fixes for 32 bit builds/non x86 archs

12 files changed, 441 insertions, 211 deletions

diff --git a/include/asm/unaligned.h b/include/asm/unaligned.h
index ced1a290..e695bede 100644
--- a/include/asm/unaligned.h
+++ b/include/asm/unaligned.h
@@ -1,13 +1,13 @@
 #ifndef _ASM_UNALIGNED_H
 #define _ASM_UNALIGNED_H
 
-#if defined(__LITTLE_ENDIAN)
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
 # include <linux/unaligned/le_struct.h>
 # include <linux/unaligned/be_byteshift.h>
 # include <linux/unaligned/generic.h>
 # define get_unaligned	__get_unaligned_le
 # define put_unaligned	__put_unaligned_le
-#elif defined(__BIG_ENDIAN)
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
 # include <linux/unaligned/be_struct.h>
 # include <linux/unaligned/le_byteshift.h>
 # include <linux/unaligned/generic.h>
diff --git a/include/linux/atomic.h b/include/linux/atomic.h
index ad33ad3b..7471bd97 100644
--- a/include/linux/atomic.h
+++ b/include/linux/atomic.h
@@ -4,11 +4,58 @@
 #include <linux/compiler.h>
 #include <linux/types.h>
 
-#define xchg(p, v)						\
-	__atomic_exchange_n(p, v, __ATOMIC_SEQ_CST)
+typedef struct {
+	int		counter;
+} atomic_t;
+
+typedef struct {
+	long		counter;
+} atomic_long_t;
+
+typedef struct {
+	u64		counter;
+} atomic64_t;
+
+#ifndef C11_ATOMICS
+
+#include <urcu/uatomic.h>
+
+#if (CAA_BITS_PER_LONG != 64)
+#define ATOMIC64_SPINLOCK
+#endif
+
+#define __ATOMIC_READ(p)		uatomic_read(p)
+#define __ATOMIC_SET(p, v)		uatomic_set(p, v)
+#define __ATOMIC_ADD_RETURN(v, p)	uatomic_add_return(p, v)
+#define __ATOMIC_SUB_RETURN(v, p)	uatomic_sub_return(p, v)
+#define __ATOMIC_ADD(v, p)		uatomic_add(p, v)
+#define __ATOMIC_SUB(v, p)		uatomic_sub(p, v)
+#define __ATOMIC_INC(p)			uatomic_inc(p)
+#define __ATOMIC_DEC(p)			uatomic_dec(p)
+
+#define xchg(p, v)			uatomic_xchg(p, v)
+#define xchg_acquire(p, v)		uatomic_xchg(p, v)
+#define cmpxchg(p, old, new)		uatomic_cmpxchg(p, old, new)
+#define cmpxchg_acquire(p, old, new)	uatomic_cmpxchg(p, old, new)
 
-#define xchg_acquire(p, v)					\
-	__atomic_exchange_n(p, v, __ATOMIC_ACQUIRE)
+#define smp_mb__before_atomic()		cmm_smp_mb__before_uatomic_add()
+#define smp_mb__after_atomic()		cmm_smp_mb__after_uatomic_add()
+#define smp_wmb()			cmm_smp_wmb()
+#define smp_rmb()			cmm_smp_rmb()
+#define smp_mb()			cmm_smp_mb()
+#define smp_read_barrier_depends()	cmm_smp_read_barrier_depends()
+
+#else /* C11_ATOMICS */
+
+#define __ATOMIC_READ(p)		__atomic_load_n(p,	__ATOMIC_RELAXED)
+#define __ATOMIC_SET(p, v)		__atomic_store_n(p, v,	__ATOMIC_RELAXED)
+#define __ATOMIC_ADD_RETURN(v, p)	__atomic_add_fetch(p, v, __ATOMIC_RELAXED)
+#define __ATOMIC_ADD_RETURN_RELEASE(v, p)				\
+					__atomic_add_fetch(p, v, __ATOMIC_RELEASE)
+#define __ATOMIC_SUB_RETURN(v, p)	__atomic_sub_fetch(p, v, __ATOMIC_RELAXED)
+
+#define xchg(p, v)			__atomic_exchange_n(p, v, __ATOMIC_SEQ_CST)
+#define xchg_acquire(p, v)		__atomic_exchange_n(p, v, __ATOMIC_ACQUIRE)
 
