Merge branch 'master' into for-next

Pull linus#master to merge PER_CPU_DEF_ATTRIBUTES and alpha build fix
changes.  As alpha in percpu tree uses 'weak' attribute instead of
inline assembly, there's no need for __used attribute.

Conflicts:
	arch/alpha/include/asm/percpu.h
	arch/mn10300/kernel/vmlinux.lds.S
	include/linux/percpu-defs.h

17 files changed, 845 insertions, 187 deletions

diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
new file mode 100644
index 000000000000..ad3c28578961
--- /dev/null
+++ b/tools/perf/util/callchain.c
@@ -0,0 +1,174 @@
+/*
+ * Copyright (C) 2009, Frederic Weisbecker <fweisbec@gmail.com>
+ *
+ * Handle the callchains from the stream in an ad-hoc radix tree and then
+ * sort them in an rbtree.
+ *
+ */
+
+#include <stdlib.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <errno.h>
+
+#include "callchain.h"
+
+
+static void rb_insert_callchain(struct rb_root *root, struct callchain_node *chain)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_node *parent = NULL;
+	struct callchain_node *rnode;
+
+	while (*p) {
+		parent = *p;
+		rnode = rb_entry(parent, struct callchain_node, rb_node);
+
+		if (rnode->hit < chain->hit)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	rb_link_node(&chain->rb_node, parent, p);
+	rb_insert_color(&chain->rb_node, root);
+}
+
+/*
+ * Once we get every callchains from the stream, we can now
+ * sort them by hit
+ */
+void sort_chain_to_rbtree(struct rb_root *rb_root, struct callchain_node *node)
+{
+	struct callchain_node *child;
+
+	list_for_each_entry(child, &node->children, brothers)
+		sort_chain_to_rbtree(rb_root, child);
+
+	if (node->hit)
+		rb_insert_callchain(rb_root, node);
+}
+
+static struct callchain_node *create_child(struct callchain_node *parent)
+{
+	struct callchain_node *new;
+
+	new = malloc(sizeof(*new));
+	if (!new) {
+		perror("not enough memory to create child for code path tree");
+		return NULL;
+	}
+	new->parent = parent;
+	INIT_LIST_HEAD(&new->children);
+	INIT_LIST_HEAD(&new->val);
+	list_add_tail(&new->brothers, &parent->children);
+
+	return new;
+}
+
+static void
+fill_node(struct callchain_node *node, struct ip_callchain *chain, int start)
+{
+	int i;
+
+	for (i = start; i < chain->nr; i++) {
+		struct callchain_list *call;
+
+		call = malloc(sizeof(*chain));
+		if (!call) {
+			perror("not enough memory for the code path tree");
+			return;
+		}
+		call->ip = chain->ips[i];
+		list_add_tail(&call->list, &node->val);
+	}
+	node->val_nr = i - start;
+}
+
+static void add_child(struct callchain_node *parent, struct ip_callchain *chain)
+{
+	struct callchain_node *new;
+
+	new = create_child(parent);
+	fill_node(new, chain, parent->val_nr);
+
+	new->hit = 1;
+}
+
+static void
+split_add_child(struct callchain_node *parent, struct ip_callchain *chain,
+		struct callchain_list *to_split, int idx)
+{
+	struct callchain_node *new;
+
+	/* split */
+	new = create_child(parent);
+	list_move_tail(&to_split->list, &new->val);
+	new->hit = parent->hit;
+	parent->hit = 0;
+	parent->val_nr = idx;
+
+	/* create the new one */
+	add_child(parent, chain);
+}
+
+static int
+__append_chain(struct callchain_node *root, struct ip_callchain *chain,
+		int start);
+
+static int
+__append_chain_children(struct callchain_node *root, struct ip_callchain *chain)
+{
+	struct callchain_node *rnode;
+
+	/* lookup in childrens */
+	list_for_each_entry(rnode, &root->children, brothers) {
+		int ret = __append_chain(rnode, chain, root->val_nr);
+		if (!ret)
+			return 0;
+	}
+	return -1;
+}
+
+static int
+__append_chain(struct callchain_node *root, struct ip_callchain *chain,
+		int start)
+{
+	struct callchain_list *cnode;
+	int i = start;
+	bool found = false;
+
+	/* lookup in the current node */
+	list_for_each_entry(cnode, &root->val, list) {
+		if (cnode->ip != chain->ips[i++])
+			break;
+		if (!found)
+			found = true;
+		if (i == chain->nr)
+			break;
+	}
+
+	/* matches not, relay on the parent */
+	if (!found)
+		return -1;
+
+	/* we match only a part of the node. Split it and add the new chain */
+	if (i < root->val_nr) {
+		split_add_child(root, chain, cnode, i);
+		return 0;
+	}
+
+	/* we match 100% of the path, increment the hit */
+	if (i == root->val_nr) {
+		root->hit++;
+		return 0;
+	}
+
+	return __append_chain_children(root, chain);
+}
+
+void append_chain(struct callchain_node *root, struct ip_callchain *chain)
+{
+	if (__append_chain_children(root, chain) == -1)
+		add_child(root, chain);
+}
diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h
new file mode 100644
index 000000000000..fa1cd2f71fd3
--- /dev/null
+++ b/tools/perf/util/callchain.h
@@ -0,0 +1,33 @@
+#ifndef __PERF_CALLCHAIN_H
+#define __PERF_CALLCHAIN_H
+
+#include "../perf.h"
+#include "list.h"
+#include "rbtree.h"
+
+
+struct callchain_node {
+	struct callchain_node	*parent;
+	struct list_head	brothers;
+	struct list_head 	children;
+	struct list_head 	val;
+	struct rb_node		rb_node;
+	int			val_nr;
+	int			hit;
+};
+
+struct callchain_list {
+	unsigned long		ip;
+	struct list_head	list;
+};
+
+static inline void callchain_init(struct callchain_node *node)
+{
+	INIT_LIST_HEAD(&node->brothers);
+	INIT_LIST_HEAD(&node->children);
+	INIT_LIST_HEAD(&node->val);
+}
+
+void append_chain(struct callchain_node *root, struct ip_callchain *chain);
+void sort_chain_to_rbtree(struct rb_root *rb_root, struct callchain_node *node);
+#endif
diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c
new file mode 100644
index 000000000000..450384b3bbe5
--- /dev/null
+++ b/tools/perf/util/header.c
@@ -0,0 +1,242 @@
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+#include "util.h"
+#include "header.h"
+
+/*
+ *
+ */
+
+struct perf_header_attr *perf_header_attr__new(struct perf_counter_attr *attr)
+{
