locking: remove spin_lock_prefetch

The only remaining consumer is new_inode, where it showed up in 2001 as
commit c37fa164f793 ("v2.4.9.9 -> v2.4.9.10") in a historical repo [1]
with a changelog which does not mention it.

Since then the line got only touched up to keep compiling.

While it may have been of benefit back in the day, it is guaranteed to
at best not get in the way in the multicore setting -- as the code
performs *a lot* of work between the prefetch and actual lock acquire,
any contention means the cacheline is already invalid by the time the
routine calls spin_lock().  It adds spurious traffic, for short.

On top of it prefetch is notoriously tricky to use for single-threaded
purposes, making it questionable from the get go.

As such, remove it.

I admit upfront I did not see value in benchmarking this change, but I
can do it if that is deemed appropriate.

Removal from new_inode and of the entire thing are in the same patch as
requested by Linus, so whatever weird looks can be directed at that guy.

Link: https://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git/commit/fs/inode.c?id=c37fa164f793735b32aa3f53154ff1a7659e6442 [1]
Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

9 files changed, 1 insertions, 47 deletions

diff --git a/arch/alpha/include/asm/processor.h b/arch/alpha/include/asm/processor.h
index 714abe494e5fd..55bb1c09fd39d 100644
--- a/arch/alpha/include/asm/processor.h
+++ b/arch/alpha/include/asm/processor.h
@@ -47,12 +47,6 @@ unsigned long __get_wchan(struct task_struct *p);
 
 #define ARCH_HAS_PREFETCH
 #define ARCH_HAS_PREFETCHW
-#define ARCH_HAS_SPINLOCK_PREFETCH
-
-#ifndef CONFIG_SMP
-/* Nothing to prefetch. */
-#define spin_lock_prefetch(lock)  	do { } while (0)
-#endif
 
 extern inline void prefetch(const void *ptr)  
 { 
@@ -64,11 +58,4 @@ extern inline void prefetchw(const void *ptr)
 	__builtin_prefetch(ptr, 1, 3);
 }
 
-#ifdef CONFIG_SMP
-extern inline void spin_lock_prefetch(const void *ptr)  
-{
-	__builtin_prefetch(ptr, 1, 3);
-}
-#endif
-
 #endif /* __ASM_ALPHA_PROCESSOR_H */
diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
index 3918f2a679707..e5bc54522e711 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h
@@ -359,14 +359,6 @@ static inline void prefetchw(const void *ptr)
 	asm volatile("prfm pstl1keep, %a0\n" : : "p" (ptr));
 }
 
-#define ARCH_HAS_SPINLOCK_PREFETCH
-static inline void spin_lock_prefetch(const void *ptr)
-{
-	asm volatile(ARM64_LSE_ATOMIC_INSN(
-		     "prfm pstl1strm, %a0",
-		     "nop") : : "p" (ptr));
-}
-
 extern unsigned long __ro_after_init signal_minsigstksz; /* sigframe size */
 extern void __init minsigstksz_setup(void);
 
diff --git a/arch/ia64/include/asm/processor.h b/arch/ia64/include/asm/processor.h
index d1978e0040548..47e3801b526a1 100644
--- a/arch/ia64/include/asm/processor.h
+++ b/arch/ia64/include/asm/processor.h
@@ -634,7 +634,6 @@ ia64_imva (void *addr)
 
 #define ARCH_HAS_PREFETCH
 #define ARCH_HAS_PREFETCHW
-#define ARCH_HAS_SPINLOCK_PREFETCH
 #define PREFETCH_STRIDE			L1_CACHE_BYTES
 
 static inline void
@@ -649,8 +648,6 @@ prefetchw (const void *x)
 	ia64_lfetch_excl(ia64_lfhint_none, x);
 }
 
-#define spin_lock_prefetch(x)	prefetchw(x)
-
 extern unsigned long boot_option_idle_override;
 
 enum idle_boot_override {IDLE_NO_OVERRIDE=0, IDLE_HALT, IDLE_FORCE_MWAIT,
diff --git a/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h b/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h
index 9151dcd9d0d5c..af9cea21c8535 100644
--- a/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h
+++ b/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h
@@ -58,8 +58,6 @@
 
 #define cpu_has_rixi		(cpu_data[0].cputype != CPU_CAVIUM_OCTEON)
 
-#define ARCH_HAS_SPINLOCK_PREFETCH 1
-#define spin_lock_prefetch(x) prefetch(x)
 #define PREFETCH_STRIDE 128
 
