mcount and latency tracing utility -v3

RFC PATCH In review Steven Rostedt 2 participants 35 messages Jan 15, 2008

Subsystems (11)

x86 PowerPC Kbuild Library code +7 more subsystems

lore

Share & tools

lore

Steven Rostedt

author reviewer

2008-01-15 20:54 UTC #0 lore raw

v3 cover mcount and latency tracing utility -v3

[
version 3 of mcount patches:

changes include:

Made mcount be able to register more than one function to call. If only one function is registered, then it is called directly. If more than one is registered, then a loop function is called to call all registered functions.

Add schedule context switch tracing.

Added preemption off tracing.

Removed recording of command line of task at each trace, and placed it into the scheduling trace, or at time of max preemption off.

Renamed irqs off files to preempt prefixes.

Suggested by Jan Kiszka:
cleaned up calls to mcount (following glibc more)

always call mcount_trace_function directly from assembly

Suggested by Sam Ravnborg:

Created CONFIG_HAVE_MCOUNT to be selected by archs instead of each are defining a new config option.

Added time keeping fixes by John Stultz
]

All released version of these patches can be found at:

http://people.redhat.com/srostedt/tracing/

The following patch series brings to vanilla Linux a bit of the RT kernel trace facility. This incorporates the "-pg" profiling option of gcc that will call the "mcount" function for all functions called in the kernel.

Note: I did investigate using -finstrument-functions but that adds a call to both start and end of a function. Using mcount only does the beginning of the function. mcount alone adds ~13% overhead. The -finstrument-functions added ~19%. Also it caused me to do tricks with inline, because it adds the function calls to inline functions as well.

This patch series implements the code for x86 (32 and 64 bit), but other archs can easily be implemented as well (note: ARM and PPC are already implemented in -rt)

Some Background:
----------------

A while back, Ingo Molnar and William Lee Irwin III created a latency tracer to find problem latency areas in the kernel for the RT patch. This tracer became a very integral part of the RT kernel in solving where latency hot spots were. One of the features that the latency tracer added was a function trace. This function tracer would record all functions that were called (implemented by the gcc "-pg" option) and would show what was called when interrupts or preemption was turned off.

This feature is also very helpful in normal debugging. So it's been talked about taking bits and pieces from the RT latency tracer and bring them to LKML. But no one had the time to do it.

Arnaldo Carvalho de Melo took a crack at it. He pulled out the mcount as well as part of the tracing code and made it generic from the point of the tracing code. I'm not sure why this stopped. Probably because Arnaldo is a very busy man, and his efforts had to be utilized elsewhere.

While I still maintain my own Logdev utility:

http://rostedt.homelinux.com/logdev

I came across a need to do the mcount with logdev too. I was successful but found that it became very dependent on a lot of code. One thing that I liked about my logdev utility was that it was very non-intrusive, and has been easy to port from the Linux 2.0 days. I did not want to burden the logdev patch with the intrusiveness of mcount (not really that intrusive, it just needs to add a "notrace" annotation to functions in the kernel that will cause more conflicts in applying patches for me).

Being close to the holidays, I grabbed Arnaldos old patches and started massaging them into something that could be useful for logdev, and what I found out (and talking this over with Arnaldo too) that this can be much more useful for others as well.

The main thing I changed, was that I made the mcount function itself generic, and not the dependency on the tracing code. That is I added

register_mcount_function()
and
clear_mcount_function()

So when ever mcount is enabled and a function is registered that function is called for all functions in the kernel that is not labeled with the "notrace" annotation.

The Simple Tracer:
------------------

To show the power of this I also massaged the tracer code that Arnaldo pulled from the RT patch and made it be a nice example of what can be done with this.

The function that is registered to mcount has the prototype:

void func(unsigned long ip, unsigned long parent_ip);

The ip is the address of the function and parent_ip is the address of the parent function that called it.

The x86_64 version has the assembly call the registered function directly to save having to do a double function call.

To enable mcount, a sysctl is added:

/proc/sys/kernel/mcount_enabled

Once mcount is enabled, when a function is registed, it will be called by all functions. The tracer in this patch series shows how this is done. It adds a directory in the debugfs, called mctracer. With a ctrl file that will allow the user have the tracer register its function. Note, the order of enabling mcount and registering a function is not important, but both must be done to initiate the tracing. That is, you can disable tracing by either disabling mcount or by clearing the registered function.

Only one function may be registered at a time. If another function is registered, it will simply override what ever was there previously.

Here's a simple example of the tracer output:

CPU 2: hackbench:11867 preempt_schedule+0xc/0x84 <-- avc_has_perm_noaudit+0x45d/0x52c CPU 1: hackbench:12052 selinux_file_permission+0x10/0x11c <-- security_file_permission+0x16/0x18 CPU 3: hackbench:12017 update_curr+0xe/0x8b <-- put_prev_task_fair+0x24/0x4c
CPU 2: hackbench:11867 avc_audit+0x16/0x9e3 <-- avc_has_perm+0x51/0x63
CPU 0: hackbench:12019 socket_has_perm+0x16/0x7c <-- selinux_socket_sendmsg+0x27/0x3e CPU 1: hackbench:12052 file_has_perm+0x16/0xbb <-- selinux_file_permission+0x104/0x11c

This is formated like:

CPU <CPU#>: <task-comm>:<task-pid> <function> <-- <parent-function>

Latency Tracer Format:
----------------------

The format used by the RT patch is a bit more complex. It is designed to record a lot of information quickly and dump out a lot too.

There's two versions of the format. Verbose and non-vebose.

verbose:

preemption latency trace v1.1.5 on 2.6.24-rc7-tst
--------------------------------------------------------------------
latency: 89 us, #3/3, CPU#1 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4)

    -----------------
    | task: kjournald-600 (uid:0 nice:-5 policy:0 rt_prio:0)
    -----------------

=> started at: _spin_lock_irqsave+0x2a/0x63 <c06310d2>
=> ended at: _spin_unlock_irqrestore+0x32/0x41 <c0631245>

       kjournald   600 1 1 00000000 00000000 [397408f1] 0.003ms (+0.079ms): _spin_lock_irqsave+0x2a/0x63 <c06310d2> (scsi_dispatch_cmd+0x155/0x234 [scsi_mod] <f8867c19>)
       kjournald   600 1 1 00000000 00000001 [39740940] 0.081ms (+0.005ms): _spin_unlock_irqrestore+0x32/0x41 <c0631245> (scsi_dispatch_cmd+0x1be/0x234 [scsi_mod] <f8867c82>)
       kjournald   600 1 1 00000000 00000002 [39740945] 0.087ms (+0.000ms): trace_hardirqs_on_caller+0x74/0x86 <c0508bdc> (_spin_unlock_irqrestore+0x32/0x41 <c0631245>)

non-verbose:

preemption latency trace v1.1.5 on 2.6.24-rc7-tst
--------------------------------------------------------------------
latency: 89 us, #3/3, CPU#2 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4)

    -----------------
    | task: kjournald-600 (uid:0 nice:-5 policy:0 rt_prio:0)
    -----------------

=> started at: _spin_lock_irqsave+0x2a/0x63 <c06310d2>
=> ended at: _spin_unlock_irqrestore+0x32/0x41 <c0631245>

                 _------=> CPU#            
                / _-----=> irqs-off        
               | / _----=> need-resched    
               || / _---=> hardirq/softirq 
               ||| / _--=> preempt-depth   
               |||| /                      
               |||||     delay             
   cmd     pid ||||| time  |   caller      
      \   /    |||||   \   |   /           
kjournal-600   1d...    3us+: _spin_lock_irqsave+0x2a/0x63 <c06310d2> (scsi_dispatch_cmd+0x155/0x234 [scsi_mod] <f8867c19>)
kjournal-600   1d...   81us+: _spin_unlock_irqrestore+0x32/0x41 <c0631245> (scsi_dispatch_cmd+0x1be/0x234 [scsi_mod] <f8867c82>)
kjournal-600   1d...   87us : trace_hardirqs_on_caller+0x74/0x86 <c0508bdc> (_spin_unlock_irqrestore+0x32/0x41 <c0631245>)

Debug FS:
---------

Although enabling and disabling mcount is done through the sysctl:

/proc/sys/kernel/mcount_enabled

The rest of the tracing uses debugfs.

