[PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[Eric Chan]

This series makes barrier-related macro more neat and clear.
This is a follow-up to [0-3], change to multiple patches,
for readability, create new message thread.

v5 -> v6: [PATCH 1/4] let this change to pass review by checkpatch.pl
instead of overwriting again in [PATCH 4/4]. for [PATCH 4/4] change
the error message example in commit message to make it more relevant

v4 -> v5: [PATCH 3/4] __atomic_acquire_fence and __atomic_release_fence
omit-the-fence-on-uniprocessor optimization, and fix the typo of
RISCV_RELEASE_BARRIER when spliting the patch in v3.

v3 -> v4: fix [PATCH 1/4] commit message weird line breaks and let
[PATCH 3/4] fix the form that can pass the checking of checkpatch.pl.

v2 -> v3: split the patch into multiple patches for one problem per patch.
Also review the changelog to make the description more precise.

v1 -> v2: makes compilation pass with allyesconfig instead of
defconfig only, also satisfy scripts/checkpatch.pl.
- (__asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"))
+ ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })

[0](v1/v2) https://lore.kernel.org/lkml/20240209125048.4078639-1-ericchancf@google.com/
[1] (v3) https://lore.kernel.org/lkml/20240213142856.2416073-1-ericchancf@google.com/
[2] (v4) https://lore.kernel.org/lkml/20240213200923.2547570-1-ericchancf@google.com/
[4] (v5) https://lore.kernel.org/lkml/20240213223810.2595804-1-ericchancf@google.com/

Eric Chan (4):
  riscv/barrier: Define __{mb,rmb,wmb}
  riscv/barrier: Define RISCV_FULL_BARRIER
  riscv/barrier: Consolidate fence definitions
  riscv/barrier: Resolve checkpatch.pl error

 arch/riscv/include/asm/atomic.h  | 17 ++++++++---------
 arch/riscv/include/asm/barrier.h | 21 ++++++++++-----------
 arch/riscv/include/asm/cmpxchg.h |  5 ++---
 arch/riscv/include/asm/fence.h   | 10 ++++++++--
 arch/riscv/include/asm/io.h      |  8 ++++----
 arch/riscv/include/asm/mmio.h    |  5 +++--
 arch/riscv/include/asm/mmiowb.h  |  2 +-
 7 files changed, 36 insertions(+), 32 deletions(-)

--
2.44.0.rc0.258.g7320e95886-goog

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[Eric Chan]

This series makes barrier-related macro more neat and clear.
This is a follow-up to [0-3], change to multiple patches,
for readability, create new message thread.

v5 -> v6: [PATCH 1/4] let this change to pass review by checkpatch.pl
instead of overwriting again in [PATCH 4/4]. for [PATCH 4/4] change
the error message example in commit message to make it more relevant

v4 -> v5: [PATCH 3/4] __atomic_acquire_fence and __atomic_release_fence
omit-the-fence-on-uniprocessor optimization, and fix the typo of
RISCV_RELEASE_BARRIER when spliting the patch in v3.

v3 -> v4: fix [PATCH 1/4] commit message weird line breaks and let
[PATCH 3/4] fix the form that can pass the checking of checkpatch.pl.

v2 -> v3: split the patch into multiple patches for one problem per patch.
Also review the changelog to make the description more precise.

v1 -> v2: makes compilation pass with allyesconfig instead of
defconfig only, also satisfy scripts/checkpatch.pl.
- (__asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"))
+ ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })

[0](v1/v2) https://lore.kernel.org/lkml/20240209125048.4078639-1-ericchancf@google.com/
[1] (v3) https://lore.kernel.org/lkml/20240213142856.2416073-1-ericchancf@google.com/
[2] (v4) https://lore.kernel.org/lkml/20240213200923.2547570-1-ericchancf@google.com/
[4] (v5) https://lore.kernel.org/lkml/20240213223810.2595804-1-ericchancf@google.com/

Eric Chan (4):
  riscv/barrier: Define __{mb,rmb,wmb}
  riscv/barrier: Define RISCV_FULL_BARRIER
  riscv/barrier: Consolidate fence definitions
  riscv/barrier: Resolve checkpatch.pl error

 arch/riscv/include/asm/atomic.h  | 17 ++++++++---------
 arch/riscv/include/asm/barrier.h | 21 ++++++++++-----------
 arch/riscv/include/asm/cmpxchg.h |  5 ++---
 arch/riscv/include/asm/fence.h   | 10 ++++++++--
 arch/riscv/include/asm/io.h      |  8 ++++----
 arch/riscv/include/asm/mmio.h    |  5 +++--
 arch/riscv/include/asm/mmiowb.h  |  2 +-
 7 files changed, 36 insertions(+), 32 deletions(-)

--
2.44.0.rc0.258.g7320e95886-goog

[PATCH v6 1/4] riscv/barrier: Define __{mb,rmb,wmb}

[Eric Chan]

Introduce __{mb,rmb,wmb}, and rely on the generic definitions for
{mb,rmb,wmb}. Although KCSAN is not supported yet, the definitions can
be made more consistent with generic instrumentation. Also add a space
to make the changes pass check by checkpatch.pl.
Without the space, the error message is as below:
ERROR: space required after that ',' (ctx:VxV)
26: FILE: arch/riscv/include/asm/barrier.h:23:
+#define __mb()         RISCV_FENCE(iorw,iorw)
                                        ^

Signed-off-by: Eric Chan <ericchancf@google.com>
---
v5 -> v6: let this change to pass review by checkpatch.pl instead
of overwriting again in [PATCH 4/4]

v3 -> v4: fix commit message weird line breaks

 arch/riscv/include/asm/barrier.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 110752594228..173b44a989f8 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/arch/riscv/include/asm/barrier.h
@@ -20,9 +20,9 @@
 	__asm__ __volatile__ ("fence " #p "," #s : : : "memory")

