LKML Archive on lore.kernel.org
 help / color / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: will.deacon@arm.com, mingo@kernel.org
Cc: linux-kernel@vger.kernel.org, longman@redhat.com,
	andrea.parri@amarulasolutions.com, tglx@linutronix.de,
	bigeasy@linutronix.de, Peter Zijlstra <peterz@infradead.org>,
	hpa@linux.intel.com, JBeulich@suse.com, bp@alien8.de
Subject: [PATCH v2 3/4] x86/asm: Simplify GEN_*_RMWcc() macros
Date: Wed, 03 Oct 2018 15:03:00 +0200
Message-ID: <20181003130957.108960094@infradead.org> (raw)
In-Reply-To: <20181003130257.156322446@infradead.org>

Currently the GEN_*_RMWcc() macros include a return statement, which
pretty much mandates we directly wrap them in a (inline) function.

Macros with return statements are tricky and, as per the above, limit
use, so remove the return statement and make them
statement-expressions. This allows them to be used more widely.

Also, shuffle the arguments a bit. Place the @cc argument as 3rd, this
makes it consistent between UNARY and BINARY, but more importantly, it
makes the @arg0 argument last.

Since the @arg0 argument is now last, we can do CPP trickery and make
it an optional argument, simplifying the users; 17 out of 18
occurences do not need this argument.

Finally, change to asm symbolic names, instead of the numeric ordering
of operands, which allows us to get rid of __BINARY_RMWcc_ARG and get
cleaner code overall.

Cc: hpa@linux.intel.com
Cc: JBeulich@suse.com
Cc: bp@alien8.de
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 arch/x86/include/asm/atomic.h      |    8 ++--
 arch/x86/include/asm/atomic64_64.h |    8 ++--
 arch/x86/include/asm/bitops.h      |    9 +----
 arch/x86/include/asm/local.h       |    8 ++--
 arch/x86/include/asm/preempt.h     |    2 -
 arch/x86/include/asm/refcount.h    |   18 +++++-----
 arch/x86/include/asm/rmwcc.h       |   66 +++++++++++++++++++++----------------
 7 files changed, 64 insertions(+), 55 deletions(-)

--- a/arch/x86/include/asm/atomic.h
+++ b/arch/x86/include/asm/atomic.h
@@ -82,7 +82,7 @@ static __always_inline void arch_atomic_
  */
 static __always_inline bool arch_atomic_sub_and_test(int i, atomic_t *v)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX "subl", v->counter, "er", i, "%0", e);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX "subl", v->counter, e, "er", i);
 }
 #define arch_atomic_sub_and_test arch_atomic_sub_and_test
 
@@ -122,7 +122,7 @@ static __always_inline void arch_atomic_
  */
 static __always_inline bool arch_atomic_dec_and_test(atomic_t *v)
 {
-	GEN_UNARY_RMWcc(LOCK_PREFIX "decl", v->counter, "%0", e);
+	return GEN_UNARY_RMWcc(LOCK_PREFIX "decl", v->counter, e);
 }
 #define arch_atomic_dec_and_test arch_atomic_dec_and_test
 
@@ -136,7 +136,7 @@ static __always_inline bool arch_atomic_
  */
 static __always_inline bool arch_atomic_inc_and_test(atomic_t *v)
 {
-	GEN_UNARY_RMWcc(LOCK_PREFIX "incl", v->counter, "%0", e);
+	return GEN_UNARY_RMWcc(LOCK_PREFIX "incl", v->counter, e);
 }
 #define arch_atomic_inc_and_test arch_atomic_inc_and_test
 
@@ -151,7 +151,7 @@ static __always_inline bool arch_atomic_
  */
 static __always_inline bool arch_atomic_add_negative(int i, atomic_t *v)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX "addl", v->counter, "er", i, "%0", s);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX "addl", v->counter, s, "er", i);
 }
 #define arch_atomic_add_negative arch_atomic_add_negative
 