/debugfs/tracing

Here's the available files:

fn_trace_ctrl
echo 1 to this enables mcount tracing (if mcount_enabled is set) echo 0 to disable the function trace tracing.

function_trace
Outputs the function trace in latency_trace format.

preempt_fn_trace_ctrl
echo 1 to enable function tracing in critical sections timings echo 0 to disable

preempt_trace
Outputs the critical section latency

preempt_thresh
echo a number (in usecs) into this to record all traces that are greater than threshold.

iter_ctrl
echo "symonly" to not show the instruction pointers in the trace echo "nosymonly" to disable symonly.
echo "verbose" for verbose output from latency format. echo "noverbose" to disable verbose ouput.
cat iter_ctrl to see the current settings.

preempt_max_latency
Holds the current max critical latency.
echo 0 to reset and start tracing.

trace
simple output format of the function trace code.

Overhead:
---------

Note that having mcount compiled in seems to show a little overhead.

Here's 3 runs of hackbench 50 without the patches:
Time: 2.137
Time: 2.283
Time: 2.245

Avg: 2.221

and here's 3 runs with the patches (without tracing on):
Time: 2.738
Time: 2.469
Time: 2.388

Avg: 2.531

So it is a 13% overhead when enabled (according to hackbench).

But full tracing can cause a bit more problems:

# hackbench 50
Time: 113.350

113.350!!!!!

But this is tracing *every* function call!

Future:
-------
The way the mcount hook is done here, other utilities can easily add their own functions. Just care needs to be made not to call anything that is not marked with notrace, or you will crash the box with recursion. But even the simple tracer adds a "disabled" feature so in case it happens to call something that is not marked with notrace, it is a safety net not to kill the box.

I was originally going to use the relay system to record the data, but that had a chance of calling functions not marked with notrace. But, if for example LTTng wanted to use this, it could disable tracing on a CPU when doing the calls, and this will protect from recusion.

SystemTap:
----------
One thing that Arnaldo and I discussed last year was using systemtap to add hooks into the kernel to start and stop tracing. kprobes is too heavy to do on all funtion calls, but it would be perfect to add to non hot paths to start the tracer and stop the tracer.

So when debugging the kernel, instead of recompiling with printks or other markers, you could simply use systemtap to place a trace start and stop locations and trace the problem areas to see what is happening.

Latency Tracing:
----------------

We can also add trace points to record the time the highest priority task needs to wait before running. This too is currently done in the RT patch.

These are just some of the ideas we have with this. And we are sure others could come up with more.

These patches are for the underlining work. We'll see what happens next.

Steven Rostedt

author reviewer

2008-01-15 21:01 UTC #1 lore raw

RFC PATCH v3 1/30 Add basic support for gcc profiler instrumentation

If CONFIG_MCOUNT is selected and /proc/sys/kernel/mcount_enabled is set to a non-zero value the mcount routine will be called everytime we enter a kernel function that is not marked with the "notrace" attribute.

The mcount routine will then call a registered function if a function happens to be registered.

[This code has been highly hacked by Steven Rostedt, so don't
blame Arnaldo for all of this ;-) ]

Update:
It is now possible to register more than one mcount function. If only one mcount function is registered, that will be the function that mcount calls directly. If more than one function is registered, then mcount will call a function that will loop through the functions to call.

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 Makefile                   |    3 
 arch/x86/Kconfig           |    1 
 arch/x86/kernel/entry_32.S |   22 +++++++
 arch/x86/kernel/entry_64.S |   33 ++++++++++
 include/linux/linkage.h    |    2 
 include/linux/mcount.h     |   38 ++++++++++++
 kernel/sysctl.c            |   11 +++
 lib/Kconfig.debug          |    2 
 lib/Makefile               |    2 
 lib/tracing/Kconfig        |   10 +++
 lib/tracing/Makefile       |    3 
 lib/tracing/mcount.c       |  141 +++++++++++++++++++++++++++++++++++++++++++++
 12 files changed, 268 insertions(+)

Makefile +3 -0

Index: linux-compile.git/Makefile
===================================================================
@@ -509,6 +509,9 @@ endif
 
 include $(srctree)/arch/$(SRCARCH)/Makefile
 
+ifdef CONFIG_MCOUNT
+KBUILD_CFLAGS	+= -pg
+endif
 ifdef CONFIG_FRAME_POINTER
 KBUILD_CFLAGS	+= -fno-omit-frame-pointer -fno-optimize-sibling-calls
 else

arch/x86/Kconfig +1 -0

Index: linux-compile.git/arch/x86/Kconfig
===================================================================
@@ -19,6 +19,7 @@ config X86_64
 config X86
 	bool
 	default y
+	select HAVE_MCOUNT
 
 config GENERIC_TIME
 	bool

arch/x86/kernel/entry_32.S +22 -0

Index: linux-compile.git/arch/x86/kernel/entry_32.S
===================================================================
@@ -75,6 +75,28 @@ DF_MASK		= 0x00000400 
 NT_MASK		= 0x00004000
 VM_MASK		= 0x00020000
 
+#ifdef CONFIG_MCOUNT
+.globl mcount
+mcount:
+	cmpl $0, mcount_enabled
+	jz out
+
+	/* taken from glibc */
+	pushl %eax
+	pushl %ecx
+	pushl %edx
+	movl 0xc(%esp), %edx
+	movl 0x4(%ebp), %eax
+
+	call   *mcount_trace_function
+
+	popl %edx
+	popl %ecx
+	popl %eax
+out:
+	ret
+#endif
+
 #ifdef CONFIG_PREEMPT
 #define preempt_stop(clobbers)	DISABLE_INTERRUPTS(clobbers); TRACE_IRQS_OFF
 #else

arch/x86/kernel/entry_64.S +33 -0

Index: linux-compile.git/arch/x86/kernel/entry_64.S
===================================================================
@@ -53,6 +53,39 @@
 
