS		((mm_segment_t) {TASK_SIZE - 1})
-
-#define get_fs()	(current_thread_info()->addr_limit)
-#define user_addr_max	get_fs
-#define uaccess_kernel() (get_fs() == KERNEL_DS)
-
-static inline void set_fs(mm_segment_t fs)
-{
-	current_thread_info()->addr_limit = fs;
-}
-
-#include <asm-generic/access_ok.h>
-
 /*
  * Single-value transfer routines.  They automatically use the right
  * size if we just have the right pointer type.  Note that the functions
diff --git a/arch/nds32/kernel/process.c b/arch/nds32/kernel/process.c
index 49fab9e39cbf..d35c1f63fa11 100644
--- a/arch/nds32/kernel/process.c
+++ b/arch/nds32/kernel/process.c
@@ -119,9 +119,8 @@ void show_regs(struct pt_regs *regs)
 		regs->uregs[7], regs->uregs[6], regs->uregs[5], regs->uregs[4]);
 	pr_info("r3 : %08lx  r2 : %08lx  r1 : %08lx  r0 : %08lx\n",
 		regs->uregs[3], regs->uregs[2], regs->uregs[1], regs->uregs[0]);
-	pr_info("  IRQs o%s  Segment %s\n",
-		interrupts_enabled(regs) ? "n" : "ff",
-		uaccess_kernel() ? "kernel" : "user");
+	pr_info("  IRQs o%s  Segment user\n",
+		interrupts_enabled(regs) ? "n" : "ff");
 }
 
 EXPORT_SYMBOL(show_regs);
diff --git a/arch/nds32/mm/alignment.c b/arch/nds32/mm/alignment.c
index 1eb7ded6992b..9c2c0a454da8 100644
--- a/arch/nds32/mm/alignment.c
+++ b/arch/nds32/mm/alignment.c
@@ -512,7 +512,6 @@ int do_unaligned_access(unsigned long addr, struct pt_regs *regs)
 {
 	unsigned long inst;
 	int ret = -EFAULT;
-	mm_segment_t seg;
 
 	inst = get_inst(regs->ipc);
 
@@ -520,12 +519,10 @@ int do_unaligned_access(unsigned long addr, struct pt_regs *regs)
 	      "Faulting addr: 0x%08lx, pc: 0x%08lx [inst: 0x%08lx ]\n", addr,
 	      regs->ipc, inst);
 
-	seg = force_uaccess_begin();
 	if (inst & NDS32_16BIT_INSTRUCTION)
 		ret = do_16((inst >> 16) & 0xffff, regs);
 	else
 		ret = do_32(inst, regs);
-	force_uaccess_end(seg);
 
 	return ret;
 }
diff --git a/arch/nios2/Kconfig b/arch/nios2/Kconfig
index 33fd06f5fa41..4167f1eb4cd8 100644
--- a/arch/nios2/Kconfig
+++ b/arch/nios2/Kconfig
@@ -24,7 +24,6 @@ config NIOS2
 	select USB_ARCH_HAS_HCD if USB_SUPPORT
 	select CPU_NO_EFFICIENT_FFS
 	select MMU_GATHER_NO_RANGE if MMU
-	select SET_FS
 
 config GENERIC_CSUM
 	def_bool y
diff --git a/arch/nios2/include/asm/thread_info.h b/arch/nios2/include/asm/thread_info.h
index 272d2c72a727..bcc0e9915ebd 100644
--- a/arch/nios2/include/asm/thread_info.h
+++ b/arch/nios2/include/asm/thread_info.h
@@ -26,10 +26,6 @@
 
 #ifndef __ASSEMBLY__
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
 /*
  * low level task data that entry.S needs immediate access to
  * - this struct should fit entirely inside of one cache line
@@ -42,10 +38,6 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;		/* current CPU */
 	int			preempt_count;	/* 0 => preemptable,<0 => BUG */
-	mm_segment_t		addr_limit;	/* thread address space:
-						  0-0x7FFFFFFF for user-thead
-						  0-0xFFFFFFFF for kernel-thread
-						*/
 	struct pt_regs		*regs;
 };
 
@@ -60,7 +52,6 @@ struct thread_info {
 	.flags		= 0,			\
 	.cpu		= 0,			\
 	.preempt_count	= INIT_PREEMPT_COUNT,	\
-	.addr_limit	= KERNEL_DS,		\
 }
 
 /* how to get the thread information struct from C */
diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
index 6664ddc0e8e5..b8299082adbe 100644
--- a/arch/nios2/include/asm/uaccess.h
+++ b/arch/nios2/include/asm/uaccess.h
@@ -18,18 +18,6 @@
 #include <asm/page.h>
 
 #include <asm/extable.h>
-
-/*
- * Segment stuff
- */
-#define MAKE_MM_SEG(s)		((mm_segment_t) { (s) })
-#define USER_DS			MAKE_MM_SEG(0x80000000UL)
-#define KERNEL_DS		MAKE_MM_SEG(0)
-
-
-#define get_fs()		(current_thread_info()->addr_limit)
-#define set_fs(seg)		(current_thread_info()->addr_limit = (seg))
-
 #include <asm-generic/access_ok.h>
 
 # define __EX_TABLE_SECTION	".section __ex_table,\"a\"\n"
diff --git a/arch/openrisc/Kconfig b/arch/openrisc/Kconfig
index f724b3f1aeed..0d68adf6e02b 100644
--- a/arch/openrisc/Kconfig
+++ b/arch/openrisc/Kconfig
@@ -36,7 +36,6 @@ config OPENRISC
 	select ARCH_WANT_FRAME_POINTERS
 	select GENERIC_IRQ_MULTI_HANDLER
 	select MMU_GATHER_NO_RANGE if MMU
-	select SET_FS
 	select TRACE_IRQFLAGS_SUPPORT
 
 config CPU_BIG_ENDIAN
diff --git a/arch/openrisc/include/asm/thread_info.h b/arch/openrisc/include/asm/thread_info.h
index 659834ab87fa..4af3049c34c2 100644
--- a/arch/openrisc/include/asm/thread_info.h
+++ b/arch/openrisc/include/asm/thread_info.h
@@ -40,18 +40,12 @@
  */
 #ifndef __ASSEMBLY__
 
-typedef unsigned long mm_segment_t;
-
 struct thread_info {
 	struct task_struct	*task;		/* main task structure */
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;		/* current CPU */
 	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
 
-	mm_segment_t		addr_limit; /* thread address space:
-					       0-0x7FFFFFFF for user-thead
-					       0-0xFFFFFFFF for kernel-thread
-					     */
 	__u8			supervisor_stack[0];
 
 	/* saved context data */
@@ -71,7 +65,6 @@ struct thread_info {
 	.flags		= 0,				\
 	.cpu		= 0,				\
 	.preempt_count	= INIT_PREEMPT_COUNT,		\
-	.addr_limit	= KERNEL_DS,			\
 	.ksp            = 0,                            \
 }
 
diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h
index 8f049ec99b3e..d6500a374e18 100644
--- a/arch/openrisc/include/asm/uaccess.h
+++ b/arch/openrisc/include/asm/uaccess.h
@@ -22,29 +22,6 @@
 #include <linux/string.h>
 #include <asm/page.h>
 #include <asm/extable.h>
-
-/*
- * The fs value determines whether argument validity checking should be
- * performed or not.  If get_fs() == USER_DS, checking is performed, with
- * get_fs() == KERNEL_DS, checking is bypassed.
- *
- * For historical reasons, these macros are grossly misnamed.
- */
-
-/* addr_limit is the maximum accessible address for the task. we misuse
- * the KERNEL_DS and USER_DS values to both assign and compare the
- * addr_limit values through the equally misnamed get/set_fs macros.
- * (see above)
- */
-
-#define KERNEL_DS	(~0UL)
-
-#define USER_DS		(TASK_SIZE)
-#define get_fs()	(current_thread_info()->addr_limit)
-#define set_fs(x)	(current_thread_info()->addr_limit = (x))
-
-#define uaccess_kernel()	(get_fs() == KERNEL_DS)
-
 #include <asm-generic/access_ok.h>
 
 /*
diff --git a/arch/parisc/include/asm/futex.h b/arch/parisc/include/asm/futex.h
index b5835325d44b..3222206cb3ea 100644
--- a/arch/parisc/include/asm/futex.h
+++ b/arch/parisc/include/asm/futex.h
@@ -96,12 +96,6 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
 	u32 val;
 	unsigned long flags;
 
-	/* futex.c wants to do a cmpxchg_inatomic on kernel NULL, which is
-	 * our gateway page, and causes no end of trouble...
-	 */
-	if (uaccess_kernel() && !uaddr)
-		return -EFAULT;
-
 	if (!access_ok(uaddr, sizeof(u32)))
 		return -EFAULT;
 
diff --git a/arch/parisc/lib/memcpy.c b/arch/parisc/lib/memcpy.c
index ea70a0e08321..468704ce8a1c 100644
--- a/arch/parisc/lib/memcpy.c
+++ b/arch/parisc/lib/memcpy.c
@@ -13,7 +13,7 @@
 #include <linux/compiler.h>
 #include <linux/uaccess.h>
 
-#define get_user_space() (uaccess_kernel() ? 0 : mfsp(3))
+#define get_user_space() (mfsp(3))
 #define get_kernel_space() (0)
 
 /* Returns 0 for success, otherwise, returns number of bytes not transferred. */
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 9f6f9bce5292..9276f321b3e3 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -46,7 +46,6 @@ config SPARC
 	select LOCKDEP_SMALL if LOCKDEP
 	select NEED_DMA_MAP_STATE
 	select NEED_SG_DMA_LENGTH
-	select SET_FS
 	select TRACE_IRQFLAGS_SUPPORT
 
 config SPARC32
@@ -101,6 +100,7 @@ config SPARC64
 	select HAVE_SETUP_PER_CPU_AREA
 	select NEED_PER_CPU_EMBED_FIRST_CHUNK
 	select NEED_PER_CPU_PAGE_FIRST_CHUNK
+	select SET_FS
 
 config ARCH_PROC_KCORE_TEXT
 	def_bool y
diff --git a/arch/sparc/include/asm/processor_32.h b/arch/sparc/include/asm/processor_32.h
index 647bf0ac7beb..b26c35336b51 100644
--- a/arch/sparc/include/asm/processor_32.h
+++ b/arch/sparc/include/asm/processor_32.h
@@ -32,10 +32,6 @@ struct fpq {
 };
 #endif
 
-typedef struct {
-	int seg;
-} mm_segment_t;
-
 /* The Sparc processor specific thread struct. */
 struct thread_struct {
 	struct pt_regs *kregs;
@@ -50,11 +46,9 @@ struct thread_struct {
 	unsigned long   fsr;
 	unsigned long   fpqdepth;
 	struct fpq	fpqueue[16];
-	mm_segment_t current_ds;
 };
 
 #define INIT_THREAD  { \
-	.current_ds = KERNEL_DS, \
 	.kregs = (struct pt_regs *)(init_stack+THREAD_SIZE)-1 \
 }
 
diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h
index 367747116260..9fd6c53644b6 100644
--- a/arch/sparc/include/asm/uaccess_32.h
+++ b/arch/sparc/include/asm/uaccess_32.h
@@ -12,19 +12,6 @@
 #include <linux/string.h>
 
 #include <asm/processor.h>
-
-/* Sparc is not segmented, however we need to be able to fool access_ok()
- * when doing system calls from kernel mode legitimately.
- *
- * "For historical reasons, these macros are grossly misnamed." -Linus
- */
-
-#define KERNEL_DS   ((mm_segment_t) { 0 })
-#define USER_DS     ((mm_segment_t) { -1 })
-
-#define get_fs()	(current->thread.current_ds)
-#define set_fs(val)	((current->thread.current_ds) = (val))
-
 #include <asm-generic/access_ok.h>
 
 /* Uh, these should become the main single-value transfer routines..
diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c
index 2dc0bf9fe62e..88c0c14aaff0 100644
--- a/arch/sparc/kernel/process_32.c
+++ b/arch/sparc/kernel/process_32.c
@@ -300,7 +300,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg,
 		extern int nwindows;
 		unsigned long psr;
 		memset(new_stack, 0, STACKFRAME_SZ + TRACEREG_SZ);
-		p->thread.current_ds = KERNEL_DS;
 		ti->kpc = (((unsigned long) ret_from_kernel_thread) - 0x8);
 		childregs->u_regs[UREG_G1] = sp; /* function */
 		childregs->u_regs[UREG_G2] = arg;
@@ -311,7 +310,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg,
 	}
 	memcpy(new_stack, (char *)regs - STACKFRAME_SZ, STACKFRAME_SZ + TRACEREG_SZ);
 	childregs->u_regs[UREG_FP] = sp;
-	p->thread.current_ds = USER_DS;
 	ti->kpc = (((unsigned long) ret_from_fork) - 0x8);
 	ti->kpsr = current->thread.fork_kpsr | PSR_PIL;
 	ti->kwim = current->thread.fork_kwim;
diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig
index 8ac599aa6d99..09f7616a0b46 100644
--- a/arch/xtensa/Kconfig
+++ b/arch/xtensa/Kconfig
@@ -40,7 +40,6 @@ config XTENSA
 	select IRQ_DOMAIN
 	select MODULES_USE_ELF_RELA
 	select PERF_USE_VMALLOC
-	select SET_FS
 	select TRACE_IRQFLAGS_SUPPORT
 	select VIRT_TO_BUS
 	help
diff --git a/arch/xtensa/include/asm/asm-uaccess.h b/arch/xtensa/include/asm/asm-uaccess.h
index 7f6cf4151843..7cec869136e3 100644
--- a/arch/xtensa/include/asm/asm-uaccess.h
+++ b/arch/xtensa/include/asm/asm-uaccess.h
@@ -23,76 +23,6 @@
 #include <asm/asm-offsets.h>
 #include <asm/processor.h>
 
-/*
- * These assembly macros mirror the C macros in asm/uaccess.h.  They
- * should always have identical functionality.  See
- * arch/xtensa/kernel/sys.S for usage.
- */
-
-#define KERNEL_DS	0
-#define USER_DS		1
-
-/*
- * get_fs reads current->thread.current_ds into a register.
- * On Entry:
- * 	<ad>	anything
- * 	<sp>	stack
- * On Exit:
- * 	<ad>	contains current->thread.current_ds
- */
-	.macro	get_fs	ad, sp
-	GET_CURRENT(\ad,\sp)
-#if THREAD_CURRENT_DS > 1020
-	addi	\ad, \ad, TASK_THREAD
-	l32i	\ad, \ad, THREAD_CURRENT_DS - TASK_THREAD
-#else
-	l32i	\ad, \ad, THREAD_CURRENT_DS
-#endif
-	.endm
-
-/*
- * set_fs sets current->thread.current_ds to some value.
- * On Entry:
- *	<at>	anything (temp register)
- *	<av>	value to write
- *	<sp>	stack
- * On Exit:
- *	<at>	destroyed (actually, current)
- *	<av>	preserved, value to write
- */
-	.macro	set_fs	at, av, sp
-	GET_CURRENT(\at,\sp)
-	s32i	\av, \at, THREAD_CURRENT_DS
-	.endm
-
-/*
- * kernel_ok determines whether we should bypass addr/size checking.
- * See the equivalent C-macro version below for clarity.
- * On success, kernel_ok branches to a label indicated by parameter
- * <success>.  This implies that the macro falls through to the next
- * insruction on an error.
- *
- * Note that while this macro can be used independently, we designed
- * in for optimal use in the access_ok macro below (i.e., we fall
- * through on error).
- *
- * On Entry:
- * 	<at>		anything (temp register)
- * 	<success>	label to branch to on success; implies
- * 			fall-through macro on error
- * 	<sp>		stack pointer
- * On Exit:
- * 	<at>		destroyed (actually, current->thread.current_ds)
- */
-
-#if ((KERNEL_DS != 0) || (USER_DS == 0))
-# error Assembly macro kernel_ok fails
-#endif
-	.macro	kernel_ok  at, sp, success
-	get_fs	\at, \sp
-	beqz	\at, \success
-	.endm
-
 /*
  * user_ok determines whether the access to user-space memory is allowed.
  * See the equivalent C-macro version below for clarity.
@@ -147,7 +77,6 @@
  * 	<at>	destroyed
  */
 	.macro	access_ok  aa, as, at, sp, error
-	kernel_ok  \at, \sp, .Laccess_ok_\@
 	user_ok    \aa, \as, \at, \error
 .Laccess_ok_\@:
 	.endm
diff --git a/arch/xtensa/include/asm/processor.h b/arch/xtensa/include/asm/processor.h
index 37d3e9887fe7..abad7c3df46f 100644
--- a/arch/xtensa/include/asm/processor.h
+++ b/arch/xtensa/include/asm/processor.h
@@ -152,18 +152,12 @@
  */
 #define SPILL_SLOT_CALL12(sp, reg) (*(((unsigned long *)(sp)) - 16 + (reg)))
 
-typedef struct {
-	unsigned long seg;
-} mm_segment_t;
-
 struct thread_struct {
 
 	/* kernel's return address and stack pointer for context switching */
 	unsigned long ra; /* kernel's a0: return address and window call size */
 	unsigned long sp; /* kernel's a1: stack pointer */
 