 #define cmpxchg(p, old, new)					\
 ({								\
@@ -37,6 +84,8 @@
 #define smp_mb()		__atomic_thread_fence(__ATOMIC_SEQ_CST)
 #define smp_read_barrier_depends()
 
+#endif
+
 #define smp_store_mb(var, value)  do { WRITE_ONCE(var, value); smp_mb(); } while (0)
 
 #define smp_load_acquire(p)						\
@@ -52,199 +101,179 @@ do {									\
 	WRITE_ONCE(*p, v);						\
 } while (0)
 
-typedef struct {
-	int		counter;
-} atomic_t;
-
-static inline int atomic_read(const atomic_t *v)
-{
-	return __atomic_load_n(&v->counter, __ATOMIC_RELAXED);
-}
-
-static inline void atomic_set(atomic_t *v, int i)
-{
-	__atomic_store_n(&v->counter, i, __ATOMIC_RELAXED);
+/* atomic interface: */
+
+#ifndef __ATOMIC_ADD
+#define __ATOMIC_ADD(i, v) __ATOMIC_ADD_RETURN(i, v)
+#endif
+
+#ifndef __ATOMIC_ADD_RETURN_RELEASE
+#define __ATOMIC_ADD_RETURN_RELEASE(i, v)				\
+	({ smp_mb__before_atomic(); __ATOMIC_ADD_RETURN(i, v); })
+#endif
+
+#ifndef __ATOMIC_SUB
+#define __ATOMIC_SUB(i, v) __ATOMIC_SUB_RETURN(i, v)
+#endif
+
+#ifndef __ATOMIC_INC_RETURN
+#define __ATOMIC_INC_RETURN(v) __ATOMIC_ADD_RETURN(1, v)
+#endif
+
+#ifndef __ATOMIC_DEC_RETURN
+#define __ATOMIC_DEC_RETURN(v) __ATOMIC_SUB_RETURN(1, v)
+#endif
+
+#ifndef __ATOMIC_INC
+#define __ATOMIC_INC(v) __ATOMIC_ADD(1, v)
+#endif
+
+#ifndef __ATOMIC_DEC
+#define __ATOMIC_DEC(v) __ATOMIC_SUB(1, v)
+#endif
+
+#define DEF_ATOMIC_OPS(a_type, i_type)					\
+static inline i_type a_type##_read(const a_type##_t *v)			\
+{									\
+	return __ATOMIC_READ(&v->counter);				\
+}									\
+									\
+static inline void a_type##_set(a_type##_t *v, i_type i)		\
+{									\
+	return __ATOMIC_SET(&v->counter, i);				\
+}									\
+									\
+static inline i_type a_type##_add_return(i_type i, a_type##_t *v)	\
+{									\
+	return __ATOMIC_ADD_RETURN(i, &v->counter);			\
+}									\
+									\
+static inline i_type a_type##_add_return_release(i_type i, a_type##_t *v)\
+{									\
+	return __ATOMIC_ADD_RETURN_RELEASE(i, &v->counter);		\
+}									\
+									\
+static inline i_type a_type##_sub_return(i_type i, a_type##_t *v)	\
+{									\
+	return __ATOMIC_SUB_RETURN(i, &v->counter);			\
+}									\
+									\
+static inline void a_type##_add(i_type i, a_type##_t *v)		\
+{									\
+	__ATOMIC_ADD(i, &v->counter);					\
+}									\
+									\
+static inline void a_type##_sub(i_type i, a_type##_t *v)		\
+{									\
+	__ATOMIC_SUB(i, &v->counter);					\
+}									\
+									\
+static inline i_type a_type##_inc_return(a_type##_t *v)			\
+{									\
+	return __ATOMIC_INC_RETURN(&v->counter);			\
+}									\
+									\
+static inline i_type a_type##_dec_return(a_type##_t *v)			\
+{									\
+	return __ATOMIC_DEC_RETURN(&v->counter);			\
+}									\
+									\
+static inline void a_type##_inc(a_type##_t *v)				\
+{									\
+	__ATOMIC_INC(&v->counter);					\
+}									\
+									\
+static inline void a_type##_dec(a_type##_t *v)				\
+{									\
+	__ATOMIC_DEC(&v->counter);					\
+}									\
+									\
+static inline bool a_type##_add_negative(i_type i, a_type##_t *v)	\
+{									\
+	return __ATOMIC_ADD_RETURN(i, &v->counter) < 0;			\
+}									\
+									\
+static inline bool a_type##_sub_and_test(i_type i, a_type##_t *v)	\
+{									\
+	return __ATOMIC_SUB_RETURN(i, &v->counter) == 0;		\
+}									\
+									\
+static inline bool a_type##_inc_and_test(a_type##_t *v)			\
+{									\
+	return __ATOMIC_INC_RETURN(&v->counter) == 0;			\
+}									\
+									\
+static inline bool a_type##_dec_and_test(a_type##_t *v)			\
+{									\
+	return __ATOMIC_DEC_RETURN(&v->counter) == 0;			\
+}									\
+									\
+static inline i_type a_type##_add_unless(a_type##_t *v, i_type a, i_type u)\
+{									\
+	i_type old, c = __ATOMIC_READ(&v->counter);			\
+	while (c != u && (old = cmpxchg(&v->counter, c, c + a)) != c)	\
+		c = old;						\
+	return c;							\
+}									\
+									\
+static inline bool a_type##_inc_not_zero(a_type##_t *v)			\
+{									\
+	return a_type##_add_unless(v, 1, 0);				\
+}									\
+									\
+static inline i_type a_type##_xchg(a_type##_t *v, i_type i)		\
+{									\
+	return xchg(&v->counter, i);					\
+}									\
+									\
+static inline i_type a_type##_cmpxchg(a_type##_t *v, i_type old, i_type new)\
+{									\
+	return cmpxchg(&v->counter, old, new);				\
+}									\
+									\
+static inline i_type a_type##_cmpxchg_acquire(a_type##_t *v, i_type old, i_type new)\
+{									\
+	return cmpxchg_acquire(&v->counter, old, new);			\
 }
 