 	.code64
 
+#ifdef CONFIG_MCOUNT
+
+ENTRY(mcount)
+	cmpl $0, mcount_enabled
+	jz out
+
+	/* taken from glibc */
+	subq $0x38, %rsp
+	movq %rax, (%rsp)
+	movq %rcx, 8(%rsp)
+	movq %rdx, 16(%rsp)
+	movq %rsi, 24(%rsp)
+	movq %rdi, 32(%rsp)
+	movq %r8, 40(%rsp)
+	movq %r9, 48(%rsp)
+
+	movq 0x38(%rsp), %rsi
+	movq 8(%rbp), %rdi
+
+	call   *mcount_trace_function
+
+	movq 48(%rsp), %r9
+	movq 40(%rsp), %r8
+	movq 32(%rsp), %rdi
+	movq 24(%rsp), %rsi
+	movq 16(%rsp), %rdx
+	movq 8(%rsp), %rcx
+	movq (%rsp), %rax
+	addq $0x38, %rsp
+out:
+	retq
+#endif
+
 #ifndef CONFIG_PREEMPT
 #define retint_kernel retint_restore_args
 #endif

include/linux/linkage.h +2 -0

Index: linux-compile.git/include/linux/linkage.h
===================================================================
@@ -3,6 +3,8 @@
 
 #include <asm/linkage.h>
 
+#define notrace __attribute__((no_instrument_function))
+
 #ifdef __cplusplus
 #define CPP_ASMLINKAGE extern "C"
 #else

include/linux/mcount.h new file +38 -0

Index: linux-compile.git/include/linux/mcount.h
===================================================================
@@ -0,0 +1,38 @@
+#ifndef _LINUX_MCOUNT_H
+#define _LINUX_MCOUNT_H
+
+#ifdef CONFIG_MCOUNT
+extern int mcount_enabled;
+
+#include <linux/linkage.h>
+
+#define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
+#define CALLER_ADDR1 ((unsigned long)__builtin_return_address(1))
+#define CALLER_ADDR2 ((unsigned long)__builtin_return_address(2))
+
+typedef void (*mcount_func_t)(unsigned long ip, unsigned long parent_ip);
+
+struct mcount_ops {
+	mcount_func_t func;
+	struct mcount_ops *next;
+};
+
+/*
+ * The mcount_ops must be a static and should also
+ * be read_mostly.  These functions do modify read_mostly variables
+ * so use them sparely. Never free an mcount_op or modify the
+ * next pointer after it has been registered. Even after unregistering
+ * it, the next pointer may still be used internally.
+ */
+int register_mcount_function(struct mcount_ops *ops);
+int unregister_mcount_function(struct mcount_ops *ops);
+void clear_mcount_function(void);
+
+extern void mcount(void);
+
+#else /* !CONFIG_MCOUNT */
+# define register_mcount_function(ops) do { } while (0)
+# define unregister_mcount_function(ops) do { } while (0)
+# define clear_mcount_function(ops) do { } while (0)
+#endif /* CONFIG_MCOUNT */
+#endif /* _LINUX_MCOUNT_H */

kernel/sysctl.c +11 -0

Index: linux-compile.git/kernel/sysctl.c
===================================================================
@@ -46,6 +46,7 @@
 #include <linux/nfs_fs.h>
 #include <linux/acpi.h>
 #include <linux/reboot.h>
+#include <linux/mcount.h>
 
 #include <asm/uaccess.h>
 #include <asm/processor.h>

@@ -470,6 +471,16 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+#ifdef CONFIG_MCOUNT
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "mcount_enabled",
+		.data		= &mcount_enabled,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+#endif
 #ifdef CONFIG_KMOD
 	{
 		.ctl_name	= KERN_MODPROBE,

lib/Kconfig.debug +2 -0

Index: linux-compile.git/lib/Kconfig.debug
===================================================================
@@ -517,4 +517,6 @@ config FAULT_INJECTION_STACKTRACE_FILTER
 	help
 	  Provide stacktrace filter for fault-injection capabilities
 
+source lib/tracing/Kconfig
+
 source "samples/Kconfig"

lib/Makefile +2 -0

Index: linux-compile.git/lib/Makefile
===================================================================
@@ -66,6 +66,8 @@ obj-$(CONFIG_AUDIT_GENERIC) += audit.o
 obj-$(CONFIG_SWIOTLB) += swiotlb.o
 obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o
 
+obj-$(CONFIG_MCOUNT) += tracing/
+
 lib-$(CONFIG_GENERIC_BUG) += bug.o
 
 hostprogs-y	:= gen_crc32table

lib/tracing/Kconfig new file +10 -0

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -0,0 +1,10 @@
+
+# Archs that enable MCOUNT should select HAVE_MCOUNT
+config HAVE_MCOUNT
+       bool
+
+# MCOUNT itself is useless, or will just be added overhead.
+# It needs something to register a function with it.
+config MCOUNT
+	bool
+	select FRAME_POINTER

lib/tracing/Makefile new file +3 -0

Index: linux-compile.git/lib/tracing/Makefile
===================================================================
@@ -0,0 +1,3 @@
+obj-$(CONFIG_MCOUNT) += libmcount.o
+
+libmcount-y := mcount.o

lib/tracing/mcount.c new file +141 -0

141 lines not on this page · raw message

Steven Rostedt

author reviewer

2008-01-15 20:55 UTC #2 lore raw

RFC PATCH v3 2/30 Annotate core code that should not be traced

In reply to Steven Rostedt

Mark with "notrace" functions in core code that should not be traced. The "notrace" attribute will prevent gcc from adding a call to mcount on the annotated funtions.

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/smp_processor_id.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

lib/smp_processor_id.c +1 -1

Index: linux-compile.git/lib/smp_processor_id.c
===================================================================
@@ -7,7 +7,7 @@
 #include <linux/kallsyms.h>
 #include <linux/sched.h>
 
-unsigned int debug_smp_processor_id(void)
+notrace unsigned int debug_smp_processor_id(void)
 {
 	unsigned long preempt_count = preempt_count();
 	int this_cpu = raw_smp_processor_id();

Steven Rostedt

author reviewer

2008-01-15 20:52 UTC #3 lore raw

RFC PATCH v3 3/30 x86_64: notrace annotations

In reply to Steven Rostedt

Add "notrace" annotation to x86_64 specific files.

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/head64.c     |    2 +-
 arch/x86/kernel/setup64.c    |    4 ++--
 arch/x86/kernel/smpboot_64.c |    2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

arch/x86/kernel/head64.c +1 -1

Index: linux-compile.git/arch/x86/kernel/head64.c
===================================================================
@@ -46,7 +46,7 @@ static void __init copy_bootdata(char *r
 	}
 }
 
-void __init x86_64_start_kernel(char * real_mode_data)
+notrace void __init x86_64_start_kernel(char *real_mode_data)
 {
 	int i;

arch/x86/kernel/setup64.c +2 -2

Index: linux-compile.git/arch/x86/kernel/setup64.c
===================================================================
@@ -114,7 +114,7 @@ void __init setup_per_cpu_areas(void)
 	}
 } 
 
-void pda_init(int cpu)
+notrace void pda_init(int cpu)
 { 
 	struct x8664_pda *pda = cpu_pda(cpu);

@@ -197,7 +197,7 @@ DEFINE_PER_CPU(struct orig_ist, orig_ist
  * 'CPU state barrier', nothing should get across.
  * A lot of state is already set up in PDA init.
  */
-void __cpuinit cpu_init (void)
+notrace void __cpuinit cpu_init(void)
 {
 	int cpu = stack_smp_processor_id();
 	struct tss_struct *t = &per_cpu(init_tss, cpu);

arch/x86/kernel/smpboot_64.c +1 -1

Index: linux-compile.git/arch/x86/kernel/smpboot_64.c
===================================================================
@@ -317,7 +317,7 @@ static inline void set_cpu_sibling_map(i
 /*
  * Setup code on secondary processor (after comming out of the trampoline)
  */
-void __cpuinit start_secondary(void)
+notrace __cpuinit void start_secondary(void)
 {
 	/*
 	 * Dont put anything before smp_callin(), SMP

Steven Rostedt

author reviewer

2008-01-15 20:57 UTC #4 lore raw

RFC PATCH v3 4/30 add notrace annotations to vsyscall.