-	mm_segment_t current_ds;    /* see uaccess.h for example uses */
-
 	/* struct xtensa_cpuinfo info; */
 
 	unsigned long bad_vaddr; /* last user fault */
@@ -186,7 +180,6 @@ struct thread_struct {
 {									\
 	ra:		0, 						\
 	sp:		sizeof(init_stack) + (long) &init_stack,	\
-	current_ds:	{0},						\
 	/*info:		{0}, */						\
 	bad_vaddr:	0,						\
 	bad_uaddr:	0,						\
diff --git a/arch/xtensa/include/asm/thread_info.h b/arch/xtensa/include/asm/thread_info.h
index a312333a9add..f6fcbba1d02f 100644
--- a/arch/xtensa/include/asm/thread_info.h
+++ b/arch/xtensa/include/asm/thread_info.h
@@ -52,8 +52,6 @@ struct thread_info {
 	__u32			cpu;		/* current CPU */
 	__s32			preempt_count;	/* 0 => preemptable,< 0 => BUG*/
 
-	mm_segment_t		addr_limit;	/* thread address space */
-
 	unsigned long		cpenable;
 #if XCHAL_HAVE_EXCLUSIVE
 	/* result of the most recent exclusive store */
@@ -81,7 +79,6 @@ struct thread_info {
 	.flags		= 0,			\
 	.cpu		= 0,			\
 	.preempt_count	= INIT_PREEMPT_COUNT,	\
-	.addr_limit	= KERNEL_DS,		\
 }
 
 /* how to get the thread information struct from C */
diff --git a/arch/xtensa/include/asm/uaccess.h b/arch/xtensa/include/asm/uaccess.h
index 0edd9e4b23d0..56aec6d504fe 100644
--- a/arch/xtensa/include/asm/uaccess.h
+++ b/arch/xtensa/include/asm/uaccess.h
@@ -19,22 +19,6 @@
 #include <linux/prefetch.h>
 #include <asm/types.h>
 #include <asm/extable.h>
-
-/*
- * The fs value determines whether argument validity checking should
- * be performed or not.  If get_fs() == USER_DS, checking is
- * performed, with get_fs() == KERNEL_DS, checking is bypassed.
- *
- * For historical reasons (Data Segment Register?), these macros are
- * grossly misnamed.
- */
-
-#define KERNEL_DS	((mm_segment_t) { 0 })
-#define USER_DS		((mm_segment_t) { 1 })
-
-#define get_fs()	(current->thread.current_ds)
-#define set_fs(val)	(current->thread.current_ds = (val))
-
 #include <asm-generic/access_ok.h>
 
 /*
diff --git a/arch/xtensa/kernel/asm-offsets.c b/arch/xtensa/kernel/asm-offsets.c
index dc5c83cad9be..f1fd1390d069 100644
--- a/arch/xtensa/kernel/asm-offsets.c
+++ b/arch/xtensa/kernel/asm-offsets.c
@@ -87,7 +87,6 @@ int main(void)
 	OFFSET(TI_STSTUS, thread_info, status);
 	OFFSET(TI_CPU, thread_info, cpu);
 	OFFSET(TI_PRE_COUNT, thread_info, preempt_count);
-	OFFSET(TI_ADDR_LIMIT, thread_info, addr_limit);
 
 	/* struct thread_info (offset from start_struct) */
 	DEFINE(THREAD_RA, offsetof (struct task_struct, thread.ra));
@@ -108,8 +107,6 @@ int main(void)
 #endif
 	DEFINE(THREAD_XTREGS_USER, offsetof (struct thread_info, xtregs_user));
 	DEFINE(XTREGS_USER_SIZE, sizeof(xtregs_user_t));
-	DEFINE(THREAD_CURRENT_DS, offsetof (struct task_struct, \
-	       thread.current_ds));
 
 	/* struct mm_struct */
 	DEFINE(MM_USERS, offsetof(struct mm_struct, mm_users));
diff --git a/drivers/hid/uhid.c b/drivers/hid/uhid.c
index 614adb510dbd..2a918aeb0af1 100644
--- a/drivers/hid/uhid.c
+++ b/drivers/hid/uhid.c
@@ -747,7 +747,7 @@ static ssize_t uhid_char_write(struct file *file, const char __user *buffer,
 		 * copied from, so it's unsafe to allow this with elevated
 		 * privileges (e.g. from a setuid binary) or via kernel_write().
 		 */
-		if (file->f_cred != current_cred() || uaccess_kernel()) {
+		if (file->f_cred != current_cred()) {
 			pr_err_once("UHID_CREATE from different security context by process %d (%s), this is not allowed.\n",
 				    task_tgid_vnr(current), current->comm);
 			ret = -EACCES;
diff --git a/drivers/scsi/sg.c b/drivers/scsi/sg.c
index 6b43e97bd417..aaa2376b9d34 100644
--- a/drivers/scsi/sg.c
+++ b/drivers/scsi/sg.c
@@ -224,11 +224,6 @@ static int sg_check_file_access(struct file *filp, const char *caller)
 			caller, task_tgid_vnr(current), current->comm);
 		return -EPERM;
 	}
-	if (uaccess_kernel()) {
-		pr_err_once("%s: process %d (%s) called from kernel context, this is not allowed.\n",
-			caller, task_tgid_vnr(current), current->comm);
-		return -EACCES;
-	}
 	return 0;
 }
 
diff --git a/fs/exec.c b/fs/exec.c
index 79f2c9483302..bc68a0c089ac 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1303,12 +1303,6 @@ int begin_new_exec(struct linux_binprm * bprm)
 	if (retval)
 		goto out_unlock;
 
-	/*
-	 * Ensure that the uaccess routines can actually operate on userspace
-	 * pointers:
-	 */
-	force_uaccess_begin();
-
 	if (me->flags & PF_KTHREAD)
 		free_kthread_struct(me);
 	me->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD |
diff --git a/include/asm-generic/access_ok.h b/include/asm-generic/access_ok.h
index 1aad8964d2ed..88a7cb5d9aad 100644
--- a/include/asm-generic/access_ok.h
+++ b/include/asm-generic/access_ok.h
@@ -16,16 +16,8 @@
 #define TASK_SIZE_MAX			TASK_SIZE
 #endif
 
-#ifndef uaccess_kernel
-#ifdef CONFIG_SET_FS
-#define uaccess_kernel()		(get_fs().seg == KERNEL_DS.seg)
-#else
-#define uaccess_kernel()		(0)
-#endif
-#endif
-
 #ifndef user_addr_max
-#define user_addr_max()			(uaccess_kernel() ? ~0UL : TASK_SIZE_MAX)
+#define user_addr_max()			TASK_SIZE_MAX
 #endif
 
 #ifndef __access_ok
diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
index ebc685dc8d74..a5be9e61a2a2 100644
--- a/include/asm-generic/uaccess.h
+++ b/include/asm-generic/uaccess.h
@@ -8,6 +8,7 @@
  * address space, e.g. all NOMMU machines.
  */
 #include <linux/string.h>
+#include <asm-generic/access_ok.h>
 
 #ifdef CONFIG_UACCESS_MEMCPY
 #include <asm/unaligned.h>
@@ -94,30 +95,6 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 #define INLINE_COPY_TO_USER
 #endif /* CONFIG_UACCESS_MEMCPY */
 
-#ifdef CONFIG_SET_FS
-#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-
-#ifndef KERNEL_DS
-#define KERNEL_DS	MAKE_MM_SEG(~0UL)
-#endif
-
-#ifndef USER_DS
-#define USER_DS		MAKE_MM_SEG(TASK_SIZE - 1)
-#endif
-
-#ifndef get_fs
-#define get_fs()	(current_thread_info()->addr_limit)
-
-static inline void set_fs(mm_segment_t fs)
-{
-	current_thread_info()->addr_limit = fs;
-}
-#endif
-
-#endif /* CONFIG_SET_FS */
-
-#include <asm-generic/access_ok.h>
-
 /*
  * These are the main single-value transfer routines.  They automatically
  * use the right size if we just have the right pointer type.
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 819c0cb00b6d..a34b0f9a9972 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -290,10 +290,6 @@ static inline void addr_limit_user_check(void)
 		return;
 #endif
 
-	if (CHECK_DATA_CORRUPTION(uaccess_kernel(),
-				  "Invalid address limit on user-mode return"))
-		force_sig(SIGKILL);
-
 #ifdef TIF_FSCHECK
 	clear_thread_flag(TIF_FSCHECK);
 #endif
diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 2c31667e62e0..2421a41f3a8e 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -10,39 +10,6 @@
 
 #include <asm/uaccess.h>
 
-#ifdef CONFIG_SET_FS
-/*
- * Force the uaccess routines to be wired up for actual userspace access,
- * overriding any possible set_fs(KERNEL_DS) still lingering around.  Undone
- * using force_uaccess_end below.
- */
-static inline mm_segment_t force_uaccess_begin(void)
-{
-	mm_segment_t fs = get_fs();
-
-	set_fs(USER_DS);
-	return fs;
-}
-
-static inline void force_uaccess_end(mm_segment_t oldfs)
-{
-	set_fs(oldfs);
-}
-#else /* CONFIG_SET_FS */
-typedef struct {
-	/* empty dummy */
-} mm_segment_t;
-
-static inline mm_segment_t force_uaccess_begin(void)
-{
-	return (mm_segment_t) { };
-}
-
-static inline void force_uaccess_end(mm_segment_t oldfs)
-{
-}
-#endif /* CONFIG_SET_FS */
-
 /*
  * Architectures should provide two primitives (raw_copy_{to,from}_user())
  * and get rid of their private instances of copy_{to,from}_user() and
diff --git a/include/rdma/ib.h b/include/rdma/ib.h
index 83139b9ce409..f7c185ff7a11 100644
--- a/include/rdma/ib.h
+++ b/include/rdma/ib.h
@@ -75,7 +75,7 @@ struct sockaddr_ib {
  */
 static inline bool ib_safe_file_access(struct file *filp)
 {
-	return filp->f_cred == current_cred() && !uaccess_kernel();
+	return filp->f_cred == current_cred();
 }
 
 #endif /* _RDMA_IB_H */
diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c
index 58cbe357fb2b..1273be84392c 100644
--- a/kernel/events/callchain.c
+++ b/kernel/events/callchain.c
@@ -209,17 +209,13 @@ get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user,
 		}
 
 		if (regs) {
-			mm_segment_t fs;
-
 			if (crosstask)
 				goto exit_put;
 
 			if (add_mark)
 				perf_callchain_store_context(&ctx, PERF_CONTEXT_USER);
 
-			fs = force_uaccess_begin();
 			perf_callchain_user(&ctx, regs);
-			force_uaccess_end(fs);
 		}
 	}
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 57c7197838db..11ca7303d6df 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -6746,7 +6746,6 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 		unsigned long sp;
 		unsigned int rem;
 		u64 dyn_size;
-		mm_segment_t fs;
 
 		/*
 		 * We dump:
@@ -6764,9 +6763,7 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
 
 		/* Data. */
 		sp = perf_user_stack_pointer(regs);
-		fs = force_uaccess_begin();
 		rem = __output_copy_user(handle, (void *) sp, dump_size);
-		force_uaccess_end(fs);
 		dyn_size = dump_size - rem;
 
 		perf_output_skip(handle, rem);
diff --git a/kernel/exit.c b/kernel/exit.c
index b00a25bb4ab9..0884a75bc2f8 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -737,20 +737,6 @@ void __noreturn do_exit(long code)
 
 	WARN_ON(blk_needs_flush_plug(tsk));
 
-	/*
-	 * If do_dead is called because this processes oopsed, it's possible
-	 * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before
-	 * continuing. Amongst other possible reasons, this is to prevent
-	 * mm_release()->clear_child_tid() from writing to a user-controlled
-	 * kernel address.
-	 *
-	 * On uptodate architectures force_uaccess_begin is a noop.  On
-	 * architectures that still have set_fs/get_fs in addition to handling
-	 * oopses handles kernel threads that run as set_fs(KERNEL_DS) by
-	 * default.
-	 */
-	force_uaccess_begin();
-
 	kcov_task_exit(tsk);
 
 	coredump_task_exit(tsk);
diff --git a/kernel/kthread.c b/kernel/kthread.c
index 38c6dd822da8..16c2275d4b50 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -55,7 +55,6 @@ struct kthread {
 	int result;
 	int (*threadfn)(void *);
 	void *data;
-	mm_segment_t oldfs;
 	struct completion parked;
 	struct completion exited;
 #ifdef CONFIG_BLK_CGROUP
@@ -1441,8 +1440,6 @@ void kthread_use_mm(struct mm_struct *mm)
 		mmdrop(active_mm);
 	else
 		smp_mb();
-
-	to_kthread(tsk)->oldfs = force_uaccess_begin();
 }
 EXPORT_SYMBOL_GPL(kthread_use_mm);
 
@@ -1457,8 +1454,6 @@ void kthread_unuse_mm(struct mm_struct *mm)
 	WARN_ON_ONCE(!(tsk->flags & PF_KTHREAD));
 	WARN_ON_ONCE(!tsk->mm);
 
-	force_uaccess_end(to_kthread(tsk)->oldfs);
-
 	task_lock(tsk);
 	/*
 	 * When a kthread stops operating on an address space, the loop
diff --git a/kernel/stacktrace.c b/kernel/stacktrace.c
index 9c625257023d..9ed5ce989415 100644
--- a/kernel/stacktrace.c
+++ b/kernel/stacktrace.c
@@ -226,15 +226,12 @@ unsigned int stack_trace_save_user(unsigned long *store, unsigned int size)
 		.store	= store,
 		.size	= size,
 	};
-	mm_segment_t fs;
 
 	/* Trace user stack if not a kernel thread */
 	if (current->flags & PF_KTHREAD)
 		return 0;
 
-	fs = force_uaccess_begin();
 	arch_stack_walk_user(consume_entry, &c, task_pt_regs(current));
-	force_uaccess_end(fs);
 
 	return c.len;
 }
diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
index 21aa30644219..8115fff17018 100644
--- a/kernel/trace/bpf_trace.c
+++ b/kernel/trace/bpf_trace.c
@@ -332,8 +332,6 @@ BPF_CALL_3(bpf_probe_write_user, void __user *, unsafe_ptr, const void *, src,
 	if (unlikely(in_interrupt() ||
 		     current->flags & (PF_KTHREAD | PF_EXITING)))
 		return -EPERM;
-	if (unlikely(uaccess_kernel()))
-		return -EPERM;
 	if (unlikely(!nmi_uaccess_okay()))
 		return -EPERM;
 
@@ -835,8 +833,6 @@ static int bpf_send_signal_common(u32 sig, enum pid_type type)
 	 */
 	if (unlikely(current->flags & (PF_KTHREAD | PF_EXITING)))
 		return -EPERM;
-	if (unlikely(uaccess_kernel()))
-		return -EPERM;
 	if (unlikely(!nmi_uaccess_okay()))
 		return -EPERM;
 
diff --git a/mm/maccess.c b/mm/maccess.c
index cbd1b3959af2..106820b33a2b 100644
--- a/mm/maccess.c
+++ b/mm/maccess.c
@@ -113,14 +113,11 @@ long strncpy_from_kernel_nofault(char *dst, const void *unsafe_addr, long count)
 long copy_from_user_nofault(void *dst, const void __user *src, size_t size)
 {
 	long ret = -EFAULT;
-	mm_segment_t old_fs = force_uaccess_begin();
-
 	if (access_ok(src, size)) {
 		pagefault_disable();
 		ret = __copy_from_user_inatomic(dst, src, size);
 		pagefault_enable();
 	}
-	force_uaccess_end(old_fs);
 
 	if (ret)
 		return -EFAULT;
@@ -140,14 +137,12 @@ EXPORT_SYMBOL_GPL(copy_from_user_nofault);
 long copy_to_user_nofault(void __user *dst, const void *src, size_t size)
 {
 	long ret = -EFAULT;
-	mm_segment_t old_fs = force_uaccess_begin();
 
 	if (access_ok(dst, size)) {
 		pagefault_disable();
 		ret = __copy_to_user_inatomic(dst, src, size);
 		pagefault_enable();
 	}
-	force_uaccess_end(old_fs);
 
 	if (ret)
 		return -EFAULT;
@@ -176,17 +171,14 @@ EXPORT_SYMBOL_GPL(copy_to_user_nofault);
 long strncpy_from_user_nofault(char *dst, const void __user *unsafe_addr,
 			      long count)
 {
-	mm_segment_t old_fs;
 	long ret;
 
 	if (unlikely(count <= 0))
 		return 0;
 
-	old_fs = force_uaccess_begin();
 	pagefault_disable();
 	ret = strncpy_from_user(dst, unsafe_addr, count);
 	pagefault_enable();
-	force_uaccess_end(old_fs);
 
 	if (ret >= count) {
 		ret = count;
@@ -216,14 +208,11 @@ long strncpy_from_user_nofault(char *dst, const void __user *unsafe_addr,
  */
 long strnlen_user_nofault(const void __user *unsafe_addr, long count)
 {
-	mm_segment_t old_fs;
 	int ret;
 
-	old_fs = force_uaccess_begin();
 	pagefault_disable();
 	ret = strnlen_user(unsafe_addr, count);
 	pagefault_enable();
-	force_uaccess_end(old_fs);
 
 	return ret;
 }
diff --git a/mm/memory.c b/mm/memory.c
index c125c4969913..9a6ebf68a846 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -5256,14 +5256,6 @@ void print_vma_addr(char *prefix, unsigned long ip)
 #if defined(CONFIG_PROVE_LOCKING) || defined(CONFIG_DEBUG_ATOMIC_SLEEP)
 void __might_fault(const char *file, int line)
 {
-	/*
-	 * Some code (nfs/sunrpc) uses socket ops on kernel memory while
-	 * holding the mmap_lock, this is safe because kernel memory doesn't
-	 * get paged out, therefore we'll never actually fault, and the
-	 * below annotations will generate false positives.
-	 */
-	if (uaccess_kernel())
-		return;
 	if (pagefault_disabled())
 		return;
 	__might_sleep(file, line);
diff --git a/net/bpfilter/bpfilter_kern.c b/net/bpfilter/bpfilter_kern.c
index 51a941b56ec3..422ec6e7ccff 100644
--- a/net/bpfilter/bpfilter_kern.c
+++ b/net/bpfilter/bpfilter_kern.c
@@ -70,7 +70,7 @@ static int bpfilter_process_sockopt(struct sock *sk, int optname,
 		.addr		= (uintptr_t)optval.user,
 		.len		= optlen,
 	};
-	if (uaccess_kernel() || sockptr_is_kernel(optval)) {
+	if (sockptr_is_kernel(optval)) {
 		pr_err("kernel access not supported\n");
 		return -EFAULT;
 	}
-- 
2.29.2

RE: [PATCH v2 02/18] uaccess: fix nios2 and microblaze get_user_8()

[David Laight]

From: Arnd Bergmann
> Sent: 16 February 2022 13:13
> 
> These two architectures implement 8-byte get_user() through
> a memcpy() into a four-byte variable, which won't fit.
> 
> Use a temporary 64-bit variable instead here, and use a double
> cast the way that risc-v and openrisc do to avoid compile-time
> warnings.
> 
...
>  	case 4:								\
> -		__get_user_asm("lw", (ptr), __gu_val, __gu_err);	\
> +		__get_user_asm("lw", (ptr), x, __gu_err);		\
>  		break;							\
> -	case 8:								\
> -		__gu_err = __copy_from_user(&__gu_val, ptr, 8);		\
> -		if (__gu_err)						\
> -			__gu_err = -EFAULT;				\
> +	case 8: {							\
> +		__u64 __x = 0;						\
> +		__gu_err = raw_copy_from_user(&__x, ptr, 8) ?		\
> +							-EFAULT : 0;	\
> +		(x) = (typeof(x))(typeof((x) - (x)))__x;		\
>  		break;							\

Wouldn't it be better to just fetch two 32bit values:
Something like (for LE - nios2 is definitely LE:
		__u32 val_lo, val_hi;
		__get_user_asm("lw", (ptr), val_lo, __gu_err);
		__get_user_asm("lw", (ptr) + 4, val_hi, __gu_err);
		x = val_lo | val_hi << 32;
		break;

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

Re: [PATCH v2 15/18] sparc64: remove CONFIG_SET_FS support

[Sam Ravnborg]

Hi Arnd.