-static inline int atomic_add_return(int i, atomic_t *v)
-{
-	return __atomic_add_fetch(&v->counter, i, __ATOMIC_RELAXED);
-}
+DEF_ATOMIC_OPS(atomic,		int)
+DEF_ATOMIC_OPS(atomic_long,	long)
+
+#ifndef ATOMIC64_SPINLOCK
+DEF_ATOMIC_OPS(atomic64,	s64)
+#else
+s64 atomic64_read(const atomic64_t *v);
+void atomic64_set(atomic64_t *v, s64);
+
+s64 atomic64_add_return(s64, atomic64_t *);
+s64 atomic64_sub_return(s64, atomic64_t *);
+void atomic64_add(s64, atomic64_t *);
+void atomic64_sub(s64, atomic64_t *);
+
+s64 atomic64_xchg(atomic64_t *, s64);
+s64 atomic64_cmpxchg(atomic64_t *, s64, s64);
+
+#define atomic64_add_negative(a, v)	(atomic64_add_return((a), (v)) < 0)
+#define atomic64_inc(v)			atomic64_add(1LL, (v))
+#define atomic64_inc_return(v)		atomic64_add_return(1LL, (v))
+#define atomic64_inc_and_test(v)	(atomic64_inc_return(v) == 0)
+#define atomic64_sub_and_test(a, v)	(atomic64_sub_return((a), (v)) == 0)
+#define atomic64_dec(v)			atomic64_sub(1LL, (v))
+#define atomic64_dec_return(v)		atomic64_sub_return(1LL, (v))
+#define atomic64_dec_and_test(v)	(atomic64_dec_return((v)) == 0)
+#define atomic64_inc_not_zero(v)	atomic64_add_unless((v), 1LL, 0LL)
 