In reply to Steven Rostedt

Add the notrace annotations to some of the vsyscall functions.

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/vsyscall_64.c  |    3 ++-
 arch/x86/vdso/vclock_gettime.c |   15 ++++++++-------
 arch/x86/vdso/vgetcpu.c        |    3 ++-
 include/asm-x86/vsyscall.h     |    3 ++-
 4 files changed, 14 insertions(+), 10 deletions(-)

arch/x86/kernel/vsyscall_64.c +2 -1

Index: linux-compile.git/arch/x86/kernel/vsyscall_64.c
===================================================================
@@ -42,7 +42,8 @@
 #include <asm/topology.h>
 #include <asm/vgtod.h>
 
-#define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
+#define __vsyscall(nr) \
+		__attribute__ ((unused, __section__(".vsyscall_" #nr))) notrace
 #define __syscall_clobber "r11","rcx","memory"
 #define __pa_vsymbol(x)			\
 	({unsigned long v;  		\

arch/x86/vdso/vclock_gettime.c +8 -7

Index: linux-compile.git/arch/x86/vdso/vclock_gettime.c
===================================================================
@@ -24,7 +24,7 @@
 
 #define gtod vdso_vsyscall_gtod_data
 
-static long vdso_fallback_gettime(long clock, struct timespec *ts)
+notrace static long vdso_fallback_gettime(long clock, struct timespec *ts)
 {
 	long ret;
 	asm("syscall" : "=a" (ret) :

@@ -32,7 +32,7 @@ static long vdso_fallback_gettime(long c
 	return ret;
 }
 
-static inline long vgetns(void)
+notrace static inline long vgetns(void)
 {
 	long v;
 	cycles_t (*vread)(void);

@@ -41,7 +41,7 @@ static inline long vgetns(void)
 	return (v * gtod->clock.mult) >> gtod->clock.shift;
 }
 
-static noinline int do_realtime(struct timespec *ts)
+notrace static noinline int do_realtime(struct timespec *ts)
 {
 	unsigned long seq, ns;
 	do {

@@ -55,7 +55,8 @@ static noinline int do_realtime(struct t
 }
 
 /* Copy of the version in kernel/time.c which we cannot directly access */
-static void vset_normalized_timespec(struct timespec *ts, long sec, long nsec)
+notrace static void
+vset_normalized_timespec(struct timespec *ts, long sec, long nsec)
 {
 	while (nsec >= NSEC_PER_SEC) {
 		nsec -= NSEC_PER_SEC;

@@ -69,7 +70,7 @@ static void vset_normalized_timespec(str
 	ts->tv_nsec = nsec;
 }
 
-static noinline int do_monotonic(struct timespec *ts)
+notrace static noinline int do_monotonic(struct timespec *ts)
 {
 	unsigned long seq, ns, secs;
 	do {

@@ -83,7 +84,7 @@ static noinline int do_monotonic(struct 
 	return 0;
 }
 
-int __vdso_clock_gettime(clockid_t clock, struct timespec *ts)
+notrace int __vdso_clock_gettime(clockid_t clock, struct timespec *ts)
 {
 	if (likely(gtod->sysctl_enabled && gtod->clock.vread))
 		switch (clock) {

@@ -97,7 +98,7 @@ int __vdso_clock_gettime(clockid_t clock
 int clock_gettime(clockid_t, struct timespec *)
 	__attribute__((weak, alias("__vdso_clock_gettime")));
 
-int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz)
+notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz)
 {
 	long ret;
 	if (likely(gtod->sysctl_enabled && gtod->clock.vread)) {

arch/x86/vdso/vgetcpu.c +2 -1

Index: linux-compile.git/arch/x86/vdso/vgetcpu.c
===================================================================
@@ -13,7 +13,8 @@
 #include <asm/vgtod.h>
 #include "vextern.h"
 
-long __vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused)
+notrace long
+__vdso_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *unused)
 {
 	unsigned int dummy, p;

include/asm-x86/vsyscall.h +2 -1

Index: linux-compile.git/include/asm-x86/vsyscall.h
===================================================================
@@ -24,7 +24,8 @@ enum vsyscall_num {
 	((unused, __section__ (".vsyscall_gtod_data"),aligned(16)))
 #define __section_vsyscall_clock __attribute__ \
 	((unused, __section__ (".vsyscall_clock"),aligned(16)))
-#define __vsyscall_fn __attribute__ ((unused,__section__(".vsyscall_fn")))
+#define __vsyscall_fn \
+	__attribute__ ((unused, __section__(".vsyscall_fn"))) notrace
 
 #define VGETCPU_RDTSCP	1
 #define VGETCPU_LSL	2

Steven Rostedt

author reviewer

2008-01-15 20:54 UTC #5 lore raw

RFC PATCH v3 5/30 add notrace annotations for NMI routines

In reply to Steven Rostedt

This annotates NMI functions with notrace. Some tracers may be able to live with this, but some cannot. So we turn off NMI tracing.

One solution might be to make a notrace_nmi which would only turn off NMI tracing if a trace utility needed it off.

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/nmi_32.c   |    2 +-
 arch/x86/kernel/nmi_64.c   |    2 +-
 arch/x86/kernel/traps_32.c |    4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)

arch/x86/kernel/nmi_32.c +1 -1

Index: linux-compile.git/arch/x86/kernel/nmi_32.c
===================================================================
@@ -323,7 +323,7 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-__kprobes int nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
+notrace __kprobes int nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 {
 
 	/*

arch/x86/kernel/nmi_64.c +1 -1

Index: linux-compile.git/arch/x86/kernel/nmi_64.c
===================================================================
@@ -314,7 +314,7 @@ void touch_nmi_watchdog(void)
  	touch_softlockup_watchdog();
 }
 
-int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
+notrace __kprobes int nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;

arch/x86/kernel/traps_32.c +2 -2

Index: linux-compile.git/arch/x86/kernel/traps_32.c
===================================================================
@@ -722,7 +722,7 @@ void __kprobes die_nmi(struct pt_regs *r
 	do_exit(SIGSEGV);
 }
 
-static __kprobes void default_do_nmi(struct pt_regs * regs)
+static notrace __kprobes void default_do_nmi(struct pt_regs *regs)
 {
 	unsigned char reason = 0;

@@ -762,7 +762,7 @@ static __kprobes void default_do_nmi(str
 
 static int ignore_nmis;
 
-fastcall __kprobes void do_nmi(struct pt_regs * regs, long error_code)
+notrace fastcall __kprobes void do_nmi(struct pt_regs *regs, long error_code)
 {
 	int cpu;

Steven Rostedt

author reviewer

2008-01-15 20:56 UTC #6 lore raw

RFC PATCH v3 6/30 mcount based trace in the form of a header file library

In reply to Steven Rostedt

The design is for mcount based tracers to be added thru the lib/tracing/tracer_interface.h file, just like mcount users should add themselves to lib/tracing/mcount.h. A Kconfig rule chooses the right MCOUNT and MCOUNT_TRACER user.