On Wed, Feb 16, 2022 at 02:13:29PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> sparc64 uses address space identifiers to differentiate between kernel
> and user space, using ASI_P for kernel threads but ASI_AIUS for normal
> user space, with the option of changing between them.
> 
> As nothing really changes the ASI any more, just hardcode ASI_AIUS
> everywhere. Kernel threads are not allowed to access __user pointers
> anyway.
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  arch/sparc/include/asm/processor_64.h   |  4 ----
>  arch/sparc/include/asm/switch_to_64.h   |  4 +---
>  arch/sparc/include/asm/thread_info_64.h |  4 +---
>  arch/sparc/include/asm/uaccess_64.h     | 20 +-------------------
>  arch/sparc/kernel/process_64.c          | 12 ------------
>  arch/sparc/kernel/traps_64.c            |  2 --
>  arch/sparc/lib/NGmemcpy.S               |  3 +--
>  arch/sparc/mm/init_64.c                 |  7 ++++---
>  8 files changed, 8 insertions(+), 48 deletions(-)

I think you somehow missed the Kconfig change, and also the related
sparc32 change which continue to have set_fs() after this patch.


I did not manage to review the patch - as I am too unfamiliar with the
code paths and the set_fs() removal changes.

	Sam

Re: [PATCH v2 15/18] sparc64: remove CONFIG_SET_FS support

[Sam Ravnborg]

Hi Arnd,

On Wed, Feb 16, 2022 at 07:34:59PM +0100, Sam Ravnborg wrote:
> Hi Arnd.
> 
> On Wed, Feb 16, 2022 at 02:13:29PM +0100, Arnd Bergmann wrote:
> > From: Arnd Bergmann <arnd@arndb.de>
> > 
> > sparc64 uses address space identifiers to differentiate between kernel
> > and user space, using ASI_P for kernel threads but ASI_AIUS for normal
> > user space, with the option of changing between them.
> > 
> > As nothing really changes the ASI any more, just hardcode ASI_AIUS
> > everywhere. Kernel threads are not allowed to access __user pointers
> > anyway.
> > 
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > ---
> >  arch/sparc/include/asm/processor_64.h   |  4 ----
> >  arch/sparc/include/asm/switch_to_64.h   |  4 +---
> >  arch/sparc/include/asm/thread_info_64.h |  4 +---
> >  arch/sparc/include/asm/uaccess_64.h     | 20 +-------------------
> >  arch/sparc/kernel/process_64.c          | 12 ------------
> >  arch/sparc/kernel/traps_64.c            |  2 --
> >  arch/sparc/lib/NGmemcpy.S               |  3 +--
> >  arch/sparc/mm/init_64.c                 |  7 ++++---
> >  8 files changed, 8 insertions(+), 48 deletions(-)
> 
> I think you somehow missed the Kconfig change, and also the related
> sparc32 change which continue to have set_fs() after this patch.
I now notice the sparc32 bits are in the last patch.
To avoid breaking bisect-ability on sparc64 I think you need to merge
the sparc32 changes with this patch, unless the sparc64 changes can
coexist with CONFIG_SET_FS continue to be set.

	Sam

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Sam Ravnborg]

Hi Arnd,

Fix spelling in $subject...

sparc/Kconfig b/arch/sparc/Kconfig
> index 9f6f9bce5292..9276f321b3e3 100644
> --- a/arch/sparc/Kconfig
> +++ b/arch/sparc/Kconfig
> @@ -46,7 +46,6 @@ config SPARC
>  	select LOCKDEP_SMALL if LOCKDEP
>  	select NEED_DMA_MAP_STATE
>  	select NEED_SG_DMA_LENGTH
> -	select SET_FS
>  	select TRACE_IRQFLAGS_SUPPORT
>  
>  config SPARC32
> @@ -101,6 +100,7 @@ config SPARC64
>  	select HAVE_SETUP_PER_CPU_AREA
>  	select NEED_PER_CPU_EMBED_FIRST_CHUNK
>  	select NEED_PER_CPU_PAGE_FIRST_CHUNK
> +	select SET_FS
This looks wrong - looks like some merge went wrong here.

>  
>  config ARCH_PROC_KCORE_TEXT
>  	def_bool y
> diff --git a/arch/sparc/include/asm/processor_32.h b/arch/sparc/include/asm/processor_32.h
> index 647bf0ac7beb..b26c35336b51 100644
> --- a/arch/sparc/include/asm/processor_32.h
> +++ b/arch/sparc/include/asm/processor_32.h
> @@ -32,10 +32,6 @@ struct fpq {
>  };
>  #endif
>  
> -typedef struct {
> -	int seg;
> -} mm_segment_t;
> -
>  /* The Sparc processor specific thread struct. */
>  struct thread_struct {
>  	struct pt_regs *kregs;
> @@ -50,11 +46,9 @@ struct thread_struct {
>  	unsigned long   fsr;
>  	unsigned long   fpqdepth;
>  	struct fpq	fpqueue[16];
> -	mm_segment_t current_ds;
>  };
>  
>  #define INIT_THREAD  { \
> -	.current_ds = KERNEL_DS, \
>  	.kregs = (struct pt_regs *)(init_stack+THREAD_SIZE)-1 \
>  }
>  
> diff --git a/arch/sparc/include/asm/uaccess_32.h b/arch/sparc/include/asm/uaccess_32.h
> index 367747116260..9fd6c53644b6 100644
> --- a/arch/sparc/include/asm/uaccess_32.h
> +++ b/arch/sparc/include/asm/uaccess_32.h
> @@ -12,19 +12,6 @@
>  #include <linux/string.h>
>  
>  #include <asm/processor.h>
> -
> -/* Sparc is not segmented, however we need to be able to fool access_ok()
> - * when doing system calls from kernel mode legitimately.
> - *
> - * "For historical reasons, these macros are grossly misnamed." -Linus
> - */
> -
> -#define KERNEL_DS   ((mm_segment_t) { 0 })
> -#define USER_DS     ((mm_segment_t) { -1 })
> -
> -#define get_fs()	(current->thread.current_ds)
> -#define set_fs(val)	((current->thread.current_ds) = (val))
> -
>  #include <asm-generic/access_ok.h>
>  
>  /* Uh, these should become the main single-value transfer routines..
> diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c
> index 2dc0bf9fe62e..88c0c14aaff0 100644
> --- a/arch/sparc/kernel/process_32.c
> +++ b/arch/sparc/kernel/process_32.c
> @@ -300,7 +300,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg,
>  		extern int nwindows;
>  		unsigned long psr;
>  		memset(new_stack, 0, STACKFRAME_SZ + TRACEREG_SZ);
> -		p->thread.current_ds = KERNEL_DS;
>  		ti->kpc = (((unsigned long) ret_from_kernel_thread) - 0x8);
>  		childregs->u_regs[UREG_G1] = sp; /* function */
>  		childregs->u_regs[UREG_G2] = arg;
> @@ -311,7 +310,6 @@ int copy_thread(unsigned long clone_flags, unsigned long sp, unsigned long arg,
>  	}
>  	memcpy(new_stack, (char *)regs - STACKFRAME_SZ, STACKFRAME_SZ + TRACEREG_SZ);
>  	childregs->u_regs[UREG_FP] = sp;
> -	p->thread.current_ds = USER_DS;
>  	ti->kpc = (((unsigned long) ret_from_fork) - 0x8);
>  	ti->kpsr = current->thread.fork_kpsr | PSR_PIL;
>  	ti->kwim = current->thread.fork_kwim;

Other than the above the sparc32 changes looks fine, and with the Kconf
stuff fixed:
Acked-by: Sam Ravnborg <sam@ravnborg.org> # for sparc32 changes

Re: [PATCH v2 15/18] sparc64: remove CONFIG_SET_FS support

[Arnd Bergmann]

On Wed, Feb 16, 2022 at 7:41 PM Sam Ravnborg <sam@ravnborg.org> wrote:
> On Wed, Feb 16, 2022 at 07:34:59PM +0100, Sam Ravnborg wrote:

> >
> > I think you somehow missed the Kconfig change, and also the related
> > sparc32 change which continue to have set_fs() after this patch.

Right, thanks for pointing out the issue.

> I now notice the sparc32 bits are in the last patch.
> To avoid breaking bisect-ability on sparc64 I think you need to merge
> the sparc32 changes with this patch, unless the sparc64 changes can
> coexist with CONFIG_SET_FS continue to be set.

I originally had them in the reverse order and broke bisectability during my
rebase. The end result is still fine, but now I need to move the 'select
SET_FS' from CONFIG_SPARC to CONFIG_SPARC32 in this patch
and then remove it again from there in the last step.

I've done that in my local copy now.

         Arnd

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Arnd Bergmann]

On Wed, Feb 16, 2022 at 7:44 PM Sam Ravnborg <sam@ravnborg.org> wrote:
>
> Hi Arnd,
>
> Fix spelling in $subject...

done

> sparc/Kconfig b/arch/sparc/Kconfig
> > index 9f6f9bce5292..9276f321b3e3 100644
> > --- a/arch/sparc/Kconfig
> > +++ b/arch/sparc/Kconfig
> > @@ -46,7 +46,6 @@ config SPARC
> >       select LOCKDEP_SMALL if LOCKDEP
> >       select NEED_DMA_MAP_STATE
> >       select NEED_SG_DMA_LENGTH
> > -     select SET_FS
> >       select TRACE_IRQFLAGS_SUPPORT
> >
> >  config SPARC32
> > @@ -101,6 +100,7 @@ config SPARC64
> >       select HAVE_SETUP_PER_CPU_AREA
> >       select NEED_PER_CPU_EMBED_FIRST_CHUNK
> >       select NEED_PER_CPU_PAGE_FIRST_CHUNK
> > +     select SET_FS
> This looks wrong - looks like some merge went wrong here.

Fixed now.

>
> Other than the above the sparc32 changes looks fine, and with the Kconf
> stuff fixed:
> Acked-by: Sam Ravnborg <sam@ravnborg.org> # for sparc32 changes

Thanks!

      Arnd

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Christophe Leroy]

Le 16/02/2022 à 14:13, Arnd Bergmann a écrit :
> From: Arnd Bergmann <arnd@arndb.de>
> 
> Christoph Hellwig and a few others spent a huge effort on removing
> set_fs() from most of the important architectures, but about half the
> other architectures were never completed even though most of them don't
> actually use set_fs() at all.
> 
> I did a patch for microblaze at some point, which turned out to be fairly
> generic, and now ported it to most other architectures, using new generic
> implementations of access_ok() and __{get,put}_kernel_nocheck().
> 
> Three architectures (sparc64, ia64, and sh) needed some extra work,
> which I also completed.
> 
> The final series contains extra cleanup changes that touch all
> architectures. Please review and test these, so we can merge them
> for v5.18.

As a further cleanup, have you thought about making a generic version of 
clear_user() ? On almost all architectures, clear_user() does an 
access_ok() then calls __clear_user() or similar.

Maybe also the same with put_user() and get_user() ? After all it is 
just access_ok() followed by __put_user() or __get_user() ? It seems 
more tricky though, as some architectures seems to have less trivial 
stuff there.

I also see all architectures have a prototype for strncpy_from_user() 
and strnlen_user(). Could be a common prototype instead when we have 
GENERIC_STRNCPY_FROM_USER / GENERIC_STRNLEN_USER

And we have also 
user_access_begin()/user_read_access_begin()/user_write_access_begin() 
which call access_ok() then do the real work. Could be made generic with 
call to some arch specific __user_access_begin() and friends after the 
access_ok() and eventually the might_fault().

Christophe

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Arnd Bergmann]

On Thu, Feb 17, 2022 at 8:20 AM Christophe Leroy
<christophe.leroy@csgroup.eu> wrote:
> Le 16/02/2022 à 14:13, Arnd Bergmann a écrit :
> >
> > Christoph Hellwig and a few others spent a huge effort on removing
> > set_fs() from most of the important architectures, but about half the
> > other architectures were never completed even though most of them don't
> > actually use set_fs() at all.
> >
> > I did a patch for microblaze at some point, which turned out to be fairly
> > generic, and now ported it to most other architectures, using new generic
> > implementations of access_ok() and __{get,put}_kernel_nocheck().
> >
> > Three architectures (sparc64, ia64, and sh) needed some extra work,
> > which I also completed.
> >
> > The final series contains extra cleanup changes that touch all
> > architectures. Please review and test these, so we can merge them
> > for v5.18.
>
> As a further cleanup, have you thought about making a generic version of
> clear_user() ? On almost all architectures, clear_user() does an
> access_ok() then calls __clear_user() or similar.

This already exists in include/asm-generic/uaccess.h, but that file is
currently not as easy to use as it should be. I've previously looked into
what it would take to get more architectures to use common code
in that file, but I currently have no plans to work on that.

> Maybe also the same with put_user() and get_user() ? After all it is
> just access_ok() followed by __put_user() or __get_user() ? It seems
> more tricky though, as some architectures seems to have less trivial
> stuff there.

Same here: architectures can already provide a __put_user_fn()
and __get_user_fn(), to get the generic versions of the interface,
but few architectures use that. You can actually get all the interfaces
by just providing raw_copy_from_user() and raw_copy_to_user(),
but the get_user/put_user versions you get from that are fairly
inefficient.

> I also see all architectures have a prototype for strncpy_from_user()
> and strnlen_user(). Could be a common prototype instead when we have
> GENERIC_STRNCPY_FROM_USER / GENERIC_STRNLEN_USER
>
> And we have also
> user_access_begin()/user_read_access_begin()/user_write_access_begin()
> which call access_ok() then do the real work. Could be made generic with
> call to some arch specific __user_access_begin() and friends after the
> access_ok() and eventually the might_fault().

In my opinion, the biggest win would be to move the type-agnostic part of
get_user/put_user into completely generic code, this is what architectures
get wrong the most, see patch 02/18 in this series for instance.

What I'd like to see is that architectures only provide fixed-length
versions of unsafe_get_user()/unsafe_put_user(), with the type-agnostic
versions (get_user(), __get_user(), unsafe_get_user() and their put
versions) all defined once in include/linux/uaccess.h based on those.

I tried implementing this in the past, but unfortunately the resulting
object code from my generalized implementation was worse than
what we have today, so I did not continue that work.

      Arnd

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Arnd Bergmann]

On Wed, Feb 16, 2022 at 2:13 PM Arnd Bergmann <arnd@kernel.org> wrote:

> + * limit and catch all possible overflows.
> + * On architectures with separate user address space (m68k, s390, parisc,
> + * sparc64) or those without an MMU, this should always return true.
...
> +static inline int __access_ok(const void __user *ptr, unsigned long size)
> +{
> +       unsigned long limit = user_addr_max();
> +       unsigned long addr = (unsigned long)ptr;
> +
> +       if (IS_ENABLED(CONFIG_ALTERNATE_USER_ADDRESS_SPACE))
> +               return true;

I noticed that I'm missing the check for !CONFIG_MMU here, despite
mentioning that in the comment above it. I've added it now.

        Arnd

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Arnd Bergmann]

On Wed, Feb 16, 2022 at 2:13 PM Arnd Bergmann <arnd@kernel.org> wrote:
>
> From: Arnd Bergmann <arnd@arndb.de>
>
> Christoph Hellwig and a few others spent a huge effort on removing
> set_fs() from most of the important architectures, but about half the
> other architectures were never completed even though most of them don't
> actually use set_fs() at all.
>
> I did a patch for microblaze at some point, which turned out to be fairly
> generic, and now ported it to most other architectures, using new generic
> implementations of access_ok() and __{get,put}_kernel_nocheck().
>
> Three architectures (sparc64, ia64, and sh) needed some extra work,
> which I also completed.
>
> The final series contains extra cleanup changes that touch all
> architectures. Please review and test these, so we can merge them
> for v5.18.
>
> The series is available at
> https://git.kernel.org/pub/scm/linux/kernel/git/arnd/playground.git/log/?h=set_fs-2
> for testing.