-static inline int atomic_sub_return(int i, atomic_t *v)
-{
-	return __atomic_sub_fetch(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline int atomic_add_negative(int i, atomic_t *v)
-{
-	return atomic_add_return(i, v) < 0;
-}
-
-static inline void atomic_add(int i, atomic_t *v)
-{
-	atomic_add_return(i, v);
-}
-
-static inline void atomic_sub(int i, atomic_t *v)
-{
-	atomic_sub_return(i, v);
-}
-
-static inline void atomic_inc(atomic_t *v)
-{
-	atomic_add(1, v);
-}
-
-static inline void atomic_dec(atomic_t *v)
-{
-	atomic_sub(1, v);
-}
-
-#define atomic_dec_return(v)		atomic_sub_return(1, (v))
-#define atomic_inc_return(v)		atomic_add_return(1, (v))
-
-#define atomic_sub_and_test(i, v)	(atomic_sub_return((i), (v)) == 0)
-#define atomic_dec_and_test(v)		(atomic_dec_return(v) == 0)
-#define atomic_inc_and_test(v)		(atomic_inc_return(v) == 0)
-
-#define atomic_xchg(ptr, v)		(xchg(&(ptr)->counter, (v)))
-#define atomic_cmpxchg(v, old, new)	(cmpxchg(&((v)->counter), (old), (new)))
-
-static inline int atomic_add_unless(atomic_t *v, int a, int u)
-{
-	int c, old;
-	c = atomic_read(v);
-	while (c != u && (old = atomic_cmpxchg(v, c, c + a)) != c)
-		c = old;
-	return c;
-}
-
-#define atomic_inc_not_zero(v)		atomic_add_unless((v), 1, 0)
-
-typedef struct {
-	long		counter;
-} atomic_long_t;
-
-static inline long atomic_long_read(const atomic_long_t *v)
-{
-	return __atomic_load_n(&v->counter, __ATOMIC_RELAXED);
-}
-
-static inline void atomic_long_set(atomic_long_t *v, long i)
-{
-	__atomic_store_n(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline long atomic_long_add_return(long i, atomic_long_t *v)
-{
-	return __atomic_add_fetch(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline long atomic_long_sub_return(long i, atomic_long_t *v)
-{
-	return __atomic_sub_fetch(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline void atomic_long_add(long i, atomic_long_t *v)
-{
-	atomic_long_add_return(i, v);
-}
-
-static inline void atomic_long_sub(long i, atomic_long_t *v)
-{
-	atomic_long_sub_return(i, v);
-}
-
-static inline void atomic_long_inc(atomic_long_t *v)
-{
-	atomic_long_add(1, v);
-}
-
-static inline void atomic_long_dec(atomic_long_t *v)
-{
-	atomic_long_sub(1, v);
-}
-
-static inline long atomic_long_cmpxchg(atomic_long_t *v, long old, long new)
-{
-	return cmpxchg(&v->counter, old, new);
-}
-
-static inline bool atomic_long_inc_not_zero(atomic_long_t *i)
-{
-	long old, v = atomic_long_read(i);
-
-	do {
-		if (!(old = v))
-			return false;
-	} while ((v = atomic_long_cmpxchg(i, old, old + 1)) != old);
-
-	return true;
-}
-
-#define atomic_long_sub_and_test(i, v)	(atomic_long_sub_return((i), (v)) == 0)
-
-typedef struct {
-	u64		counter;
-} atomic64_t;
-
-static inline s64 atomic64_read(const atomic64_t *v)
-{
-	return __atomic_load_n(&v->counter, __ATOMIC_RELAXED);
-}
-
-static inline void atomic64_set(atomic64_t *v, s64 i)
-{
-	__atomic_store_n(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline s64 atomic64_add_return(s64 i, atomic64_t *v)
-{
-	return __atomic_add_fetch(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline s64 atomic64_sub_return(s64 i, atomic64_t *v)
-{
-	return __atomic_sub_fetch(&v->counter, i, __ATOMIC_RELAXED);
-}
-
-static inline void atomic64_add(s64 i, atomic64_t *v)
-{
-	atomic64_add_return(i, v);
-}
-
-static inline void atomic64_sub(s64 i, atomic64_t *v)
-{
-	atomic64_sub_return(i, v);
-}
-
-static inline void atomic64_inc(atomic64_t *v)
-{
-	atomic64_add(1, v);
-}
-
-static inline void atomic64_dec(atomic64_t *v)
-{
-	atomic64_sub(1, v);
-}
-
-#define atomic64_dec_return(v)		atomic64_sub_return(1, (v))
-#define atomic64_inc_return(v)		atomic64_add_return(1, (v))
-
-static inline s64 atomic64_cmpxchg(atomic64_t *v, s64 old, s64 new)
+static inline s64 atomic64_add_return_release(s64 i, atomic64_t *v)
 {
-	return cmpxchg(&v->counter, old, new);
+	smp_mb__before_atomic();
+	return atomic64_add_return(i, v);
 }
 
 static inline s64 atomic64_cmpxchg_acquire(atomic64_t *v, s64 old, s64 new)
 {
-	return cmpxchg_acquire(&v->counter, old, new);
+	return atomic64_cmpxchg(v, old, new);
 }
 