+	struct perf_header_attr *self = malloc(sizeof(*self));
+
+	if (!self)
+		die("nomem");
+
+	self->attr = *attr;
+	self->ids = 0;
+	self->size = 1;
+	self->id = malloc(sizeof(u64));
+
+	if (!self->id)
+		die("nomem");
+
+	return self;
+}
+
+void perf_header_attr__add_id(struct perf_header_attr *self, u64 id)
+{
+	int pos = self->ids;
+
+	self->ids++;
+	if (self->ids > self->size) {
+		self->size *= 2;
+		self->id = realloc(self->id, self->size * sizeof(u64));
+		if (!self->id)
+			die("nomem");
+	}
+	self->id[pos] = id;
+}
+
+/*
+ *
+ */
+
+struct perf_header *perf_header__new(void)
+{
+	struct perf_header *self = malloc(sizeof(*self));
+
+	if (!self)
+		die("nomem");
+
+	self->frozen = 0;
+
+	self->attrs = 0;
+	self->size = 1;
+	self->attr = malloc(sizeof(void *));
+
+	if (!self->attr)
+		die("nomem");
+
+	self->data_offset = 0;
+	self->data_size = 0;
+
+	return self;
+}
+
+void perf_header__add_attr(struct perf_header *self,
+			   struct perf_header_attr *attr)
+{
+	int pos = self->attrs;
+
+	if (self->frozen)
+		die("frozen");
+
+	self->attrs++;
+	if (self->attrs > self->size) {
+		self->size *= 2;
+		self->attr = realloc(self->attr, self->size * sizeof(void *));
+		if (!self->attr)
+			die("nomem");
+	}
+	self->attr[pos] = attr;
+}
+
+static const char *__perf_magic = "PERFFILE";
+
+#define PERF_MAGIC	(*(u64 *)__perf_magic)
+
+struct perf_file_section {
+	u64 offset;
+	u64 size;
+};
+
+struct perf_file_attr {
+	struct perf_counter_attr	attr;
+	struct perf_file_section	ids;
+};
+
+struct perf_file_header {
+	u64				magic;
+	u64				size;
+	u64				attr_size;
+	struct perf_file_section	attrs;
+	struct perf_file_section	data;
+};
+
+static void do_write(int fd, void *buf, size_t size)
+{
+	while (size) {
+		int ret = write(fd, buf, size);
+
+		if (ret < 0)
+			die("failed to write");
+
+		size -= ret;
+		buf += ret;
+	}
+}
+
+void perf_header__write(struct perf_header *self, int fd)
+{
+	struct perf_file_header f_header;
+	struct perf_file_attr   f_attr;
+	struct perf_header_attr	*attr;
+	int i;
+
+	lseek(fd, sizeof(f_header), SEEK_SET);
+
+
+	for (i = 0; i < self->attrs; i++) {
+		attr = self->attr[i];
+
+		attr->id_offset = lseek(fd, 0, SEEK_CUR);
+		do_write(fd, attr->id, attr->ids * sizeof(u64));
+	}
+
+
+	self->attr_offset = lseek(fd, 0, SEEK_CUR);
+
+	for (i = 0; i < self->attrs; i++) {
+		attr = self->attr[i];
+
+		f_attr = (struct perf_file_attr){
+			.attr = attr->attr,
+			.ids  = {
+				.offset = attr->id_offset,
+				.size   = attr->ids * sizeof(u64),
+			}
+		};
+		do_write(fd, &f_attr, sizeof(f_attr));
+	}
+
+
+	self->data_offset = lseek(fd, 0, SEEK_CUR);
+
+	f_header = (struct perf_file_header){
+		.magic	   = PERF_MAGIC,
+		.size	   = sizeof(f_header),
+		.attr_size = sizeof(f_attr),
+		.attrs = {
+			.offset = self->attr_offset,
+			.size   = self->attrs * sizeof(f_attr),
+		},
+		.data = {
+			.offset = self->data_offset,
+			.size	= self->data_size,
+		},
+	};
+
+	lseek(fd, 0, SEEK_SET);
+	do_write(fd, &f_header, sizeof(f_header));
+	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+
+	self->frozen = 1;
+}
+
+static void do_read(int fd, void *buf, size_t size)
+{
+	while (size) {
+		int ret = read(fd, buf, size);
+
+		if (ret < 0)
+			die("failed to read");
+
+		size -= ret;
+		buf += ret;
+	}
+}
+
+struct perf_header *perf_header__read(int fd)
+{
+	struct perf_header	*self = perf_header__new();
+	struct perf_file_header f_header;
+	struct perf_file_attr	f_attr;
+	u64			f_id;
+
+	int nr_attrs, nr_ids, i, j;
+
+	lseek(fd, 0, SEEK_SET);
+	do_read(fd, &f_header, sizeof(f_header));
+
+	if (f_header.magic	!= PERF_MAGIC		||
+	    f_header.size	!= sizeof(f_header)	||
+	    f_header.attr_size	!= sizeof(f_attr))
+		die("incompatible file format");
+
+	nr_attrs = f_header.attrs.size / sizeof(f_attr);
+	lseek(fd, f_header.attrs.offset, SEEK_SET);
+
+	for (i = 0; i < nr_attrs; i++) {
+		struct perf_header_attr *attr;
+		off_t tmp = lseek(fd, 0, SEEK_CUR);
+
+		do_read(fd, &f_attr, sizeof(f_attr));
+
+		attr = perf_header_attr__new(&f_attr.attr);
+
+		nr_ids = f_attr.ids.size / sizeof(u64);
+		lseek(fd, f_attr.ids.offset, SEEK_SET);
+
+		for (j = 0; j < nr_ids; j++) {
+			do_read(fd, &f_id, sizeof(f_id));
+
+			perf_header_attr__add_id(attr, f_id);
+		}
+		perf_header__add_attr(self, attr);
+		lseek(fd, tmp, SEEK_SET);
+	}
+
+	self->data_offset = f_header.data.offset;
+	self->data_size   = f_header.data.size;
+
+	lseek(fd, self->data_offset + self->data_size, SEEK_SET);
+
+	self->frozen = 1;
+
+	return self;
+}
diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h
new file mode 100644
index 000000000000..b5ef53ad4c7a
--- /dev/null
+++ b/tools/perf/util/header.h
@@ -0,0 +1,37 @@
+#ifndef _PERF_HEADER_H
+#define _PERF_HEADER_H
+
+#include "../../../include/linux/perf_counter.h"
+#include <sys/types.h>
+#include "types.h"
+
+struct perf_header_attr {
+	struct perf_counter_attr attr;
+	int ids, size;
+	u64 *id;
+	off_t id_offset;
+};
+
+struct perf_header {
+	int frozen;
+	int attrs, size;
+	struct perf_header_attr **attr;
+	off_t attr_offset;
+	u64 data_offset;
+	u64 data_size;
+};
+
+struct perf_header *perf_header__read(int fd);
+void perf_header__write(struct perf_header *self, int fd);
+
+void perf_header__add_attr(struct perf_header *self,
+			   struct perf_header_attr *attr);
+
+struct perf_header_attr *
+perf_header_attr__new(struct perf_counter_attr *attr);
+void perf_header_attr__add_id(struct perf_header_attr *self, u64 id);
+
+
+struct perf_header *perf_header__new(void);
+
+#endif /* _PERF_HEADER_H */
diff --git a/tools/perf/util/help.c b/tools/perf/util/help.c
index 6653f7dd1d78..17a00e0df2c4 100644
--- a/tools/perf/util/help.c
+++ b/tools/perf/util/help.c
@@ -126,21 +126,6 @@ static int is_executable(const char *name)
 	    !S_ISREG(st.st_mode))
 		return 0;
 