 /* These barriers need to enforce ordering on both devices or memory. */
-#define mb()		RISCV_FENCE(iorw,iorw)
-#define rmb()		RISCV_FENCE(ir,ir)
-#define wmb()		RISCV_FENCE(ow,ow)
+#define __mb()		RISCV_FENCE(iorw, iorw)
+#define __rmb()		RISCV_FENCE(ir, ir)
+#define __wmb()		RISCV_FENCE(ow, ow)

 /* These barriers do not need to enforce ordering on devices, just memory. */
 #define __smp_mb()	RISCV_FENCE(rw,rw)
--
2.44.0.rc0.258.g7320e95886-goog

[PATCH v6 1/4] riscv/barrier: Define __{mb,rmb,wmb}

[Eric Chan]

Introduce __{mb,rmb,wmb}, and rely on the generic definitions for
{mb,rmb,wmb}. Although KCSAN is not supported yet, the definitions can
be made more consistent with generic instrumentation. Also add a space
to make the changes pass check by checkpatch.pl.
Without the space, the error message is as below:
ERROR: space required after that ',' (ctx:VxV)
26: FILE: arch/riscv/include/asm/barrier.h:23:
+#define __mb()         RISCV_FENCE(iorw,iorw)
                                        ^

Signed-off-by: Eric Chan <ericchancf@google.com>
---
v5 -> v6: let this change to pass review by checkpatch.pl instead
of overwriting again in [PATCH 4/4]

v3 -> v4: fix commit message weird line breaks

 arch/riscv/include/asm/barrier.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 110752594228..173b44a989f8 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/arch/riscv/include/asm/barrier.h
@@ -20,9 +20,9 @@
 	__asm__ __volatile__ ("fence " #p "," #s : : : "memory")

 /* These barriers need to enforce ordering on both devices or memory. */
-#define mb()		RISCV_FENCE(iorw,iorw)
-#define rmb()		RISCV_FENCE(ir,ir)
-#define wmb()		RISCV_FENCE(ow,ow)
+#define __mb()		RISCV_FENCE(iorw, iorw)
+#define __rmb()		RISCV_FENCE(ir, ir)
+#define __wmb()		RISCV_FENCE(ow, ow)

 /* These barriers do not need to enforce ordering on devices, just memory. */
 #define __smp_mb()	RISCV_FENCE(rw,rw)
--
2.44.0.rc0.258.g7320e95886-goog

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH v6 2/4] riscv/barrier: Define RISCV_FULL_BARRIER

[Eric Chan]

Introduce RISCV_FULL_BARRIER and use in arch_atomic* function.
like RISCV_ACQUIRE_BARRIER and RISCV_RELEASE_BARRIER, the fence
instruction can be eliminated When SMP is not enabled.

Signed-off-by: Eric Chan <ericchancf@google.com>
---
 arch/riscv/include/asm/atomic.h  | 16 ++++++++--------
 arch/riscv/include/asm/cmpxchg.h |  4 ++--
 arch/riscv/include/asm/fence.h   |  2 ++
 3 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/arch/riscv/include/asm/atomic.h b/arch/riscv/include/asm/atomic.h
index f5dfef6c2153..31e6e2e7cc18 100644
--- a/arch/riscv/include/asm/atomic.h
+++ b/arch/riscv/include/asm/atomic.h
@@ -207,7 +207,7 @@ static __always_inline int arch_atomic_fetch_add_unless(atomic_t *v, int a, int
 		"	add      %[rc], %[p], %[a]\n"
 		"	sc.w.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
@@ -228,7 +228,7 @@ static __always_inline s64 arch_atomic64_fetch_add_unless(atomic64_t *v, s64 a,
 		"	add      %[rc], %[p], %[a]\n"
 		"	sc.d.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
@@ -248,7 +248,7 @@ static __always_inline bool arch_atomic_inc_unless_negative(atomic_t *v)
 		"	addi      %[rc], %[p], 1\n"
 		"	sc.w.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -268,7 +268,7 @@ static __always_inline bool arch_atomic_dec_unless_positive(atomic_t *v)
 		"	addi      %[rc], %[p], -1\n"
 		"	sc.w.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -288,7 +288,7 @@ static __always_inline int arch_atomic_dec_if_positive(atomic_t *v)
 		"	bltz     %[rc], 1f\n"
 		"	sc.w.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -310,7 +310,7 @@ static __always_inline bool arch_atomic64_inc_unless_negative(atomic64_t *v)
 		"	addi      %[rc], %[p], 1\n"
 		"	sc.d.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -331,7 +331,7 @@ static __always_inline bool arch_atomic64_dec_unless_positive(atomic64_t *v)
 		"	addi      %[rc], %[p], -1\n"
 		"	sc.d.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -352,7 +352,7 @@ static __always_inline s64 arch_atomic64_dec_if_positive(atomic64_t *v)
 		"	bltz     %[rc], 1f\n"
 		"	sc.d.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index 2f4726d3cfcc..a608e4d1a0a4 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -313,7 +313,7 @@
 			"	bne  %0, %z3, 1f\n"			\
 			"	sc.w.rl %1, %z4, %2\n"			\
 			"	bnez %1, 0b\n"				\
-			"	fence rw, rw\n"				\
+			RISCV_FULL_BARRIER				\
 			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" ((long)__old), "rJ" (__new)		\
@@ -325,7 +325,7 @@
 			"	bne %0, %z3, 1f\n"			\
 			"	sc.d.rl %1, %z4, %2\n"			\
 			"	bnez %1, 0b\n"				\
-			"	fence rw, rw\n"				\
+			RISCV_FULL_BARRIER				\
 			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
diff --git a/arch/riscv/include/asm/fence.h b/arch/riscv/include/asm/fence.h
index 2b443a3a487f..6c26c44dfcd6 100644
--- a/arch/riscv/include/asm/fence.h
+++ b/arch/riscv/include/asm/fence.h
@@ -4,9 +4,11 @@
 #ifdef CONFIG_SMP
 #define RISCV_ACQUIRE_BARRIER		"\tfence r , rw\n"
 #define RISCV_RELEASE_BARRIER		"\tfence rw,  w\n"
+#define RISCV_FULL_BARRIER		"\tfence rw, rw\n"
 #else
 #define RISCV_ACQUIRE_BARRIER
 #define RISCV_RELEASE_BARRIER
+#define RISCV_FULL_BARRIER
 #endif
 