-static inline s64 atomic64_add_return_release(s64 i, atomic64_t *v)
-{
-	return __atomic_add_fetch(&v->counter, i, __ATOMIC_RELEASE);
-}
+#endif
 
 #endif /* __TOOLS_LINUX_ATOMIC_H */
diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h
index 01b2c153..d8a86b45 100644
--- a/include/linux/backing-dev.h
+++ b/include/linux/backing-dev.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_BACKING_DEV_H
 #define _LINUX_BACKING_DEV_H
 
+#include <linux/list.h>
+
 typedef int (congested_fn)(void *, int);
 
 enum wb_congested_state {
diff --git a/include/linux/cache.h b/include/linux/cache.h
index 4ee609ae..c61167ca 100644
--- a/include/linux/cache.h
+++ b/include/linux/cache.h
@@ -1,7 +1,8 @@
 #ifndef __TOOLS_LINUX_CACHE_H
 #define __TOOLS_LINUX_CACHE_H
 
-#define L1_CACHE_BYTES		64
+#define L1_CACHE_SHIFT		6
+#define L1_CACHE_BYTES		(1 << L1_CACHE_SHIFT)
 #define SMP_CACHE_BYTES		L1_CACHE_BYTES
 
 #define L1_CACHE_ALIGN(x)	__ALIGN_KERNEL(x, L1_CACHE_BYTES)
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index b569b2c1..15b803ea 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -5,7 +5,7 @@ struct super_block;
 struct inode;
 
 /* The hash is always the low bits of hash_len */
-#ifdef __LITTLE_ENDIAN
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
  #define HASH_LEN_DECLARE u32 hash; u32 len
 #else
  #define HASH_LEN_DECLARE u32 len; u32 hash
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index dc778f1c..b6afea43 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -111,17 +111,11 @@
 #define cpu_relax()		do {} while (0)
 #define cpu_relax_lowlatency()	do {} while (0)
 
-__printf(1, 2)
-static inline void panic(const char *fmt, ...)
-{
-       va_list args;
-
-       va_start(args, fmt);
-       vprintf(fmt, args);
-       va_end(args);
-
-       BUG();
-}
+#define panic(fmt, ...)					\
+do {							\
+	printf(fmt, ##__VA_ARGS__);			\
+	BUG();						\
+} while (0)
 
 unsigned long simple_strtoul(const char *,char **,unsigned int);
 long simple_strtol(const char *,char **,unsigned int);
diff --git a/include/linux/page.h b/include/linux/page.h
index 8d6413ce..87be064f 100644
--- a/include/linux/page.h
+++ b/include/linux/page.h
@@ -5,6 +5,13 @@
 
 struct page;
 
+#ifndef PAGE_SIZE
+
+#define PAGE_SIZE   4096UL
+#define PAGE_MASK   (~(PAGE_SIZE - 1))
+
+#endif
+
 #define virt_to_page(p)							\
 	((struct page *) (((unsigned long) (p)) & PAGE_MASK))
 #define offset_in_page(p)		((unsigned long) (p) & ~PAGE_MASK)
diff --git a/libbcachefs.h b/libbcachefs.h
index 8537a048..7f914d22 100644
--- a/libbcachefs.h
+++ b/libbcachefs.h
@@ -100,7 +100,7 @@ struct bchfs_handle bchu_fs_open_by_dev(const char *, unsigned *);
 
 static inline void bchu_disk_add(struct bchfs_handle fs, char *dev)
 {
-	struct bch_ioctl_disk i = { .dev = (__u64) dev, };
+	struct bch_ioctl_disk i = { .dev = (unsigned long) dev, };
 
 	xioctl(fs.ioctl_fd, BCH_IOCTL_DISK_ADD, &i);
 }
@@ -118,7 +118,7 @@ static inline void bchu_disk_remove(struct bchfs_handle fs, unsigned dev_idx,
 
 static inline void bchu_disk_online(struct bchfs_handle fs, char *dev)
 {
-	struct bch_ioctl_disk i = { .dev = (__u64) dev, };
+	struct bch_ioctl_disk i = { .dev = (unsigned long) dev, };
 
 	xioctl(fs.ioctl_fd, BCH_IOCTL_DISK_ONLINE, &i);
 }
@@ -173,7 +173,7 @@ static inline struct bch_sb *bchu_read_super(struct bchfs_handle fs, unsigned id
 		sb = xrealloc(sb, size);
 		struct bch_ioctl_read_super i = {
 			.size	= size,
-			.sb	= (u64) sb,
+			.sb	= (unsigned long) sb,
 		};
 