-#ifdef __MINGW32__
-	/* cannot trust the executable bit, peek into the file instead */
-	char buf[3] = { 0 };
-	int n;
-	int fd = open(name, O_RDONLY);
-	st.st_mode &= ~S_IXUSR;
-	if (fd >= 0) {
-		n = read(fd, buf, 2);
-		if (n == 2)
-			/* DOS executables start with "MZ" */
-			if (!strcmp(buf, "#!") || !strcmp(buf, "MZ"))
-				st.st_mode |= S_IXUSR;
-		close(fd);
-	}
-#endif
 	return st.st_mode & S_IXUSR;
 }
 
diff --git a/tools/perf/util/pager.c b/tools/perf/util/pager.c
index a28bccae5458..1915de20dcac 100644
--- a/tools/perf/util/pager.c
+++ b/tools/perf/util/pager.c
@@ -9,7 +9,6 @@
 
 static int spawned_pager;
 
-#ifndef __MINGW32__
 static void pager_preexec(void)
 {
 	/*
@@ -24,7 +23,6 @@ static void pager_preexec(void)
 
 	setenv("LESS", "FRSX", 0);
 }
-#endif
 
 static const char *pager_argv[] = { "sh", "-c", NULL, NULL };
 static struct child_process pager_process;
@@ -70,9 +68,8 @@ void setup_pager(void)
 	pager_argv[2] = pager;
 	pager_process.argv = pager_argv;
 	pager_process.in = -1;
-#ifndef __MINGW32__
 	pager_process.preexec_cb = pager_preexec;
-#endif
+
 	if (start_command(&pager_process))
 		return;
 
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 35d04da38d6a..4d042f104cdc 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -16,32 +16,28 @@ struct event_symbol {
 	u8	type;
 	u64	config;
 	char	*symbol;
+	char	*alias;
 };
 