This is to avoid function call costs for something that is supposed to be used only in a debug kernel and that has to reduce to the bare minimum the per function call overhead of mcount based tracing.

Signed-off-by Arnaldo Carvalho de Melo <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/Kconfig            |   11 +++
 lib/tracing/Makefile           |    2 
 lib/tracing/tracer.c           |  128 +++++++++++++++++++++++++++++++++++++++++
 lib/tracing/tracer.h           |   21 ++++++
 lib/tracing/tracer_interface.h |   14 ++++
 5 files changed, 176 insertions(+)

lib/tracing/Kconfig +11 -0

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -8,3 +8,14 @@ config HAVE_MCOUNT
 config MCOUNT
 	bool
 	select FRAME_POINTER
+
+config MCOUNT_TRACER
+	bool "Profiler instrumentation based tracer"
+	depends on DEBUG_KERNEL && HAVE_MCOUNT
+	default n
+	select MCOUNT
+	help
+	  Use profiler instrumentation, adding -pg to CFLAGS. This will
+	  insert a call to an architecture specific __mcount routine,
+	  that the debugging mechanism using this facility will hook by
+	  providing a set of inline routines.

lib/tracing/Makefile +2 -0

Index: linux-compile.git/lib/tracing/Makefile
===================================================================
@@ -1,3 +1,5 @@
 obj-$(CONFIG_MCOUNT) += libmcount.o
 
+obj-$(CONFIG_MCOUNT_TRACER) += tracer.o
+
 libmcount-y := mcount.o

lib/tracing/tracer.c new file +128 -0

128 lines not on this page · raw message

lib/tracing/tracer.h new file +21 -0

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -0,0 +1,21 @@
+#ifndef _LINUX_MCOUNT_TRACER_H
+#define _LINUX_MCOUNT_TRACER_H
+
+#include <asm/atomic.h>
+
+struct mctracer_entry {
+	unsigned long idx;
+	unsigned long ip;
+	unsigned long parent_ip;
+};
+
+struct mctracer_trace {
+	void	      *trace[NR_CPUS];
+	unsigned long trace_idx[NR_CPUS];
+	unsigned long entries;
+	atomic_t      cnt;
+	atomic_t      disabled[NR_CPUS];
+	atomic_t      underrun[NR_CPUS];
+};
+
+#endif /* _LINUX_MCOUNT_TRACER_H */

lib/tracing/tracer_interface.h new file +14 -0

Index: linux-compile.git/lib/tracing/tracer_interface.h
===================================================================
@@ -0,0 +1,14 @@
+#ifndef _LINUX_MCTRACER_INTERFACE_H
+#define _LINUX_MCTRACER_INTERFACE_H
+
+#include "tracer.h"
+
+/*
+ * Will be at least sizeof(struct mctracer_entry), but callers can request more
+ * space for private stuff, such as a timestamp, preempt_count, etc.
+ */
+#define MCTRACER_ENTRY_SIZE sizeof(struct mctracer_entry)
+
+#define MCTRACER_NR_ENTRIES (65536UL)
+
+#endif /* _LINUX_MCTRACER_INTERFACE_H */

Steven Rostedt

author reviewer

2008-01-15 20:56 UTC #7 lore raw

RFC PATCH v3 7/30 tracer add debugfs interface

In reply to Steven Rostedt

This patch adds an interface into debugfs.

/debugfs/tracing/ctrl

echoing 1 into the ctrl file turns on the tracer,
and echoing 0 turns it off.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |   87 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 lib/tracing/tracer.h |    1 
 2 files changed, 87 insertions(+), 1 deletion(-)

lib/tracing/tracer.c +86 -1

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -15,6 +15,8 @@
 #include <linux/init.h>
 #include <linux/linkage.h>
 #include <linux/seq_file.h>
+#include <linux/debugfs.h>
+#include <linux/uaccess.h>
 #include <linux/mcount.h>
 
 #include "tracer.h"

@@ -58,9 +60,12 @@ static notrace void trace_function(const
 	int cpu;
 
 	raw_local_irq_save(flags);
-	cpu = raw_smp_processor_id();
 
 	tr = &mctracer_trace;
+	if (!tr->ctrl)
+		goto out;
+
+	cpu = raw_smp_processor_id();
 
 	atomic_inc(&tr->disabled[cpu]);
 	if (likely(atomic_read(&tr->disabled[cpu]) == 1))

@@ -68,6 +73,7 @@ static notrace void trace_function(const
 
 	atomic_dec(&tr->disabled[cpu]);
 
+ out:
 	raw_local_irq_restore(flags);
 }

@@ -76,6 +82,83 @@ static struct mcount_ops trace_ops __rea
 	.func = trace_function,
 };
 
+#ifdef CONFIG_DEBUG_FS
+static int mctracer_open_generic(struct inode *inode, struct file *filp)
+{
+	filp->private_data = inode->i_private;
+	return 0;
+}
+
+
+static ssize_t mctracer_ctrl_read(struct file *filp, char __user *ubuf,
+				  size_t cnt, loff_t *ppos)
+{
+	struct mctracer_trace *tr = filp->private_data;
+	char buf[16];
+	int r;
+
+	r = sprintf(buf, "%ld\n", tr->ctrl);
+	return simple_read_from_buffer(ubuf, cnt, ppos,
+				       buf, r);
+}
+
+static ssize_t mctracer_ctrl_write(struct file *filp,
+				   const char __user *ubuf,
+				   size_t cnt, loff_t *ppos)
+{
+	struct mctracer_trace *tr = filp->private_data;
+	long val;
+	char buf[16];
+
+	if (cnt > 15)
+		cnt = 15;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	buf[cnt] = 0;
+
+	val = !!simple_strtoul(buf, NULL, 10);
+
+	tr->ctrl = val;
+
+	filp->f_pos += cnt;
+
+	return cnt;
+}
+
+static struct file_operations mctracer_ctrl_fops = {
+	.open = mctracer_open_generic,
+	.read = mctracer_ctrl_read,
+	.write = mctracer_ctrl_write,
+};
+
+static void mctrace_init_debugfs(void)
+{
+	struct dentry *d_mctracer;
+	struct dentry *entry;
+
+	d_mctracer = debugfs_create_dir("tracing", NULL);
+	if (!d_mctracer) {
+		pr_warning("Could not create debugfs directory mctracer\n");
+		return;
+	}
+
+	entry = debugfs_create_file("ctrl", 0644, d_mctracer,
+				    &mctracer_trace, &mctracer_ctrl_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs 'ctrl' entry\n");
+}
+#else /* CONFIG_DEBUG_FS */
+static void mctrace_init_debugfs(void)
+{
+	/*
+	 * No way to turn on or off the trace function
+	 * without debugfs.
+	 */
+}
+#endif /* CONFIG_DEBUG_FS */
+
 static notrace int page_order(const unsigned long size)
 {
 	const unsigned long nr_pages = DIV_ROUND_UP(size, PAGE_SIZE);

@@ -112,6 +195,8 @@ static notrace int mctracer_alloc_buffer
 
 	register_mcount_function(&trace_ops);
 
+	mctrace_init_debugfs();
+
 	return 0;
 
  free_buffers:

lib/tracing/tracer.h +1 -0

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -13,6 +13,7 @@ struct mctracer_trace {
 	void	      *trace[NR_CPUS];
 	unsigned long trace_idx[NR_CPUS];
 	unsigned long entries;
+	long	      ctrl;
 	atomic_t      cnt;
 	atomic_t      disabled[NR_CPUS];
 	atomic_t      underrun[NR_CPUS];

Steven Rostedt

author reviewer

2008-01-15 20:51 UTC #8 lore raw

RFC PATCH v3 8/30 mcount tracer output file

In reply to Steven Rostedt

Add /debugfs/tracing/trace to output trace output.

