[PATCH] spinlock consolidation

This patch (written by me and also containing many suggestions of Arjan van
de Ven) does a major cleanup of the spinlock code.  It does the following
things:

 - consolidates and enhances the spinlock/rwlock debugging code

 - simplifies the asm/spinlock.h files

 - encapsulates the raw spinlock type and moves generic spinlock
   features (such as ->break_lock) into the generic code.

 - cleans up the spinlock code hierarchy to get rid of the spaghetti.

Most notably there's now only a single variant of the debugging code,
located in lib/spinlock_debug.c.  (previously we had one SMP debugging
variant per architecture, plus a separate generic one for UP builds)

Also, i've enhanced the rwlock debugging facility, it will now track
write-owners.  There is new spinlock-owner/CPU-tracking on SMP builds too.
All locks have lockup detection now, which will work for both soft and hard
spin/rwlock lockups.

The arch-level include files now only contain the minimally necessary
subset of the spinlock code - all the rest that can be generalized now
lives in the generic headers:

 include/asm-i386/spinlock_types.h       |   16
 include/asm-x86_64/spinlock_types.h     |   16

I have also split up the various spinlock variants into separate files,
making it easier to see which does what. The new layout is:

   SMP                         |  UP
   ----------------------------|-----------------------------------
   asm/spinlock_types_smp.h    |  linux/spinlock_types_up.h
   linux/spinlock_types.h      |  linux/spinlock_types.h
   asm/spinlock_smp.h          |  linux/spinlock_up.h
   linux/spinlock_api_smp.h    |  linux/spinlock_api_up.h
   linux/spinlock.h            |  linux/spinlock.h

/*
 * here's the role of the various spinlock/rwlock related include files:
 *
 * on SMP builds:
 *
 *  asm/spinlock_types.h: contains the raw_spinlock_t/raw_rwlock_t and the
 *                        initializers
 *
 *  linux/spinlock_types.h:
 *                        defines the generic type and initializers
 *
 *  asm/spinlock.h:       contains the __raw_spin_*()/etc. lowlevel
 *                        implementations, mostly inline assembly code
 *
 *   (also included on UP-debug builds:)
 *
 *  linux/spinlock_api_smp.h:
 *                        contains the prototypes for the _spin_*() APIs.
 *
 *  linux/spinlock.h:     builds the final spin_*() APIs.
 *
 * on UP builds:
 *
 *  linux/spinlock_type_up.h:
 *                        contains the generic, simplified UP spinlock type.
 *                        (which is an empty structure on non-debug builds)
 *
 *  linux/spinlock_types.h:
 *                        defines the generic type and initializers
 *
 *  linux/spinlock_up.h:
 *                        contains the __raw_spin_*()/etc. version of UP
 *                        builds. (which are NOPs on non-debug, non-preempt
 *                        builds)
 *
 *   (included on UP-non-debug builds:)
 *
 *  linux/spinlock_api_up.h:
 *                        builds the _spin_*() APIs.
 *
 *  linux/spinlock.h:     builds the final spin_*() APIs.
 */

All SMP and UP architectures are converted by this patch.

arm, i386, ia64, ppc, ppc64, s390/s390x, x64 was build-tested via
crosscompilers.  m32r, mips, sh, sparc, have not been tested yet, but should
be mostly fine.

From: Grant Grundler <grundler@parisc-linux.org>

  Booted and lightly tested on a500-44 (64-bit, SMP kernel, dual CPU).
  Builds 32-bit SMP kernel (not booted or tested).  I did not try to build
  non-SMP kernels.  That should be trivial to fix up later if necessary.

  I converted bit ops atomic_hash lock to raw_spinlock_t.  Doing so avoids
  some ugly nesting of linux/*.h and asm/*.h files.  Those particular locks
  are well tested and contained entirely inside arch specific code.  I do NOT
  expect any new issues to arise with them.

 If someone does ever need to use debug/metrics with them, then they will
  need to unravel this hairball between spinlocks, atomic ops, and bit ops
  that exist only because parisc has exactly one atomic instruction: LDCW
  (load and clear word).