-#define C(x, y) .type = PERF_TYPE_##x, .config = PERF_COUNT_##y
-#define CR(x, y) .type = PERF_TYPE_##x, .config = y
+#define CHW(x) .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_##x
+#define CSW(x) .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_##x
 
 static struct event_symbol event_symbols[] = {
-  { C(HARDWARE, HW_CPU_CYCLES),		"cpu-cycles",		},
-  { C(HARDWARE, HW_CPU_CYCLES),		"cycles",		},
-  { C(HARDWARE, HW_INSTRUCTIONS),	"instructions",		},
-  { C(HARDWARE, HW_CACHE_REFERENCES),	"cache-references",	},
-  { C(HARDWARE, HW_CACHE_MISSES),	"cache-misses",		},
-  { C(HARDWARE, HW_BRANCH_INSTRUCTIONS),"branch-instructions",	},
-  { C(HARDWARE, HW_BRANCH_INSTRUCTIONS),"branches",		},
-  { C(HARDWARE, HW_BRANCH_MISSES),	"branch-misses",	},
-  { C(HARDWARE, HW_BUS_CYCLES),		"bus-cycles",		},
-
-  { C(SOFTWARE, SW_CPU_CLOCK),		"cpu-clock",		},
-  { C(SOFTWARE, SW_TASK_CLOCK),		"task-clock",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS),	"page-faults",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS),	"faults",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS_MIN),	"minor-faults",		},
-  { C(SOFTWARE, SW_PAGE_FAULTS_MAJ),	"major-faults",		},
-  { C(SOFTWARE, SW_CONTEXT_SWITCHES),	"context-switches",	},
-  { C(SOFTWARE, SW_CONTEXT_SWITCHES),	"cs",			},
-  { C(SOFTWARE, SW_CPU_MIGRATIONS),	"cpu-migrations",	},
-  { C(SOFTWARE, SW_CPU_MIGRATIONS),	"migrations",		},
+  { CHW(CPU_CYCLES),		"cpu-cycles",		"cycles"	},
+  { CHW(INSTRUCTIONS),		"instructions",		""		},
+  { CHW(CACHE_REFERENCES),	"cache-references",	""		},
+  { CHW(CACHE_MISSES),		"cache-misses",		""		},
+  { CHW(BRANCH_INSTRUCTIONS),	"branch-instructions",	"branches"	},
+  { CHW(BRANCH_MISSES),		"branch-misses",	""		},
+  { CHW(BUS_CYCLES),		"bus-cycles",		""		},
+
+  { CSW(CPU_CLOCK),		"cpu-clock",		""		},
+  { CSW(TASK_CLOCK),		"task-clock",		""		},
+  { CSW(PAGE_FAULTS),		"page-faults",		"faults"	},
+  { CSW(PAGE_FAULTS_MIN),	"minor-faults",		""		},
+  { CSW(PAGE_FAULTS_MAJ),	"major-faults",		""		},
+  { CSW(CONTEXT_SWITCHES),	"context-switches",	"cs"		},
+  { CSW(CPU_MIGRATIONS),	"cpu-migrations",	"migrations"	},
 };
 
 #define __PERF_COUNTER_FIELD(config, name) \
@@ -74,26 +70,70 @@ static char *sw_event_names[] = {
 
 #define MAX_ALIASES 8
 
-static char *hw_cache [][MAX_ALIASES] = {
-	{ "L1-data"		, "l1-d", "l1d"					},
-	{ "L1-instruction"	, "l1-i", "l1i"					},
-	{ "L2"			, "l2"						},
-	{ "Data-TLB"		, "dtlb", "d-tlb"				},
-	{ "Instruction-TLB"	, "itlb", "i-tlb"				},
-	{ "Branch"		, "bpu" , "btb", "bpc"				},
+static char *hw_cache[][MAX_ALIASES] = {
+ { "L1-d$",	"l1-d",		"l1d",		"L1-data",		},
+ { "L1-i$",	"l1-i",		"l1i",		"L1-instruction",	},
+ { "LLC",	"L2"							},
+ { "dTLB",	"d-tlb",	"Data-TLB",				},
+ { "iTLB",	"i-tlb",	"Instruction-TLB",			},
+ { "branch",	"branches",	"bpu",		"btb",		"bpc",	},
 };
 
-static char *hw_cache_op [][MAX_ALIASES] = {
-	{ "Load"		, "read"					},
-	{ "Store"		, "write"					},
-	{ "Prefetch"		, "speculative-read", "speculative-load"	},
+static char *hw_cache_op[][MAX_ALIASES] = {
+ { "load",	"loads",	"read",					},
+ { "store",	"stores",	"write",				},
+ { "prefetch",	"prefetches",	"speculative-read", "speculative-load",	},
 };
 
-static char *hw_cache_result [][MAX_ALIASES] = {
-	{ "Reference"		, "ops", "access"				},
-	{ "Miss"								},
+static char *hw_cache_result[][MAX_ALIASES] = {
+ { "refs",	"Reference",	"ops",		"access",		},
+ { "misses",	"miss",							},
 };
 
