linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: alex.shi@linaro.org (Alex Shi)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 09/39] arm64: uaccess: Don't bother eliding access_ok checks in __{get, put}_user
Date: Fri,  9 Mar 2018 17:06:52 +0800	[thread overview]
Message-ID: <20180309090722.26279-10-alex.shi@linaro.org> (raw)
In-Reply-To: <20180309090722.26279-1-alex.shi@linaro.org>

From: Will Deacon <will.deacon@arm.com>

commit 84624087dd7e upstream.

access_ok isn't an expensive operation once the addr_limit for the current
thread has been loaded into the cache. Given that the initial access_ok
check preceding a sequence of __{get,put}_user operations will take
the brunt of the miss, we can make the __* variants identical to the
full-fat versions, which brings with it the benefits of address masking.

The likely cost in these sequences will be from toggling PAN/UAO, which
we can address later by implementing the *_unsafe versions.

Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Alex Shi <alex.shi@linaro.org>
---
 arch/arm64/include/asm/uaccess.h | 54 +++++++++++++++++++-------------
 1 file changed, 32 insertions(+), 22 deletions(-)

diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index fc11c50af558..73f02e39ab1f 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -200,30 +200,35 @@ do {									\
 			CONFIG_ARM64_PAN));				\
 } while (0)
 
-#define __get_user(x, ptr)						\
+#define __get_user_check(x, ptr, err)					\
 ({									\
-	int __gu_err = 0;						\
-	__get_user_err((x), (ptr), __gu_err);				\
-	__gu_err;							\
+	__typeof__(*(ptr)) __user *__p = (ptr);				\
+	might_fault();							\
+	if (access_ok(VERIFY_READ, __p, sizeof(*__p))) {		\
+		__p = uaccess_mask_ptr(__p);				\
+		__get_user_err((x), __p, (err));			\
+	} else {							\
+		(x) = 0; (err) = -EFAULT;				\
+	}								\
 })
 
 #define __get_user_error(x, ptr, err)					\
 ({									\
-	__get_user_err((x), (ptr), (err));				\
+	__get_user_check((x), (ptr), (err));				\
 	(void)0;							\
 })
 
 #define __get_user_unaligned __get_user
 
-#define get_user(x, ptr)						\
+#define __get_user(x, ptr)						\
 ({									\
-	__typeof__(*(ptr)) __user *__p = (ptr);				\
-	might_fault();							\
-	access_ok(VERIFY_READ, __p, sizeof(*__p)) ?			\
-		__p = uaccess_mask_ptr(__p), __get_user((x), __p) :	\
-		((x) = 0, -EFAULT);					\
+	int __gu_err = 0;						\
+	__get_user_check((x), (ptr), __gu_err);				\
+	__gu_err;							\
 })
 
+#define get_user	__get_user
+
 #define __put_user_asm(instr, reg, x, addr, err)			\
 	asm volatile(							\
 	"1:	" instr "	" reg "1, [%2]\n"			\
@@ -266,30 +271,35 @@ do {									\
 			CONFIG_ARM64_PAN));				\
 } while (0)
 
-#define __put_user(x, ptr)						\
+#define __put_user_check(x, ptr, err)					\
 ({									\
-	int __pu_err = 0;						\
-	__put_user_err((x), (ptr), __pu_err);				\
-	__pu_err;							\
+	__typeof__(*(ptr)) __user *__p = (ptr);				\
+	might_fault();							\
+	if (access_ok(VERIFY_WRITE, __p, sizeof(*__p))) {		\
+		__p = uaccess_mask_ptr(__p);				\
+		__put_user_err((x), __p, (err));			\
+	} else	{							\
+		(err) = -EFAULT;					\
+	}								\
 })
 
 #define __put_user_error(x, ptr, err)					\
 ({									\
-	__put_user_err((x), (ptr), (err));				\
+	__put_user_check((x), (ptr), (err));				\
 	(void)0;							\
 })
 
 #define __put_user_unaligned __put_user
 
-#define put_user(x, ptr)						\
+#define __put_user(x, ptr)						\
 ({									\
-	__typeof__(*(ptr)) __user *__p = (ptr);				\
-	might_fault();							\
-	access_ok(VERIFY_WRITE, __p, sizeof(*__p)) ?			\
-		__p = uaccess_mask_ptr(__p), __put_user((x), __p) :	\
-		-EFAULT;						\
+	int __pu_err = 0;						\
+	__put_user_check((x), (ptr), __pu_err);				\
+	__pu_err;							\
 })
 