 #endif	/* _ASM_RISCV_FENCE_H */
-- 
2.44.0.rc0.258.g7320e95886-goog

[PATCH v6 2/4] riscv/barrier: Define RISCV_FULL_BARRIER

[Eric Chan]

Introduce RISCV_FULL_BARRIER and use in arch_atomic* function.
like RISCV_ACQUIRE_BARRIER and RISCV_RELEASE_BARRIER, the fence
instruction can be eliminated When SMP is not enabled.

Signed-off-by: Eric Chan <ericchancf@google.com>
---
 arch/riscv/include/asm/atomic.h  | 16 ++++++++--------
 arch/riscv/include/asm/cmpxchg.h |  4 ++--
 arch/riscv/include/asm/fence.h   |  2 ++
 3 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/arch/riscv/include/asm/atomic.h b/arch/riscv/include/asm/atomic.h
index f5dfef6c2153..31e6e2e7cc18 100644
--- a/arch/riscv/include/asm/atomic.h
+++ b/arch/riscv/include/asm/atomic.h
@@ -207,7 +207,7 @@ static __always_inline int arch_atomic_fetch_add_unless(atomic_t *v, int a, int
 		"	add      %[rc], %[p], %[a]\n"
 		"	sc.w.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
@@ -228,7 +228,7 @@ static __always_inline s64 arch_atomic64_fetch_add_unless(atomic64_t *v, s64 a,
 		"	add      %[rc], %[p], %[a]\n"
 		"	sc.d.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		: [a]"r" (a), [u]"r" (u)
@@ -248,7 +248,7 @@ static __always_inline bool arch_atomic_inc_unless_negative(atomic_t *v)
 		"	addi      %[rc], %[p], 1\n"
 		"	sc.w.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -268,7 +268,7 @@ static __always_inline bool arch_atomic_dec_unless_positive(atomic_t *v)
 		"	addi      %[rc], %[p], -1\n"
 		"	sc.w.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -288,7 +288,7 @@ static __always_inline int arch_atomic_dec_if_positive(atomic_t *v)
 		"	bltz     %[rc], 1f\n"
 		"	sc.w.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -310,7 +310,7 @@ static __always_inline bool arch_atomic64_inc_unless_negative(atomic64_t *v)
 		"	addi      %[rc], %[p], 1\n"
 		"	sc.d.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -331,7 +331,7 @@ static __always_inline bool arch_atomic64_dec_unless_positive(atomic64_t *v)
 		"	addi      %[rc], %[p], -1\n"
 		"	sc.d.rl   %[rc], %[rc], %[c]\n"
 		"	bnez      %[rc], 0b\n"
-		"	fence     rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
@@ -352,7 +352,7 @@ static __always_inline s64 arch_atomic64_dec_if_positive(atomic64_t *v)
 		"	bltz     %[rc], 1f\n"
 		"	sc.d.rl  %[rc], %[rc], %[c]\n"
 		"	bnez     %[rc], 0b\n"
-		"	fence    rw, rw\n"
+		RISCV_FULL_BARRIER
 		"1:\n"
 		: [p]"=&r" (prev), [rc]"=&r" (rc), [c]"+A" (v->counter)
 		:
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index 2f4726d3cfcc..a608e4d1a0a4 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -313,7 +313,7 @@
 			"	bne  %0, %z3, 1f\n"			\
 			"	sc.w.rl %1, %z4, %2\n"			\
 			"	bnez %1, 0b\n"				\
-			"	fence rw, rw\n"				\
+			RISCV_FULL_BARRIER				\
 			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" ((long)__old), "rJ" (__new)		\
@@ -325,7 +325,7 @@
 			"	bne %0, %z3, 1f\n"			\
 			"	sc.d.rl %1, %z4, %2\n"			\
 			"	bnez %1, 0b\n"				\
-			"	fence rw, rw\n"				\
+			RISCV_FULL_BARRIER				\
 			"1:\n"						\
 			: "=&r" (__ret), "=&r" (__rc), "+A" (*__ptr)	\
 			: "rJ" (__old), "rJ" (__new)			\
diff --git a/arch/riscv/include/asm/fence.h b/arch/riscv/include/asm/fence.h
index 2b443a3a487f..6c26c44dfcd6 100644
--- a/arch/riscv/include/asm/fence.h
+++ b/arch/riscv/include/asm/fence.h
@@ -4,9 +4,11 @@
 #ifdef CONFIG_SMP
 #define RISCV_ACQUIRE_BARRIER		"\tfence r , rw\n"
 #define RISCV_RELEASE_BARRIER		"\tfence rw,  w\n"
+#define RISCV_FULL_BARRIER		"\tfence rw, rw\n"
 #else
 #define RISCV_ACQUIRE_BARRIER
 #define RISCV_RELEASE_BARRIER
+#define RISCV_FULL_BARRIER
 #endif
 
 #endif	/* _ASM_RISCV_FENCE_H */
-- 
2.44.0.rc0.258.g7320e95886-goog


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH v6 3/4] riscv/barrier: Consolidate fence definitions

[Eric Chan]

Disparate fence implementations are consolidated into fence.h.
Also introduce RISCV_FENCE_ASM to make fence macro more reusable.