From: "Luck, Tony" <tony.luck@intel.com>

   ia64 fix

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjanv@infradead.org>
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Cc: Matthew Wilcox <willy@debian.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Mikael Pettersson <mikpe@csd.uu.se>
Signed-off-by: Benoit Boissinot <benoit.boissinot@ens-lyon.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

4 files changed, 0 insertions, 377 deletions

diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c
index 66255434128a..7d10b0397091 100644
--- a/arch/sparc64/kernel/process.c
+++ b/arch/sparc64/kernel/process.c
@@ -607,11 +607,6 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
 	struct thread_info *t = p->thread_info;
 	char *child_trap_frame;
 
-#ifdef CONFIG_DEBUG_SPINLOCK
-	p->thread.smp_lock_count = 0;
-	p->thread.smp_lock_pc = 0;
-#endif
-
 	/* Calculate offset to stack_frame & pt_regs */
 	child_trap_frame = ((char *)t) + (THREAD_SIZE - (TRACEREG_SZ+STACKFRAME_SZ));
 	memcpy(child_trap_frame, (((struct sparc_stackf *)regs)-1), (TRACEREG_SZ+STACKFRAME_SZ));
diff --git a/arch/sparc64/kernel/sparc64_ksyms.c b/arch/sparc64/kernel/sparc64_ksyms.c
index 7d9a0f6c437d..cbb5e59824e5 100644
--- a/arch/sparc64/kernel/sparc64_ksyms.c
+++ b/arch/sparc64/kernel/sparc64_ksyms.c
@@ -115,17 +115,12 @@ EXPORT_PER_CPU_SYMBOL(__cpu_data);
 
 /* used by various drivers */
 #ifdef CONFIG_SMP
-#ifndef CONFIG_DEBUG_SPINLOCK
 /* Out of line rw-locking implementation. */
 EXPORT_SYMBOL(__read_lock);
 EXPORT_SYMBOL(__read_unlock);
 EXPORT_SYMBOL(__write_lock);
 EXPORT_SYMBOL(__write_unlock);
 EXPORT_SYMBOL(__write_trylock);
-/* Out of line spin-locking implementation. */
-EXPORT_SYMBOL(_raw_spin_lock);
-EXPORT_SYMBOL(_raw_spin_lock_flags);
-#endif
 
 /* Hard IRQ locking */
 EXPORT_SYMBOL(synchronize_irq);
diff --git a/arch/sparc64/lib/Makefile b/arch/sparc64/lib/Makefile
index 40dbeec7e5d6..d968aebe83b2 100644
--- a/arch/sparc64/lib/Makefile
+++ b/arch/sparc64/lib/Makefile
@@ -14,7 +14,6 @@ lib-y := PeeCeeI.o copy_page.o clear_page.o strlen.o strncmp.o \
 	 copy_in_user.o user_fixup.o memmove.o \
 	 mcount.o ipcsum.o rwsem.o xor.o find_bit.o delay.o
 
-lib-$(CONFIG_DEBUG_SPINLOCK) += debuglocks.o
 lib-$(CONFIG_HAVE_DEC_LOCK) += dec_and_lock.o
 