--- a/arch/x86/include/asm/atomic64_64.h
+++ b/arch/x86/include/asm/atomic64_64.h
@@ -73,7 +73,7 @@ static inline void arch_atomic64_sub(lon
  */
 static inline bool arch_atomic64_sub_and_test(long i, atomic64_t *v)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX "subq", v->counter, "er", i, "%0", e);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX "subq", v->counter, e, "er", i);
 }
 #define arch_atomic64_sub_and_test arch_atomic64_sub_and_test
 
@@ -115,7 +115,7 @@ static __always_inline void arch_atomic6
  */
 static inline bool arch_atomic64_dec_and_test(atomic64_t *v)
 {
-	GEN_UNARY_RMWcc(LOCK_PREFIX "decq", v->counter, "%0", e);
+	return GEN_UNARY_RMWcc(LOCK_PREFIX "decq", v->counter, e);
 }
 #define arch_atomic64_dec_and_test arch_atomic64_dec_and_test
 
@@ -129,7 +129,7 @@ static inline bool arch_atomic64_dec_and
  */
 static inline bool arch_atomic64_inc_and_test(atomic64_t *v)
 {
-	GEN_UNARY_RMWcc(LOCK_PREFIX "incq", v->counter, "%0", e);
+	return GEN_UNARY_RMWcc(LOCK_PREFIX "incq", v->counter, e);
 }
 #define arch_atomic64_inc_and_test arch_atomic64_inc_and_test
 
@@ -144,7 +144,7 @@ static inline bool arch_atomic64_inc_and
  */
 static inline bool arch_atomic64_add_negative(long i, atomic64_t *v)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX "addq", v->counter, "er", i, "%0", s);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX "addq", v->counter, s, "er", i);
 }
 #define arch_atomic64_add_negative arch_atomic64_add_negative
 
--- a/arch/x86/include/asm/bitops.h
+++ b/arch/x86/include/asm/bitops.h
@@ -217,8 +217,7 @@ static __always_inline void change_bit(l
  */
 static __always_inline bool test_and_set_bit(long nr, volatile unsigned long *addr)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX __ASM_SIZE(bts),
-	                 *addr, "Ir", nr, "%0", c);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX __ASM_SIZE(bts), *addr, c, "Ir", nr);
 }
 
 /**
@@ -264,8 +263,7 @@ static __always_inline bool __test_and_s
  */
 static __always_inline bool test_and_clear_bit(long nr, volatile unsigned long *addr)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX __ASM_SIZE(btr),
-	                 *addr, "Ir", nr, "%0", c);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX __ASM_SIZE(btr), *addr, c, "Ir", nr);
 }
 
 /**
@@ -318,8 +316,7 @@ static __always_inline bool __test_and_c
  */
 static __always_inline bool test_and_change_bit(long nr, volatile unsigned long *addr)
 {
-	GEN_BINARY_RMWcc(LOCK_PREFIX __ASM_SIZE(btc),
-	                 *addr, "Ir", nr, "%0", c);
+	return GEN_BINARY_RMWcc(LOCK_PREFIX __ASM_SIZE(btc), *addr, c, "Ir", nr);
 }
 
 static __always_inline bool constant_test_bit(long nr, const volatile unsigned long *addr)
--- a/arch/x86/include/asm/local.h
+++ b/arch/x86/include/asm/local.h
@@ -53,7 +53,7 @@ static inline void local_sub(long i, loc
  */
 static inline bool local_sub_and_test(long i, local_t *l)
 {
-	GEN_BINARY_RMWcc(_ASM_SUB, l->a.counter, "er", i, "%0", e);
+	return GEN_BINARY_RMWcc(_ASM_SUB, l->a.counter, e, "er", i);
 }
 
 /**
@@ -66,7 +66,7 @@ static inline bool local_sub_and_test(lo
  */
 static inline bool local_dec_and_test(local_t *l)
 {
-	GEN_UNARY_RMWcc(_ASM_DEC, l->a.counter, "%0", e);
+	return GEN_UNARY_RMWcc(_ASM_DEC, l->a.counter, e);
 }
 