Signed-off-by: Eric Chan <ericchancf@google.com>
---
v4 -> v5: __atomic_acquire_fence and __atomic_release_fence
omit-the-fence-on-uniprocessor optimization, and fix the typo of
RISCV_RELEASE_BARRIER when spliting the patch in v3.

v3 -> v4 fix the form that can pass the checking of checkpatch.pl.

v1 -> v2: makes compilation pass with allyesconfig instead of
defconfig only, also satisfy scripts/checkpatch.pl.
- (__asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"))
+ ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })

 arch/riscv/include/asm/atomic.h  |  1 -
 arch/riscv/include/asm/barrier.h |  3 +--
 arch/riscv/include/asm/cmpxchg.h |  1 -
 arch/riscv/include/asm/fence.h   | 10 +++++++---
 arch/riscv/include/asm/io.h      |  8 ++++----
 arch/riscv/include/asm/mmio.h    |  5 +++--
 arch/riscv/include/asm/mmiowb.h  |  2 +-
 7 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/arch/riscv/include/asm/atomic.h b/arch/riscv/include/asm/atomic.h
index 31e6e2e7cc18..0e0522e588ca 100644
--- a/arch/riscv/include/asm/atomic.h
+++ b/arch/riscv/include/asm/atomic.h
@@ -17,7 +17,6 @@
 #endif

 #include <asm/cmpxchg.h>
-#include <asm/barrier.h>

 #define __atomic_acquire_fence()					\
 	__asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory")
diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 173b44a989f8..15857dbc2279 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/arch/riscv/include/asm/barrier.h
@@ -11,13 +11,12 @@
 #define _ASM_RISCV_BARRIER_H

 #ifndef __ASSEMBLY__
+#include <asm/fence.h>

 #define nop()		__asm__ __volatile__ ("nop")
 #define __nops(n)	".rept	" #n "\nnop\n.endr\n"
 #define nops(n)		__asm__ __volatile__ (__nops(n))

-#define RISCV_FENCE(p, s) \
-	__asm__ __volatile__ ("fence " #p "," #s : : : "memory")

 /* These barriers need to enforce ordering on both devices or memory. */
 #define __mb()		RISCV_FENCE(iorw, iorw)
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index a608e4d1a0a4..2fee65cc8443 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -8,7 +8,6 @@

 #include <linux/bug.h>

-#include <asm/barrier.h>
 #include <asm/fence.h>

 #define __xchg_relaxed(ptr, new, size)					\
diff --git a/arch/riscv/include/asm/fence.h b/arch/riscv/include/asm/fence.h
index 6c26c44dfcd6..6bcd80325dfc 100644
--- a/arch/riscv/include/asm/fence.h
+++ b/arch/riscv/include/asm/fence.h
@@ -1,10 +1,14 @@
 #ifndef _ASM_RISCV_FENCE_H
 #define _ASM_RISCV_FENCE_H

+#define RISCV_FENCE_ASM(p, s)		"\tfence " #p "," #s "\n"
+#define RISCV_FENCE(p, s) \
+	({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })
+
 #ifdef CONFIG_SMP
-#define RISCV_ACQUIRE_BARRIER		"\tfence r , rw\n"
-#define RISCV_RELEASE_BARRIER		"\tfence rw,  w\n"
-#define RISCV_FULL_BARRIER		"\tfence rw, rw\n"
+#define RISCV_ACQUIRE_BARRIER		RISCV_FENCE_ASM(r, rw)
+#define RISCV_RELEASE_BARRIER		RISCV_FENCE_ASM(rw, w)
+#define RISCV_FULL_BARRIER		RISCV_FENCE_ASM(rw, rw)
 #else
 #define RISCV_ACQUIRE_BARRIER
 #define RISCV_RELEASE_BARRIER
diff --git a/arch/riscv/include/asm/io.h b/arch/riscv/include/asm/io.h
index 42497d487a17..1c5c641075d2 100644
--- a/arch/riscv/include/asm/io.h
+++ b/arch/riscv/include/asm/io.h
@@ -47,10 +47,10 @@
  * sufficient to ensure this works sanely on controllers that support I/O
  * writes.
  */
-#define __io_pbr()	__asm__ __volatile__ ("fence io,i"  : : : "memory");
-#define __io_par(v)	__asm__ __volatile__ ("fence i,ior" : : : "memory");
-#define __io_pbw()	__asm__ __volatile__ ("fence iow,o" : : : "memory");
-#define __io_paw()	__asm__ __volatile__ ("fence o,io"  : : : "memory");
+#define __io_pbr()	RISCV_FENCE(io, i)
+#define __io_par(v)	RISCV_FENCE(i, ior)
+#define __io_pbw()	RISCV_FENCE(iow, o)
+#define __io_paw()	RISCV_FENCE(o, io)

 /*
  * Accesses from a single hart to a single I/O address must be ordered.  This
diff --git a/arch/riscv/include/asm/mmio.h b/arch/riscv/include/asm/mmio.h
index 4c58ee7f95ec..06cadfd7a237 100644
--- a/arch/riscv/include/asm/mmio.h
+++ b/arch/riscv/include/asm/mmio.h
@@ -12,6 +12,7 @@
 #define _ASM_RISCV_MMIO_H