 		if (idx != -1) {
diff --git a/linux/atomic64.c b/linux/atomic64.c
new file mode 100644
index 00000000..4654d092
--- /dev/null
+++ b/linux/atomic64.c
@@ -0,0 +1,188 @@
+/*
+ * Generic implementation of 64-bit atomics using spinlocks,
+ * useful on processors that don't have 64-bit atomic instructions.
+ *
+ * Copyright © 2009 Paul Mackerras, IBM Corp. <paulus@au1.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <linux/types.h>
+#include <linux/cache.h>
+#include <linux/spinlock.h>
+#include <linux/atomic.h>
+
+#ifdef ATOMIC64_SPINLOCK
+
+/*
+ * We use a hashed array of spinlocks to provide exclusive access
+ * to each atomic64_t variable.  Since this is expected to used on
+ * systems with small numbers of CPUs (<= 4 or so), we use a
+ * relatively small array of 16 spinlocks to avoid wasting too much
+ * memory on the spinlock array.
+ */
+#define NR_LOCKS	16
+
+/*
+ * Ensure each lock is in a separate cacheline.
+ */
+static union {
+	raw_spinlock_t lock;
+	char pad[L1_CACHE_BYTES];
+} atomic64_lock[NR_LOCKS] ____cacheline_aligned_in_smp = {
+	[0 ... (NR_LOCKS - 1)] = {
+		.lock =  __RAW_SPIN_LOCK_UNLOCKED(atomic64_lock.lock),
+	},
+};
+
+static inline raw_spinlock_t *lock_addr(const atomic64_t *v)
+{
+	unsigned long addr = (unsigned long) v;
+
+	addr >>= L1_CACHE_SHIFT;
+	addr ^= (addr >> 8) ^ (addr >> 16);
+	return &atomic64_lock[addr & (NR_LOCKS - 1)].lock;
+}
+
+long long atomic64_read(const atomic64_t *v)
+{
+	unsigned long flags;
+	raw_spinlock_t *lock = lock_addr(v);
+	long long val;
+
+	raw_spin_lock_irqsave(lock, flags);
+	val = v->counter;
+	raw_spin_unlock_irqrestore(lock, flags);
+	return val;
+}
+
+void atomic64_set(atomic64_t *v, long long i)
+{
+	unsigned long flags;
+	raw_spinlock_t *lock = lock_addr(v);
+
+	raw_spin_lock_irqsave(lock, flags);
+	v->counter = i;
+	raw_spin_unlock_irqrestore(lock, flags);
+}
+
+#define ATOMIC64_OP(op, c_op)						\
+void atomic64_##op(long long a, atomic64_t *v)				\
+{									\
+	unsigned long flags;						\
+	raw_spinlock_t *lock = lock_addr(v);				\
+									\
+	raw_spin_lock_irqsave(lock, flags);				\
+	v->counter c_op a;						\
+	raw_spin_unlock_irqrestore(lock, flags);			\
+}
+
+#define ATOMIC64_OP_RETURN(op, c_op)					\
+long long atomic64_##op##_return(long long a, atomic64_t *v)		\
+{									\
+	unsigned long flags;						\
+	raw_spinlock_t *lock = lock_addr(v);				\
+	long long val;							\
+									\
+	raw_spin_lock_irqsave(lock, flags);				\
+	val = (v->counter c_op a);					\
+	raw_spin_unlock_irqrestore(lock, flags);			\
+	return val;							\
+}
+
+#define ATOMIC64_FETCH_OP(op, c_op)					\
+long long atomic64_fetch_##op(long long a, atomic64_t *v)		\
+{									\
+	unsigned long flags;						\
+	raw_spinlock_t *lock = lock_addr(v);				\
+	long long val;							\
+									\
+	raw_spin_lock_irqsave(lock, flags);				\
+	val = v->counter;						\
+	v->counter c_op a;						\
+	raw_spin_unlock_irqrestore(lock, flags);			\
+	return val;							\
+}
+
+#define ATOMIC64_OPS(op, c_op)						\
+	ATOMIC64_OP(op, c_op)						\
+	ATOMIC64_OP_RETURN(op, c_op)					\
+	ATOMIC64_FETCH_OP(op, c_op)
+
+ATOMIC64_OPS(add, +=)
+ATOMIC64_OPS(sub, -=)
+
+#undef ATOMIC64_OPS
+#define ATOMIC64_OPS(op, c_op)						\
+	ATOMIC64_OP(op, c_op)						\
+	ATOMIC64_OP_RETURN(op, c_op)					\
+	ATOMIC64_FETCH_OP(op, c_op)
+
+ATOMIC64_OPS(and, &=)
+ATOMIC64_OPS(or, |=)
+ATOMIC64_OPS(xor, ^=)
+
+#undef ATOMIC64_OPS
+#undef ATOMIC64_FETCH_OP
+#undef ATOMIC64_OP_RETURN
+#undef ATOMIC64_OP
+
+long long atomic64_dec_if_positive(atomic64_t *v)
+{
+	unsigned long flags;
+	raw_spinlock_t *lock = lock_addr(v);
+	long long val;
+
+	raw_spin_lock_irqsave(lock, flags);
+	val = v->counter - 1;
+	if (val >= 0)
+		v->counter = val;
+	raw_spin_unlock_irqrestore(lock, flags);
+	return val;
+}
+
+long long atomic64_cmpxchg(atomic64_t *v, long long o, long long n)
+{
+	unsigned long flags;
+	raw_spinlock_t *lock = lock_addr(v);
+	long long val;
+
+	raw_spin_lock_irqsave(lock, flags);
+	val = v->counter;
+	if (val == o)
+		v->counter = n;
+	raw_spin_unlock_irqrestore(lock, flags);
+	return val;
+}
+
+long long atomic64_xchg(atomic64_t *v, long long new)
+{
+	unsigned long flags;
+	raw_spinlock_t *lock = lock_addr(v);
+	long long val;
+
+	raw_spin_lock_irqsave(lock, flags);
+	val = v->counter;
+	v->counter = new;
+	raw_spin_unlock_irqrestore(lock, flags);
+	return val;
+}
+
+int atomic64_add_unless(atomic64_t *v, long long a, long long u)
+{
+	unsigned long flags;
+	raw_spinlock_t *lock = lock_addr(v);
+	int ret = 0;
+
+	raw_spin_lock_irqsave(lock, flags);
+	if (v->counter != u) {
+		v->counter += a;
+		ret = 1;
+	}
+	raw_spin_unlock_irqrestore(lock, flags);
+	return ret;
+}
+
+#endif
diff --git a/linux/lz4defs.h b/linux/lz4defs.h
index 42045272..80018ea6 100644
--- a/linux/lz4defs.h
+++ b/linux/lz4defs.h
@@ -62,9 +62,9 @@ typedef uintptr_t uptrval;
 #define LZ4_ARCH64 0
 #endif
 