 #ifdef __OCTEON__
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h
index 8a6754ffdc7ea..a6c7069bec5d4 100644
--- a/arch/powerpc/include/asm/processor.h
+++ b/arch/powerpc/include/asm/processor.h
@@ -393,7 +393,6 @@ int validate_sp_size(unsigned long sp, struct task_struct *p,
  */
 #define ARCH_HAS_PREFETCH
 #define ARCH_HAS_PREFETCHW
-#define ARCH_HAS_SPINLOCK_PREFETCH
 
 static inline void prefetch(const void *x)
 {
@@ -411,8 +410,6 @@ static inline void prefetchw(const void *x)
 	__asm__ __volatile__ ("dcbtst 0,%0" : : "r" (x));
 }
 
-#define spin_lock_prefetch(x)	prefetchw(x)
-
 /* asm stubs */
 extern unsigned long isa300_idle_stop_noloss(unsigned long psscr_val);
 extern unsigned long isa300_idle_stop_mayloss(unsigned long psscr_val);
diff --git a/arch/sparc/include/asm/processor_64.h b/arch/sparc/include/asm/processor_64.h
index 2667f35d5ea56..0a0d5c3d184c7 100644
--- a/arch/sparc/include/asm/processor_64.h
+++ b/arch/sparc/include/asm/processor_64.h
@@ -213,7 +213,6 @@ unsigned long __get_wchan(struct task_struct *task);
  */
 #define ARCH_HAS_PREFETCH
 #define ARCH_HAS_PREFETCHW
-#define ARCH_HAS_SPINLOCK_PREFETCH
 
 static inline void prefetch(const void *x)
 {
@@ -239,8 +238,6 @@ static inline void prefetchw(const void *x)
 			     : "r" (x));
 }
 
-#define spin_lock_prefetch(x)	prefetchw(x)
-
 #define HAVE_ARCH_PICK_MMAP_LAYOUT
 
 int do_mathemu(struct pt_regs *regs, struct fpustate *f, bool illegal_insn_trap);
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index 4ae2773b873da..fd750247ca891 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -586,7 +586,6 @@ extern char			ignore_fpu_irq;
 
 #define HAVE_ARCH_PICK_MMAP_LAYOUT 1
 #define ARCH_HAS_PREFETCHW
-#define ARCH_HAS_SPINLOCK_PREFETCH
 
 #ifdef CONFIG_X86_32
 # define BASE_PREFETCH		""
@@ -620,11 +619,6 @@ static __always_inline void prefetchw(const void *x)
 			  "m" (*(const char *)x));
 }
 
-static inline void spin_lock_prefetch(const void *x)
-{
-	prefetchw(x);
-}
-
 #define TOP_OF_INIT_STACK ((unsigned long)&init_stack + sizeof(init_stack) - \
 			   TOP_OF_KERNEL_STACK_PADDING)
 
diff --git a/fs/inode.c b/fs/inode.c
index 8fefb69e1f84a..67611a360031b 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -16,7 +16,6 @@
 #include <linux/fsnotify.h>
 #include <linux/mount.h>
 #include <linux/posix_acl.h>
-#include <linux/prefetch.h>
 #include <linux/buffer_head.h> /* for inode_has_buffers */
 #include <linux/ratelimit.h>
 #include <linux/list_lru.h>
@@ -1041,8 +1040,6 @@ struct inode *new_inode(struct super_block *sb)
 {
 	struct inode *inode;
 
-	spin_lock_prefetch(&sb->s_inode_list_lock);
-
 	inode = new_inode_pseudo(sb);
 	if (inode)
 		inode_sb_list_add(inode);
diff --git a/include/linux/prefetch.h b/include/linux/prefetch.h
index b83a3f944f287..b068e2e609393 100644
--- a/include/linux/prefetch.h
+++ b/include/linux/prefetch.h
@@ -25,11 +25,10 @@ struct page;
 	prefetch() should be defined by the architecture, if not, the 
 	#define below provides a no-op define.	
 	
-	There are 3 prefetch() macros:
+	There are 2 prefetch() macros:
 	
 	prefetch(x)  	- prefetches the cacheline at "x" for read
 	prefetchw(x)	- prefetches the cacheline at "x" for write
-	spin_lock_prefetch(x) - prefetches the spinlock *x for taking
 	
 	there is also PREFETCH_STRIDE which is the architecure-preferred 
 	"lookahead" size for prefetching streamed operations.
@@ -44,10 +43,6 @@ struct page;
 #define prefetchw(x) __builtin_prefetch(x,1)
 #endif
 
-#ifndef ARCH_HAS_SPINLOCK_PREFETCH
-#define spin_lock_prefetch(x) prefetchw(x)
-#endif
-
 #ifndef PREFETCH_STRIDE
 #define PREFETCH_STRIDE (4*L1_CACHE_BYTES)
 #endif