+#define put_user	__put_user
+
 extern unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n);
 extern unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n);
 extern unsigned long __must_check __copy_in_user(void __user *to, const void __user *from, unsigned long n);
-- 
2.16.2.440.gc6284da

  parent reply	other threads:[~2018-03-09  9:06 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-09  9:06 spectre backport for LTS 4.4 review Alex Shi
2018-03-09  9:06 ` [PATCH 01/39] mm: Introduce lm_alias Alex Shi
2018-03-09  9:06 ` [PATCH 02/39] arm64: barrier: Add CSDB macros to control data-value prediction Alex Shi
2018-03-09  9:06 ` [PATCH 03/39] arm64: Implement array_index_mask_nospec() Alex Shi
2018-03-09  9:06 ` [PATCH 04/39] arm64: move TASK_* definitions to <asm/processor.h> Alex Shi
2018-03-09  9:06 ` [PATCH 05/39] arm64: Make USER_DS an inclusive limit Alex Shi
2018-03-09  9:06 ` [PATCH 06/39] arm64: entry: Ensure branch through syscall table is bounded under speculation Alex Shi
2018-03-09  9:06 ` [PATCH 07/39] arm64: Use pointer masking to limit uaccess speculation Alex Shi
2018-03-09  9:06 ` [PATCH 08/39] arm64: uaccess: Prevent speculative use of the current addr_limit Alex Shi
2018-03-09  9:06 ` Alex Shi [this message]
2018-03-09  9:06 ` [PATCH 10/39] arm64: futex: Mask __user pointers prior to dereference Alex Shi
2018-03-09  9:06 ` [PATCH 11/39] drivers/firmware: Expose psci_get_version through psci_ops structure Alex Shi
2018-03-09  9:06 ` [PATCH 12/39] arm64: Move post_ttbr_update_workaround to C code Alex Shi
2018-03-09  9:06 ` [PATCH 13/39] arm64: Add skeleton to harden the branch predictor against aliasing attacks Alex Shi
2018-03-09  9:06 ` [PATCH 14/39] arm64: Move BP hardening to check_and_switch_context Alex Shi
2018-03-09  9:06 ` [PATCH 15/39] arm64: KVM: Use per-CPU vector when BP hardening is enabled Alex Shi
2018-03-09  9:06 ` [PATCH 16/39] arm64: entry: Apply BP hardening for high-priority synchronous exceptions Alex Shi
2018-03-09  9:07 ` [PATCH 17/39] arm64: entry: Apply BP hardening for suspicious interrupts from EL0 Alex Shi
2018-03-09  9:07 ` [PATCH 18/39] arm64: cpu_errata: Allow an erratum to be match for all revisions of a core Alex Shi
2018-03-09  9:07 ` [PATCH 19/39] arm64: prefetch: add alternative pattern for CPUs without a prefetcher Alex Shi
2018-03-09  9:07 ` [PATCH 20/39] arm64: cputype: Add missing MIDR values for Cortex-A72 and Cortex-A75 Alex Shi
2018-03-09  9:07 ` [PATCH 21/39] arm64: Implement branch predictor hardening for affected Cortex-A CPUs Alex Shi
2018-03-09  9:07 ` [PATCH 22/39] arm64: KVM: Increment PC after handling an SMC trap Alex Shi
2018-03-09  9:07 ` [PATCH 23/39] arm/arm64: KVM: Consolidate the PSCI include files Alex Shi
2018-03-09  9:07 ` [PATCH 24/39] arm/arm64: KVM: Add PSCI_VERSION helper Alex Shi
2018-03-09  9:07 ` [PATCH 25/39] arm/arm64: KVM: Add smccc accessors to PSCI code Alex Shi
2018-03-09  9:07 ` [PATCH 26/39] arm/arm64: KVM: Implement PSCI 1.0 support Alex Shi
2018-03-09  9:07 ` [PATCH 27/39] ARM: 8478/2: arm/arm64: add arm-smccc Alex Shi
2018-03-09  9:07 ` [PATCH 28/39] ARM: 8479/2: add implementation for arm-smccc Alex Shi
2018-03-09  9:07 ` [PATCH 29/39] ARM: 8480/2: arm64: " Alex Shi
2018-03-09  9:07 ` [PATCH 30/39] ARM: 8481/2: drivers: psci: replace psci firmware calls Alex Shi
2018-03-09  9:07 ` [PATCH 31/39] arm/arm64: KVM: Advertise SMCCC v1.1 Alex Shi
2018-03-09  9:07 ` [PATCH 32/39] arm/arm64: KVM: Turn kvm_psci_version into a static inline Alex Shi
2018-03-09  9:07 ` [PATCH 33/39] arm64: KVM: Report SMCCC_ARCH_WORKAROUND_1 BP hardening support Alex Shi
2018-03-09  9:07 ` [PATCH 34/39] arm64: KVM: Add SMCCC_ARCH_WORKAROUND_1 fast handling Alex Shi
2018-03-09  9:07 ` [PATCH 35/39] firmware/psci: Expose PSCI conduit Alex Shi
2018-03-09  9:07 ` [PATCH 36/39] firmware/psci: Expose SMCCC version through psci_ops Alex Shi
2018-03-09  9:07 ` [PATCH 37/39] arm/arm64: smccc: Make function identifiers an unsigned quantity Alex Shi
2018-03-09  9:07 ` [PATCH 38/39] arm/arm64: smccc: Implement SMCCC v1.1 inline primitive Alex Shi
2018-03-09  9:07 ` [PATCH 39/39] arm64: Add ARM_SMCCC_ARCH_WORKAROUND_1 BP hardening support Alex Shi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180309090722.26279-10-alex.shi@linaro.org \
    --to=alex.shi@linaro.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).