I've added the updated contents to my asm-generic tree now to put them
into linux-next.

         Arnd

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Andy Lutomirski]

On Wed, Feb 16, 2022 at 5:19 AM Arnd Bergmann <arnd@kernel.org> wrote:
>
> From: Arnd Bergmann <arnd@arndb.de>
>
> There are many different ways that access_ok() is defined across
> architectures, but in the end, they all just compare against the
> user_addr_max() value or they accept anything.
>
> Provide one definition that works for most architectures, checking
> against TASK_SIZE_MAX for user processes or skipping the check inside
> of uaccess_kernel() sections.
>
> For architectures without CONFIG_SET_FS(), this should be the fastest
> check, as it comes down to a single comparison of a pointer against a
> compile-time constant, while the architecture specific versions tend to
> do something more complex for historic reasons or get something wrong.

This isn't actually optimal.  On x86, TASK_SIZE_MAX is a bizarre
constant that has a very specific value to work around a bug^Wdesign
error^Wfeature of Intel CPUs.  TASK_SIZE_MAX is the maximum address at
which userspace is permitted to allocate memory, but there is a huge
gap between user and kernel addresses, and any value in the gap would
be adequate for the comparison.  If we wanted to optimize this, simply
checking the high bit (which x86 can do without any immediate
constants at all) would be sufficient and, for an access known to fit
in 32 bits, one could get even fancier and completely ignore the size
of the access.  (For accesses not known to fit in 32 bits, I suspect
some creativity could still come up with a construction that's
substantially faster than the one in your patch.)

So there's plenty of room for optimization here.

(This is not in any respect a NAK -- it's just an observation that
this could be even better.)

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Eric W. Biederman]

Arnd Bergmann <arnd@kernel.org> writes:

> From: Arnd Bergmann <arnd@arndb.de>
>
> There are no remaining callers of set_fs(), so CONFIG_SET_FS
> can be removed globally, along with the thread_info field and
> any references to it.
>
> This turns access_ok() into a cheaper check against TASK_SIZE_MAX.
>
> With CONFIG_SET_FS gone, so drop all remaining references to
> set_fs()/get_fs(), mm_segment_t and uaccess_kernel().

For the bits I have looked at recently, and think I understand.

Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>

>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  fs/exec.c                                 |  6 --
>  kernel/exit.c                             | 14 -----
>  kernel/kthread.c                          |  5 --
>
> diff --git a/fs/exec.c b/fs/exec.c
> index 79f2c9483302..bc68a0c089ac 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -1303,12 +1303,6 @@ int begin_new_exec(struct linux_binprm * bprm)
>  	if (retval)
>  		goto out_unlock;
>  
> -	/*
> -	 * Ensure that the uaccess routines can actually operate on userspace
> -	 * pointers:
> -	 */
> -	force_uaccess_begin();
> -
>  	if (me->flags & PF_KTHREAD)
>  		free_kthread_struct(me);
>  	me->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD |
> diff --git a/kernel/exit.c b/kernel/exit.c
> index b00a25bb4ab9..0884a75bc2f8 100644
> --- a/kernel/exit.c
> +++ b/kernel/exit.c
> @@ -737,20 +737,6 @@ void __noreturn do_exit(long code)
>  
>  	WARN_ON(blk_needs_flush_plug(tsk));
>  
> -	/*
> -	 * If do_dead is called because this processes oopsed, it's possible
> -	 * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before
> -	 * continuing. Amongst other possible reasons, this is to prevent
> -	 * mm_release()->clear_child_tid() from writing to a user-controlled
> -	 * kernel address.
> -	 *
> -	 * On uptodate architectures force_uaccess_begin is a noop.  On
> -	 * architectures that still have set_fs/get_fs in addition to handling
> -	 * oopses handles kernel threads that run as set_fs(KERNEL_DS) by
> -	 * default.
> -	 */
> -	force_uaccess_begin();
> -
>  	kcov_task_exit(tsk);
>  
>  	coredump_task_exit(tsk);
> diff --git a/kernel/kthread.c b/kernel/kthread.c
> index 38c6dd822da8..16c2275d4b50 100644
> --- a/kernel/kthread.c
> +++ b/kernel/kthread.c
> @@ -55,7 +55,6 @@ struct kthread {
>  	int result;
>  	int (*threadfn)(void *);
>  	void *data;
> -	mm_segment_t oldfs;
>  	struct completion parked;
>  	struct completion exited;
>  #ifdef CONFIG_BLK_CGROUP
> @@ -1441,8 +1440,6 @@ void kthread_use_mm(struct mm_struct *mm)
>  		mmdrop(active_mm);
>  	else
>  		smp_mb();
> -
> -	to_kthread(tsk)->oldfs = force_uaccess_begin();
>  }
>  EXPORT_SYMBOL_GPL(kthread_use_mm);
>  
> @@ -1457,8 +1454,6 @@ void kthread_unuse_mm(struct mm_struct *mm)
>  	WARN_ON_ONCE(!(tsk->flags & PF_KTHREAD));
>  	WARN_ON_ONCE(!tsk->mm);
>  
> -	force_uaccess_end(to_kthread(tsk)->oldfs);
> -
>  	task_lock(tsk);
>  	/*
>  	 * When a kthread stops operating on an address space, the loop

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Al Viro]

On Thu, Feb 17, 2022 at 07:20:11AM +0000, Christophe Leroy wrote:

> And we have also 
> user_access_begin()/user_read_access_begin()/user_write_access_begin() 
> which call access_ok() then do the real work. Could be made generic with 
> call to some arch specific __user_access_begin() and friends after the 
> access_ok() and eventually the might_fault().

Not a good idea, considering the fact that we do not want to invite
uses of "faster" variants...

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Al Viro]

On Thu, Feb 17, 2022 at 08:49:59AM +0100, Arnd Bergmann wrote:

> Same here: architectures can already provide a __put_user_fn()
> and __get_user_fn(), to get the generic versions of the interface,
> but few architectures use that. You can actually get all the interfaces
> by just providing raw_copy_from_user() and raw_copy_to_user(),
> but the get_user/put_user versions you get from that are fairly
> inefficient.

FWIW, __{get,put}_user_{8,16,32,64} would probably make it easier to
unify.  That's where the really variable part tends to be, anyway.
IMO __get_user_fn() had been a mistake.

One thing I somewhat dislike about the series is the boilerplate in
asm/uaccess.h instances - #include <asm-generic/access-ok.h> in
a lot of them might make sense as a transitory state, but getting
stuck with those indefinitely...

	BTW, do we need user_addr_max() anymore?  The definition in
asm-generic/access-ok.h is the only one, so ifndef around it is pointless.

Re: [PATCH v2 02/18] uaccess: fix nios2 and microblaze get_user_8()

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 03/18] nds32: fix access_ok() checks in get/put_user

[Christoph Hellwig]

On Wed, Feb 16, 2022 at 02:13:17PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> The get_user()/put_user() functions are meant to check for
> access_ok(), while the __get_user()/__put_user() functions
> don't.
> 
> This broke in 4.19 for nds32, when it gained an extraneous
> check in __get_user(), but lost the check it needs in
> __put_user().

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 05/18] x86: remove __range_not_ok()

[Christoph Hellwig]

On Wed, Feb 16, 2022 at 02:13:19PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> The __range_not_ok() helper is an x86 (and sparc64) specific interface
> that does roughly the same thing as __access_ok(), but with different
> calling conventions.
> 
> Change this to use the normal interface in order for consistency as we
> clean up all access_ok() implementations.
> 
> This changes the limit from TASK_SIZE to TASK_SIZE_MAX, which Al points
> out is the right thing do do here anyway.
> 
> The callers have to use __access_ok() instead of the normal access_ok()
> though, because on x86 that contains a WARN_ON_IN_IRQ() check that cannot
> be used inside of NMI context while tracing.
> 
> Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
> Suggested-by: Christoph Hellwig <hch@infradead.org>
> Link: https://lore.kernel.org/lkml/YgsUKcXGR7r4nINj@zeniv-ca.linux.org.uk/
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  arch/x86/events/core.c         |  2 +-
>  arch/x86/include/asm/uaccess.h | 10 ++++++----
>  arch/x86/kernel/dumpstack.c    |  2 +-
>  arch/x86/kernel/stacktrace.c   |  2 +-
>  arch/x86/lib/usercopy.c        |  2 +-
>  5 files changed, 10 insertions(+), 8 deletions(-)
> 
> diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
> index e686c5e0537b..eef816fc216d 100644
> --- a/arch/x86/events/core.c
> +++ b/arch/x86/events/core.c
> @@ -2794,7 +2794,7 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re
>  static inline int
>  valid_user_frame(const void __user *fp, unsigned long size)
>  {
> -	return (__range_not_ok(fp, size, TASK_SIZE) == 0);
> +	return __access_ok(fp, size);
>  }

valid_user_frame just need to go away and the following __get_user calls
replaced with normal get_user ones.

> diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
> index 53de044e5654..da534fb7b5c6 100644
> --- a/arch/x86/kernel/dumpstack.c
> +++ b/arch/x86/kernel/dumpstack.c
> @@ -85,7 +85,7 @@ static int copy_code(struct pt_regs *regs, u8 *buf, unsigned long src,
>  	 * Make sure userspace isn't trying to trick us into dumping kernel
>  	 * memory by pointing the userspace instruction pointer at it.
>  	 */
> -	if (__chk_range_not_ok(src, nbytes, TASK_SIZE_MAX))
> +	if (!__access_ok((void __user *)src, nbytes))
>  		return -EINVAL;

This one is not needed at all as copy_from_user_nmi already checks the
access range.

> diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
> index 15b058eefc4e..ee117fcf46ed 100644
> --- a/arch/x86/kernel/stacktrace.c
> +++ b/arch/x86/kernel/stacktrace.c
> @@ -90,7 +90,7 @@ copy_stack_frame(const struct stack_frame_user __user *fp,
>  {
>  	int ret;
>  
> -	if (__range_not_ok(fp, sizeof(*frame), TASK_SIZE))
> +	if (!__access_ok(fp, sizeof(*frame)))
>  		return 0;

Just switch the __get_user calls below to get_user instead.

Re: [PATCH v2 06/18] x86: use more conventional access_ok() definition

[Christoph Hellwig]

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 07/18] nios2: drop access_ok() check from __put_user()

[Christoph Hellwig]

On Wed, Feb 16, 2022 at 02:13:21PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> Unlike other architectures, the nios2 version of __put_user() has an
> extra check for access_ok(), preventing it from being used to implement
> __put_kernel_nofault().
> 
> Split up put_user() along the same lines as __get_user()/get_user()

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 08/18] uaccess: add generic __{get,put}_kernel_nofault

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 10/18] m68k: fix access_ok for coldfire

[Christoph Hellwig]

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 12/18] uaccess: fix type mismatch warnings from access_ok()

[Christoph Hellwig]

Looks good,

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Christoph Hellwig]

> +#include <asm-generic/access_ok.h>

Instead of the asm-generic games, shouldn't we just define access_ok in
<linux/uaccess.h> if not already defined by the architecture?

Re: [PATCH v2 14/18] lib/test_lockup: fix kernel pointer check for separate address spaces

[Christoph Hellwig]

On Wed, Feb 16, 2022 at 02:13:28PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> test_kernel_ptr() uses access_ok() to figure out if a given address
> points to user space instead of kernel space. However on architectures
> that set CONFIG_ALTERNATE_USER_ADDRESS_SPACE, a pointer can be valid
> for both, and the check always fails because access_ok() returns true.
> 
> Make the check for user space pointers conditional on the type of
> address space layout.

What is this code even trying to do?  It looks extremly broken.

Re: [PATCH v2 16/18] sh: remove CONFIG_SET_FS support

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Christoph Hellwig]

s/maining/remaining/ ?

Or maybe rather:

uaccess: remove CONFIG_SET_FS

because it is all gone now.

> With CONFIG_SET_FS gone, so drop all remaining references to
> set_fs()/get_fs(), mm_segment_t and uaccess_kernel().

And this sentence does not parse.

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Arnd Bergmann]

On Fri, Feb 18, 2022 at 7:37 AM Christoph Hellwig <hch@lst.de> wrote:
>
> s/maining/remaining/ ?
>
> Or maybe rather:
>
> uaccess: remove CONFIG_SET_FS
>
> because it is all gone now.
>
> > With CONFIG_SET_FS gone, so drop all remaining references to
> > set_fs()/get_fs(), mm_segment_t and uaccess_kernel().
>
> And this sentence does not parse.

Both fixed now, thanks!

       Arnd

Re: [PATCH v2 14/18] lib/test_lockup: fix kernel pointer check for separate address spaces

[Arnd Bergmann]

On Fri, Feb 18, 2022 at 7:35 AM Christoph Hellwig <hch@lst.de> wrote:
>
> On Wed, Feb 16, 2022 at 02:13:28PM +0100, Arnd Bergmann wrote:
> > From: Arnd Bergmann <arnd@arndb.de>
> >
> > test_kernel_ptr() uses access_ok() to figure out if a given address
> > points to user space instead of kernel space. However on architectures
> > that set CONFIG_ALTERNATE_USER_ADDRESS_SPACE, a pointer can be valid
> > for both, and the check always fails because access_ok() returns true.
> >
> > Make the check for user space pointers conditional on the type of
> > address space layout.
>
> What is this code even trying to do?  It looks extremly broken.

As I understand it, this is only meant for debugging, and the module contains
intentionally broken lock usage to test whether the watchdog and lockup
detection in the kernel is able to find them.

I did not try that hard to understand how it works though.

      Arnd

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Arnd Bergmann]

On Thu, Feb 17, 2022 at 8:15 PM Andy Lutomirski <luto@amacapital.net> wrote:
>
> On Wed, Feb 16, 2022 at 5:19 AM Arnd Bergmann <arnd@kernel.org> wrote:
> >
> > From: Arnd Bergmann <arnd@arndb.de>
> >
> > There are many different ways that access_ok() is defined across
> > architectures, but in the end, they all just compare against the
> > user_addr_max() value or they accept anything.
> >
> > Provide one definition that works for most architectures, checking
> > against TASK_SIZE_MAX for user processes or skipping the check inside
> > of uaccess_kernel() sections.
> >
> > For architectures without CONFIG_SET_FS(), this should be the fastest
> > check, as it comes down to a single comparison of a pointer against a
> > compile-time constant, while the architecture specific versions tend to
> > do something more complex for historic reasons or get something wrong.
>
> This isn't actually optimal.  On x86, TASK_SIZE_MAX is a bizarre
> constant that has a very specific value to work around a bug^Wdesign
> error^Wfeature of Intel CPUs.  TASK_SIZE_MAX is the maximum address at
> which userspace is permitted to allocate memory, but there is a huge
> gap between user and kernel addresses, and any value in the gap would
> be adequate for the comparison.  If we wanted to optimize this, simply
> checking the high bit (which x86 can do without any immediate
> constants at all) would be sufficient and, for an access known to fit
> in 32 bits, one could get even fancier and completely ignore the size
> of the access.  (For accesses not known to fit in 32 bits, I suspect
> some creativity could still come up with a construction that's
> substantially faster than the one in your patch.)
>
> So there's plenty of room for optimization here.
>
> (This is not in any respect a NAK -- it's just an observation that
> this could be even better.)

Thank you for taking a look!

As you can see in the patch that changes the algorithm on x86 [1],
it was already using TASK_SIZE_MAX as the limit, only the order
in which the comparison is done, hopefully leading to better code
already. I have looked at trivial examples on x86 that showed an
improvement for constant sizes, but only looked at arm64 in detail
for the overall result.

It may be worth checking if using LONG_MAX as the limit produces
better code, but it's probably best to do the optimization in the
common code in a portable way to keep it from diverging again.

       Arnd

[1] https://lore.kernel.org/lkml/20220216131332.1489939-7-arnd@kernel.org/

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Arnd Bergmann]

On Fri, Feb 18, 2022 at 7:34 AM Christoph Hellwig <hch@lst.de> wrote:
>
> > +#include <asm-generic/access_ok.h>
>
> Instead of the asm-generic games, shouldn't we just define access_ok in
> <linux/uaccess.h> if not already defined by the architecture?

I tried, but couldn't actually make it work because asm/uaccess.h tends
to contain inline functions that rely on access_ok(). It could work once we
move all the high-level functions into linux/uaccess.h, but that would likely
require another long patch series.

One option that can work is to require architectures to have an
asm/access_ok.h header that gets included by linux/uaccess.h.
On most architectures, that would be redirected to
asm-generic/access_ok.h, as only ia64, x86, arm64 and um
need to override the definition.

      Arnd

Re: [PATCH v2 05/18] x86: remove __range_not_ok()

[Arnd Bergmann]

On Fri, Feb 18, 2022 at 7:28 AM Christoph Hellwig <hch@lst.de> wrote:
> On Wed, Feb 16, 2022 at 02:13:19PM +0100, Arnd Bergmann wrote:
> > --- a/arch/x86/events/core.c
> > +++ b/arch/x86/events/core.c
> > @@ -2794,7 +2794,7 @@ perf_callchain_kernel(struct perf_callchain_entry_ctx *entry, struct pt_regs *re
> >  static inline int
> >  valid_user_frame(const void __user *fp, unsigned long size)
> >  {
> > -     return (__range_not_ok(fp, size, TASK_SIZE) == 0);
> > +     return __access_ok(fp, size);
> >  }
>
> valid_user_frame just need to go away and the following __get_user calls
> replaced with normal get_user ones.