Here's an example of the content.

  CPU 0:  [<ffffffff80494691>] notifier_call_chain+0x16/0x60 <-- [<ffffffff80494701>] __atomic_notifier_call_chain+0x26/0x56
  CPU 0:  [<ffffffff802161c8>] mce_idle_callback+0x9/0x2f <-- [<ffffffff804946b3>] notifier_call_chain+0x38/0x60
  CPU 0:  [<ffffffff8037fb7a>] acpi_processor_idle+0x16/0x518 <-- [<ffffffff8020aee8>] cpu_idle+0xa1/0xe7
  CPU 0:  [<ffffffff8037fa98>] acpi_safe_halt+0x9/0x43 <-- [<ffffffff8037fd3a>] acpi_processor_idle+0x1d6/0x518
  CPU 1:  [<ffffffff80221db8>] smp_apic_timer_interrupt+0xc/0x58 <-- [<ffffffff8020cf06>] apic_timer_interrupt+0x66/0x70
  CPU 1:  [<ffffffff8020ac22>] exit_idle+0x9/0x22 <-- [<ffffffff80221de1>] smp_apic_timer_interrupt+0x35/0x58
  CPU 1:  [<ffffffff8020ab97>] __exit_idle+0x9/0x2e <-- [<ffffffff8020ac39>] exit_idle+0x20/0x22
  CPU 1:  [<ffffffff8049473a>] atomic_notifier_call_chain+0x9/0x16 <-- [<ffffffff8020abba>] __exit_idle+0x2c/0x2e
  CPU 1:  [<ffffffff804946e9>] __atomic_notifier_call_chain+0xe/0x56 <-- [<ffffffff80494745>] atomic_notifier_call_chain+0x14/0x16
  CPU 1:  [<ffffffff80494691>] notifier_call_chain+0x16/0x60 <-- [<ffffffff80494701>] __atomic_notifier_call_chain+0x26/0x56
  CPU 1:  [<ffffffff802161c8>] mce_idle_callback+0x9/0x2f <-- [<ffffffff804946b3>] notifier_call_chain+0x38/0x60

This is in the format of the output when KALLSYMS is defined.

CPU <CPU#>: [<IP>] <func> <-- [<Parent-IP>] <parent-func>

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |  217 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 215 insertions(+), 2 deletions(-)

lib/tracing/tracer.c +215 -2

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -13,9 +13,11 @@
 #include <linux/fs.h>
 #include <linux/gfp.h>
 #include <linux/init.h>
+#include <linux/module.h>
 #include <linux/linkage.h>
 #include <linux/seq_file.h>
 #include <linux/debugfs.h>
+#include <linux/kallsyms.h>
 #include <linux/uaccess.h>
 #include <linux/mcount.h>

@@ -23,6 +25,7 @@
 #include "tracer_interface.h"
 
 static struct mctracer_trace mctracer_trace;
+static int trace_enabled __read_mostly;
 
 static inline notrace void
 mctracer_add_trace_entry(struct mctracer_trace *tr,

@@ -62,7 +65,7 @@ static notrace void trace_function(const
 	raw_local_irq_save(flags);
 
 	tr = &mctracer_trace;
-	if (!tr->ctrl)
+	if (!trace_enabled)
 		goto out;
 
 	cpu = raw_smp_processor_id();

@@ -83,6 +86,205 @@ static struct mcount_ops trace_ops __rea
 };
 
 #ifdef CONFIG_DEBUG_FS
+struct mctracer_iterator {
+	struct mctracer_trace *tr;
+	struct mctracer_entry *ent;
+	unsigned long next_idx[NR_CPUS];
+	int cpu;
+	int idx;
+};
+
+static struct mctracer_entry *mctracer_entry_idx(struct mctracer_trace *tr,
+						 unsigned long idx,
+						 int cpu)
+{
+	struct mctracer_entry *array = tr->trace[cpu];
+	unsigned long underrun;
+
+	if (idx >= tr->entries)
+		return NULL;
+
+	underrun = atomic_read(&tr->underrun[cpu]);
+	if (underrun)
+		idx = ((underrun - 1) + idx) % tr->entries;
+	else if (idx >= tr->trace_idx[cpu])
+		return NULL;
+
+	return &array[idx];
+}
+
+static void *find_next_entry(struct mctracer_iterator *iter)
+{
+	struct mctracer_trace *tr = iter->tr;
+	struct mctracer_entry *ent;
+	struct mctracer_entry *next = NULL;
+	int next_i = -1;
+	int i;
+
+	for_each_possible_cpu(i) {
+		if (!tr->trace[i])
+			continue;
+		ent = mctracer_entry_idx(tr, iter->next_idx[i], i);
+		if (ent && (!next || next->idx > ent->idx)) {
+			next = ent;
+			next_i = i;
+		}
+	}
+	if (next) {
+		iter->next_idx[next_i]++;
+		iter->idx++;
+	}
+	iter->ent = next;
+	iter->cpu = next_i;
+
+	return next ? iter : NULL;
+}
+
+static void *s_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct mctracer_iterator *iter = m->private;
+	void *ent;
+	int i = (int)*pos;
+
+	(*pos)++;
+
+	/* can't go backwards */
+	if (iter->idx > i)
+		return NULL;
+
+	if (iter->idx < 0)
+		ent = find_next_entry(iter);
+	else
+		ent = iter;
+
+	while (ent && iter->idx < i)
+		ent = find_next_entry(iter);
+
+	return ent;
+}
+
+static void *s_start(struct seq_file *m, loff_t *pos)
+{
+	struct mctracer_iterator *iter = m->private;
+	void *p = NULL;
+	loff_t l = 0;
+	int i;
+
+	iter->ent = NULL;
+	iter->cpu = 0;
+	iter->idx = -1;
+
+	for (i = 0; i < NR_CPUS; i++)
+		iter->next_idx[i] = 0;
+
+	/* stop the trace while dumping */
+	if (iter->tr->ctrl)
+		trace_enabled = 0;
+
+	for (p = iter; p && l < *pos; p = s_next(m, p, &l))
+		;
+
+	return p;
+}
+
+static void s_stop(struct seq_file *m, void *p)
+{
+	struct mctracer_iterator *iter = m->private;
+	if (iter->tr->ctrl)
+		trace_enabled = 1;
+}
+
+#ifdef CONFIG_KALLSYMS
+static void seq_print_symbol(struct seq_file *m,
+			     const char *fmt, unsigned long address)
+{
+	char buffer[KSYM_SYMBOL_LEN];
+
+	sprint_symbol(buffer, address);
+	seq_printf(m, fmt, buffer);
+}
+#else
+# define seq_print_symbol(m, fmt, address) do { } while (0)
+#endif
+
+#ifndef CONFIG_64BIT
+# define IP_FMT "%08lx"
+#else
+# define IP_FMT "%016lx"
+#endif
+
+static void notrace seq_print_ip_sym(struct seq_file *m,
+				     unsigned long ip)
+{
+	seq_print_symbol(m, "%s", ip);
+	seq_printf(m, " <" IP_FMT ">", ip);
+}
+
+static int s_show(struct seq_file *m, void *v)
+{
+	struct mctracer_iterator *iter = v;
+
+	if (iter->ent == NULL) {
+		seq_printf(m, "mctracer:\n");
+	} else {
+		seq_printf(m, "  CPU %d:  ", iter->cpu);
+		seq_print_ip_sym(m, iter->ent->ip);
+		if (iter->ent->parent_ip) {
+			seq_printf(m, " <-- ");
+			seq_print_ip_sym(m, iter->ent->parent_ip);
+		}
+		seq_printf(m, "\n");
+	}
+
+	return 0;
+}
+
+static struct seq_operations mctrace_seq_ops = {
+	.start = s_start,
+	.next = s_next,
+	.stop = s_stop,
+	.show = s_show,
+};
+
+static int mctrace_open(struct inode *inode, struct file *file)
+{
+	struct mctracer_iterator *iter;
+	int ret;
+
+	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
+	if (!iter)
+		return -ENOMEM;
+
+	iter->tr = &mctracer_trace;
+
+	/* TODO stop tracer */
+	ret = seq_open(file, &mctrace_seq_ops);
+	if (!ret) {
+		struct seq_file *m = file->private_data;
+		m->private = iter;
+	} else
+		kfree(iter);
+
+	return ret;
+}
+
+int mctrace_release(struct inode *inode, struct file *file)
+{
+	struct seq_file *m = (struct seq_file *)file->private_data;
+	struct mctracer_iterator *iter = m->private;
+
+	seq_release(inode, file);
+	kfree(iter);
+	return 0;
+}
+
+static struct file_operations mctrace_fops = {
+	.open = mctrace_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = mctrace_release,
+};
+
 static int mctracer_open_generic(struct inode *inode, struct file *filp)
 {
 	filp->private_data = inode->i_private;