 #include <linux/types.h>
+#include <asm/fence.h>
 #include <asm/mmiowb.h>

 /* Generic IO read/write.  These perform native-endian accesses. */
@@ -131,8 +132,8 @@ static inline u64 __raw_readq(const volatile void __iomem *addr)
  * doesn't define any ordering between the memory space and the I/O space.
  */
 #define __io_br()	do {} while (0)
-#define __io_ar(v)	({ __asm__ __volatile__ ("fence i,ir" : : : "memory"); })
-#define __io_bw()	({ __asm__ __volatile__ ("fence w,o" : : : "memory"); })
+#define __io_ar(v)	RISCV_FENCE(i, ir)
+#define __io_bw()	RISCV_FENCE(w, o)
 #define __io_aw()	mmiowb_set_pending()

 #define readb(c)	({ u8  __v; __io_br(); __v = readb_cpu(c); __io_ar(__v); __v; })
diff --git a/arch/riscv/include/asm/mmiowb.h b/arch/riscv/include/asm/mmiowb.h
index 0b2333e71fdc..52ce4a399d9b 100644
--- a/arch/riscv/include/asm/mmiowb.h
+++ b/arch/riscv/include/asm/mmiowb.h
@@ -7,7 +7,7 @@
  * "o,w" is sufficient to ensure that all writes to the device have completed
  * before the write to the spinlock is allowed to commit.
  */
-#define mmiowb()	__asm__ __volatile__ ("fence o,w" : : : "memory");
+#define mmiowb()	RISCV_FENCE(o, w)

 #include <linux/smp.h>
 #include <asm-generic/mmiowb.h>
--
2.44.0.rc0.258.g7320e95886-goog

[PATCH v6 3/4] riscv/barrier: Consolidate fence definitions

[Eric Chan]

Disparate fence implementations are consolidated into fence.h.
Also introduce RISCV_FENCE_ASM to make fence macro more reusable.

Signed-off-by: Eric Chan <ericchancf@google.com>
---
v4 -> v5: __atomic_acquire_fence and __atomic_release_fence
omit-the-fence-on-uniprocessor optimization, and fix the typo of
RISCV_RELEASE_BARRIER when spliting the patch in v3.

v3 -> v4 fix the form that can pass the checking of checkpatch.pl.

v1 -> v2: makes compilation pass with allyesconfig instead of
defconfig only, also satisfy scripts/checkpatch.pl.
- (__asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"))
+ ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })

 arch/riscv/include/asm/atomic.h  |  1 -
 arch/riscv/include/asm/barrier.h |  3 +--
 arch/riscv/include/asm/cmpxchg.h |  1 -
 arch/riscv/include/asm/fence.h   | 10 +++++++---
 arch/riscv/include/asm/io.h      |  8 ++++----
 arch/riscv/include/asm/mmio.h    |  5 +++--
 arch/riscv/include/asm/mmiowb.h  |  2 +-
 7 files changed, 16 insertions(+), 14 deletions(-)

diff --git a/arch/riscv/include/asm/atomic.h b/arch/riscv/include/asm/atomic.h
index 31e6e2e7cc18..0e0522e588ca 100644
--- a/arch/riscv/include/asm/atomic.h
+++ b/arch/riscv/include/asm/atomic.h
@@ -17,7 +17,6 @@
 #endif

 #include <asm/cmpxchg.h>
-#include <asm/barrier.h>

 #define __atomic_acquire_fence()					\
 	__asm__ __volatile__(RISCV_ACQUIRE_BARRIER "" ::: "memory")
diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 173b44a989f8..15857dbc2279 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/arch/riscv/include/asm/barrier.h
@@ -11,13 +11,12 @@
 #define _ASM_RISCV_BARRIER_H

 #ifndef __ASSEMBLY__
+#include <asm/fence.h>

 #define nop()		__asm__ __volatile__ ("nop")
 #define __nops(n)	".rept	" #n "\nnop\n.endr\n"
 #define nops(n)		__asm__ __volatile__ (__nops(n))

-#define RISCV_FENCE(p, s) \
-	__asm__ __volatile__ ("fence " #p "," #s : : : "memory")

 /* These barriers need to enforce ordering on both devices or memory. */
 #define __mb()		RISCV_FENCE(iorw, iorw)
diff --git a/arch/riscv/include/asm/cmpxchg.h b/arch/riscv/include/asm/cmpxchg.h
index a608e4d1a0a4..2fee65cc8443 100644
--- a/arch/riscv/include/asm/cmpxchg.h
+++ b/arch/riscv/include/asm/cmpxchg.h
@@ -8,7 +8,6 @@

 #include <linux/bug.h>

-#include <asm/barrier.h>
 #include <asm/fence.h>

 #define __xchg_relaxed(ptr, new, size)					\
diff --git a/arch/riscv/include/asm/fence.h b/arch/riscv/include/asm/fence.h
index 6c26c44dfcd6..6bcd80325dfc 100644
--- a/arch/riscv/include/asm/fence.h
+++ b/arch/riscv/include/asm/fence.h
@@ -1,10 +1,14 @@
 #ifndef _ASM_RISCV_FENCE_H
 #define _ASM_RISCV_FENCE_H