 /**
@@ -79,7 +79,7 @@ static inline bool local_dec_and_test(lo
  */
 static inline bool local_inc_and_test(local_t *l)
 {
-	GEN_UNARY_RMWcc(_ASM_INC, l->a.counter, "%0", e);
+	return GEN_UNARY_RMWcc(_ASM_INC, l->a.counter, e);
 }
 
 /**
@@ -93,7 +93,7 @@ static inline bool local_inc_and_test(lo
  */
 static inline bool local_add_negative(long i, local_t *l)
 {
-	GEN_BINARY_RMWcc(_ASM_ADD, l->a.counter, "er", i, "%0", s);
+	return GEN_BINARY_RMWcc(_ASM_ADD, l->a.counter, s, "er", i);
 }
 
 /**
--- a/arch/x86/include/asm/preempt.h
+++ b/arch/x86/include/asm/preempt.h
@@ -88,7 +88,7 @@ static __always_inline void __preempt_co
  */
 static __always_inline bool __preempt_count_dec_and_test(void)
 {
-	GEN_UNARY_RMWcc("decl", __preempt_count, __percpu_arg(0), e);
+	return GEN_UNARY_RMWcc("decl", __preempt_count, e, __percpu_arg([var]));
 }
 
 /*
--- a/arch/x86/include/asm/refcount.h
+++ b/arch/x86/include/asm/refcount.h
@@ -17,7 +17,7 @@
  */
 #define _REFCOUNT_EXCEPTION				\
 	".pushsection .text..refcount\n"		\
-	"111:\tlea %[counter], %%" _ASM_CX "\n"		\
+	"111:\tlea %[var], %%" _ASM_CX "\n"		\
 	"112:\t" ASM_UD2 "\n"				\
 	ASM_UNREACHABLE					\
 	".popsection\n"					\
@@ -43,7 +43,7 @@ static __always_inline void refcount_add
 {
 	asm volatile(LOCK_PREFIX "addl %1,%0\n\t"
 		REFCOUNT_CHECK_LT_ZERO
-		: [counter] "+m" (r->refs.counter)
+		: [var] "+m" (r->refs.counter)
 		: "ir" (i)
 		: "cc", "cx");
 }
@@ -52,7 +52,7 @@ static __always_inline void refcount_inc
 {
 	asm volatile(LOCK_PREFIX "incl %0\n\t"
 		REFCOUNT_CHECK_LT_ZERO
-		: [counter] "+m" (r->refs.counter)
+		: [var] "+m" (r->refs.counter)
 		: : "cc", "cx");
 }
 
@@ -60,21 +60,21 @@ static __always_inline void refcount_dec
 {
 	asm volatile(LOCK_PREFIX "decl %0\n\t"
 		REFCOUNT_CHECK_LE_ZERO
-		: [counter] "+m" (r->refs.counter)
+		: [var] "+m" (r->refs.counter)
 		: : "cc", "cx");
 }
 
 static __always_inline __must_check
 bool refcount_sub_and_test(unsigned int i, refcount_t *r)
 {
-	GEN_BINARY_SUFFIXED_RMWcc(LOCK_PREFIX "subl", REFCOUNT_CHECK_LT_ZERO,
-				  r->refs.counter, "er", i, "%0", e, "cx");
+	return GEN_BINARY_SUFFIXED_RMWcc(LOCK_PREFIX "subl",
+		REFCOUNT_CHECK_LT_ZERO, r->refs.counter, e, "er", i, "cx");
 }
 
 static __always_inline __must_check bool refcount_dec_and_test(refcount_t *r)
 {
-	GEN_UNARY_SUFFIXED_RMWcc(LOCK_PREFIX "decl", REFCOUNT_CHECK_LT_ZERO,
-				 r->refs.counter, "%0", e, "cx");
+	return GEN_UNARY_SUFFIXED_RMWcc(LOCK_PREFIX "decl",
+		REFCOUNT_CHECK_LT_ZERO, r->refs.counter, e, "cx");
 }
 
