riscv: xchg: Prefetch the destination word for sc.w

The cost of changing a cacheline from shared to exclusive state can be
significant, especially when this is triggered by an exclusive store,
since it may result in having to retry the transaction.

This patch makes use of prefetch.w to prefetch cachelines for write
prior to lr/sc loops when using the xchg_small atomic routine.

This patch is inspired by commit 0ea366f5e1 ("arm64: atomics:
prefetch the destination word for write prior to stxr").

Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Link: https://lore.kernel.org/r/20231231082955.16516-4-guoren@kernel.org
Tested-by: Andrea Parri <parri.andrea@gmail.com>
Link: https://lore.kernel.org/r/20250421142441.395849-5-alexghiti@rivosinc.com
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Signed-off-by: Palmer Dabbelt <palmer@dabbelt.com>

arch/riscv/include/asm/cmpxchg.h

@@ -13,6 +13,7 @@
 #include <asm/hwcap.h>
 #include <asm/insn-def.h>
 #include <asm/cpufeature-macros.h>
+#include <asm/processor.h>
 
 #define __arch_xchg_masked(sc_sfx, swap_sfx, prepend, sc_append,		\
 			   swap_append, r, p, n)				\
@@ -37,6 +38,7 @@
 										\
 		__asm__ __volatile__ (						\
 		       prepend							\
+		       PREFETCHW_ASM(%5)					\
 		       "0:	lr.w %0, %2\n"					\
 		       "	and  %1, %0, %z4\n"				\
 		       "	or   %1, %1, %z3\n"				\
@@ -44,7 +46,7 @@
 		       "	bnez %1, 0b\n"					\
 		       sc_append						\
 		       : "=&r" (__retx), "=&r" (__rc), "+A" (*(__ptr32b))	\
-		       : "rJ" (__newx), "rJ" (~__mask)				\
+		       : "rJ" (__newx), "rJ" (~__mask), "rJ" (__ptr32b)		\
 		       : "memory");						\
 										\
 		r = (__typeof__(*(p)))((__retx & __mask) >> __s);		\