Skip to contentSkip to conversation

FAILED: patch "[PATCH] arm64: compat: Avoid sending SIGILL for unallocated syscall" failed to apply to 4.14-stable tree

PATCH [email protected] 2 participants 2 messages Jan 7, 2019

Subsystems (2)
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 4.14-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 169113ece0f29ebe884a6cfcf57c1ace04d8a36a Mon Sep 17 00:00:00 2001
From: Will Deacon <[email protected]>
Date: Thu, 3 Jan 2019 17:45:07 +0000
Subject: [PATCH] arm64: compat: Avoid sending SIGILL for unallocated syscall
numbers

The ARM Linux kernel handles the EABI syscall numbers as follows:

0 - NR_SYSCALLS-1 : Invoke syscall via syscall table
NR_SYSCALLS - 0xeffff : -ENOSYS (to be allocated in future)
0xf0000 - 0xf07ff : Private syscall or -ENOSYS if not allocated

Quoted 1 line

0xf07ff : SIGILL

Our compat code gets this wrong and ends up sending SIGILL in response to all syscalls greater than NR_SYSCALLS which have a value greater than 0x7ff in the bottom 16 bits.

Fix this by defining the end of the ARM private syscall region and checking the syscall number against that directly. Update the comment while we're at it.

arch/arm64/include/asm/unistd.h +3 -2
index b13ca091f833..85d5c1026204 100644
@@ -40,8 +40,9 @@
  * The following SVCs are ARM private.
  */
 #define __ARM_NR_COMPAT_BASE		0x0f0000
-#define __ARM_NR_compat_cacheflush	(__ARM_NR_COMPAT_BASE+2)
-#define __ARM_NR_compat_set_tls		(__ARM_NR_COMPAT_BASE+5)
+#define __ARM_NR_compat_cacheflush	(__ARM_NR_COMPAT_BASE + 2)
+#define __ARM_NR_compat_set_tls		(__ARM_NR_COMPAT_BASE + 5)
+#define __ARM_NR_COMPAT_END		(__ARM_NR_COMPAT_BASE + 0x800)
 
 #define __NR_compat_syscalls		399
 #endif
arch/arm64/kernel/sys_compat.c +2 -2
index 32653d156747..a79db4e485a6 100644
@@ -102,12 +102,12 @@ long compat_arm_syscall(struct pt_regs *regs)
 
 	default:
 		/*
-		 * Calls 9f00xx..9f07ff are defined to return -ENOSYS
+		 * Calls 0xf0xxx..0xf07ff are defined to return -ENOSYS
 		 * if not implemented, rather than raising SIGILL. This
 		 * way the calling program can gracefully determine whether
 		 * a feature is supported.
 		 */
-		if ((no & 0xffff) <= 0x7ff)
+		if (no < __ARM_NR_COMPAT_END)
 			return -ENOSYS;
 		break;
 	}
#1 lore raw
Quoted 25 lines · On Mon, Jan 07, 2019 at 10:39:41AM +0100, [email protected] wrote:

The patch below does not apply to the 4.14-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 169113ece0f29ebe884a6cfcf57c1ace04d8a36a Mon Sep 17 00:00:00 2001
From: Will Deacon <[email protected]>
Date: Thu, 3 Jan 2019 17:45:07 +0000
Subject: [PATCH] arm64: compat: Avoid sending SIGILL for unallocated syscall
numbers

The ARM Linux kernel handles the EABI syscall numbers as follows:

0 - NR_SYSCALLS-1 : Invoke syscall via syscall table
NR_SYSCALLS - 0xeffff : -ENOSYS (to be allocated in future)
0xf0000 - 0xf07ff : Private syscall or -ENOSYS if not allocated

Quoted 1 line

0xf07ff : SIGILL

Quoted 7 lines

Our compat code gets this wrong and ends up sending SIGILL in response to all syscalls greater than NR_SYSCALLS which have a value greater than 0x7ff in the bottom 16 bits.

Fix this by defining the end of the ARM private syscall region and checking the syscall number against that directly. Update the comment while we're at it.

Aha, looks like 532826f3712b ("arm64: Mirror arm for unimplemented compat syscalls") wasn't CC'd for stable so the breakage it caused didn't propagate to 4.14 or earlier.

I'll send a rebased version of the other patch in this series for 4.19 stable separately.

Will