 static __always_inline __must_check
@@ -92,7 +92,7 @@ bool refcount_add_not_zero(unsigned int
 		/* Did we try to increment from/to an undesirable state? */
 		if (unlikely(c < 0 || c == INT_MAX || result < c)) {
 			asm volatile(REFCOUNT_ERROR
-				     : : [counter] "m" (r->refs.counter)
+				     : : [var] "m" (r->refs.counter)
 				     : "cc", "cx");
 			break;
 		}
--- a/arch/x86/include/asm/rmwcc.h
+++ b/arch/x86/include/asm/rmwcc.h
@@ -2,56 +2,68 @@
 #ifndef _ASM_X86_RMWcc
 #define _ASM_X86_RMWcc
 
+/* This counts to 12. Any more, it will return 13th argument. */
+#define __RMWcc_ARGS(_0, _1, _2, _3, _4, _5, _6, _7, _8, _9, _10, _11, _12, _n, X...) _n
+#define RMWcc_ARGS(X...) __RMWcc_ARGS(, ##X, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0)
+
+#define __RMWcc_CONCAT(a, b) a ## b
+#define RMWcc_CONCAT(a, b) __RMWcc_CONCAT(a, b)
+
 #define __CLOBBERS_MEM(clb...)	"memory", ## clb
 
 #if !defined(__GCC_ASM_FLAG_OUTPUTS__) && defined(CC_HAVE_ASM_GOTO)
 
 /* Use asm goto */
 
-#define __GEN_RMWcc(fullop, var, cc, clobbers, ...)			\
-do {									\
+#define __GEN_RMWcc(fullop, _var, cc, clobbers, ...)			\
+({									\
+	__label__ cc_label;						\
 	asm_volatile_goto (fullop "; j" #cc " %l[cc_label]"		\
-			: : [counter] "m" (var), ## __VA_ARGS__		\
+			: : [var] "m" (_var), ## __VA_ARGS__		\
 			: clobbers : cc_label);				\
-	return 0;							\
+	0;								\
 cc_label:								\
-	return 1;							\
-} while (0)
-
-#define __BINARY_RMWcc_ARG	" %1, "
-
+	1;								\
+})
 
 #else /* defined(__GCC_ASM_FLAG_OUTPUTS__) || !defined(CC_HAVE_ASM_GOTO) */
 
 /* Use flags output or a set instruction */
 
-#define __GEN_RMWcc(fullop, var, cc, clobbers, ...)			\
-do {									\
+#define __GEN_RMWcc(fullop, _var, cc, clobbers, ...)			\
+({									\
 	bool c;								\
 	asm volatile (fullop CC_SET(cc)					\
-			: [counter] "+m" (var), CC_OUT(cc) (c)		\
+			: [var] "+m" (_var), CC_OUT(cc) (c)		\
 			: __VA_ARGS__ : clobbers);			\
-	return c;							\
-} while (0)
-
-#define __BINARY_RMWcc_ARG	" %2, "
+	c;								\
+})
 
 #endif /* defined(__GCC_ASM_FLAG_OUTPUTS__) || !defined(CC_HAVE_ASM_GOTO) */
 
-#define GEN_UNARY_RMWcc(op, var, arg0, cc)				\
+#define GEN_UNARY_RMWcc_4(op, var, cc, arg0)				\
 	__GEN_RMWcc(op " " arg0, var, cc, __CLOBBERS_MEM())
 
