[PATCH] x86/mpx: Fix off-by-one comparison with nr_registers

[PATCH] x86/mpx: Fix off-by-one comparison with nr_registers

[Colin King]

From: Colin Ian King <colin.king@canonical.com>

In the unlikely event that regno == nr_registers then we get an array
overrun on regoff because the invalid register check is currently
off-by-one. Fix this with a check that regno is >= nr_registers instead.

Detected with static analysis using CoverityScan.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
---
 arch/x86/mm/mpx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/mm/mpx.c b/arch/x86/mm/mpx.c
index 5584397..8047687 100644
--- a/arch/x86/mm/mpx.c
+++ b/arch/x86/mm/mpx.c
@@ -123,7 +123,7 @@ static int get_reg_offset(struct insn *insn, struct pt_regs *regs,
 		break;
 	}
 
-	if (regno > nr_registers) {
+	if (regno >= nr_registers) {
 		WARN_ONCE(1, "decoded an instruction with an invalid register");
 		return -EINVAL;
 	}
-- 
2.7.0

Re: [PATCH] x86/mpx: Fix off-by-one comparison with nr_registers

[Dave Hansen]

On 02/26/2016 10:55 AM, Colin King wrote:
> In the unlikely event that regno == nr_registers then we get an array
> overrun on regoff because the invalid register check is currently
> off-by-one. Fix this with a check that regno is >= nr_registers instead.
> 
> Detected with static analysis using CoverityScan.

Oh, the irony of an overrun in the MPX code!  Thanks for catching this.

Acked-by: Dave Hansen <dave.hansen@intel.com>

[tip:x86/urgent] x86/mpx: Fix off-by-one comparison with nr_registers

[tip-bot for Colin Ian King]

Commit-ID:  9bf148cb0812595bfdf5100bd2c07e9bec9c6ef5
Gitweb:     http://git.kernel.org/tip/9bf148cb0812595bfdf5100bd2c07e9bec9c6ef5
Author:     Colin Ian King <colin.king@canonical.com>
AuthorDate: Fri, 26 Feb 2016 18:55:31 +0000
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Fri, 26 Feb 2016 22:12:47 +0100

x86/mpx: Fix off-by-one comparison with nr_registers

In the unlikely event that regno == nr_registers then we get an array
overrun on regoff because the invalid register check is currently
off-by-one. Fix this with a check that regno is >= nr_registers instead.

Detected with static analysis using CoverityScan.

Fixes: fcc7ffd67991 "x86, mpx: Decode MPX instruction to get bound violation information"
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/1456512931-3388-1-git-send-email-colin.king@canonical.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/mm/mpx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/mm/mpx.c b/arch/x86/mm/mpx.c
index b2fd67d..ef05755 100644
--- a/arch/x86/mm/mpx.c
+++ b/arch/x86/mm/mpx.c
@@ -123,7 +123,7 @@ static int get_reg_offset(struct insn *insn, struct pt_regs *regs,
 		break;
 	}
 
-	if (regno > nr_registers) {
+	if (regno >= nr_registers) {
 		WARN_ONCE(1, "decoded an instruction with an invalid register");
 		return -EINVAL;
 	}