@@ -120,7 +322,13 @@ static ssize_t mctracer_ctrl_write(struc
 
 	val = !!simple_strtoul(buf, NULL, 10);
 
-	tr->ctrl = val;
+	if (tr->ctrl ^ val) {
+		if (val)
+			trace_enabled = 1;
+		else
+			trace_enabled = 0;
+		tr->ctrl = val;
+	}
 
 	filp->f_pos += cnt;

@@ -148,6 +356,11 @@ static void mctrace_init_debugfs(void)
 				    &mctracer_trace, &mctracer_ctrl_fops);
 	if (!entry)
 		pr_warning("Could not create debugfs 'ctrl' entry\n");
+
+	entry = debugfs_create_file("trace", 0444, d_mctracer,
+				    &mctracer_trace, &mctrace_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs 'trace' entry\n");
 }
 #else /* CONFIG_DEBUG_FS */
 static void mctrace_init_debugfs(void)

Steven Rostedt

author reviewer

2008-01-15 20:51 UTC #9 lore raw

RFC PATCH v3 9/30 mcount tracer show task comm and pid

In reply to Steven Rostedt

This adds the task comm and pid to the trace output. This gives the output like:

CPU 0: sshd:2605 [<ffffffff80251858>] remove_wait_queue+0xc/0x4a <-- [<ffffffff802ad7be>] free_poll_entry+0x1e/0x2a CPU 2: bash:2610 [<ffffffff8038c3aa>] tty_check_change+0x9/0xb6 <-- [<ffffffff8038d295>] tty_ioctl+0x59f/0xcdd CPU 0: sshd:2605 [<ffffffff80491ec6>] _spin_lock_irqsave+0xe/0x81 <-- [<ffffffff80251863>] remove_wait_queue+0x17/0x4a CPU 2: bash:2610 [<ffffffff8024e2f7>] find_vpid+0x9/0x24 <-- [<ffffffff8038d325>] tty_ioctl+0x62f/0xcdd CPU 0: sshd:2605 [<ffffffff804923ec>] _spin_unlock_irqrestore+0x9/0x3a <-- [<ffffffff80251891>] remove_wait_queue+0x45/0x4a CPU 0: sshd:2605 [<ffffffff802a18b3>] fput+0x9/0x1b <-- [<ffffffff802ad7c6>] free_poll_entry+0x26/0x2a

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |    6 +++++-
 lib/tracing/tracer.h |    3 +++
 2 files changed, 8 insertions(+), 1 deletion(-)

lib/tracing/tracer.c +5 -1

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -35,6 +35,7 @@ mctracer_add_trace_entry(struct mctracer
 {
 	unsigned long idx, idx_next;
 	struct mctracer_entry *entry;
+	struct task_struct *tsk = current;
 
 	idx = tr->trace_idx[cpu];
 	idx_next = idx + 1;

@@ -53,6 +54,8 @@ mctracer_add_trace_entry(struct mctracer
 	entry->idx	 = atomic_inc_return(&tr->cnt);
 	entry->ip	 = ip;
 	entry->parent_ip = parent_ip;
+	entry->pid	 = tsk->pid;
+	memcpy(entry->comm, tsk->comm, TASK_COMM_LEN);
 }
 
 static notrace void trace_function(const unsigned long ip,

@@ -227,7 +230,8 @@ static int s_show(struct seq_file *m, vo
 	if (iter->ent == NULL) {
 		seq_printf(m, "mctracer:\n");
 	} else {
-		seq_printf(m, "  CPU %d:  ", iter->cpu);
+		seq_printf(m, "CPU %d: ", iter->cpu);
+		seq_printf(m, "%s:%d ", iter->ent->comm, iter->ent->pid);
 		seq_print_ip_sym(m, iter->ent->ip);
 		if (iter->ent->parent_ip) {
 			seq_printf(m, " <-- ");

lib/tracing/tracer.h +3 -0

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -2,11 +2,14 @@
 #define _LINUX_MCOUNT_TRACER_H
 
 #include <asm/atomic.h>
+#include <linux/sched.h>
 
 struct mctracer_entry {
 	unsigned long idx;
 	unsigned long ip;
 	unsigned long parent_ip;
+	char comm[TASK_COMM_LEN];
+	pid_t pid;
 };
 
 struct mctracer_trace {

Steven Rostedt

author reviewer

2008-01-15 20:57 UTC #10 lore raw

RFC PATCH v3 10/30 Add a symbol only trace output

In reply to Steven Rostedt

The trace output is very verbose with outputing both the IP address (Instruction Pointer not Internet Protocol!) and the kallsyms symbol. So if kallsyms is configured into the kernel, another file is created in the debugfs system.

A new file is added to the debugfs. iter_ctrl

echo symonly > /debugfs/tracing/iter_ctrl

will turn off printing of instruction pointers.

echo nosymonly > /debugfs/tracing/iter_ctrl

will turn them back on.