-#define GEN_UNARY_SUFFIXED_RMWcc(op, suffix, var, arg0, cc, clobbers...)\
-	__GEN_RMWcc(op " " arg0 "\n\t" suffix, var, cc,			\
+#define GEN_UNARY_RMWcc_3(op, var, cc)					\
+	GEN_UNARY_RMWcc_4(op, var, cc, "%[var]")
+
+#define GEN_UNARY_RMWcc(X...) RMWcc_CONCAT(GEN_UNARY_RMWcc_, RMWcc_ARGS(X))(X)
+
+#define GEN_BINARY_RMWcc_6(op, var, cc, vcon, _val, arg0)		\
+	__GEN_RMWcc(op " %[val], " arg0, var, cc,			\
+		    __CLOBBERS_MEM(), [val] vcon (_val))
+
+#define GEN_BINARY_RMWcc_5(op, var, cc, vcon, val)			\
+	GEN_BINARY_RMWcc_6(op, var, cc, vcon, val, "%[var]")
+
+#define GEN_BINARY_RMWcc(X...) RMWcc_CONCAT(GEN_BINARY_RMWcc_, RMWcc_ARGS(X))(X)
+
+#define GEN_UNARY_SUFFIXED_RMWcc(op, suffix, var, cc, clobbers...)	\
+	__GEN_RMWcc(op " %[var]\n\t" suffix, var, cc,			\
 		    __CLOBBERS_MEM(clobbers))
 
-#define GEN_BINARY_RMWcc(op, var, vcon, val, arg0, cc)			\
-	__GEN_RMWcc(op __BINARY_RMWcc_ARG arg0, var, cc,		\
-		    __CLOBBERS_MEM(), vcon (val))
-
-#define GEN_BINARY_SUFFIXED_RMWcc(op, suffix, var, vcon, val, arg0, cc,	\
-				  clobbers...)				\
-	__GEN_RMWcc(op __BINARY_RMWcc_ARG arg0 "\n\t" suffix, var, cc,	\
-		    __CLOBBERS_MEM(clobbers), vcon (val))
+#define GEN_BINARY_SUFFIXED_RMWcc(op, suffix, var, cc, vcon, _val, clobbers...)\
+	__GEN_RMWcc(op " %[val], %[var]\n\t" suffix, var, cc,		\
+		    __CLOBBERS_MEM(clobbers), [val] vcon (_val))
 
 #endif /* _ASM_X86_RMWcc */



  parent reply index

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-03 13:02 [PATCH v2 0/4] locking/qspinlock, x86: Improve determinism for x86 Peter Zijlstra
2018-10-03 13:02 ` [PATCH v2 1/4] locking/qspinlock: Re-order code Peter Zijlstra
2018-10-03 13:02 ` [PATCH v2 2/4] locking/qspinlock: Rework some comments Peter Zijlstra
2018-10-10 16:13   ` Will Deacon
2018-10-03 13:03 ` Peter Zijlstra [this message]
2018-10-04  9:18   ` [PATCH v2 3/4] x86/asm: Simplify GEN_*_RMWcc() macros Peter Zijlstra
2018-10-16 16:05   ` [tip:locking/core] x86/asm: 'Simplify' " tip-bot for Peter Zijlstra
2018-10-03 13:03 ` [PATCH v2 4/4] locking/qspinlock, x86: Provide liveness guarantee Peter Zijlstra
2018-10-10 16:12   ` Will Deacon
2018-10-12  9:22     ` Will Deacon
2018-10-16 16:06   ` [tip:locking/core] " tip-bot for Peter Zijlstra
2018-10-16 16:04 ` [tip:locking/core] locking/qspinlock: Re-order code tip-bot for Peter Zijlstra
2018-10-16 16:04 ` [tip:locking/core] locking/qspinlock: Rework some comments tip-bot for Peter Zijlstra

Reply instructions:

You may reply publically 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=20181003130957.108960094@infradead.org \
    --to=peterz@infradead.org \
    --cc=JBeulich@suse.com \
    --cc=andrea.parri@amarulasolutions.com \
    --cc=bigeasy@linutronix.de \
    --cc=bp@alien8.de \
    --cc=hpa@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=longman@redhat.com \
    --cc=mingo@kernel.org \
    --cc=tglx@linutronix.de \
    --cc=will.deacon@arm.com \
    /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

LKML Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git
	git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git
	git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git
	git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git
	git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git
	git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git
	git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git
	git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \
		linux-kernel@vger.kernel.org linux-kernel@archiver.kernel.org
	public-inbox-index lkml


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel


AGPL code for this site: git clone https://public-inbox.org/ public-inbox