+#define C(x)		PERF_COUNT_HW_CACHE_##x
+#define CACHE_READ	(1 << C(OP_READ))
+#define CACHE_WRITE	(1 << C(OP_WRITE))
+#define CACHE_PREFETCH	(1 << C(OP_PREFETCH))
+#define COP(x)		(1 << x)
+
+/*
+ * cache operartion stat
+ * L1I : Read and prefetch only
+ * ITLB and BPU : Read-only
+ */
+static unsigned long hw_cache_stat[C(MAX)] = {
+ [C(L1D)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(L1I)]	= (CACHE_READ | CACHE_PREFETCH),
+ [C(LL)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(DTLB)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
+ [C(ITLB)]	= (CACHE_READ),
+ [C(BPU)]	= (CACHE_READ),
+};
+
+static int is_cache_op_valid(u8 cache_type, u8 cache_op)
+{
+	if (hw_cache_stat[cache_type] & COP(cache_op))
+		return 1;	/* valid */
+	else
+		return 0;	/* invalid */
+}
+
+static char *event_cache_name(u8 cache_type, u8 cache_op, u8 cache_result)
+{
+	static char name[50];
+
+	if (cache_result) {
+		sprintf(name, "%s-%s-%s", hw_cache[cache_type][0],
+			hw_cache_op[cache_op][0],
+			hw_cache_result[cache_result][0]);
+	} else {
+		sprintf(name, "%s-%s", hw_cache[cache_type][0],
+			hw_cache_op[cache_op][1]);
+	}
+
+	return name;
+}
+
 char *event_name(int counter)
 {
 	u64 config = attrs[counter].config;
@@ -113,7 +153,6 @@ char *event_name(int counter)
 
 	case PERF_TYPE_HW_CACHE: {
 		u8 cache_type, cache_op, cache_result;
-		static char name[100];
 
 		cache_type   = (config >>  0) & 0xff;
 		if (cache_type > PERF_COUNT_HW_CACHE_MAX)
@@ -127,12 +166,10 @@ char *event_name(int counter)
 		if (cache_result > PERF_COUNT_HW_CACHE_RESULT_MAX)
 			return "unknown-ext-hardware-cache-result";
 
-		sprintf(name, "%s-Cache-%s-%ses",
-			hw_cache[cache_type][0],
-			hw_cache_op[cache_op][0],
-			hw_cache_result[cache_result][0]);
+		if (!is_cache_op_valid(cache_type, cache_op))
+			return "invalid-cache";
 
-		return name;
+		return event_cache_name(cache_type, cache_op, cache_result);
 	}
 
 	case PERF_TYPE_SOFTWARE:
@@ -163,7 +200,8 @@ static int parse_aliases(const char *str, char *names[][MAX_ALIASES], int size)
 	return -1;
 }
 
-static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
+static int
+parse_generic_hw_symbols(const char *str, struct perf_counter_attr *attr)
 {
 	int cache_type = -1, cache_op = 0, cache_result = 0;
 
@@ -182,6 +220,9 @@ static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *a
 	if (cache_op == -1)
 		cache_op = PERF_COUNT_HW_CACHE_OP_READ;
 
+	if (!is_cache_op_valid(cache_type, cache_op))
+		return -EINVAL;
+
 	cache_result = parse_aliases(str, hw_cache_result,
 					PERF_COUNT_HW_CACHE_RESULT_MAX);
 	/*
@@ -196,6 +237,19 @@ static int parse_generic_hw_symbols(const char *str, struct perf_counter_attr *a
 	return 0;
 }
 
+static int check_events(const char *str, unsigned int i)
+{
+	if (!strncmp(str, event_symbols[i].symbol,
+		     strlen(event_symbols[i].symbol)))
+		return 1;
+
+	if (strlen(event_symbols[i].alias))
+		if (!strncmp(str, event_symbols[i].alias,
+			     strlen(event_symbols[i].alias)))
+			return 1;
+	return 0;
+}
+
 /*
  * Each event can have multiple symbolic names.
  * Symbolic names are (almost) exactly matched.
@@ -235,9 +289,7 @@ static int parse_event_symbols(const char *str, struct perf_counter_attr *attr)
 	}
 
 	for (i = 0; i < ARRAY_SIZE(event_symbols); i++) {
-		if (!strncmp(str, event_symbols[i].symbol,
-			     strlen(event_symbols[i].symbol))) {
-
+		if (check_events(str, i)) {
 			attr->type = event_symbols[i].type;
 			attr->config = event_symbols[i].config;
 
@@ -289,6 +341,7 @@ void print_events(void)
 {
 	struct event_symbol *syms = event_symbols;
 	unsigned int i, type, prev_type = -1;
+	char name[40];
 
 	fprintf(stderr, "\n");
 	fprintf(stderr, "List of pre-defined events (to be used in -e):\n");
@@ -301,14 +354,18 @@ void print_events(void)
 		if (type != prev_type)
 			fprintf(stderr, "\n");
 
-		fprintf(stderr, "  %-30s [%s]\n", syms->symbol,
+		if (strlen(syms->alias))
+			sprintf(name, "%s OR %s", syms->symbol, syms->alias);
+		else
+			strcpy(name, syms->symbol);
+		fprintf(stderr, "  %-40s [%s]\n", name,
 			event_type_descriptors[type]);
 
 		prev_type = type;
 	}
 
 	fprintf(stderr, "\n");
-	fprintf(stderr, "  %-30s [raw hardware event descriptor]\n",
+	fprintf(stderr, "  %-40s [raw hardware event descriptor]\n",
 		"rNNN");
 	fprintf(stderr, "\n");
 
diff --git a/tools/perf/util/run-command.c b/tools/perf/util/run-command.c
index b2f5e854f40a..a3935343091a 100644
--- a/tools/perf/util/run-command.c
+++ b/tools/perf/util/run-command.c
@@ -65,7 +65,6 @@ int start_command(struct child_process *cmd)
 		cmd->err = fderr[0];
 	}
 