 obj-y += iomap.o
diff --git a/arch/sparc64/lib/debuglocks.c b/arch/sparc64/lib/debuglocks.c
deleted file mode 100644
index f5f0b5586f01..000000000000
--- a/arch/sparc64/lib/debuglocks.c
+++ /dev/null
@@ -1,366 +0,0 @@
-/* $Id: debuglocks.c,v 1.9 2001/11/17 00:10:48 davem Exp $
- * debuglocks.c: Debugging versions of SMP locking primitives.
- *
- * Copyright (C) 1998 David S. Miller (davem@redhat.com)
- */
-
-#include <linux/config.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/spinlock.h>
-#include <asm/system.h>
-
-#ifdef CONFIG_SMP
-
-static inline void show (char *str, spinlock_t *lock, unsigned long caller)
-{
-	int cpu = smp_processor_id();
-
-	printk("%s(%p) CPU#%d stuck at %08x, owner PC(%08x):CPU(%x)\n",
-	       str, lock, cpu, (unsigned int) caller,
-	       lock->owner_pc, lock->owner_cpu);
-}
-
-static inline void show_read (char *str, rwlock_t *lock, unsigned long caller)
-{
-	int cpu = smp_processor_id();
-
-	printk("%s(%p) CPU#%d stuck at %08x, writer PC(%08x):CPU(%x)\n",
-	       str, lock, cpu, (unsigned int) caller,
-	       lock->writer_pc, lock->writer_cpu);
-}
-
-static inline void show_write (char *str, rwlock_t *lock, unsigned long caller)
-{
-	int cpu = smp_processor_id();
-	int i;
-
-	printk("%s(%p) CPU#%d stuck at %08x\n",
-	       str, lock, cpu, (unsigned int) caller);
-	printk("Writer: PC(%08x):CPU(%x)\n",
-	       lock->writer_pc, lock->writer_cpu);
-	printk("Readers:");
-	for (i = 0; i < NR_CPUS; i++)
-		if (lock->reader_pc[i])
-			printk(" %d[%08x]", i, lock->reader_pc[i]);
-	printk("\n");
-}
-
-#undef INIT_STUCK
-#define INIT_STUCK 100000000
-
-void _do_spin_lock(spinlock_t *lock, char *str, unsigned long caller)
-{
-	unsigned long val;
-	int stuck = INIT_STUCK;
-	int cpu = get_cpu();
-	int shown = 0;
-
-again:
-	__asm__ __volatile__("ldstub [%1], %0"
-			     : "=r" (val)
-			     : "r" (&(lock->lock))
-			     : "memory");
-	membar_storeload_storestore();
-	if (val) {
-		while (lock->lock) {
-			if (!--stuck) {
-				if (shown++ <= 2)
-					show(str, lock, caller);
-				stuck = INIT_STUCK;
-			}
-			rmb();
-		}
-		goto again;
-	}
-	lock->owner_pc = ((unsigned int)caller);
-	lock->owner_cpu = cpu;
-	current->thread.smp_lock_count++;
-	current->thread.smp_lock_pc = ((unsigned int)caller);
-
-	put_cpu();
-}
-
-int _do_spin_trylock(spinlock_t *lock, unsigned long caller)
-{
-	unsigned long val;
-	int cpu = get_cpu();
-
-	__asm__ __volatile__("ldstub [%1], %0"
-			     : "=r" (val)
-			     : "r" (&(lock->lock))
-			     : "memory");
-	membar_storeload_storestore();
-	if (!val) {
-		lock->owner_pc = ((unsigned int)caller);
-		lock->owner_cpu = cpu;
-		current->thread.smp_lock_count++;
-		current->thread.smp_lock_pc = ((unsigned int)caller);
-	}
-
-	put_cpu();
-
-	return val == 0;
-}
-
-void _do_spin_unlock(spinlock_t *lock)
-{
-	lock->owner_pc = 0;
-	lock->owner_cpu = NO_PROC_ID;