As I understand it, that would not work here because get_user() calls
access_ok() rather than __access_ok(), and on x86 that can not be
called in NMI context.

It is a bit odd that x86 is the only architecture that has this check,
but adding
it was clearly intentional, see 7c4788950ba5 ("x86/uaccess, sched/preempt:
Verify access_ok() context").

> > diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c
> > index 53de044e5654..da534fb7b5c6 100644
> > --- a/arch/x86/kernel/dumpstack.c
> > +++ b/arch/x86/kernel/dumpstack.c
> > @@ -85,7 +85,7 @@ static int copy_code(struct pt_regs *regs, u8 *buf, unsigned long src,
> >        * Make sure userspace isn't trying to trick us into dumping kernel
> >        * memory by pointing the userspace instruction pointer at it.
> >        */
> > -     if (__chk_range_not_ok(src, nbytes, TASK_SIZE_MAX))
> > +     if (!__access_ok((void __user *)src, nbytes))
> >               return -EINVAL;
>
> This one is not needed at all as copy_from_user_nmi already checks the
> access range.

Ok, removing this.

> > diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
> > index 15b058eefc4e..ee117fcf46ed 100644
> > --- a/arch/x86/kernel/stacktrace.c
> > +++ b/arch/x86/kernel/stacktrace.c
> > @@ -90,7 +90,7 @@ copy_stack_frame(const struct stack_frame_user __user *fp,
> >  {
> >       int ret;
> >
> > -     if (__range_not_ok(fp, sizeof(*frame), TASK_SIZE))
> > +     if (!__access_ok(fp, sizeof(*frame)))
> >               return 0;
>
> Just switch the __get_user calls below to get_user instead.

Same as the first one, I think we can't do this in NMI context.

         Arnd

Re: [PATCH v2 08/18] uaccess: add generic __{get,put}_kernel_nofault

[Geert Uytterhoeven]

On Wed, Feb 16, 2022 at 2:17 PM Arnd Bergmann <arnd@kernel.org> wrote:
> From: Arnd Bergmann <arnd@arndb.de>
>
> Nine architectures are still missing __{get,put}_kernel_nofault:
> alpha, ia64, microblaze, nds32, nios2, openrisc, sh, sparc32, xtensa.
>
> Add a generic version that lets everything use the normal
> copy_{from,to}_kernel_nofault() code based on these, removing the last
> use of get_fs()/set_fs() from architecture-independent code.
>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

>  arch/m68k/include/asm/uaccess.h     |   2 -

Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH v2 10/18] m68k: fix access_ok for coldfire

[Geert Uytterhoeven]

Hi Arnd,

On Wed, Feb 16, 2022 at 2:17 PM Arnd Bergmann <arnd@kernel.org> wrote:
> From: Arnd Bergmann <arnd@arndb.de>
>
> While most m68k platforms use separate address spaces for user
> and kernel space, at least coldfire does not, and the other
> ones have a TASK_SIZE that is less than the entire 4GB address
> range.
>
> Using the default implementation of __access_ok() stops coldfire
> user space from trivially accessing kernel memory.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Thanks for your patch!

> --- a/arch/m68k/include/asm/uaccess.h
> +++ b/arch/m68k/include/asm/uaccess.h
> @@ -12,14 +12,21 @@
>  #include <asm/extable.h>
>
>  /* We let the MMU do all checking */
> -static inline int access_ok(const void __user *addr,
> +static inline int access_ok(const void __user *ptr,
>                             unsigned long size)
>  {
> +       unsigned long limit = TASK_SIZE;
> +       unsigned long addr = (unsigned long)ptr;
> +
>         /*
>          * XXX: for !CONFIG_CPU_HAS_ADDRESS_SPACES this really needs to check
>          * for TASK_SIZE!
> +        * Removing this helper is probably sufficient.
>          */

Shouldn't the above comment block be removed completely,
as this is now implemented below?

> -       return 1;
> +       if (IS_ENABLED(CONFIG_CPU_HAS_ADDRESS_SPACES))
> +               return 1;
> +
> +       return (size <= limit) && (addr <= (limit - size));
>  }

Any pesky compilers that warn (or worse with -Werror) about
"condition always true" for TASK_SIZE = 0xFFFFFFFFUL?

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Geert Uytterhoeven]

On Wed, Feb 16, 2022 at 2:17 PM Arnd Bergmann <arnd@kernel.org> wrote:
> From: Arnd Bergmann <arnd@arndb.de>
>
> There are many different ways that access_ok() is defined across
> architectures, but in the end, they all just compare against the
> user_addr_max() value or they accept anything.
>
> Provide one definition that works for most architectures, checking
> against TASK_SIZE_MAX for user processes or skipping the check inside
> of uaccess_kernel() sections.
>
> For architectures without CONFIG_SET_FS(), this should be the fastest
> check, as it comes down to a single comparison of a pointer against a
> compile-time constant, while the architecture specific versions tend to
> do something more complex for historic reasons or get something wrong.
>
> Type checking for __user annotations is handled inconsistently across
> architectures, but this is easily simplified as well by using an inline
> function that takes a 'const void __user *' argument. A handful of
> callers need an extra __user annotation for this.
>
> Some architectures had trick to use 33-bit or 65-bit arithmetic on the
> addresses to calculate the overflow, however this simpler version uses
> fewer registers, which means it can produce better object code in the
> end despite needing a second (statically predicted) branch.
>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Mark Rutland <mark.rutland@arm.com> [arm64, asm-generic]
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

>  arch/m68k/Kconfig.cpu                 |  1 +
>  arch/m68k/include/asm/uaccess.h       | 19 +--------

Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Arnd Bergmann]

n Fri, Feb 18, 2022 at 3:21 AM Al Viro <viro@zeniv.linux.org.uk> wrote:
>
> On Thu, Feb 17, 2022 at 08:49:59AM +0100, Arnd Bergmann wrote:
>
> > Same here: architectures can already provide a __put_user_fn()
> > and __get_user_fn(), to get the generic versions of the interface,
> > but few architectures use that. You can actually get all the interfaces
> > by just providing raw_copy_from_user() and raw_copy_to_user(),
> > but the get_user/put_user versions you get from that are fairly
> > inefficient.
>
> FWIW, __{get,put}_user_{8,16,32,64} would probably make it easier to
> unify.  That's where the really variable part tends to be, anyway.
> IMO __get_user_fn() had been a mistake.

I've prototyped this now, to see what this might look like, see
https://git.kernel.org/pub/scm/linux/kernel/git/arnd/playground.git/commit/?h=generic-get_user-prototype

This adds generic inline version of {__get,get,__put,put}_user()
and converts x86 to (optionally) use it. This builds with gcc-5
through gcc-11 on 32-bit and 64-bit x86, using asm-goto with
outputs where possible, and requiring a minimum set of macro
definitions from the architecture. Compiling with clang produces
no warnings but does cause a linker issue at the moment, so
there is probably at least one bug in it.

Aside from compile-testing, I have not tried to verify if this
is correct or efficient, but let me know if you think this is headed
in the right direction.

> One thing I somewhat dislike about the series is the boilerplate in
> asm/uaccess.h instances - #include <asm-generic/access-ok.h> in
> a lot of them might make sense as a transitory state, but getting
> stuck with those indefinitely...

Christoph also complained about it, the problem for now is that
asm-generic/access_ok.h must first see the macro definitions for
architectures that override any of the contents, but access_ok()
itself is used at least in some of the asm/uaccess.h files as well,
so it must be included in the middle of it, until more of the uaccess.h
implementation is moved to linux/uaccess.h in an architecture
independent way.

Would you prefer having an asm/access_ok.h that falls back to
the asm-generic version but can have an architecture specific
override when needed (ia64, arm64, x86, um)?

>         BTW, do we need user_addr_max() anymore?  The definition in
> asm-generic/access-ok.h is the only one, so ifndef around it is pointless.

Right, the v2 changes got rid of the last override, so it could get
hardcoded to TASK_SIZE_MAX, or we can convert the five
references to just use that instead and remove it altogether:

arch/arm64/kernel/traps.c:      if (address >= user_addr_max()) {
                 \
arch/parisc/kernel/signal.c:    if (start >= user_addr_max() - sigframe_size)
arch/parisc/kernel/signal.c:            if (A(&usp[0]) >=
user_addr_max() - 5 * sizeof(int))
lib/strncpy_from_user.c:        max_addr = user_addr_max();
lib/strnlen_user.c:     max_addr = user_addr_max();

user_addr_max() first showed up in architecture-independent code in
c5389831cda3 ("sparc: Fix user_addr_max() definition."), and from that
I think the original intent is no longer useful.

          Arnd

Re: [PATCH v2 10/18] m68k: fix access_ok for coldfire

[Arnd Bergmann]

On Fri, Feb 18, 2022 at 10:00 AM Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> >  /* We let the MMU do all checking */
> > -static inline int access_ok(const void __user *addr,
> > +static inline int access_ok(const void __user *ptr,
> >                             unsigned long size)
> >  {
> > +       unsigned long limit = TASK_SIZE;
> > +       unsigned long addr = (unsigned long)ptr;
> > +
> >         /*
> >          * XXX: for !CONFIG_CPU_HAS_ADDRESS_SPACES this really needs to check
> >          * for TASK_SIZE!
> > +        * Removing this helper is probably sufficient.
> >          */
>
> Shouldn't the above comment block be removed completely,
> as this is now implemented below?

Yes, obviously. Fixed now.

> > -       return 1;
> > +       if (IS_ENABLED(CONFIG_CPU_HAS_ADDRESS_SPACES))
> > +               return 1;

I just noticed this should have the same change that I made for the
generic version, changed it now to

+       if (IS_ENABLED(CONFIG_CPU_HAS_ADDRESS_SPACES) ||
+           !IS_ENABLED(CONFIG_MMU))

This is gone again after the cleanup patch, when the generic version
is used instead.

> > +       return (size <= limit) && (addr <= (limit - size));
> >  }
>
> Any pesky compilers that warn (or worse with -Werror) about
> "condition always true" for TASK_SIZE = 0xFFFFFFFFUL?

No, using a local variable avoids this warning.

        Arnd

RE: [PATCH v2 13/18] uaccess: generalize access_ok()

[David Laight]

From: Andy Lutomirski
> Sent: 17 February 2022 19:15
...
> This isn't actually optimal.  On x86, TASK_SIZE_MAX is a bizarre
> constant that has a very specific value to work around a bug^Wdesign
> error^Wfeature of Intel CPUs.  TASK_SIZE_MAX is the maximum address at
> which userspace is permitted to allocate memory, but there is a huge
> gap between user and kernel addresses, and any value in the gap would
> be adequate for the comparison.  If we wanted to optimize this, simply
> checking the high bit (which x86 can do without any immediate
> constants at all) would be sufficient and, for an access known to fit
> in 32 bits, one could get even fancier and completely ignore the size
> of the access.  (For accesses not known to fit in 32 bits, I suspect
> some creativity could still come up with a construction that's
> substantially faster than the one in your patch.)
> 
> So there's plenty of room for optimization here.
> 
> (This is not in any respect a NAK -- it's just an observation that
> this could be even better.)

For 64bit arch that use the top bit to separate user/kernel
you can test '(addr | size) >> 62)'.
The compiler optimises out constant sizes.

This has all been mentioned a lot of times.
You do get different fault types.

OTOH an explicit check for constant size (less than something big)
can use the cheaper test of the sign bit.
Big constant sizes could be compile time errors.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

Re: [PATCH v2 00/18] clean up asm/uaccess.h, kill set_fs for good

[Christophe Leroy]

Le 18/02/2022 à 02:50, Al Viro a écrit :
> On Thu, Feb 17, 2022 at 07:20:11AM +0000, Christophe Leroy wrote:
> 
>> And we have also
>> user_access_begin()/user_read_access_begin()/user_write_access_begin()
>> which call access_ok() then do the real work. Could be made generic with
>> call to some arch specific __user_access_begin() and friends after the
>> access_ok() and eventually the might_fault().
> 
> Not a good idea, considering the fact that we do not want to invite
> uses of "faster" variants...

I'm not sure I understand your concern.

Today in powerpc we have:

	static __must_check inline bool
	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;
	}

We could instead have a generic

	static __must_check inline bool
	user_read_access_begin(const void __user *ptr, size_t len)
	{
		if (unlikely(!access_ok(ptr, len)))
			return false;

		might_fault();

		return arch_user_read_access_begin(ptr, len);
	}

And then a powerpc specific

	static __must_check __always_inline bool
	arch_user_read_access_begin(const void __user *ptr, size_t len)
	{
		allow_read_from_user(ptr, len);
		return true;
	}
	#define arch_user_read_access_begin arch_user_read_access_begin

And a generic fallback for arch_user_read_access_begin() that does 
nothing at all.

Do you mean that in that case people might be tempted to use 
arch_user_read_access_begin() instead of using user_read_access_begin() ?

If that's the case isn't it something we could verify via checkpatch.pl ?

Today it seems to be problematic that functions in asm/uaccess.h use 
access_ok(). Such an approach would allow to get rid of access_ok() use 
in architecture's uaccess.h

Christophe

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Sergey Matyukevich]

Hi Arnd,

> From: Arnd Bergmann <arnd@arndb.de>
> 
> There are no remaining callers of set_fs(), so CONFIG_SET_FS
> can be removed globally, along with the thread_info field and
> any references to it.
> 
> This turns access_ok() into a cheaper check against TASK_SIZE_MAX.
> 
> With CONFIG_SET_FS gone, so drop all remaining references to
> set_fs()/get_fs(), mm_segment_t and uaccess_kernel().
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Tested-by: Sergey Matyukevich <sergey.matyukevich@synopsys.com> # for arc changes

Regards,
Sergey

RE: [PATCH v2 05/18] x86: remove __range_not_ok()

[David Laight]

From: Christoph Hellwig
> Sent: 18 February 2022 06:29
...
> 
> > diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c
> > index 15b058eefc4e..ee117fcf46ed 100644
> > --- a/arch/x86/kernel/stacktrace.c
> > +++ b/arch/x86/kernel/stacktrace.c
> > @@ -90,7 +90,7 @@ copy_stack_frame(const struct stack_frame_user __user *fp,
> >  {
> >  	int ret;
> >
> > -	if (__range_not_ok(fp, sizeof(*frame), TASK_SIZE))
> > +	if (!__access_ok(fp, sizeof(*frame)))
> >  		return 0;
> 
> Just switch the __get_user calls below to get_user instead.

Is this worth doing at all?
How much userspace code is actually compiled with stack frames?

Won't work well for a 32bit process on a 64bit kernel either.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Andy Lutomirski]

On Fri, Feb 18, 2022 at 1:30 AM David Laight <David.Laight@aculab.com> wrote:
>
> From: Andy Lutomirski
> > Sent: 17 February 2022 19:15
> ...
> > This isn't actually optimal.  On x86, TASK_SIZE_MAX is a bizarre
> > constant that has a very specific value to work around a bug^Wdesign
> > error^Wfeature of Intel CPUs.  TASK_SIZE_MAX is the maximum address at
> > which userspace is permitted to allocate memory, but there is a huge
> > gap between user and kernel addresses, and any value in the gap would
> > be adequate for the comparison.  If we wanted to optimize this, simply
> > checking the high bit (which x86 can do without any immediate
> > constants at all) would be sufficient and, for an access known to fit
> > in 32 bits, one could get even fancier and completely ignore the size
> > of the access.  (For accesses not known to fit in 32 bits, I suspect
> > some creativity could still come up with a construction that's
> > substantially faster than the one in your patch.)
> >
> > So there's plenty of room for optimization here.
> >
> > (This is not in any respect a NAK -- it's just an observation that
> > this could be even better.)
>
> For 64bit arch that use the top bit to separate user/kernel
> you can test '(addr | size) >> 62)'.
> The compiler optimises out constant sizes.
>
> This has all been mentioned a lot of times.
> You do get different fault types.
>
> OTOH an explicit check for constant size (less than something big)
> can use the cheaper test of the sign bit.
> Big constant sizes could be compile time errors.

The different fault type issue may well be a real problem.  Right now
the core x86 fault code reserves the right to grouch if we get #GP
instead of #PF.  We could change that.

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Thomas Bogendoerfer]

On Wed, Feb 16, 2022 at 02:13:23PM +0100, Arnd Bergmann wrote:
> 
> diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
> index db9a8e002b62..d7c89dc3426c 100644
> --- a/arch/mips/include/asm/uaccess.h
> +++ b/arch/mips/include/asm/uaccess.h
> @@ -19,6 +19,7 @@
>  #ifdef CONFIG_32BIT
>  
>  #define __UA_LIMIT 0x80000000UL
> +#define TASK_SIZE_MAX	__UA_LIMIT
>  
>  #define __UA_ADDR	".word"
>  #define __UA_LA		"la"
> @@ -33,6 +34,7 @@
>  extern u64 __ua_limit;
>  
>  #define __UA_LIMIT	__ua_limit
> +#define TASK_SIZE_MAX	XKSSEG

this doesn't work. For every access above maximum implemented virtual address
space of the CPU an address error will be issued, but not a TLB miss.
And address error isn't able to handle this situation.

With this patch

diff --git a/arch/mips/kernel/unaligned.c b/arch/mips/kernel/unaligned.c
index df4b708c04a9..3911f1481f3d 100644
--- a/arch/mips/kernel/unaligned.c
+++ b/arch/mips/kernel/unaligned.c
@@ -1480,6 +1480,13 @@ asmlinkage void do_ade(struct pt_regs *regs)
 	prev_state = exception_enter();
 	perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS,
 			1, regs, regs->cp0_badvaddr);
+
+	/* Are we prepared to handle this kernel fault?	 */
+	if (fixup_exception(regs)) {
+		current->thread.cp0_baduaddr = regs->cp0_badvaddr;
+		return;
+	}
+
 	/*
 	 * Did we catch a fault trying to load an instruction?
 	 */

I at least get my simple test cases fixed, but I'm not sure this is
correct.

