linux-kernel - [PATCH] locking/atomic/x86: Merge __arch{,_try}_cmpxchg64_emu_local() with __arch{,_try}_cmpxchg64

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20240417175830.161561-1-ubizjak@gmail.com>
Date: Wed, 17 Apr 2024 19:58:12 +0200
From: Uros Bizjak <ubizjak@...il.com>
To: x86@...nel.org,
	linux-kernel@...r.kernel.org
Cc: Uros Bizjak <ubizjak@...il.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...nel.org>,
	Borislav Petkov <bp@...en8.de>,
	Dave Hansen <dave.hansen@...ux.intel.com>,
	"H. Peter Anvin" <hpa@...or.com>,
	Peter Zijlstra <peterz@...radead.org>
Subject: [PATCH] locking/atomic/x86: Merge __arch{,_try}_cmpxchg64_emu_local() with __arch{,_try}_cmpxchg64_emu()

Macros __arch{,_try}_cmpxchg64_emu() are almost identical to their
local variants __arch{,_try}_cmpxchg64_emu_local(), differing only
by lock prefixes.

Merge these two macros by introducing additional macro parameters
to pass lock location and lock prefix from their respective static
inline functions.

No functional change intended.

Signed-off-by: Uros Bizjak <ubizjak@...il.com>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Ingo Molnar <mingo@...nel.org>
Cc: Borislav Petkov <bp@...en8.de>
Cc: Dave Hansen <dave.hansen@...ux.intel.com>
Cc: "H. Peter Anvin" <hpa@...or.com>
Cc: Peter Zijlstra <peterz@...radead.org>
---
 arch/x86/include/asm/cmpxchg_32.h | 56 ++++++-------------------------
 1 file changed, 10 insertions(+), 46 deletions(-)

diff --git a/arch/x86/include/asm/cmpxchg_32.h b/arch/x86/include/asm/cmpxchg_32.h
index 9dedc13d5a77..ed2797f132ce 100644
--- a/arch/x86/include/asm/cmpxchg_32.h
+++ b/arch/x86/include/asm/cmpxchg_32.h
@@ -86,14 +86,14 @@ static __always_inline bool __try_cmpxchg64_local(volatile u64 *ptr, u64 *oldp,
  * to simulate the cmpxchg8b on the 80386 and 80486 CPU.
  */
 
-#define __arch_cmpxchg64_emu(_ptr, _old, _new)				\
+#define __arch_cmpxchg64_emu(_ptr, _old, _new, _lock_loc, _lock)	\
 ({									\
 	union __u64_halves o = { .full = (_old), },			\
 			   n = { .full = (_new), };			\
 									\
-	asm volatile(ALTERNATIVE(LOCK_PREFIX_HERE			\
+	asm volatile(ALTERNATIVE(_lock_loc				\
 				 "call cmpxchg8b_emu",			\
-				 "lock; cmpxchg8b %[ptr]", X86_FEATURE_CX8) \
+				 _lock "cmpxchg8b %[ptr]", X86_FEATURE_CX8) \
 		     : [ptr] "+m" (*(_ptr)),				\
 		       "+a" (o.low), "+d" (o.high)			\
 		     : "b" (n.low), "c" (n.high), "S" (_ptr)		\
@@ -104,40 +104,25 @@ static __always_inline bool __try_cmpxchg64_local(volatile u64 *ptr, u64 *oldp,
 
 static __always_inline u64 arch_cmpxchg64(volatile u64 *ptr, u64 old, u64 new)
 {
-	return __arch_cmpxchg64_emu(ptr, old, new);
+	return __arch_cmpxchg64_emu(ptr, old, new, LOCK_PREFIX_HERE, "lock; ");
 }
 #define arch_cmpxchg64 arch_cmpxchg64
 
-#define __arch_cmpxchg64_emu_local(_ptr, _old, _new)			\
-({									\
-	union __u64_halves o = { .full = (_old), },			\
-			   n = { .full = (_new), };			\
-									\
-	asm volatile(ALTERNATIVE("call cmpxchg8b_emu",			\
-				 "cmpxchg8b %[ptr]", X86_FEATURE_CX8)	\
-		     : [ptr] "+m" (*(_ptr)),				\
-		       "+a" (o.low), "+d" (o.high)			\
-		     : "b" (n.low), "c" (n.high), "S" (_ptr)		\
-		     : "memory");					\
-									\
-	o.full;								\
-})
-
 static __always_inline u64 arch_cmpxchg64_local(volatile u64 *ptr, u64 old, u64 new)
 {
-	return __arch_cmpxchg64_emu_local(ptr, old, new);
+	return __arch_cmpxchg64_emu(ptr, old, new, ,);
 }
 #define arch_cmpxchg64_local arch_cmpxchg64_local
 
-#define __arch_try_cmpxchg64_emu(_ptr, _oldp, _new)			\
+#define __arch_try_cmpxchg64_emu(_ptr, _oldp, _new, _lock_loc, _lock)	\
 ({									\
 	union __u64_halves o = { .full = *(_oldp), },			\
 			   n = { .full = (_new), };			\
 	bool ret;							\
 									\
-	asm volatile(ALTERNATIVE(LOCK_PREFIX_HERE			\
+	asm volatile(ALTERNATIVE(_lock_loc				\
 				 "call cmpxchg8b_emu",			\
-				 "lock; cmpxchg8b %[ptr]", X86_FEATURE_CX8) \
+				 _lock "cmpxchg8b %[ptr]", X86_FEATURE_CX8) \
 		     CC_SET(e)						\
 		     : CC_OUT(e) (ret),					\
 		       [ptr] "+m" (*(_ptr)),				\
@@ -153,34 +138,13 @@ static __always_inline u64 arch_cmpxchg64_local(volatile u64 *ptr, u64 old, u64
 
 static __always_inline bool arch_try_cmpxchg64(volatile u64 *ptr, u64 *oldp, u64 new)
 {
-	return __arch_try_cmpxchg64_emu(ptr, oldp, new);
+	return __arch_try_cmpxchg64_emu(ptr, oldp, new, LOCK_PREFIX_HERE, "lock; ");
 }
 #define arch_try_cmpxchg64 arch_try_cmpxchg64
 
-#define __arch_try_cmpxchg64_emu_local(_ptr, _oldp, _new)		\
-({									\
-	union __u64_halves o = { .full = *(_oldp), },			\
-			   n = { .full = (_new), };			\
-	bool ret;							\
-									\
-	asm volatile(ALTERNATIVE("call cmpxchg8b_emu",			\
-				 "cmpxchg8b %[ptr]", X86_FEATURE_CX8)	\
-		     CC_SET(e)						\
-		     : CC_OUT(e) (ret),					\
-		       [ptr] "+m" (*(_ptr)),				\
-		       "+a" (o.low), "+d" (o.high)			\
-		     : "b" (n.low), "c" (n.high), "S" (_ptr)		\
-		     : "memory");					\
-									\
-	if (unlikely(!ret))						\
-		*(_oldp) = o.full;					\
-									\
-	likely(ret);							\
-})
-
 static __always_inline bool arch_try_cmpxchg64_local(volatile u64 *ptr, u64 *oldp, u64 new)
 {
-	return __arch_try_cmpxchg64_emu_local(ptr, oldp, new);
+	return __arch_try_cmpxchg64_emu(ptr, oldp, new, ,);
 }
 #define arch_try_cmpxchg64_local arch_try_cmpxchg64_local
 
-- 
2.42.0