+#define RISCV_FENCE_ASM(p, s)		"\tfence " #p "," #s "\n"
+#define RISCV_FENCE(p, s) \
+	({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })
+
 #ifdef CONFIG_SMP
-#define RISCV_ACQUIRE_BARRIER		"\tfence r , rw\n"
-#define RISCV_RELEASE_BARRIER		"\tfence rw,  w\n"
-#define RISCV_FULL_BARRIER		"\tfence rw, rw\n"
+#define RISCV_ACQUIRE_BARRIER		RISCV_FENCE_ASM(r, rw)
+#define RISCV_RELEASE_BARRIER		RISCV_FENCE_ASM(rw, w)
+#define RISCV_FULL_BARRIER		RISCV_FENCE_ASM(rw, rw)
 #else
 #define RISCV_ACQUIRE_BARRIER
 #define RISCV_RELEASE_BARRIER
diff --git a/arch/riscv/include/asm/io.h b/arch/riscv/include/asm/io.h
index 42497d487a17..1c5c641075d2 100644
--- a/arch/riscv/include/asm/io.h
+++ b/arch/riscv/include/asm/io.h
@@ -47,10 +47,10 @@
  * sufficient to ensure this works sanely on controllers that support I/O
  * writes.
  */
-#define __io_pbr()	__asm__ __volatile__ ("fence io,i"  : : : "memory");
-#define __io_par(v)	__asm__ __volatile__ ("fence i,ior" : : : "memory");
-#define __io_pbw()	__asm__ __volatile__ ("fence iow,o" : : : "memory");
-#define __io_paw()	__asm__ __volatile__ ("fence o,io"  : : : "memory");
+#define __io_pbr()	RISCV_FENCE(io, i)
+#define __io_par(v)	RISCV_FENCE(i, ior)
+#define __io_pbw()	RISCV_FENCE(iow, o)
+#define __io_paw()	RISCV_FENCE(o, io)

 /*
  * Accesses from a single hart to a single I/O address must be ordered.  This
diff --git a/arch/riscv/include/asm/mmio.h b/arch/riscv/include/asm/mmio.h
index 4c58ee7f95ec..06cadfd7a237 100644
--- a/arch/riscv/include/asm/mmio.h
+++ b/arch/riscv/include/asm/mmio.h
@@ -12,6 +12,7 @@
 #define _ASM_RISCV_MMIO_H

 #include <linux/types.h>
+#include <asm/fence.h>
 #include <asm/mmiowb.h>

 /* Generic IO read/write.  These perform native-endian accesses. */
@@ -131,8 +132,8 @@ static inline u64 __raw_readq(const volatile void __iomem *addr)
  * doesn't define any ordering between the memory space and the I/O space.
  */
 #define __io_br()	do {} while (0)
-#define __io_ar(v)	({ __asm__ __volatile__ ("fence i,ir" : : : "memory"); })
-#define __io_bw()	({ __asm__ __volatile__ ("fence w,o" : : : "memory"); })
+#define __io_ar(v)	RISCV_FENCE(i, ir)
+#define __io_bw()	RISCV_FENCE(w, o)
 #define __io_aw()	mmiowb_set_pending()

 #define readb(c)	({ u8  __v; __io_br(); __v = readb_cpu(c); __io_ar(__v); __v; })
diff --git a/arch/riscv/include/asm/mmiowb.h b/arch/riscv/include/asm/mmiowb.h
index 0b2333e71fdc..52ce4a399d9b 100644
--- a/arch/riscv/include/asm/mmiowb.h
+++ b/arch/riscv/include/asm/mmiowb.h
@@ -7,7 +7,7 @@
  * "o,w" is sufficient to ensure that all writes to the device have completed
  * before the write to the spinlock is allowed to commit.
  */
-#define mmiowb()	__asm__ __volatile__ ("fence o,w" : : : "memory");
+#define mmiowb()	RISCV_FENCE(o, w)

 #include <linux/smp.h>
 #include <asm-generic/mmiowb.h>
--
2.44.0.rc0.258.g7320e95886-goog

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

[PATCH v6 4/4] riscv/barrier: Add missing space after ','

[Eric Chan]

The past form of RISCV_FENCE would cause checkpatch.pl to issue
error messages, the example is as follows:
ERROR: space required after that ',' (ctx:VxV)
26: FILE: arch/riscv/include/asm/barrier.h:27:
+#define __smp_mb()         RISCV_FENCE(rw,rw)
                                          ^
fix the remaining of RISCV_FENCE.

Signed-off-by: Eric Chan <ericchancf@google.com>
---
v5 -> v6: change the error message example in commit message to make it
more relevant

 arch/riscv/include/asm/barrier.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 15857dbc2279..880b56d8480d 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/arch/riscv/include/asm/barrier.h
@@ -24,14 +24,14 @@
 #define __wmb()		RISCV_FENCE(ow, ow)

 /* These barriers do not need to enforce ordering on devices, just memory. */
-#define __smp_mb()	RISCV_FENCE(rw,rw)
-#define __smp_rmb()	RISCV_FENCE(r,r)
-#define __smp_wmb()	RISCV_FENCE(w,w)
+#define __smp_mb()	RISCV_FENCE(rw, rw)
+#define __smp_rmb()	RISCV_FENCE(r, r)
+#define __smp_wmb()	RISCV_FENCE(w, w)

 #define __smp_store_release(p, v)					\
 do {									\
 	compiletime_assert_atomic_type(*p);				\
-	RISCV_FENCE(rw,w);						\
+	RISCV_FENCE(rw, w);						\
 	WRITE_ONCE(*p, v);						\
 } while (0)

@@ -39,7 +39,7 @@ do {									\
 ({									\
 	typeof(*p) ___p1 = READ_ONCE(*p);				\
 	compiletime_assert_atomic_type(*p);				\
-	RISCV_FENCE(r,rw);						\
+	RISCV_FENCE(r, rw);						\
 	___p1;								\
 })

@@ -68,7 +68,7 @@ do {									\
  * instances the scheduler pairs this with an mb(), so nothing is necessary on
  * the new hart.
  */
-#define smp_mb__after_spinlock()	RISCV_FENCE(iorw,iorw)
+#define smp_mb__after_spinlock()	RISCV_FENCE(iorw, iorw)

 #include <asm-generic/barrier.h>

--
2.44.0.rc0.258.g7320e95886-goog

[PATCH v6 4/4] riscv/barrier: Add missing space after ','

[Eric Chan]

The past form of RISCV_FENCE would cause checkpatch.pl to issue
error messages, the example is as follows:
ERROR: space required after that ',' (ctx:VxV)
26: FILE: arch/riscv/include/asm/barrier.h:27:
+#define __smp_mb()         RISCV_FENCE(rw,rw)
                                          ^
fix the remaining of RISCV_FENCE.