Is there a reason to not also #define TASK_SIZE_MAX   __UA_LIMIT like
for the 32bit case ?

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Arnd Bergmann]

On Mon, Feb 21, 2022 at 2:24 PM Thomas Bogendoerfer
<tsbogend@alpha.franken.de> wrote:
> On Wed, Feb 16, 2022 at 02:13:23PM +0100, Arnd Bergmann wrote:
> >
> > diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
> > index db9a8e002b62..d7c89dc3426c 100644
>
> this doesn't work. For every access above maximum implemented virtual address
> space of the CPU an address error will be issued, but not a TLB miss.
> And address error isn't able to handle this situation.

Ah, so the __ex_table entry only catches TLB misses?

Does this mean it also traps for kernel memory accesses, or do those
work again? If the addresses on mips64 are separate like on
sparc64 or s390, the entire access_ok() step could be replaced
by a fixup code in the exception handler. I suppose this depends on
CONFIG_EVA and you still need a limit check at least when EVA is
disabled.

> With this patch
>
> diff --git a/arch/mips/kernel/unaligned.c b/arch/mips/kernel/unaligned.c
> index df4b708c04a9..3911f1481f3d 100644
> --- a/arch/mips/kernel/unaligned.c
> +++ b/arch/mips/kernel/unaligned.c
> @@ -1480,6 +1480,13 @@ asmlinkage void do_ade(struct pt_regs *regs)
>         prev_state = exception_enter();
>         perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS,
>                         1, regs, regs->cp0_badvaddr);
> +
> +       /* Are we prepared to handle this kernel fault?  */
> +       if (fixup_exception(regs)) {
> +               current->thread.cp0_baduaddr = regs->cp0_badvaddr;
> +               return;
> +       }
> +
>         /*
>          * Did we catch a fault trying to load an instruction?
>          */
>
> I at least get my simple test cases fixed, but I'm not sure this is
> correct.
>
> Is there a reason to not also #define TASK_SIZE_MAX   __UA_LIMIT like
> for the 32bit case ?
>

For 32-bit, the __UA_LIMIT is a compile-time constant, so the check
ends up being trivial. On all other architectures, the same thing can
be done after the set_fs removal, so I was hoping it would work here
as well.

I suspect doing the generic (size <= limit) && (addr <= (limit - size))
check on mips64 with the runtime limit ends up slightly slower
than the current code that checks a bit mask instead. If you like,
I'll update it this way, otherwise I'd need help in form of a patch
that changes the exception handling so __get_user/__put_user
also return -EFAULT for an address error.

       Arnd

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Thomas Bogendoerfer]

On Mon, Feb 21, 2022 at 03:31:23PM +0100, Arnd Bergmann wrote:
> On Mon, Feb 21, 2022 at 2:24 PM Thomas Bogendoerfer
> <tsbogend@alpha.franken.de> wrote:
> > On Wed, Feb 16, 2022 at 02:13:23PM +0100, Arnd Bergmann wrote:
> > >
> > > diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
> > > index db9a8e002b62..d7c89dc3426c 100644
> >
> > this doesn't work. For every access above maximum implemented virtual address
> > space of the CPU an address error will be issued, but not a TLB miss.
> > And address error isn't able to handle this situation.
> 
> Ah, so the __ex_table entry only catches TLB misses?

no, but there is no __ex_table handling in address error hanlder (yet).

> Does this mean it also traps for kernel memory accesses, or do those
> work again?

it will trap for every access.


> If the addresses on mips64 are separate like on
> sparc64 or s390, the entire access_ok() step could be replaced
> by a fixup code in the exception handler. I suppose this depends on
> CONFIG_EVA and you still need a limit check at least when EVA is
> disabled.

only EVA has seperate address spaces for kernel/user.

> > Is there a reason to not also #define TASK_SIZE_MAX   __UA_LIMIT like
> > for the 32bit case ?
> >
> 
> For 32-bit, the __UA_LIMIT is a compile-time constant, so the check
> ends up being trivial. On all other architectures, the same thing can
> be done after the set_fs removal, so I was hoping it would work here
> as well.

ic

> I suspect doing the generic (size <= limit) && (addr <= (limit - size))
> check on mips64 with the runtime limit ends up slightly slower
> than the current code that checks a bit mask instead. If you like,
> I'll update it this way, otherwise I'd need help in form of a patch
> that changes the exception handling so __get_user/__put_user
> also return -EFAULT for an address error.

that's what the patch does. For aligned accesses the patch should
do the right thing, but it breaks unaligned get_user/put_user.
Checking if the trapping vaddr is between end of CPU VM space and
TASK_MAX_SIZE before exception handling should do the trick. I'll
send a patch, if this works.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Thomas Bogendoerfer]

On Mon, Feb 21, 2022 at 03:31:23PM +0100, Arnd Bergmann wrote:
> I'll update it this way, otherwise I'd need help in form of a patch
> that changes the exception handling so __get_user/__put_user
> also return -EFAULT for an address error.

https://lore.kernel.org/all/20220222155345.138861-1-tsbogend@alpha.franken.de/

That does the trick.

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Thomas Bogendoerfer]

On Wed, Feb 16, 2022 at 02:13:23PM +0100, Arnd Bergmann wrote:
> diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
> index db9a8e002b62..d7c89dc3426c 100644
> --- a/arch/mips/include/asm/uaccess.h
> +++ b/arch/mips/include/asm/uaccess.h
> @@ -19,6 +19,7 @@
>  #ifdef CONFIG_32BIT
>  
>  #define __UA_LIMIT 0x80000000UL
> +#define TASK_SIZE_MAX	__UA_LIMIT

using KSEG0 instead would IMHO be the better choice. This gives the
chance to remove __UA_LIMIT completly after cleaning up ptrace.c

Thomas.

-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Arnd Bergmann]

On Wed, Feb 23, 2022 at 8:41 AM Thomas Bogendoerfer
<tsbogend@alpha.franken.de> wrote:
>
> On Wed, Feb 16, 2022 at 02:13:23PM +0100, Arnd Bergmann wrote:
> > diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
> > index db9a8e002b62..d7c89dc3426c 100644
> > --- a/arch/mips/include/asm/uaccess.h
> > +++ b/arch/mips/include/asm/uaccess.h
> > @@ -19,6 +19,7 @@
> >  #ifdef CONFIG_32BIT
> >
> >  #define __UA_LIMIT 0x80000000UL
> > +#define TASK_SIZE_MAX        __UA_LIMIT
>
> using KSEG0 instead would IMHO be the better choice. This gives the
> chance to remove __UA_LIMIT completly after cleaning up ptrace.c

Ok, changed now.

      Arnd

Re: [PATCH v2 09/18] mips: use simpler access_ok()

[Linus Torvalds]

On Mon, Feb 21, 2022 at 5:25 AM Thomas Bogendoerfer
<tsbogend@alpha.franken.de> wrote:
>
> With this patch
[ .. snip snip ..]
> I at least get my simple test cases fixed, but I'm not sure this is
> correct.

I think you really want to do that anyway, just to get things like
wild kernel pointers right (ie think get_kernel_nofault() and friends
for ftrace etc).

They shouldn't happen in any normal situation, but those kinds of
unverified pointers is why we _have_ get_kernel_nofault() in the first
place.

On x86-64, the roughly equivalent situation is that addresses that
aren't in canonical format do not take a #PF (page fault), they take a
#GP (general protection) fault.

So I think you want to do that fixup_exception() for any possible addresses.

> Is there a reason to not also #define TASK_SIZE_MAX   __UA_LIMIT like
> for the 32bit case ?

I would suggest against using a non-constant TASK_SIZE_MAX. Being
constant is literally one reason why it exists, when TASK_SIZE itself
has often been about other things (ie "32-bit process").

Having to load variables for things like get_user() is annoying, if
you could do it with a simple constant instead (where that "simple"
part is to avoid having to load big values from a constant pool -
often constants like "high bit set" can be loaded and compared against
more efficiently).

               Linus

Re: [PATCH v2 07/18] nios2: drop access_ok() check from __put_user()

[Dinh Nguyen]

On 2/16/22 07:13, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> Unlike other architectures, the nios2 version of __put_user() has an
> extra check for access_ok(), preventing it from being used to implement
> __put_kernel_nofault().
> 
> Split up put_user() along the same lines as __get_user()/get_user()
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>   arch/nios2/include/asm/uaccess.h | 56 +++++++++++++++++++-------------
>   1 file changed, 33 insertions(+), 23 deletions(-)
> 
> diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
> index ca9285a915ef..a5cbe07cf0da 100644
> --- a/arch/nios2/include/asm/uaccess.h
> +++ b/arch/nios2/include/asm/uaccess.h
> @@ -167,34 +167,44 @@ do {									\
>   	: "r" (val), "r" (ptr), "i" (-EFAULT));				\
>   }
>   
> -#define put_user(x, ptr)						\
> +#define __put_user_common(__pu_val, __pu_ptr)				\
>   ({									\
>   	long __pu_err = -EFAULT;					\
> -	__typeof__(*(ptr)) __user *__pu_ptr = (ptr);			\
> -	__typeof__(*(ptr)) __pu_val = (__typeof(*ptr))(x);		\
> -	if (access_ok(__pu_ptr, sizeof(*__pu_ptr))) {	\
> -		switch (sizeof(*__pu_ptr)) {				\
> -		case 1:							\
> -			__put_user_asm(__pu_val, "stb", __pu_ptr, __pu_err); \
> -			break;						\
> -		case 2:							\
> -			__put_user_asm(__pu_val, "sth", __pu_ptr, __pu_err); \
> -			break;						\
> -		case 4:							\
> -			__put_user_asm(__pu_val, "stw", __pu_ptr, __pu_err); \
> -			break;						\
> -		default:						\
> -			/* XXX: This looks wrong... */			\
> -			__pu_err = 0;					\
> -			if (copy_to_user(__pu_ptr, &(__pu_val),		\
> -				sizeof(*__pu_ptr)))			\
> -				__pu_err = -EFAULT;			\
> -			break;						\
> -		}							\
> +	switch (sizeof(*__pu_ptr)) {					\
> +	case 1:								\
> +		__put_user_asm(__pu_val, "stb", __pu_ptr, __pu_err);	\
> +		break;							\
> +	case 2:								\
> +		__put_user_asm(__pu_val, "sth", __pu_ptr, __pu_err);	\
> +		break;							\
> +	case 4:								\
> +		__put_user_asm(__pu_val, "stw", __pu_ptr, __pu_err);	\
> +		break;							\
> +	default:							\
> +		/* XXX: This looks wrong... */				\
> +		__pu_err = 0;						\
> +		if (__copy_to_user(__pu_ptr, &(__pu_val),		\
> +			sizeof(*__pu_ptr)))				\
> +			__pu_err = -EFAULT;				\
> +		break;							\
>   	}								\
>   	__pu_err;							\
>   })
>   
> -#define __put_user(x, ptr) put_user(x, ptr)
> +#define __put_user(x, ptr)						\
> +({									\
> +	__auto_type __pu_ptr = (ptr);					\
> +	typeof(*__pu_ptr) __pu_val = (typeof(*__pu_ptr))(x);		\
> +	__put_user_common(__pu_val, __pu_ptr);				\
> +})
> +
> +#define put_user(x, ptr)						\
> +({									\
> +	__auto_type __pu_ptr = (ptr);					\
> +	typeof(*__pu_ptr) __pu_val = (typeof(*__pu_ptr))(x);		\
> +	access_ok(__pu_ptr, sizeof(*__pu_ptr)) ?			\
> +		__put_user_common(__pu_val, __pu_ptr) :			\
> +		-EFAULT;						\
> +})
>   
>   #endif /* _ASM_NIOS2_UACCESS_H */

Acked-by: Dinh Nguyen <dinguyen@kernel.org>

Re: [PATCH v2 07/18] nios2: drop access_ok() check from __put_user()

[Arnd Bergmann]

On Thu, Feb 24, 2022 at 12:30 AM Dinh Nguyen <dinguyen@kernel.org> wrote:
> On 2/16/22 07:13, Arnd Bergmann wrote: From: Arnd Bergmann <arnd@arndb.de>
> >
> > Unlike other architectures, the nios2 version of __put_user() has an
> > extra check for access_ok(), preventing it from being used to implement
> > __put_kernel_nofault().
> >
> > Split up put_user() along the same lines as __get_user()/get_user()
> >
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>
> Acked-by: Dinh Nguyen <dinguyen@kernel.org>

Thanks! Could you also have a look at patch 2 (uaccess: fix nios2 and
microblaze get_user_8)? That one is actually more critical, and should
be backported to stable kernels.

       Arnd

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Stafford Horne]

On Wed, Feb 16, 2022 at 02:13:27PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> There are many different ways that access_ok() is defined across
> architectures, but in the end, they all just compare against the
> user_addr_max() value or they accept anything.
> 
> Provide one definition that works for most architectures, checking
> against TASK_SIZE_MAX for user processes or skipping the check inside
> of uaccess_kernel() sections.
> 
> For architectures without CONFIG_SET_FS(), this should be the fastest
> check, as it comes down to a single comparison of a pointer against a
> compile-time constant, while the architecture specific versions tend to
> do something more complex for historic reasons or get something wrong.
> 
> Type checking for __user annotations is handled inconsistently across
> architectures, but this is easily simplified as well by using an inline
> function that takes a 'const void __user *' argument. A handful of
> callers need an extra __user annotation for this.
> 
> Some architectures had trick to use 33-bit or 65-bit arithmetic on the
> addresses to calculate the overflow, however this simpler version uses
> fewer registers, which means it can produce better object code in the
> end despite needing a second (statically predicted) branch.
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Mark Rutland <mark.rutland@arm.com> [arm64, asm-generic]
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
...
>  arch/openrisc/include/asm/uaccess.h   | 19 +--------
...
>  include/asm-generic/access_ok.h       | 59 +++++++++++++++++++++++++++
>  include/asm-generic/uaccess.h         | 21 +---------
>  include/linux/uaccess.h               |  7 ----
>  32 files changed, 109 insertions(+), 366 deletions(-)
> 
...
> diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h
> index 120f5005461b..8f049ec99b3e 100644
> --- a/arch/openrisc/include/asm/uaccess.h
> +++ b/arch/openrisc/include/asm/uaccess.h
> @@ -45,21 +45,7 @@
>  
>  #define uaccess_kernel()	(get_fs() == KERNEL_DS)
>  
> -/* Ensure that the range from addr to addr+size is all within the process'
> - * address space
> - */
> -static inline int __range_ok(unsigned long addr, unsigned long size)
> -{
> -	const mm_segment_t fs = get_fs();
> -
> -	return size <= fs && addr <= (fs - size);
> -}
> -
> -#define access_ok(addr, size)						\
> -({ 									\
> -	__chk_user_ptr(addr);						\
> -	__range_ok((unsigned long)(addr), (size));			\
> -})
> +#include <asm-generic/access_ok.h>

I was going to ask why we are missing __chk_user_ptr in the generic version.
But this is basically now a no-op so I think its OK.

>  /*
>   * These are the main single-value transfer routines.  They automatically
> @@ -268,9 +254,6 @@ clear_user(void __user *addr, unsigned long size)
>  	return size;
>  }
>  
> -#define user_addr_max() \
> -	(uaccess_kernel() ? ~0UL : TASK_SIZE)
> -
>  extern long strncpy_from_user(char *dest, const char __user *src, long count);
>  
>  extern __must_check long strnlen_user(const char __user *str, long n);

...
> diff --git a/include/asm-generic/access_ok.h b/include/asm-generic/access_ok.h
> new file mode 100644
> index 000000000000..1aad8964d2ed
> --- /dev/null
> +++ b/include/asm-generic/access_ok.h
> @@ -0,0 +1,59 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef __ASM_GENERIC_ACCESS_OK_H__
> +#define __ASM_GENERIC_ACCESS_OK_H__
> +
> +/*
> + * Checking whether a pointer is valid for user space access.
> + * These definitions work on most architectures, but overrides can
> + * be used where necessary.
> + */
> +
> +/*
> + * architectures with compat tasks have a variable TASK_SIZE and should
> + * override this to a constant.
> + */
> +#ifndef TASK_SIZE_MAX
> +#define TASK_SIZE_MAX			TASK_SIZE
> +#endif
> +
> +#ifndef uaccess_kernel
> +#ifdef CONFIG_SET_FS
> +#define uaccess_kernel()		(get_fs().seg == KERNEL_DS.seg)
> +#else
> +#define uaccess_kernel()		(0)
> +#endif
> +#endif
> +
> +#ifndef user_addr_max
> +#define user_addr_max()			(uaccess_kernel() ? ~0UL : TASK_SIZE_MAX)
> +#endif
> +
> +#ifndef __access_ok
> +/*
> + * 'size' is a compile-time constant for most callers, so optimize for
> + * this case to turn the check into a single comparison against a constant
> + * limit and catch all possible overflows.
> + * On architectures with separate user address space (m68k, s390, parisc,
> + * sparc64) or those without an MMU, this should always return true.
> + *
> + * This version was originally contributed by Jonas Bonn for the
> + * OpenRISC architecture, and was found to be the most efficient
> + * for constant 'size' and 'limit' values.
> + */
> +static inline int __access_ok(const void __user *ptr, unsigned long size)
> +{
> +	unsigned long limit = user_addr_max();
> +	unsigned long addr = (unsigned long)ptr;
> +
> +	if (IS_ENABLED(CONFIG_ALTERNATE_USER_ADDRESS_SPACE))
> +		return true;
> +
> +	return (size <= limit) && (addr <= (limit - size));
> +}
> +#endif
> +
> +#ifndef access_ok
> +#define access_ok(addr, size) likely(__access_ok(addr, size))
> +#endif
> +
> +#endif
> diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
> index 0870fa11a7c5..ebc685dc8d74 100644
> --- a/include/asm-generic/uaccess.h
> +++ b/include/asm-generic/uaccess.h
> @@ -114,28 +114,9 @@ static inline void set_fs(mm_segment_t fs)
>  }
>  #endif
>  
> -#ifndef uaccess_kernel
> -#define uaccess_kernel() (get_fs().seg == KERNEL_DS.seg)
> -#endif
> -
> -#ifndef user_addr_max
> -#define user_addr_max() (uaccess_kernel() ? ~0UL : TASK_SIZE)
> -#endif
> -
>  #endif /* CONFIG_SET_FS */
>  
> -#define access_ok(addr, size) __access_ok((unsigned long)(addr),(size))
> -
> -/*
> - * The architecture should really override this if possible, at least
> - * doing a check on the get_fs()
> - */
> -#ifndef __access_ok
> -static inline int __access_ok(unsigned long addr, unsigned long size)
> -{
> -	return 1;
> -}
> -#endif
> +#include <asm-generic/access_ok.h>
>  
>  /*
>   * These are the main single-value transfer routines.  They automatically
> diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
> index 67e9bc94dc40..2c31667e62e0 100644
> --- a/include/linux/uaccess.h
> +++ b/include/linux/uaccess.h
> @@ -33,13 +33,6 @@ typedef struct {
>  	/* empty dummy */
>  } mm_segment_t;
>  
> -#ifndef TASK_SIZE_MAX
> -#define TASK_SIZE_MAX			TASK_SIZE
> -#endif
> -
> -#define uaccess_kernel()		(false)
> -#define user_addr_max()			(TASK_SIZE_MAX)
> -
>  static inline mm_segment_t force_uaccess_begin(void)
>  {
>  	return (mm_segment_t) { };


Acked-by: Stafford Horne <shorne@gmail.com> [openrisc, asm-generic]

Thanks!

