Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1620279

[PATCH 4.4 20/32] metag/usercopy: Fix src fixup in from user rapf loops

Show key headers only | View raw

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: James Hogan <james.hogan@imgtec.com>

commit 2c0b1df88b987a12d95ea1d6beaf01894f3cc725 upstream.

The fixup code to rewind the source pointer in
__asm_copy_from_user_{32,64}bit_rapf_loop() always rewound the source by
a single unit (4 or 8 bytes), however this is insufficient if the fault
didn't occur on the first load in the loop, as the source pointer will
have been incremented but nothing will have been stored until all 4
register [pairs] are loaded.

Read the LSM_STEP field of TXSTATUS (which is already loaded into a
register), a bit like the copy_to_user versions, to determine how many
iterations of MGET[DL] have taken place, all of which need rewinding.

Fixes: 373cd784d0fc ("metag: Memory handling")
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: linux-metag@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/metag/lib/usercopy.c |   36 ++++++++++++++++++++++++++++--------
 1 file changed, 28 insertions(+), 8 deletions(-)

--- a/arch/metag/lib/usercopy.c
+++ b/arch/metag/lib/usercopy.c
@@ -687,29 +687,49 @@ EXPORT_SYMBOL(__copy_user);
  *
  *	Rationale:
  *		A fault occurs while reading from user buffer, which is the
- *		source. Since the fault is at a single address, we only
- *		need to rewind by 8 bytes.
+ *		source.
  *		Since we don't write to kernel buffer until we read first,
  *		the kernel buffer is at the right state and needn't be
- *		corrected.
+ *		corrected, but the source must be rewound to the beginning of
+ *		the block, which is LSM_STEP*8 bytes.
+ *		LSM_STEP is bits 10:8 in TXSTATUS which is already read
+ *		and stored in D0Ar2
+ *
+ *		NOTE: If a fault occurs at the last operation in M{G,S}ETL
+ *			LSM_STEP will be 0. ie: we do 4 writes in our case, if
+ *			a fault happens at the 4th write, LSM_STEP will be 0
+ *			instead of 4. The code copes with that.
  */
 #define __asm_copy_from_user_64bit_rapf_loop(to, from, ret, n, id)	\
 	__asm_copy_user_64bit_rapf_loop(to, from, ret, n, id,		\
-		"SUB	%1, %1, #8\n")
+		"LSR	D0Ar2, D0Ar2, #5\n"				\
+		"ANDS	D0Ar2, D0Ar2, #0x38\n"				\
+		"ADDZ	D0Ar2, D0Ar2, #32\n"				\
+		"SUB	%1, %1, D0Ar2\n")
 
 /*	rewind 'from' pointer when a fault occurs
  *
  *	Rationale:
  *		A fault occurs while reading from user buffer, which is the
- *		source. Since the fault is at a single address, we only
- *		need to rewind by 4 bytes.
+ *		source.
  *		Since we don't write to kernel buffer until we read first,
  *		the kernel buffer is at the right state and needn't be
- *		corrected.
+ *		corrected, but the source must be rewound to the beginning of
+ *		the block, which is LSM_STEP*4 bytes.
+ *		LSM_STEP is bits 10:8 in TXSTATUS which is already read
+ *		and stored in D0Ar2
+ *
+ *		NOTE: If a fault occurs at the last operation in M{G,S}ETL
+ *			LSM_STEP will be 0. ie: we do 4 writes in our case, if
+ *			a fault happens at the 4th write, LSM_STEP will be 0
+ *			instead of 4. The code copes with that.
  */
 #define __asm_copy_from_user_32bit_rapf_loop(to, from, ret, n, id)	\
 	__asm_copy_user_32bit_rapf_loop(to, from, ret, n, id,		\
-		"SUB	%1, %1, #4\n")
+		"LSR	D0Ar2, D0Ar2, #6\n"				\
+		"ANDS	D0Ar2, D0Ar2, #0x1c\n"				\
+		"ADDZ	D0Ar2, D0Ar2, #16\n"				\
+		"SUB	%1, %1, D0Ar2\n")
 
 
 /*

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 4.4 00/32] 4.4.61-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:40 +0200
  [PATCH 4.4 05/32] drm/vmwgfx: Remove getparam error message Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:40 +0200
  [PATCH 4.4 01/32] drm/vmwgfx: Type-check lookups of fence objects Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:40 +0200
  [PATCH 4.4 06/32] drm/vmwgfx: fix integer overflow in vmw_surface_define_ioctl() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:40 +0200
  [PATCH 4.4 28/32] MIPS: ralink: Fix typos in rt3883 pinctrl Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 09/32] arm/arm64: KVM: Take mmap_sem in stage2_unmap_vm Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 16/32] metag/usercopy: Fix alignment error checking Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 30/32] MIPS: Lantiq: fix missing xbar kernel panic Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 32/32] mm/mempolicy.c: fix error handling in set_mempolicy and mbind. Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 23/32] powerpc: Dont try to fix up misaligned load-with-reservation instructions Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 20/32] metag/usercopy: Fix src fixup in from user rapf loops Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 08/32] staging: android: ashmem: lseek failed due to no FMODE_LSEEK. Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 29/32] MIPS: End spinlocks with .insn Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 27/32] MIPS: Force o32 fp64 support on 32bit MIPS64r6 kernels Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
    Re: [PATCH 4.4 27/32] MIPS: Force o32 fp64 support on 32bit MIPS64r6  kernels "Maciej W. Rozycki" <macro@linux-mips.org> - 2017-04-15 01:50 +0200
      Re: [PATCH 4.4 27/32] MIPS: Force o32 fp64 support on 32bit MIPS64r6  kernels Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-19 15:20 +0200
  [PATCH 4.4 26/32] s390/uaccess: get_user() should zero on failure (again) Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 31/32] MIPS: Flush wrong invalid FTLB entry for huge page Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 14/32] ring-buffer: Fix return value check in test_ringbuffer() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 17/32] metag/usercopy: Add early abort to copy_to_user Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 22/32] powerpc/mm: Add missing global TLB invalidate if cxl is active Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 04/32] drm/ttm, drm/vmwgfx: Relax permission checking when opening surfaces Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 25/32] s390/decompressor: fix initrd corruption caused by bss clear Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 15/32] metag/usercopy: Drop unused macros Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 24/32] nios2: reserve boot memory for device tree Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 18:50 +0200
  [PATCH 4.4 21/32] metag/usercopy: Add missing fixups Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 11/32] iio: bmg160: reset chip when probing Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 19/32] metag/usercopy: Set flags before ADDZ Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 13/32] ptrace: fix PTRACE_LISTEN race corrupting task->state Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 18/32] metag/usercopy: Zero rest of buffer from copy_from_user Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 10/32] arm/arm64: KVM: Take mmap_sem in kvm_arch_prepare_memory_region Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 12/32] Reset TreeId to zero on SMB2 TREE_CONNECT Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 02/32] drm/vmwgfx: NULL pointer dereference in vmw_surface_define_ioctl() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  [PATCH 4.4 07/32] sysfs: be careful of error returns from ops->show() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2017-04-10 20:00 +0200
  Re: [PATCH 4.4 00/32] 4.4.61-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2017-04-10 22:40 +0200

csiph-web