Signed-off-by: Eric Chan <ericchancf@google.com>
---
v5 -> v6: change the error message example in commit message to make it
more relevant

 arch/riscv/include/asm/barrier.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/riscv/include/asm/barrier.h b/arch/riscv/include/asm/barrier.h
index 15857dbc2279..880b56d8480d 100644
--- a/arch/riscv/include/asm/barrier.h
+++ b/arch/riscv/include/asm/barrier.h
@@ -24,14 +24,14 @@
 #define __wmb()		RISCV_FENCE(ow, ow)

 /* These barriers do not need to enforce ordering on devices, just memory. */
-#define __smp_mb()	RISCV_FENCE(rw,rw)
-#define __smp_rmb()	RISCV_FENCE(r,r)
-#define __smp_wmb()	RISCV_FENCE(w,w)
+#define __smp_mb()	RISCV_FENCE(rw, rw)
+#define __smp_rmb()	RISCV_FENCE(r, r)
+#define __smp_wmb()	RISCV_FENCE(w, w)

 #define __smp_store_release(p, v)					\
 do {									\
 	compiletime_assert_atomic_type(*p);				\
-	RISCV_FENCE(rw,w);						\
+	RISCV_FENCE(rw, w);						\
 	WRITE_ONCE(*p, v);						\
 } while (0)

@@ -39,7 +39,7 @@ do {									\
 ({									\
 	typeof(*p) ___p1 = READ_ONCE(*p);				\
 	compiletime_assert_atomic_type(*p);				\
-	RISCV_FENCE(r,rw);						\
+	RISCV_FENCE(r, rw);						\
 	___p1;								\
 })

@@ -68,7 +68,7 @@ do {									\
  * instances the scheduler pairs this with an mb(), so nothing is necessary on
  * the new hart.
  */
-#define smp_mb__after_spinlock()	RISCV_FENCE(iorw,iorw)
+#define smp_mb__after_spinlock()	RISCV_FENCE(iorw, iorw)

 #include <asm-generic/barrier.h>

--
2.44.0.rc0.258.g7320e95886-goog

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[Andrea Parri]

> Eric Chan (4):
>   riscv/barrier: Define __{mb,rmb,wmb}
>   riscv/barrier: Define RISCV_FULL_BARRIER
>   riscv/barrier: Consolidate fence definitions
>   riscv/barrier: Resolve checkpatch.pl error

LGTM.  For the series,

Reviewed-by: Andrea Parri <parri.andrea@gmail.com>

  Andrea

Re: [PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[Andrea Parri]

> Eric Chan (4):
>   riscv/barrier: Define __{mb,rmb,wmb}
>   riscv/barrier: Define RISCV_FULL_BARRIER
>   riscv/barrier: Consolidate fence definitions
>   riscv/barrier: Resolve checkpatch.pl error

LGTM.  For the series,

Reviewed-by: Andrea Parri <parri.andrea@gmail.com>

  Andrea

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[Samuel Holland]

On 2024-02-17 7:12 AM, Eric Chan wrote:
> This series makes barrier-related macro more neat and clear.
> This is a follow-up to [0-3], change to multiple patches,
> for readability, create new message thread.
> 
> v5 -> v6: [PATCH 1/4] let this change to pass review by checkpatch.pl
> instead of overwriting again in [PATCH 4/4]. for [PATCH 4/4] change
> the error message example in commit message to make it more relevant
> 
> v4 -> v5: [PATCH 3/4] __atomic_acquire_fence and __atomic_release_fence
> omit-the-fence-on-uniprocessor optimization, and fix the typo of
> RISCV_RELEASE_BARRIER when spliting the patch in v3.
> 
> v3 -> v4: fix [PATCH 1/4] commit message weird line breaks and let
> [PATCH 3/4] fix the form that can pass the checking of checkpatch.pl.
> 
> v2 -> v3: split the patch into multiple patches for one problem per patch.
> Also review the changelog to make the description more precise.
> 
> v1 -> v2: makes compilation pass with allyesconfig instead of
> defconfig only, also satisfy scripts/checkpatch.pl.
> - (__asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"))
> + ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })
> 
> [0](v1/v2) https://lore.kernel.org/lkml/20240209125048.4078639-1-ericchancf@google.com/
> [1] (v3) https://lore.kernel.org/lkml/20240213142856.2416073-1-ericchancf@google.com/
> [2] (v4) https://lore.kernel.org/lkml/20240213200923.2547570-1-ericchancf@google.com/
> [4] (v5) https://lore.kernel.org/lkml/20240213223810.2595804-1-ericchancf@google.com/
> 
> Eric Chan (4):
>   riscv/barrier: Define __{mb,rmb,wmb}
>   riscv/barrier: Define RISCV_FULL_BARRIER
>   riscv/barrier: Consolidate fence definitions
>   riscv/barrier: Resolve checkpatch.pl error
> 
>  arch/riscv/include/asm/atomic.h  | 17 ++++++++---------
>  arch/riscv/include/asm/barrier.h | 21 ++++++++++-----------
>  arch/riscv/include/asm/cmpxchg.h |  5 ++---
>  arch/riscv/include/asm/fence.h   | 10 ++++++++--
>  arch/riscv/include/asm/io.h      |  8 ++++----
>  arch/riscv/include/asm/mmio.h    |  5 +++--
>  arch/riscv/include/asm/mmiowb.h  |  2 +-
>  7 files changed, 36 insertions(+), 32 deletions(-)