-#ifndef __MINGW32__
 	fflush(NULL);
 	cmd->pid = fork();
 	if (!cmd->pid) {
@@ -118,71 +117,6 @@ int start_command(struct child_process *cmd)
 		}
 		exit(127);
 	}
-#else
-	int s0 = -1, s1 = -1, s2 = -1;	/* backups of stdin, stdout, stderr */
-	const char **sargv = cmd->argv;
-	char **env = environ;
-
-	if (cmd->no_stdin) {
-		s0 = dup(0);
-		dup_devnull(0);
-	} else if (need_in) {
-		s0 = dup(0);
-		dup2(fdin[0], 0);
-	} else if (cmd->in) {
-		s0 = dup(0);
-		dup2(cmd->in, 0);
-	}
-
-	if (cmd->no_stderr) {
-		s2 = dup(2);
-		dup_devnull(2);
-	} else if (need_err) {
-		s2 = dup(2);
-		dup2(fderr[1], 2);
-	}
-
-	if (cmd->no_stdout) {
-		s1 = dup(1);
-		dup_devnull(1);
-	} else if (cmd->stdout_to_stderr) {
-		s1 = dup(1);
-		dup2(2, 1);
-	} else if (need_out) {
-		s1 = dup(1);
-		dup2(fdout[1], 1);
-	} else if (cmd->out > 1) {
-		s1 = dup(1);
-		dup2(cmd->out, 1);
-	}
-
-	if (cmd->dir)
-		die("chdir in start_command() not implemented");
-	if (cmd->env) {
-		env = copy_environ();
-		for (; *cmd->env; cmd->env++)
-			env = env_setenv(env, *cmd->env);
-	}
-
-	if (cmd->perf_cmd) {
-		cmd->argv = prepare_perf_cmd(cmd->argv);
-	}
-
-	cmd->pid = mingw_spawnvpe(cmd->argv[0], cmd->argv, env);
-
-	if (cmd->env)
-		free_environ(env);
-	if (cmd->perf_cmd)
-		free(cmd->argv);
-
-	cmd->argv = sargv;
-	if (s0 >= 0)
-		dup2(s0, 0), close(s0);
-	if (s1 >= 0)
-		dup2(s1, 1), close(s1);
-	if (s2 >= 0)
-		dup2(s2, 2), close(s2);
-#endif
 
 	if (cmd->pid < 0) {
 		int err = errno;
@@ -288,14 +222,6 @@ int run_command_v_opt_cd_env(const char **argv, int opt, const char *dir, const
 	return run_command(&cmd);
 }
 
-#ifdef __MINGW32__
-static __stdcall unsigned run_thread(void *data)
-{
-	struct async *async = data;
-	return async->proc(async->fd_for_proc, async->data);
-}
-#endif
-
 int start_async(struct async *async)
 {
 	int pipe_out[2];
@@ -304,7 +230,6 @@ int start_async(struct async *async)
 		return error("cannot create pipe: %s", strerror(errno));
 	async->out = pipe_out[0];
 
-#ifndef __MINGW32__
 	/* Flush stdio before fork() to avoid cloning buffers */
 	fflush(NULL);
 
@@ -319,33 +244,17 @@ int start_async(struct async *async)
 		exit(!!async->proc(pipe_out[1], async->data));
 	}
 	close(pipe_out[1]);
-#else
-	async->fd_for_proc = pipe_out[1];
-	async->tid = (HANDLE) _beginthreadex(NULL, 0, run_thread, async, 0, NULL);
-	if (!async->tid) {
-		error("cannot create thread: %s", strerror(errno));
-		close_pair(pipe_out);
-		return -1;
-	}
-#endif
+
 	return 0;
 }
 
 int finish_async(struct async *async)
 {
-#ifndef __MINGW32__
 	int ret = 0;
 
 	if (wait_or_whine(async->pid))
 		ret = error("waitpid (async) failed");
-#else
-	DWORD ret = 0;
-	if (WaitForSingleObject(async->tid, INFINITE) != WAIT_OBJECT_0)
-		ret = error("waiting for thread failed: %lu", GetLastError());
-	else if (!GetExitCodeThread(async->tid, &ret))
-		ret = error("cannot get thread exit code: %lu", GetLastError());
-	CloseHandle(async->tid);
-#endif
+
 	return ret;
 }
 
diff --git a/tools/perf/util/run-command.h b/tools/perf/util/run-command.h
index 328289f23669..cc1837deba88 100644
--- a/tools/perf/util/run-command.h
+++ b/tools/perf/util/run-command.h
@@ -79,12 +79,7 @@ struct async {
 	int (*proc)(int fd, void *data);
 	void *data;
 	int out;	/* caller reads from here and closes it */
-#ifndef __MINGW32__
 	pid_t pid;
-#else
-	HANDLE tid;
-	int fd_for_proc;
-#endif
 };
 
 int start_async(struct async *async);
diff --git a/tools/perf/util/strbuf.c b/tools/perf/util/strbuf.c
index eaba09306802..464e7ca898cf 100644
--- a/tools/perf/util/strbuf.c
+++ b/tools/perf/util/strbuf.c
@@ -259,7 +259,7 @@ size_t strbuf_fread(struct strbuf *sb, size_t size, FILE *f)
 	res = fread(sb->buf + sb->len, 1, size, f);
 	if (res > 0)
 		strbuf_setlen(sb, sb->len + res);
-	else if (res < 0 && oldalloc == 0)
+	else if (oldalloc == 0)
 		strbuf_release(sb);
 	return res;
 }
diff --git a/tools/perf/util/string.h b/tools/perf/util/string.h
index 37b03255b425..3dca2f654cd0 100644
--- a/tools/perf/util/string.h
+++ b/tools/perf/util/string.h
@@ -1,7 +1,7 @@
 #ifndef _PERF_STRING_H_
 #define _PERF_STRING_H_
 
-#include "../types.h"
+#include "types.h"
 
 int hex2u64(const char *ptr, u64 *val);
 
diff --git a/tools/perf/util/strlist.c b/tools/perf/util/strlist.c
new file mode 100644
index 000000000000..025a78edfffe
--- /dev/null
+++ b/tools/perf/util/strlist.c
@@ -0,0 +1,184 @@
+/*
+ * (c) 2009 Arnaldo Carvalho de Melo <acme@redhat.com>
+ *
+ * Licensed under the GPLv2.
+ */
+
+#include "strlist.h"
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+static struct str_node *str_node__new(const char *s, bool dupstr)
+{
+	struct str_node *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		if (dupstr) {
+			s = strdup(s);
+			if (s == NULL)
+				goto out_delete;
+		}
+		self->s = s;
+	}
+
+	return self;
+
+out_delete:
+	free(self);
+	return NULL;
+}
+
+static void str_node__delete(struct str_node *self, bool dupstr)
+{
+	if (dupstr)
+		free((void *)self->s);
+	free(self);
+}
+
+int strlist__add(struct strlist *self, const char *new_entry)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+	struct str_node *sn;
+
+	while (*p != NULL) {
+		int rc;
+
+		parent = *p;
+		sn = rb_entry(parent, struct str_node, rb_node);
+		rc = strcmp(sn->s, new_entry);
+
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return -EEXIST;
+	}
+
+	sn = str_node__new(new_entry, self->dupstr);
+	if (sn == NULL)
+		return -ENOMEM;
+
+	rb_link_node(&sn->rb_node, parent, p);
+	rb_insert_color(&sn->rb_node, &self->entries);
+
+	return 0;
+}
+
+int strlist__load(struct strlist *self, const char *filename)
+{
+	char entry[1024];
+	int err;
+	FILE *fp = fopen(filename, "r");
+
+	if (fp == NULL)
+		return errno;
+
+	while (fgets(entry, sizeof(entry), fp) != NULL) {
+		const size_t len = strlen(entry);
+
+		if (len == 0)
+			continue;
+		entry[len - 1] = '\0';
+
+		err = strlist__add(self, entry);
+		if (err != 0)
+			goto out;
+	}
+
+	err = 0;
+out:
+	fclose(fp);
+	return err;
+}
+
+void strlist__remove(struct strlist *self, struct str_node *sn)
+{
+	rb_erase(&sn->rb_node, &self->entries);
+	str_node__delete(sn, self->dupstr);
+}
+
+bool strlist__has_entry(struct strlist *self, const char *entry)
+{
+	struct rb_node **p = &self->entries.rb_node;
+	struct rb_node *parent = NULL;
+
+	while (*p != NULL) {
+		struct str_node *sn;
+		int rc;
+
+		parent = *p;
+		sn = rb_entry(parent, struct str_node, rb_node);
+		rc = strcmp(sn->s, entry);
+
+		if (rc > 0)
+			p = &(*p)->rb_left;
+		else if (rc < 0)
+			p = &(*p)->rb_right;
+		else
+			return true;
+	}
+
+	return false;
+}
+
+static int strlist__parse_list_entry(struct strlist *self, const char *s)
+{
+	if (strncmp(s, "file://", 7) == 0)
+		return strlist__load(self, s + 7);
+
+	return strlist__add(self, s);
+}
+
+int strlist__parse_list(struct strlist *self, const char *s)
+{
+	char *sep;
+	int err;
+
+	while ((sep = strchr(s, ',')) != NULL) {
+		*sep = '\0';
+		err = strlist__parse_list_entry(self, s);
+		*sep = ',';
+		if (err != 0)
+			return err;
+		s = sep + 1;
+	}
+
+	return *s ? strlist__parse_list_entry(self, s) : 0;
+}
+
+struct strlist *strlist__new(bool dupstr, const char *slist)
+{
+	struct strlist *self = malloc(sizeof(*self));
+
+	if (self != NULL) {
+		self->entries = RB_ROOT;
+		self->dupstr = dupstr;
+		if (slist && strlist__parse_list(self, slist) != 0)
+			goto out_error;
+	}
+
+	return self;
+out_error:
+	free(self);
+	return NULL;
+}
+
+void strlist__delete(struct strlist *self)
+{
+	if (self != NULL) {
+		struct str_node *pos;
+		struct rb_node *next = rb_first(&self->entries);
+
+		while (next) {
+			pos = rb_entry(next, struct str_node, rb_node);
+			next = rb_next(&pos->rb_node);
+			strlist__remove(self, pos);
+		}
+		self->entries = RB_ROOT;
+		free(self);
+	}
+}
diff --git a/tools/perf/util/strlist.h b/tools/perf/util/strlist.h
new file mode 100644
index 000000000000..2fb117fb4b67
--- /dev/null
+++ b/tools/perf/util/strlist.h
@@ -0,0 +1,32 @@
+#ifndef STRLIST_H_
+#define STRLIST_H_
+
+#include "rbtree.h"
+#include <stdbool.h>
+
+struct str_node {
+	struct rb_node rb_node;
+	const char     *s;
+};
+
+struct strlist {
+	struct rb_root entries;
+	bool dupstr;
+};
+
+struct strlist *strlist__new(bool dupstr, const char *slist);
+void strlist__delete(struct strlist *self);
+
+void strlist__remove(struct strlist *self, struct str_node *sn);
+int strlist__load(struct strlist *self, const char *filename);
+int strlist__add(struct strlist *self, const char *str);
+
+bool strlist__has_entry(struct strlist *self, const char *entry);
+
+static inline bool strlist__empty(const struct strlist *self)
+{
+	return rb_first(&self->entries) == NULL;
+}
+
+int strlist__parse_list(struct strlist *self, const char *s);
+#endif /* STRLIST_H_ */
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index 86e14375e74e..78c2efde01b7 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -520,7 +520,9 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 	nr_syms = shdr.sh_size / shdr.sh_entsize;
 
 	memset(&sym, 0, sizeof(sym));
-
+	self->prelinked = elf_section_by_name(elf, &ehdr, &shdr,
+					      ".gnu.prelink_undo",
+					      NULL) != NULL;
 	elf_symtab__for_each_symbol(syms, nr_syms, index, sym) {
 		struct symbol *f;
 		u64 obj_start;
@@ -535,7 +537,13 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
 		gelf_getshdr(sec, &shdr);
 		obj_start = sym.st_value;
 
-		sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+		if (self->prelinked) {
+			if (verbose >= 2)
+				printf("adjusting symbol: st_value: %Lx sh_addr: %Lx sh_offset: %Lx\n",
+					(u64)sym.st_value, (u64)shdr.sh_addr, (u64)shdr.sh_offset);
+
+			sym.st_value -= shdr.sh_addr - shdr.sh_offset;
+		}
 
 		f = symbol__new(sym.st_value, sym.st_size,
 				elf_sym__name(&sym, symstrs),
@@ -569,6 +577,8 @@ int dso__load(struct dso *self, symbol_filter_t filter, int verbose)
 	if (!name)
 		return -1;
 
+	self->prelinked = 0;
+
 	if (strncmp(self->name, "/tmp/perf-", 10) == 0)
 		return dso__load_perf_map(self, filter, verbose);
 
@@ -629,7 +639,7 @@ int dso__load_kernel(struct dso *self, const char *vmlinux,
 	if (vmlinux)
 		err = dso__load_vmlinux(self, vmlinux, filter, verbose);
 
-	if (err)
+	if (err < 0)
 		err = dso__load_kallsyms(self, filter, verbose);
 
 	return err;
diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
index ea332e56e458..2c48ace8203b 100644
--- a/tools/perf/util/symbol.h
+++ b/tools/perf/util/symbol.h
@@ -2,7 +2,7 @@
 #define _PERF_SYMBOL_ 1
 
 #include <linux/types.h>
-#include "../types.h"
+#include "types.h"
 #include "list.h"
 #include "rbtree.h"
 
@@ -20,8 +20,9 @@ struct symbol {
 struct dso {
 	struct list_head node;
 	struct rb_root	 syms;
-	unsigned int	 sym_priv_size;
 	struct symbol    *(*find_symbol)(struct dso *, u64 ip);
+	unsigned int	 sym_priv_size;
+	unsigned char	 prelinked;
 	char		 name[0];
 };
 
diff --git a/tools/perf/util/types.h b/tools/perf/util/types.h
new file mode 100644
index 000000000000..5e75f9005940
--- /dev/null
+++ b/tools/perf/util/types.h
@@ -0,0 +1,17 @@
+#ifndef _PERF_TYPES_H
+#define _PERF_TYPES_H
+
+/*
+ * We define u64 as unsigned long long for every architecture
+ * so that we can print it with %Lx without getting warnings.
+ */
+typedef unsigned long long u64;
+typedef signed long long   s64;
+typedef unsigned int	   u32;
+typedef signed int	   s32;
+typedef unsigned short	   u16;
+typedef signed short	   s16;
+typedef unsigned char	   u8;
+typedef signed char	   s8;
+
+#endif /* _PERF_TYPES_H */
diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h
index b8cfed776d81..b4be6071c105 100644
--- a/tools/perf/util/util.h
+++ b/tools/perf/util/util.h
@@ -67,7 +67,6 @@
 #include <assert.h>
 #include <regex.h>
 #include <utime.h>
-#ifndef __MINGW32__
 #include <sys/wait.h>
 #include <sys/poll.h>
 #include <sys/socket.h>
@@ -81,20 +80,6 @@
 #include <netdb.h>
 #include <pwd.h>
 #include <inttypes.h>
-#if defined(__CYGWIN__)
-#undef _XOPEN_SOURCE
-#include <grp.h>
-#define _XOPEN_SOURCE 600
-#include "compat/cygwin.h"
-#else
-#undef _ALL_SOURCE /* AIX 5.3L defines a struct list with _ALL_SOURCE. */
-#include <grp.h>
-#define _ALL_SOURCE 1
-#endif
-#else 	/* __MINGW32__ */
-/* pull in Windows compatibility stuff */
-#include "compat/mingw.h"
-#endif	/* __MINGW32__ */
 
 #ifndef NO_ICONV
 #include <iconv.h>