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Arnd Bergmann]

On Thu, Feb 24, 2022 at 9:29 AM Stafford Horne <shorne@gmail.com> wrote:

> > -
> > -#define access_ok(addr, size)                                                \
> > -({                                                                   \
> > -     __chk_user_ptr(addr);                                           \
> > -     __range_ok((unsigned long)(addr), (size));                      \
> > -})
> > +#include <asm-generic/access_ok.h>
>
> I was going to ask why we are missing __chk_user_ptr in the generic version.
> But this is basically now a no-op so I think its OK.

Correct, the type checking is implied by making __access_ok() an inline
function that takes a __user pointer.

> Acked-by: Stafford Horne <shorne@gmail.com> [openrisc, asm-generic]

Thanks!

       Arnd

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Stafford Horne]

On Wed, Feb 16, 2022 at 02:13:32PM +0100, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> There are no remaining callers of set_fs(), so CONFIG_SET_FS
> can be removed globally, along with the thread_info field and
> any references to it.
> 
> This turns access_ok() into a cheaper check against TASK_SIZE_MAX.
> 
> With CONFIG_SET_FS gone, so drop all remaining references to
> set_fs()/get_fs(), mm_segment_t and uaccess_kernel().
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
...
>  arch/openrisc/Kconfig                     |  1 -
>  arch/openrisc/include/asm/thread_info.h   |  7 ---
>  arch/openrisc/include/asm/uaccess.h       | 23 --------
...
>  fs/exec.c                                 |  6 --
>  include/asm-generic/access_ok.h           | 10 +---
>  include/asm-generic/uaccess.h             | 25 +-------
>  include/linux/syscalls.h                  |  4 --
>  include/linux/uaccess.h                   | 33 -----------
>  include/rdma/ib.h                         |  2 +-
>  kernel/events/callchain.c                 |  4 --
>  kernel/events/core.c                      |  3 -
>  kernel/exit.c                             | 14 -----
>  kernel/kthread.c                          |  5 --
>  kernel/stacktrace.c                       |  3 -
>  kernel/trace/bpf_trace.c                  |  4 --
>  mm/maccess.c                              | 11 ----
>  mm/memory.c                               |  8 ---
>  net/bpfilter/bpfilter_kern.c              |  2 +-
>  72 files changed, 10 insertions(+), 522 deletions(-)
>  delete mode 100644 arch/arc/include/asm/segment.h
>  delete mode 100644 arch/csky/include/asm/segment.h
>  delete mode 100644 arch/h8300/include/asm/segment.h
> 
> diff --git a/arch/Kconfig b/arch/Kconfig
> index fa5db36bda67..99349547afed 100644
> --- a/arch/Kconfig
> +++ b/arch/Kconfig
> @@ -24,9 +24,6 @@ config KEXEC_ELF
>  config HAVE_IMA_KEXEC
>  	bool
>  
> -config SET_FS
> -	bool
> -
>  config HOTPLUG_SMT
>  	bool
>  
...
> diff --git a/arch/openrisc/Kconfig b/arch/openrisc/Kconfig
> index f724b3f1aeed..0d68adf6e02b 100644
> --- a/arch/openrisc/Kconfig
> +++ b/arch/openrisc/Kconfig
> @@ -36,7 +36,6 @@ config OPENRISC
>  	select ARCH_WANT_FRAME_POINTERS
>  	select GENERIC_IRQ_MULTI_HANDLER
>  	select MMU_GATHER_NO_RANGE if MMU
> -	select SET_FS
>  	select TRACE_IRQFLAGS_SUPPORT
>  
>  config CPU_BIG_ENDIAN
> diff --git a/arch/openrisc/include/asm/thread_info.h b/arch/openrisc/include/asm/thread_info.h
> index 659834ab87fa..4af3049c34c2 100644
> --- a/arch/openrisc/include/asm/thread_info.h
> +++ b/arch/openrisc/include/asm/thread_info.h
> @@ -40,18 +40,12 @@
>   */
>  #ifndef __ASSEMBLY__
>  
> -typedef unsigned long mm_segment_t;
> -
>  struct thread_info {
>  	struct task_struct	*task;		/* main task structure */
>  	unsigned long		flags;		/* low level flags */
>  	__u32			cpu;		/* current CPU */
>  	__s32			preempt_count; /* 0 => preemptable, <0 => BUG */
>  
> -	mm_segment_t		addr_limit; /* thread address space:
> -					       0-0x7FFFFFFF for user-thead
> -					       0-0xFFFFFFFF for kernel-thread
> -					     */
>  	__u8			supervisor_stack[0];
>  
>  	/* saved context data */
> @@ -71,7 +65,6 @@ struct thread_info {
>  	.flags		= 0,				\
>  	.cpu		= 0,				\
>  	.preempt_count	= INIT_PREEMPT_COUNT,		\
> -	.addr_limit	= KERNEL_DS,			\
>  	.ksp            = 0,                            \
>  }
>  
> diff --git a/arch/openrisc/include/asm/uaccess.h b/arch/openrisc/include/asm/uaccess.h
> index 8f049ec99b3e..d6500a374e18 100644
> --- a/arch/openrisc/include/asm/uaccess.h
> +++ b/arch/openrisc/include/asm/uaccess.h
> @@ -22,29 +22,6 @@
>  #include <linux/string.h>
>  #include <asm/page.h>
>  #include <asm/extable.h>
> -
> -/*
> - * The fs value determines whether argument validity checking should be
> - * performed or not.  If get_fs() == USER_DS, checking is performed, with
> - * get_fs() == KERNEL_DS, checking is bypassed.
> - *
> - * For historical reasons, these macros are grossly misnamed.
> - */
> -
> -/* addr_limit is the maximum accessible address for the task. we misuse
> - * the KERNEL_DS and USER_DS values to both assign and compare the
> - * addr_limit values through the equally misnamed get/set_fs macros.
> - * (see above)
> - */
> -
> -#define KERNEL_DS	(~0UL)
> -
> -#define USER_DS		(TASK_SIZE)
> -#define get_fs()	(current_thread_info()->addr_limit)
> -#define set_fs(x)	(current_thread_info()->addr_limit = (x))
> -
> -#define uaccess_kernel()	(get_fs() == KERNEL_DS)
> -
>  #include <asm-generic/access_ok.h>
>  
>  /*
...
> diff --git a/fs/exec.c b/fs/exec.c
> index 79f2c9483302..bc68a0c089ac 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -1303,12 +1303,6 @@ int begin_new_exec(struct linux_binprm * bprm)
>  	if (retval)
>  		goto out_unlock;
>  
> -	/*
> -	 * Ensure that the uaccess routines can actually operate on userspace
> -	 * pointers:
> -	 */
> -	force_uaccess_begin();
> -
>  	if (me->flags & PF_KTHREAD)
>  		free_kthread_struct(me);
>  	me->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD |
> diff --git a/include/asm-generic/access_ok.h b/include/asm-generic/access_ok.h
> index 1aad8964d2ed..88a7cb5d9aad 100644
> --- a/include/asm-generic/access_ok.h
> +++ b/include/asm-generic/access_ok.h
> @@ -16,16 +16,8 @@
>  #define TASK_SIZE_MAX			TASK_SIZE
>  #endif
>  
> -#ifndef uaccess_kernel
> -#ifdef CONFIG_SET_FS
> -#define uaccess_kernel()		(get_fs().seg == KERNEL_DS.seg)
> -#else
> -#define uaccess_kernel()		(0)
> -#endif
> -#endif
> -
>  #ifndef user_addr_max
> -#define user_addr_max()			(uaccess_kernel() ? ~0UL : TASK_SIZE_MAX)
> +#define user_addr_max()			TASK_SIZE_MAX
>  #endif
>  
>  #ifndef __access_ok
> diff --git a/include/asm-generic/uaccess.h b/include/asm-generic/uaccess.h
> index ebc685dc8d74..a5be9e61a2a2 100644
> --- a/include/asm-generic/uaccess.h
> +++ b/include/asm-generic/uaccess.h
> @@ -8,6 +8,7 @@
>   * address space, e.g. all NOMMU machines.
>   */
>  #include <linux/string.h>
> +#include <asm-generic/access_ok.h>
>  
>  #ifdef CONFIG_UACCESS_MEMCPY
>  #include <asm/unaligned.h>
> @@ -94,30 +95,6 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long n)
>  #define INLINE_COPY_TO_USER
>  #endif /* CONFIG_UACCESS_MEMCPY */
>  
> -#ifdef CONFIG_SET_FS
> -#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
> -
> -#ifndef KERNEL_DS
> -#define KERNEL_DS	MAKE_MM_SEG(~0UL)
> -#endif
> -
> -#ifndef USER_DS
> -#define USER_DS		MAKE_MM_SEG(TASK_SIZE - 1)
> -#endif
> -
> -#ifndef get_fs
> -#define get_fs()	(current_thread_info()->addr_limit)
> -
> -static inline void set_fs(mm_segment_t fs)
> -{
> -	current_thread_info()->addr_limit = fs;
> -}
> -#endif
> -
> -#endif /* CONFIG_SET_FS */
> -
> -#include <asm-generic/access_ok.h>
> -
>  /*
>   * These are the main single-value transfer routines.  They automatically
>   * use the right size if we just have the right pointer type.
> diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
> index 819c0cb00b6d..a34b0f9a9972 100644
> --- a/include/linux/syscalls.h
> +++ b/include/linux/syscalls.h
> @@ -290,10 +290,6 @@ static inline void addr_limit_user_check(void)
>  		return;
>  #endif
>  
> -	if (CHECK_DATA_CORRUPTION(uaccess_kernel(),
> -				  "Invalid address limit on user-mode return"))
> -		force_sig(SIGKILL);
> -
>  #ifdef TIF_FSCHECK
>  	clear_thread_flag(TIF_FSCHECK);
>  #endif
> diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
> index 2c31667e62e0..2421a41f3a8e 100644
> --- a/include/linux/uaccess.h
> +++ b/include/linux/uaccess.h
> @@ -10,39 +10,6 @@
>  
>  #include <asm/uaccess.h>
>  
> -#ifdef CONFIG_SET_FS
> -/*
> - * Force the uaccess routines to be wired up for actual userspace access,
> - * overriding any possible set_fs(KERNEL_DS) still lingering around.  Undone
> - * using force_uaccess_end below.
> - */
> -static inline mm_segment_t force_uaccess_begin(void)
> -{
> -	mm_segment_t fs = get_fs();
> -
> -	set_fs(USER_DS);
> -	return fs;
> -}
> -
> -static inline void force_uaccess_end(mm_segment_t oldfs)
> -{
> -	set_fs(oldfs);
> -}
> -#else /* CONFIG_SET_FS */
> -typedef struct {
> -	/* empty dummy */
> -} mm_segment_t;
> -
> -static inline mm_segment_t force_uaccess_begin(void)
> -{
> -	return (mm_segment_t) { };
> -}
> -
> -static inline void force_uaccess_end(mm_segment_t oldfs)
> -{
> -}
> -#endif /* CONFIG_SET_FS */
> -
>  /*
>   * Architectures should provide two primitives (raw_copy_{to,from}_user())
>   * and get rid of their private instances of copy_{to,from}_user() and
> diff --git a/include/rdma/ib.h b/include/rdma/ib.h
> index 83139b9ce409..f7c185ff7a11 100644
> --- a/include/rdma/ib.h
> +++ b/include/rdma/ib.h
> @@ -75,7 +75,7 @@ struct sockaddr_ib {
>   */
>  static inline bool ib_safe_file_access(struct file *filp)
>  {
> -	return filp->f_cred == current_cred() && !uaccess_kernel();
> +	return filp->f_cred == current_cred();
>  }
>  
>  #endif /* _RDMA_IB_H */
> diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c
> index 58cbe357fb2b..1273be84392c 100644
> --- a/kernel/events/callchain.c
> +++ b/kernel/events/callchain.c
> @@ -209,17 +209,13 @@ get_perf_callchain(struct pt_regs *regs, u32 init_nr, bool kernel, bool user,
>  		}
>  
>  		if (regs) {
> -			mm_segment_t fs;
> -
>  			if (crosstask)
>  				goto exit_put;
>  
>  			if (add_mark)
>  				perf_callchain_store_context(&ctx, PERF_CONTEXT_USER);
>  
> -			fs = force_uaccess_begin();
>  			perf_callchain_user(&ctx, regs);
> -			force_uaccess_end(fs);
>  		}
>  	}
>  
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 57c7197838db..11ca7303d6df 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -6746,7 +6746,6 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
>  		unsigned long sp;
>  		unsigned int rem;
>  		u64 dyn_size;
> -		mm_segment_t fs;
>  
>  		/*
>  		 * We dump:
> @@ -6764,9 +6763,7 @@ perf_output_sample_ustack(struct perf_output_handle *handle, u64 dump_size,
>  
>  		/* Data. */
>  		sp = perf_user_stack_pointer(regs);
> -		fs = force_uaccess_begin();
>  		rem = __output_copy_user(handle, (void *) sp, dump_size);
> -		force_uaccess_end(fs);
>  		dyn_size = dump_size - rem;
>  
>  		perf_output_skip(handle, rem);
> diff --git a/kernel/exit.c b/kernel/exit.c
> index b00a25bb4ab9..0884a75bc2f8 100644
> --- a/kernel/exit.c
> +++ b/kernel/exit.c
> @@ -737,20 +737,6 @@ void __noreturn do_exit(long code)
>  
>  	WARN_ON(blk_needs_flush_plug(tsk));
>  
> -	/*
> -	 * If do_dead is called because this processes oopsed, it's possible
> -	 * that get_fs() was left as KERNEL_DS, so reset it to USER_DS before
> -	 * continuing. Amongst other possible reasons, this is to prevent
> -	 * mm_release()->clear_child_tid() from writing to a user-controlled
> -	 * kernel address.
> -	 *
> -	 * On uptodate architectures force_uaccess_begin is a noop.  On
> -	 * architectures that still have set_fs/get_fs in addition to handling
> -	 * oopses handles kernel threads that run as set_fs(KERNEL_DS) by
> -	 * default.
> -	 */
> -	force_uaccess_begin();
> -
>  	kcov_task_exit(tsk);
>  
>  	coredump_task_exit(tsk);
> diff --git a/kernel/kthread.c b/kernel/kthread.c
> index 38c6dd822da8..16c2275d4b50 100644
> --- a/kernel/kthread.c
> +++ b/kernel/kthread.c
> @@ -55,7 +55,6 @@ struct kthread {
>  	int result;
>  	int (*threadfn)(void *);
>  	void *data;
> -	mm_segment_t oldfs;
>  	struct completion parked;
>  	struct completion exited;
>  #ifdef CONFIG_BLK_CGROUP
> @@ -1441,8 +1440,6 @@ void kthread_use_mm(struct mm_struct *mm)
>  		mmdrop(active_mm);
>  	else
>  		smp_mb();
> -
> -	to_kthread(tsk)->oldfs = force_uaccess_begin();
>  }
>  EXPORT_SYMBOL_GPL(kthread_use_mm);
>  
> @@ -1457,8 +1454,6 @@ void kthread_unuse_mm(struct mm_struct *mm)
>  	WARN_ON_ONCE(!(tsk->flags & PF_KTHREAD));
>  	WARN_ON_ONCE(!tsk->mm);
>  
> -	force_uaccess_end(to_kthread(tsk)->oldfs);
> -
>  	task_lock(tsk);
>  	/*
>  	 * When a kthread stops operating on an address space, the loop
> diff --git a/kernel/stacktrace.c b/kernel/stacktrace.c
> index 9c625257023d..9ed5ce989415 100644
> --- a/kernel/stacktrace.c
> +++ b/kernel/stacktrace.c
> @@ -226,15 +226,12 @@ unsigned int stack_trace_save_user(unsigned long *store, unsigned int size)
>  		.store	= store,
>  		.size	= size,
>  	};
> -	mm_segment_t fs;
>  
>  	/* Trace user stack if not a kernel thread */
>  	if (current->flags & PF_KTHREAD)
>  		return 0;
>  
> -	fs = force_uaccess_begin();
>  	arch_stack_walk_user(consume_entry, &c, task_pt_regs(current));
> -	force_uaccess_end(fs);
>  
>  	return c.len;
>  }
> diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c
> index 21aa30644219..8115fff17018 100644
> --- a/kernel/trace/bpf_trace.c
> +++ b/kernel/trace/bpf_trace.c
> @@ -332,8 +332,6 @@ BPF_CALL_3(bpf_probe_write_user, void __user *, unsafe_ptr, const void *, src,
>  	if (unlikely(in_interrupt() ||
>  		     current->flags & (PF_KTHREAD | PF_EXITING)))
>  		return -EPERM;
> -	if (unlikely(uaccess_kernel()))
> -		return -EPERM;
>  	if (unlikely(!nmi_uaccess_okay()))
>  		return -EPERM;
>  
> @@ -835,8 +833,6 @@ static int bpf_send_signal_common(u32 sig, enum pid_type type)
>  	 */
>  	if (unlikely(current->flags & (PF_KTHREAD | PF_EXITING)))
>  		return -EPERM;
> -	if (unlikely(uaccess_kernel()))
> -		return -EPERM;
>  	if (unlikely(!nmi_uaccess_okay()))
>  		return -EPERM;
>  
> diff --git a/mm/maccess.c b/mm/maccess.c
> index cbd1b3959af2..106820b33a2b 100644
> --- a/mm/maccess.c
> +++ b/mm/maccess.c
> @@ -113,14 +113,11 @@ long strncpy_from_kernel_nofault(char *dst, const void *unsafe_addr, long count)
>  long copy_from_user_nofault(void *dst, const void __user *src, size_t size)
>  {
>  	long ret = -EFAULT;
> -	mm_segment_t old_fs = force_uaccess_begin();
> -
>  	if (access_ok(src, size)) {
>  		pagefault_disable();
>  		ret = __copy_from_user_inatomic(dst, src, size);
>  		pagefault_enable();
>  	}
> -	force_uaccess_end(old_fs);
>  
>  	if (ret)
>  		return -EFAULT;
> @@ -140,14 +137,12 @@ EXPORT_SYMBOL_GPL(copy_from_user_nofault);
>  long copy_to_user_nofault(void __user *dst, const void *src, size_t size)
>  {
>  	long ret = -EFAULT;
> -	mm_segment_t old_fs = force_uaccess_begin();
>  
>  	if (access_ok(dst, size)) {
>  		pagefault_disable();
>  		ret = __copy_to_user_inatomic(dst, src, size);
>  		pagefault_enable();
>  	}
> -	force_uaccess_end(old_fs);
>  
>  	if (ret)
>  		return -EFAULT;
> @@ -176,17 +171,14 @@ EXPORT_SYMBOL_GPL(copy_to_user_nofault);
>  long strncpy_from_user_nofault(char *dst, const void __user *unsafe_addr,
>  			      long count)
>  {
> -	mm_segment_t old_fs;
>  	long ret;
>  
>  	if (unlikely(count <= 0))
>  		return 0;
>  
> -	old_fs = force_uaccess_begin();
>  	pagefault_disable();
>  	ret = strncpy_from_user(dst, unsafe_addr, count);
>  	pagefault_enable();
> -	force_uaccess_end(old_fs);
>  
>  	if (ret >= count) {
>  		ret = count;
> @@ -216,14 +208,11 @@ long strncpy_from_user_nofault(char *dst, const void __user *unsafe_addr,
>   */
>  long strnlen_user_nofault(const void __user *unsafe_addr, long count)
>  {
> -	mm_segment_t old_fs;
>  	int ret;
>  
> -	old_fs = force_uaccess_begin();
>  	pagefault_disable();
>  	ret = strnlen_user(unsafe_addr, count);
>  	pagefault_enable();
> -	force_uaccess_end(old_fs);
>  
>  	return ret;
>  }
> diff --git a/mm/memory.c b/mm/memory.c
> index c125c4969913..9a6ebf68a846 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -5256,14 +5256,6 @@ void print_vma_addr(char *prefix, unsigned long ip)
>  #if defined(CONFIG_PROVE_LOCKING) || defined(CONFIG_DEBUG_ATOMIC_SLEEP)
>  void __might_fault(const char *file, int line)
>  {
> -	/*
> -	 * Some code (nfs/sunrpc) uses socket ops on kernel memory while
> -	 * holding the mmap_lock, this is safe because kernel memory doesn't
> -	 * get paged out, therefore we'll never actually fault, and the
> -	 * below annotations will generate false positives.
> -	 */
> -	if (uaccess_kernel())
> -		return;
>  	if (pagefault_disabled())
>  		return;
>  	__might_sleep(file, line);
> diff --git a/net/bpfilter/bpfilter_kern.c b/net/bpfilter/bpfilter_kern.c
> index 51a941b56ec3..422ec6e7ccff 100644