-#if defined(__LITTLE_ENDIAN)
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
 #define LZ4_LITTLE_ENDIAN 1
-#else
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
 #define LZ4_LITTLE_ENDIAN 0
 #endif
 
diff --git a/linux/sched.c b/linux/sched.c
index de6eb142..52b741fa 100644
--- a/linux/sched.c
+++ b/linux/sched.c
@@ -3,6 +3,10 @@
 #include <string.h>
 #include <sys/mman.h>
 
+/* hack for mips: */
+#define CONFIG_RCU_HAVE_FUTEX 1
+#include <urcu/futex.h>
+
 #include <linux/math64.h>
 #include <linux/printk.h>
 #include <linux/rcupdate.h>
diff --git a/tools-util.c b/tools-util.c
index 41ab7fd3..8474ab06 100644
--- a/tools-util.c
+++ b/tools-util.c
@@ -534,8 +534,6 @@ static u32 crc32c_sse42(u32 crc, const void *buf, size_t size)
 
 static void *resolve_crc32c(void)
 {
-	__builtin_cpu_init();
-
 #ifdef __x86_64__
 	if (__builtin_cpu_supports("sse4.2"))
 		return crc32c_sse42;
@@ -548,8 +546,15 @@ static void *resolve_crc32c(void)
  */
 #ifdef HAVE_WORKING_IFUNC
 
+static void *ifunc_resolve_crc32c(void)
+{
+	__builtin_cpu_init();
+
+	return resolve_crc32c
+}
+
 u32 crc32c(u32, const void *, size_t)
-	__attribute__((ifunc("resolve_crc32c")));
+	__attribute__((ifunc("ifunc_resolve_crc32c")));
 
 #else