-	membar_storestore_loadstore();
-	lock->lock = 0;
-	current->thread.smp_lock_count--;
-}
-
-/* Keep INIT_STUCK the same... */
-
-void _do_read_lock(rwlock_t *rw, char *str, unsigned long caller)
-{
-	unsigned long val;
-	int stuck = INIT_STUCK;
-	int cpu = get_cpu();
-	int shown = 0;
-
-wlock_again:
-	/* Wait for any writer to go away.  */
-	while (((long)(rw->lock)) < 0) {
-		if (!--stuck) {
-			if (shown++ <= 2)
-				show_read(str, rw, caller);
-			stuck = INIT_STUCK;
-		}
-		rmb();
-	}
-	/* Try once to increment the counter.  */
-	__asm__ __volatile__(
-"	ldx		[%0], %%g1\n"
-"	brlz,a,pn	%%g1, 2f\n"
-"	 mov		1, %0\n"
-"	add		%%g1, 1, %%g7\n"
-"	casx		[%0], %%g1, %%g7\n"
-"	sub		%%g1, %%g7, %0\n"
-"2:"	: "=r" (val)
-	: "0" (&(rw->lock))
-	: "g1", "g7", "memory");
-	membar_storeload_storestore();
-	if (val)
-		goto wlock_again;
-	rw->reader_pc[cpu] = ((unsigned int)caller);
-	current->thread.smp_lock_count++;
-	current->thread.smp_lock_pc = ((unsigned int)caller);
-
-	put_cpu();
-}
-
-void _do_read_unlock(rwlock_t *rw, char *str, unsigned long caller)
-{
-	unsigned long val;
-	int stuck = INIT_STUCK;
-	int cpu = get_cpu();
-	int shown = 0;
-
-	/* Drop our identity _first_. */
-	rw->reader_pc[cpu] = 0;
-	current->thread.smp_lock_count--;
-runlock_again:
-	/* Spin trying to decrement the counter using casx.  */
-	__asm__ __volatile__(
-"	membar	#StoreLoad | #LoadLoad\n"
-"	ldx	[%0], %%g1\n"
-"	sub	%%g1, 1, %%g7\n"
-"	casx	[%0], %%g1, %%g7\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	sub	%%g1, %%g7, %0\n"
-	: "=r" (val)
-	: "0" (&(rw->lock))
-	: "g1", "g7", "memory");
-	if (val) {
-		if (!--stuck) {
-			if (shown++ <= 2)
-				show_read(str, rw, caller);
-			stuck = INIT_STUCK;
-		}
-		goto runlock_again;
-	}
-
-	put_cpu();
-}
-
-void _do_write_lock(rwlock_t *rw, char *str, unsigned long caller)
-{
-	unsigned long val;
-	int stuck = INIT_STUCK;
-	int cpu = get_cpu();
-	int shown = 0;
-
-wlock_again:
-	/* Spin while there is another writer. */
-	while (((long)rw->lock) < 0) {
-		if (!--stuck) {
-			if (shown++ <= 2)
-				show_write(str, rw, caller);
-			stuck = INIT_STUCK;
-		}
-		rmb();
-	}
-
-	/* Try to acuire the write bit.  */
-	__asm__ __volatile__(
-"	mov	1, %%g3\n"
-"	sllx	%%g3, 63, %%g3\n"
-"	ldx	[%0], %%g1\n"
-"	brlz,pn	%%g1, 1f\n"
-"	 or	%%g1, %%g3, %%g7\n"
-"	casx	[%0], %%g1, %%g7\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	ba,pt	%%xcc, 2f\n"
-"	 sub	%%g1, %%g7, %0\n"
-"1:	mov	1, %0\n"
-"2:"	: "=r" (val)
-	: "0" (&(rw->lock))
-	: "g3", "g1", "g7", "memory");
-	if (val) {
-		/* We couldn't get the write bit. */
-		if (!--stuck) {
-			if (shown++ <= 2)
-				show_write(str, rw, caller);
-			stuck = INIT_STUCK;
-		}
-		goto wlock_again;
-	}
-	if ((rw->lock & ((1UL<<63)-1UL)) != 0UL) {
-		/* Readers still around, drop the write
-		 * lock, spin, and try again.
-		 */
-		if (!--stuck) {
-			if (shown++ <= 2)
-				show_write(str, rw, caller);
-			stuck = INIT_STUCK;
-		}
-		__asm__ __volatile__(
-"		mov	1, %%g3\n"
-"		sllx	%%g3, 63, %%g3\n"
-"1:		ldx	[%0], %%g1\n"
-"		andn	%%g1, %%g3, %%g7\n"
-"		casx	[%0], %%g1, %%g7\n"
-"		cmp	%%g1, %%g7\n"
-"		membar	#StoreLoad | #StoreStore\n"
-"		bne,pn	%%xcc, 1b\n"
-"		 nop"
-		: /* no outputs */
-		: "r" (&(rw->lock))
-		: "g3", "g1", "g7", "cc", "memory");
-		while(rw->lock != 0) {
-			if (!--stuck) {
-				if (shown++ <= 2)
-					show_write(str, rw, caller);
-				stuck = INIT_STUCK;
-			}
-			rmb();
-		}
-		goto wlock_again;
-	}
-
-	/* We have it, say who we are. */
-	rw->writer_pc = ((unsigned int)caller);
-	rw->writer_cpu = cpu;
-	current->thread.smp_lock_count++;
-	current->thread.smp_lock_pc = ((unsigned int)caller);
-
-	put_cpu();
-}
-
-void _do_write_unlock(rwlock_t *rw, unsigned long caller)
-{
-	unsigned long val;
-	int stuck = INIT_STUCK;
-	int shown = 0;
-
-	/* Drop our identity _first_ */
-	rw->writer_pc = 0;
-	rw->writer_cpu = NO_PROC_ID;
-	current->thread.smp_lock_count--;
-wlock_again:
-	__asm__ __volatile__(
-"	membar	#StoreLoad | #LoadLoad\n"
-"	mov	1, %%g3\n"
-"	sllx	%%g3, 63, %%g3\n"
-"	ldx	[%0], %%g1\n"
-"	andn	%%g1, %%g3, %%g7\n"
-"	casx	[%0], %%g1, %%g7\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	sub	%%g1, %%g7, %0\n"
-	: "=r" (val)
-	: "0" (&(rw->lock))
-	: "g3", "g1", "g7", "memory");
-	if (val) {
-		if (!--stuck) {
-			if (shown++ <= 2)
-				show_write("write_unlock", rw, caller);
-			stuck = INIT_STUCK;
-		}
-		goto wlock_again;
-	}
-}
-
-int _do_write_trylock(rwlock_t *rw, char *str, unsigned long caller)
-{
-	unsigned long val;
-	int cpu = get_cpu();
-
-	/* Try to acuire the write bit.  */
-	__asm__ __volatile__(
-"	mov	1, %%g3\n"
-"	sllx	%%g3, 63, %%g3\n"
-"	ldx	[%0], %%g1\n"
-"	brlz,pn	%%g1, 1f\n"
-"	 or	%%g1, %%g3, %%g7\n"
-"	casx	[%0], %%g1, %%g7\n"
-"	membar	#StoreLoad | #StoreStore\n"
-"	ba,pt	%%xcc, 2f\n"
-"	 sub	%%g1, %%g7, %0\n"
-"1:	mov	1, %0\n"
-"2:"	: "=r" (val)
-	: "0" (&(rw->lock))
-	: "g3", "g1", "g7", "memory");
-
-	if (val) {
-		put_cpu();
-		return 0;
-	}
-
-	if ((rw->lock & ((1UL<<63)-1UL)) != 0UL) {
-		/* Readers still around, drop the write
-		 * lock, return failure.
-		 */
-		__asm__ __volatile__(
-"		mov	1, %%g3\n"
-"		sllx	%%g3, 63, %%g3\n"
-"1:		ldx	[%0], %%g1\n"
-"		andn	%%g1, %%g3, %%g7\n"
-"		casx	[%0], %%g1, %%g7\n"
-"		cmp	%%g1, %%g7\n"
-"		membar	#StoreLoad | #StoreStore\n"
-"		bne,pn	%%xcc, 1b\n"
-"		 nop"
-		: /* no outputs */
-		: "r" (&(rw->lock))
-		: "g3", "g1", "g7", "cc", "memory");
-
-		put_cpu();
-
-		return 0;
-	}
-
-	/* We have it, say who we are. */
-	rw->writer_pc = ((unsigned int)caller);
-	rw->writer_cpu = cpu;
-	current->thread.smp_lock_count++;
-	current->thread.smp_lock_pc = ((unsigned int)caller);
-
-	put_cpu();
-
-	return 1;
-}
-
-#endif /* CONFIG_SMP */