Acked-by: Stafford Horne <shorne@gmail.com> [openrisc, asm-generic]

Thanks!

Re: [PATCH v2 02/18] uaccess: fix nios2 and microblaze get_user_8()

[Dinh Nguyen]

On 2/16/22 07:13, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> These two architectures implement 8-byte get_user() through
> a memcpy() into a four-byte variable, which won't fit.
> 
> Use a temporary 64-bit variable instead here, and use a double
> cast the way that risc-v and openrisc do to avoid compile-time
> warnings.
> 
> Fixes: 6a090e97972d ("arch/microblaze: support get_user() of size 8 bytes")
> Fixes: 5ccc6af5e88e ("nios2: Memory management")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>   arch/microblaze/include/asm/uaccess.h | 18 +++++++++---------
>   arch/nios2/include/asm/uaccess.h      | 26 ++++++++++++++++----------
>   2 files changed, 25 insertions(+), 19 deletions(-)
> 
> diff --git a/arch/microblaze/include/asm/uaccess.h b/arch/microblaze/include/asm/uaccess.h
> index 5b6e0e7788f4..3fe96979d2c6 100644
> --- a/arch/microblaze/include/asm/uaccess.h
> +++ b/arch/microblaze/include/asm/uaccess.h
> @@ -130,27 +130,27 @@ extern long __user_bad(void);
>   
>   #define __get_user(x, ptr)						\
>   ({									\
> -	unsigned long __gu_val = 0;					\
>   	long __gu_err;							\
>   	switch (sizeof(*(ptr))) {					\
>   	case 1:								\
> -		__get_user_asm("lbu", (ptr), __gu_val, __gu_err);	\
> +		__get_user_asm("lbu", (ptr), x, __gu_err);		\
>   		break;							\
>   	case 2:								\
> -		__get_user_asm("lhu", (ptr), __gu_val, __gu_err);	\
> +		__get_user_asm("lhu", (ptr), x, __gu_err);		\
>   		break;							\
>   	case 4:								\
> -		__get_user_asm("lw", (ptr), __gu_val, __gu_err);	\
> +		__get_user_asm("lw", (ptr), x, __gu_err);		\
>   		break;							\
> -	case 8:								\
> -		__gu_err = __copy_from_user(&__gu_val, ptr, 8);		\
> -		if (__gu_err)						\
> -			__gu_err = -EFAULT;				\
> +	case 8: {							\
> +		__u64 __x = 0;						\
> +		__gu_err = raw_copy_from_user(&__x, ptr, 8) ?		\
> +							-EFAULT : 0;	\
> +		(x) = (typeof(x))(typeof((x) - (x)))__x;		\
>   		break;							\
> +	}								\
>   	default:							\
>   		/* __gu_val = 0; __gu_err = -EINVAL;*/ __gu_err = __user_bad();\
>   	}								\
> -	x = (__force __typeof__(*(ptr))) __gu_val;			\
>   	__gu_err;							\
>   })
>   
> diff --git a/arch/nios2/include/asm/uaccess.h b/arch/nios2/include/asm/uaccess.h
> index ba9340e96fd4..ca9285a915ef 100644
> --- a/arch/nios2/include/asm/uaccess.h
> +++ b/arch/nios2/include/asm/uaccess.h
> @@ -88,6 +88,7 @@ extern __must_check long strnlen_user(const char __user *s, long n);
>   /* Optimized macros */
>   #define __get_user_asm(val, insn, addr, err)				\
>   {									\
> +	unsigned long __gu_val;						\
>   	__asm__ __volatile__(						\
>   	"       movi    %0, %3\n"					\
>   	"1:   " insn " %1, 0(%2)\n"					\
> @@ -96,14 +97,20 @@ extern __must_check long strnlen_user(const char __user *s, long n);
>   	"       .section __ex_table,\"a\"\n"				\
>   	"       .word 1b, 2b\n"						\
>   	"       .previous"						\
> -	: "=&r" (err), "=r" (val)					\
> +	: "=&r" (err), "=r" (__gu_val)					\
>   	: "r" (addr), "i" (-EFAULT));					\
> +	val = (__force __typeof__(*(addr)))__gu_val;			\
>   }
>   
> -#define __get_user_unknown(val, size, ptr, err) do {			\
> +extern void __get_user_unknown(void);
> +
> +#define __get_user_8(val, ptr, err) do {				\
> +	u64 __val = 0;							\
>   	err = 0;							\
> -	if (__copy_from_user(&(val), ptr, size)) {			\
> +	if (raw_copy_from_user(&(__val), ptr, sizeof(val))) {		\
>   		err = -EFAULT;						\
> +	} else {							\
> +		val = (typeof(val))(typeof((val) - (val)))__val;	\
>   	}								\
>   	} while (0)
>   
> @@ -119,8 +126,11 @@ do {									\
>   	case 4:								\
>   		__get_user_asm(val, "ldw", ptr, err);			\
>   		break;							\
> +	case 8:								\
> +		__get_user_8(val, ptr, err);				\
> +		break;							\
>   	default:							\
> -		__get_user_unknown(val, size, ptr, err);		\
> +		__get_user_unknown();					\
>   		break;							\
>   	}								\
>   } while (0)
> @@ -129,9 +139,7 @@ do {									\
>   	({								\
>   	long __gu_err = -EFAULT;					\
>   	const __typeof__(*(ptr)) __user *__gu_ptr = (ptr);		\
> -	unsigned long __gu_val = 0;					\
> -	__get_user_common(__gu_val, sizeof(*(ptr)), __gu_ptr, __gu_err);\
> -	(x) = (__force __typeof__(x))__gu_val;				\
> +	__get_user_common(x, sizeof(*(ptr)), __gu_ptr, __gu_err);	\
>   	__gu_err;							\
>   	})
>   
> @@ -139,11 +147,9 @@ do {									\
>   ({									\
>   	long __gu_err = -EFAULT;					\
>   	const __typeof__(*(ptr)) __user *__gu_ptr = (ptr);		\
> -	unsigned long __gu_val = 0;					\
>   	if (access_ok( __gu_ptr, sizeof(*__gu_ptr)))	\
> -		__get_user_common(__gu_val, sizeof(*__gu_ptr),		\
> +		__get_user_common(x, sizeof(*__gu_ptr),			\
>   			__gu_ptr, __gu_err);				\
> -	(x) = (__force __typeof__(x))__gu_val;				\
>   	__gu_err;							\
>   })
>   

Acked-by: Dinh Nguyen <dinguyen@kernel.org>

Re: [PATCH v2 12/18] uaccess: fix type mismatch warnings from access_ok()

[Dinh Nguyen]

On 2/16/22 07:13, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> On some architectures, access_ok() does not do any argument type
> checking, so replacing the definition with a generic one causes
> a few warnings for harmless issues that were never caught before.
> 
> Fix the ones that I found either through my own test builds or
> that were reported by the 0-day bot.
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>   arch/arc/kernel/process.c           |  2 +-
>   arch/arm/kernel/swp_emulate.c       |  2 +-
>   arch/arm/kernel/traps.c             |  2 +-
>   arch/csky/kernel/signal.c           |  2 +-
>   arch/mips/sibyte/common/sb_tbprof.c |  6 +++---
>   arch/nios2/kernel/signal.c          | 20 +++++++++++---------

Acked-by: Dinh Nguyen <dinguyen@kernel.org>

Re: [PATCH v2 13/18] uaccess: generalize access_ok()

[Dinh Nguyen]

On 2/16/22 07:13, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> There are many different ways that access_ok() is defined across
> architectures, but in the end, they all just compare against the
> user_addr_max() value or they accept anything.
> 
> Provide one definition that works for most architectures, checking
> against TASK_SIZE_MAX for user processes or skipping the check inside
> of uaccess_kernel() sections.
> 
> For architectures without CONFIG_SET_FS(), this should be the fastest
> check, as it comes down to a single comparison of a pointer against a
> compile-time constant, while the architecture specific versions tend to
> do something more complex for historic reasons or get something wrong.
> 
> Type checking for __user annotations is handled inconsistently across
> architectures, but this is easily simplified as well by using an inline
> function that takes a 'const void __user *' argument. A handful of
> callers need an extra __user annotation for this.
> 
> Some architectures had trick to use 33-bit or 65-bit arithmetic on the
> addresses to calculate the overflow, however this simpler version uses
> fewer registers, which means it can produce better object code in the
> end despite needing a second (statically predicted) branch.
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Mark Rutland <mark.rutland@arm.com> [arm64, asm-generic]
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>   arch/Kconfig                          |  7 ++++
>   arch/alpha/include/asm/uaccess.h      | 34 +++------------
>   arch/arc/include/asm/uaccess.h        | 29 -------------
>   arch/arm/include/asm/uaccess.h        | 20 +--------
>   arch/arm64/include/asm/uaccess.h      | 11 ++---
>   arch/csky/include/asm/uaccess.h       |  8 ----
>   arch/hexagon/include/asm/uaccess.h    | 25 ------------
>   arch/ia64/include/asm/uaccess.h       |  5 +--
>   arch/m68k/Kconfig.cpu                 |  1 +
>   arch/m68k/include/asm/uaccess.h       | 19 +--------
>   arch/microblaze/include/asm/uaccess.h |  8 +---
>   arch/mips/include/asm/uaccess.h       | 29 +------------
>   arch/nds32/include/asm/uaccess.h      |  7 +---
>   arch/nios2/include/asm/uaccess.h      | 11 +----

Acked-by: Dinh Nguyen <dinguyen@kernel.org>

Re: [PATCH v2 18/18] uaccess: drop maining CONFIG_SET_FS users

[Dinh Nguyen]

On 2/16/22 07:13, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> There are no remaining callers of set_fs(), so CONFIG_SET_FS
> can be removed globally, along with the thread_info field and
> any references to it.
> 
> This turns access_ok() into a cheaper check against TASK_SIZE_MAX.
> 
> With CONFIG_SET_FS gone, so drop all remaining references to
> set_fs()/get_fs(), mm_segment_t and uaccess_kernel().
> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>   arch/Kconfig                              |  3 -
>   arch/alpha/Kconfig                        |  1 -
>   arch/alpha/include/asm/processor.h        |  4 --
>   arch/alpha/include/asm/thread_info.h      |  2 -
>   arch/alpha/include/asm/uaccess.h          | 19 ------
>   arch/arc/Kconfig                          |  1 -
>   arch/arc/include/asm/segment.h            | 20 -------
>   arch/arc/include/asm/thread_info.h        |  3 -
>   arch/arc/include/asm/uaccess.h            |  1 -
>   arch/arm/lib/uaccess_with_memcpy.c        | 10 ----
>   arch/csky/Kconfig                         |  1 -
>   arch/csky/include/asm/processor.h         |  2 -
>   arch/csky/include/asm/segment.h           | 10 ----
>   arch/csky/include/asm/thread_info.h       |  2 -
>   arch/csky/include/asm/uaccess.h           |  3 -
>   arch/csky/kernel/asm-offsets.c            |  1 -
>   arch/h8300/Kconfig                        |  1 -
>   arch/h8300/include/asm/processor.h        |  1 -
>   arch/h8300/include/asm/segment.h          | 40 -------------
>   arch/h8300/include/asm/thread_info.h      |  3 -
>   arch/h8300/kernel/entry.S                 |  1 -
>   arch/h8300/kernel/head_ram.S              |  1 -
>   arch/h8300/mm/init.c                      |  6 --
>   arch/h8300/mm/memory.c                    |  1 -
>   arch/hexagon/Kconfig                      |  1 -
>   arch/hexagon/include/asm/thread_info.h    |  6 --
>   arch/hexagon/kernel/process.c             |  1 -
>   arch/microblaze/Kconfig                   |  1 -
>   arch/microblaze/include/asm/thread_info.h |  6 --
>   arch/microblaze/include/asm/uaccess.h     | 24 --------
>   arch/microblaze/kernel/asm-offsets.c      |  1 -
>   arch/microblaze/kernel/process.c          |  1 -
>   arch/nds32/Kconfig                        |  1 -
>   arch/nds32/include/asm/thread_info.h      |  4 --
>   arch/nds32/include/asm/uaccess.h          | 15 +----
>   arch/nds32/kernel/process.c               |  5 +-
>   arch/nds32/mm/alignment.c                 |  3 -
>   arch/nios2/Kconfig                        |  1 -
>   arch/nios2/include/asm/thread_info.h      |  9 ---
>   arch/nios2/include/asm/uaccess.h          | 12 ----

For NIOS2:

Acked-by: Dinh Nguyen <dinguyen@kernel.org>