For the series:

Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
Tested-by: Samuel Holland <samuel.holland@sifive.com>

Re: [PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[Samuel Holland]

On 2024-02-17 7:12 AM, Eric Chan wrote:
> This series makes barrier-related macro more neat and clear.
> This is a follow-up to [0-3], change to multiple patches,
> for readability, create new message thread.
> 
> v5 -> v6: [PATCH 1/4] let this change to pass review by checkpatch.pl
> instead of overwriting again in [PATCH 4/4]. for [PATCH 4/4] change
> the error message example in commit message to make it more relevant
> 
> v4 -> v5: [PATCH 3/4] __atomic_acquire_fence and __atomic_release_fence
> omit-the-fence-on-uniprocessor optimization, and fix the typo of
> RISCV_RELEASE_BARRIER when spliting the patch in v3.
> 
> v3 -> v4: fix [PATCH 1/4] commit message weird line breaks and let
> [PATCH 3/4] fix the form that can pass the checking of checkpatch.pl.
> 
> v2 -> v3: split the patch into multiple patches for one problem per patch.
> Also review the changelog to make the description more precise.
> 
> v1 -> v2: makes compilation pass with allyesconfig instead of
> defconfig only, also satisfy scripts/checkpatch.pl.
> - (__asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"))
> + ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })
> 
> [0](v1/v2) https://lore.kernel.org/lkml/20240209125048.4078639-1-ericchancf@google.com/
> [1] (v3) https://lore.kernel.org/lkml/20240213142856.2416073-1-ericchancf@google.com/
> [2] (v4) https://lore.kernel.org/lkml/20240213200923.2547570-1-ericchancf@google.com/
> [4] (v5) https://lore.kernel.org/lkml/20240213223810.2595804-1-ericchancf@google.com/
> 
> Eric Chan (4):
>   riscv/barrier: Define __{mb,rmb,wmb}
>   riscv/barrier: Define RISCV_FULL_BARRIER
>   riscv/barrier: Consolidate fence definitions
>   riscv/barrier: Resolve checkpatch.pl error
> 
>  arch/riscv/include/asm/atomic.h  | 17 ++++++++---------
>  arch/riscv/include/asm/barrier.h | 21 ++++++++++-----------
>  arch/riscv/include/asm/cmpxchg.h |  5 ++---
>  arch/riscv/include/asm/fence.h   | 10 ++++++++--
>  arch/riscv/include/asm/io.h      |  8 ++++----
>  arch/riscv/include/asm/mmio.h    |  5 +++--
>  arch/riscv/include/asm/mmiowb.h  |  2 +-
>  7 files changed, 36 insertions(+), 32 deletions(-)

For the series:

Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
Tested-by: Samuel Holland <samuel.holland@sifive.com>


_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

Re: [PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[patchwork-bot+linux-riscv]

Hello:

This series was applied to riscv/linux.git (for-next)
by Palmer Dabbelt <palmer@rivosinc.com>:

On Sat, 17 Feb 2024 13:12:06 +0000 you wrote:
> This series makes barrier-related macro more neat and clear.
> This is a follow-up to [0-3], change to multiple patches,
> for readability, create new message thread.
> 
> v5 -> v6: [PATCH 1/4] let this change to pass review by checkpatch.pl
> instead of overwriting again in [PATCH 4/4]. for [PATCH 4/4] change
> the error message example in commit message to make it more relevant
> 
> [...]

Here is the summary with links:
  - [v6,1/4] riscv/barrier: Define __{mb,rmb,wmb}
    https://git.kernel.org/riscv/c/89f4fd7b1ab7
  - [v6,2/4] riscv/barrier: Define RISCV_FULL_BARRIER
    https://git.kernel.org/riscv/c/b3c8064ccc44
  - [v6,3/4] riscv/barrier: Consolidate fence definitions
    https://git.kernel.org/riscv/c/c85688e2b0f0
  - [v6,4/4] riscv/barrier: Add missing space after ','
    https://git.kernel.org/riscv/c/9133e6e6908d

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH v6 0/4] riscv/barrier: tidying up barrier-related macro

[patchwork-bot+linux-riscv]

Hello:

This series was applied to riscv/linux.git (for-next)
by Palmer Dabbelt <palmer@rivosinc.com>:

On Sat, 17 Feb 2024 13:12:06 +0000 you wrote:
> This series makes barrier-related macro more neat and clear.
> This is a follow-up to [0-3], change to multiple patches,
> for readability, create new message thread.
> 
> v5 -> v6: [PATCH 1/4] let this change to pass review by checkpatch.pl
> instead of overwriting again in [PATCH 4/4]. for [PATCH 4/4] change
> the error message example in commit message to make it more relevant
> 
> [...]

Here is the summary with links:
  - [v6,1/4] riscv/barrier: Define __{mb,rmb,wmb}
    https://git.kernel.org/riscv/c/89f4fd7b1ab7
  - [v6,2/4] riscv/barrier: Define RISCV_FULL_BARRIER
    https://git.kernel.org/riscv/c/b3c8064ccc44
  - [v6,3/4] riscv/barrier: Consolidate fence definitions
    https://git.kernel.org/riscv/c/c85688e2b0f0
  - [v6,4/4] riscv/barrier: Add missing space after ','
    https://git.kernel.org/riscv/c/9133e6e6908d

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html



_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv