ftrace: mmiotrace, updates

here is a patch that makes mmiotrace work almost well within the tracing
framework. The patch applies on top of my previous patch. I have my own
output formatting in place now.

Summary of changes:
- fix the NULL dereference that was due to not calling tracing_reset()
- add print_line() callback into struct tracer
- implement print_line() for mmiotrace, producing up-to-spec text
- add my output header, but that is not really called in the right place
- rewrote the main structs in mmiotrace
- added two new trace entry types: TRACE_MMIO_RW and TRACE_MMIO_MAP
- made some functions in trace.c non-static
- check current==NULL in tracing_generic_entry_update()
- fix(?) comparison in trace_seq_printf()

Things seem to work fine except a few issues. Markers (text lines injected
into mmiotrace log) are missing, I did not feel hacking them in before we
have variable length entries. My output header is printed only for 'trace'
file, but not 'trace_pipe'. For some reason, despite my quick fix,
iter->trace is NULL in print_trace_line() when called from 'trace_pipe'
file, which means I don't get proper output formatting.

I only tried by loading nouveau.ko, which just detects the card, and that
is traced fine. I didn't try further. Map, two reads and unmap. Works
perfectly.

I am missing the information about overflows, I'd prefer to have a
counter for lost events. I didn't try, but I guess currently there is no
way of knowning when it overflows?

So, not too far from being fully operational, it seems :-)
And looking at the diffstat, there also is some 700-900 lines of user space
code that just became obsolete.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

kernel/trace/trace.c

@@ -831,6 +831,40 @@ ftrace(struct trace_array *tr, struct trace_array_cpu *data,
 		trace_function(tr, data, ip, parent_ip, flags);
 }
 
+#ifdef CONFIG_MMIOTRACE
+void __trace_mmiotrace_rw(struct trace_array *tr, struct trace_array_cpu *data,
+						struct mmiotrace_rw *rw)
+{
+	struct trace_entry *entry;
+	unsigned long irq_flags;
+
+	spin_lock_irqsave(&data->lock, irq_flags);
+	entry			= tracing_get_trace_entry(tr, data);
+	tracing_generic_entry_update(entry, 0);
+	entry->type		= TRACE_MMIO_RW;
+	entry->mmiorw		= *rw;
+	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
+}
+
+void __trace_mmiotrace_map(struct trace_array *tr, struct trace_array_cpu *data,
+						struct mmiotrace_map *map)
+{
+	struct trace_entry *entry;
+	unsigned long irq_flags;
+
+	spin_lock_irqsave(&data->lock, irq_flags);
+	entry			= tracing_get_trace_entry(tr, data);
+	tracing_generic_entry_update(entry, 0);
+	entry->type		= TRACE_MMIO_MAP;
+	entry->mmiomap		= *map;
+	spin_unlock_irqrestore(&data->lock, irq_flags);
+
+	trace_wake_up();
+}
+#endif
+
 void __trace_stack(struct trace_array *tr,
 		   struct trace_array_cpu *data,
 		   unsigned long flags,

kernel/trace/trace.h

@@ -5,6 +5,7 @@
 #include <asm/atomic.h>
 #include <linux/sched.h>
 #include <linux/clocksource.h>
+#include <linux/mmiotrace.h>
 
 enum trace_type {
 	__TRACE_FIRST_TYPE = 0,
@@ -14,6 +15,8 @@ enum trace_type {
 	TRACE_WAKE,
 	TRACE_STACK,
 	TRACE_SPECIAL,
+	TRACE_MMIO_RW,
+	TRACE_MMIO_MAP,
 
 	__TRACE_LAST_TYPE
 };
@@ -75,6 +78,8 @@ struct trace_entry {
 		struct ctx_switch_entry		ctx;
 		struct special_entry		special;
 		struct stack_entry		stack;
+		struct mmiotrace_rw		mmiorw;
+		struct mmiotrace_map		mmiomap;
 	};
 };
 
@@ -255,6 +260,15 @@ extern unsigned long ftrace_update_tot_cnt;
 extern int DYN_FTRACE_TEST_NAME(void);
 #endif
 
+#ifdef CONFIG_MMIOTRACE
+extern void __trace_mmiotrace_rw(struct trace_array *tr,
+				struct trace_array_cpu *data,
+				struct mmiotrace_rw *rw);
+extern void __trace_mmiotrace_map(struct trace_array *tr,
+				struct trace_array_cpu *data,
+				struct mmiotrace_map *map);
+#endif
+
 #ifdef CONFIG_FTRACE_STARTUP_TEST
 #ifdef CONFIG_FTRACE
 extern int trace_selftest_startup_function(struct tracer *trace,

kernel/trace/trace_mmiotrace.c

@@ -11,19 +11,26 @@
 
 #include "trace.h"
 
-extern void
-__trace_special(void *__tr, void *__data,
-		unsigned long arg1, unsigned long arg2, unsigned long arg3);
-
 static struct trace_array *mmio_trace_array;
 
+static void mmio_reset_data(struct trace_array *tr)
+{
+	int cpu;
+
+	tr->time_start = ftrace_now(tr->cpu);
+
+	for_each_online_cpu(cpu)
+		tracing_reset(tr->data[cpu]);
+}
 
 static void mmio_trace_init(struct trace_array *tr)
 {
 	pr_debug("in %s\n", __func__);
 	mmio_trace_array = tr;
-	if (tr->ctrl)
+	if (tr->ctrl) {
+		mmio_reset_data(tr);
 		enable_mmiotrace();
+	}
 }
 
 static void mmio_trace_reset(struct trace_array *tr)
@@ -31,15 +38,110 @@ static void mmio_trace_reset(struct trace_array *tr)
 	pr_debug("in %s\n", __func__);
 	if (tr->ctrl)
 		disable_mmiotrace();
+	mmio_reset_data(tr);
+	mmio_trace_array = NULL;
 }
 
 static void mmio_trace_ctrl_update(struct trace_array *tr)
 {
 	pr_debug("in %s\n", __func__);
-	if (tr->ctrl)
+	if (tr->ctrl) {
+		mmio_reset_data(tr);
 		enable_mmiotrace();
-	else
+	} else {
 		disable_mmiotrace();
+	}
+}
+
+/* XXX: This is not called for trace_pipe file! */
+void mmio_print_header(struct trace_iterator *iter)
+{
+	struct trace_seq *s = &iter->seq;
+	trace_seq_printf(s, "VERSION broken 20070824\n");
+	/* TODO: print /proc/bus/pci/devices contents as PCIDEV lines */
+}
+
+static int mmio_print_rw(struct trace_iterator *iter)
+{
+	struct trace_entry *entry = iter->ent;
+	struct mmiotrace_rw *rw	= &entry->mmiorw;
+	struct trace_seq *s	= &iter->seq;
+	unsigned long long t	= ns2usecs(entry->t);
+	unsigned long usec_rem	= do_div(t, 1000000ULL);
+	unsigned secs		= (unsigned long)t;
+	int ret = 1;
+
+	switch (entry->mmiorw.opcode) {
+	case MMIO_READ:
+		ret = trace_seq_printf(s,
+			"R %d %lu.%06lu %d 0x%lx 0x%lx 0x%lx %d\n",
+			rw->width, secs, usec_rem, rw->map_id, rw->phys,
+			rw->value, rw->pc, entry->pid);
+		break;
+	case MMIO_WRITE:
+		ret = trace_seq_printf(s,
+			"W %d %lu.%06lu %d 0x%lx 0x%lx 0x%lx %d\n",
+			rw->width, secs, usec_rem, rw->map_id, rw->phys,
+			rw->value, rw->pc, entry->pid);
+		break;
+	case MMIO_UNKNOWN_OP:
+		ret = trace_seq_printf(s,
+			"UNKNOWN %lu.%06lu %d 0x%lx %02x,%02x,%02x 0x%lx %d\n",
+			secs, usec_rem, rw->map_id, rw->phys,
+			(rw->value >> 16) & 0xff, (rw->value >> 8) & 0xff,
+			(rw->value >> 0) & 0xff, rw->pc, entry->pid);
+		break;
+	default:
+		ret = trace_seq_printf(s, "rw what?\n");
+		break;
+	}
+	if (ret)
+		return 1;
+	return 0;
+}
+
+static int mmio_print_map(struct trace_iterator *iter)
+{
+	struct trace_entry *entry = iter->ent;
+	struct mmiotrace_map *m	= &entry->mmiomap;
+	struct trace_seq *s	= &iter->seq;
+	unsigned long long t	= ns2usecs(entry->t);
+	unsigned long usec_rem	= do_div(t, 1000000ULL);
+	unsigned secs		= (unsigned long)t;
+	int ret = 1;
+
+	switch (entry->mmiorw.opcode) {
+	case MMIO_PROBE:
+		ret = trace_seq_printf(s,
+			"MAP %lu.%06lu %d 0x%lx 0x%lx 0x%lx 0x%lx %d\n",
+			secs, usec_rem, m->map_id, m->phys, m->virt, m->len,
+			0UL, entry->pid);
+		break;
+	case MMIO_UNPROBE:
+		ret = trace_seq_printf(s,
+			"UNMAP %lu.%06lu %d 0x%lx %d\n",
+			secs, usec_rem, m->map_id, 0UL, entry->pid);
+		break;
+	default:
+		ret = trace_seq_printf(s, "map what?\n");
+		break;
+	}
+	if (ret)
+		return 1;
+	return 0;
+}
+
+/* return 0 to abort printing without consuming current entry in pipe mode */
+static int mmio_print_line(struct trace_iterator *iter)
+{
+	switch (iter->ent->type) {
+	case TRACE_MMIO_RW:
+		return mmio_print_rw(iter);
+	case TRACE_MMIO_MAP:
+		return mmio_print_map(iter);
+	default:
+		return 1; /* ignore unknown entries */
+	}
 }
 
 static struct tracer mmio_tracer __read_mostly =
@@ -47,38 +149,31 @@ static struct tracer mmio_tracer __read_mostly =
 	.name		= "mmiotrace",
 	.init		= mmio_trace_init,
 	.reset		= mmio_trace_reset,
+	.open		= mmio_print_header,
 	.ctrl_update	= mmio_trace_ctrl_update,
+	.print_line	= mmio_print_line,
 };
 
 __init static int init_mmio_trace(void)
 {
-	int ret = init_mmiotrace();
-	if (ret)
-		return ret;
 	return register_tracer(&mmio_tracer);
 }
 device_initcall(init_mmio_trace);
 
-void mmio_trace_record(u32 type, unsigned long addr, unsigned long arg)
+void mmio_trace_rw(struct mmiotrace_rw *rw)
 {
 	struct trace_array *tr = mmio_trace_array;
 	struct trace_array_cpu *data = tr->data[smp_processor_id()];
-
-	if (!current || current->pid == 0) {
-		/*
-		 * XXX: This is a problem. We need to able to record, no
-		 * matter what. tracing_generic_entry_update() would crash.
-		 */
-		static unsigned limit;
-		if (limit++ < 12)
-			pr_err("Error in %s: no current.\n", __func__);
-		return;
-	}
-	if (!tr || !data) {
-		static unsigned limit;
-		if (limit++ < 12)
-			pr_err("%s: no tr or data\n", __func__);
-		return;
-	}
-	__trace_special(tr, data, type, addr, arg);
+	__trace_mmiotrace_rw(tr, data, rw);
+}
+
+void mmio_trace_mapping(struct mmiotrace_map *map)
+{
+	struct trace_array *tr = mmio_trace_array;
+	struct trace_array_cpu *data;
+
+	preempt_disable();
+	data = tr->data[smp_processor_id()];
+	__trace_mmiotrace_map(tr, data, map);
+	preempt_enable();
 }