Here's an example:

CPU 1: swapper:0 smp_apic_timer_interrupt+0xc/0x58 <-- apic_timer_interrupt+0x66/0x70 CPU 1: swapper:0 exit_idle+0x9/0x22 <-- smp_apic_timer_interrupt+0x35/0x58
CPU 0: sshd:2611 _spin_unlock+0x9/0x38 <-- __qdisc_run+0xb2/0x1a1
CPU 1: swapper:0 __exit_idle+0x9/0x2e <-- exit_idle+0x20/0x22
CPU 0: sshd:2611 _spin_lock+0xe/0x7a <-- __qdisc_run+0xba/0x1a1
CPU 1: swapper:0 atomic_notifier_call_chain+0x9/0x16 <-- __exit_idle+0x2c/0x2e CPU 1: swapper:0 __atomic_notifier_call_chain+0xe/0x56 <-- atomic_notifier_call_chain+0x14/0x16

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |   65 +++++++++++++++++++++++++++++++++++++++++++++++----
 lib/tracing/tracer.h |    1 
 2 files changed, 62 insertions(+), 4 deletions(-)

lib/tracing/tracer.c +61 -4

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -89,6 +89,10 @@ static struct mcount_ops trace_ops __rea
 };
 
 #ifdef CONFIG_DEBUG_FS
+enum trace_iterator {
+	TRACE_ITER_SYM_ONLY	= 1,
+};
+
 struct mctracer_iterator {
 	struct mctracer_trace *tr;
 	struct mctracer_entry *ent;

@@ -217,25 +221,29 @@ static void seq_print_symbol(struct seq_
 #endif
 
 static void notrace seq_print_ip_sym(struct seq_file *m,
-				     unsigned long ip)
+				     unsigned long ip,
+				     int sym_only)
 {
 	seq_print_symbol(m, "%s", ip);
-	seq_printf(m, " <" IP_FMT ">", ip);
+	if (!sym_only)
+		seq_printf(m, " <" IP_FMT ">", ip);
 }
 
 static int s_show(struct seq_file *m, void *v)
 {
 	struct mctracer_iterator *iter = v;
+	int sym_only = !!(iter->tr->iter_flags & TRACE_ITER_SYM_ONLY);
 
 	if (iter->ent == NULL) {
 		seq_printf(m, "mctracer:\n");
 	} else {
 		seq_printf(m, "CPU %d: ", iter->cpu);
 		seq_printf(m, "%s:%d ", iter->ent->comm, iter->ent->pid);
-		seq_print_ip_sym(m, iter->ent->ip);
+		seq_print_ip_sym(m, iter->ent->ip, sym_only);
 		if (iter->ent->parent_ip) {
 			seq_printf(m, " <-- ");
-			seq_print_ip_sym(m, iter->ent->parent_ip);
+			seq_print_ip_sym(m, iter->ent->parent_ip,
+					 sym_only);
 		}
 		seq_printf(m, "\n");
 	}

@@ -345,6 +353,50 @@ static struct file_operations mctracer_c
 	.write = mctracer_ctrl_write,
 };
 
+static ssize_t mctracer_iter_ctrl_read(struct file *filp, char __user *ubuf,
+				       size_t cnt, loff_t *ppos)
+{
+	struct mctracer_trace *tr = filp->private_data;
+	char buf[64];
+	int r = 0;
+
+	if (tr->iter_flags & TRACE_ITER_SYM_ONLY)
+		r = sprintf(buf, "%s", "symonly ");
+	r += sprintf(buf+r, "\n");
+
+	return simple_read_from_buffer(ubuf, cnt, ppos,
+				       buf, r);
+}
+
+static ssize_t mctracer_iter_ctrl_write(struct file *filp,
+					const char __user *ubuf,
+					size_t cnt, loff_t *ppos)
+{
+	struct mctracer_trace *tr = filp->private_data;
+	char buf[64];
+
+	if (cnt > 63)
+		cnt = 63;
+
+	if (copy_from_user(&buf, ubuf, cnt))
+		return -EFAULT;
+
+	buf[cnt] = 0;
+
+	if (strncmp(buf, "symonly", 7) == 0)
+		tr->iter_flags |= TRACE_ITER_SYM_ONLY;
+
+	filp->f_pos += cnt;
+
+	return cnt;
+}
+
+static struct file_operations mctracer_iter_fops = {
+	.open = mctracer_open_generic,
+	.read = mctracer_iter_ctrl_read,
+	.write = mctracer_iter_ctrl_write,
+};
+
 static void mctrace_init_debugfs(void)
 {
 	struct dentry *d_mctracer;

@@ -361,10 +413,15 @@ static void mctrace_init_debugfs(void)
 	if (!entry)
 		pr_warning("Could not create debugfs 'ctrl' entry\n");
 
+	entry = debugfs_create_file("iter_ctrl", 0644, d_mctracer,
+				    &mctracer_trace, &mctracer_iter_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs 'iter_ctrl' entry\n");
 	entry = debugfs_create_file("trace", 0444, d_mctracer,
 				    &mctracer_trace, &mctrace_fops);
 	if (!entry)
 		pr_warning("Could not create debugfs 'trace' entry\n");
+
 }
 #else /* CONFIG_DEBUG_FS */
 static void mctrace_init_debugfs(void)

lib/tracing/tracer.h +1 -0

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -17,6 +17,7 @@ struct mctracer_trace {
 	unsigned long trace_idx[NR_CPUS];
 	unsigned long entries;
 	long	      ctrl;
+	unsigned long iter_flags;
 	atomic_t      cnt;
 	atomic_t      disabled[NR_CPUS];
 	atomic_t      underrun[NR_CPUS];

Steven Rostedt

author reviewer

2008-01-15 20:55 UTC #11 lore raw

RFC PATCH v3 11/30 Reset the tracer when started

In reply to Steven Rostedt

This patch resets the trace when it is started by the user.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

lib/tracing/tracer.c +14 -0

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -88,6 +88,16 @@ static struct mcount_ops trace_ops __rea
 	.func = trace_function,
 };
 
+static notrace void mctracer_reset(struct mctracer_trace *tr)
+{
+	int cpu;
+
+	for_each_online_cpu(cpu) {
+		tr->trace_idx[cpu] = 0;
+		atomic_set(&tr->underrun[cpu], 0);
+	}
+}
+
 #ifdef CONFIG_DEBUG_FS
 enum trace_iterator {
 	TRACE_ITER_SYM_ONLY	= 1,

@@ -334,6 +344,10 @@ static ssize_t mctracer_ctrl_write(struc
 
 	val = !!simple_strtoul(buf, NULL, 10);
 
+	/* When starting a new trace, reset the buffers */
+	if (val)
+		mctracer_reset(tr);
+
 	if (tr->ctrl ^ val) {
 		if (val)
 			trace_enabled = 1;

Steven Rostedt

author reviewer

2008-01-15 20:59 UTC #12 lore raw

RFC PATCH v3 12/30 separate out the percpu date into a percpu struct

In reply to Steven Rostedt

For better cacheline performance, this patch creates a separate struct for each CPU with the percpu data grouped together.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |   42 +++++++++++++++++++++++-------------------
 lib/tracing/tracer.h |   12 ++++++++----
 2 files changed, 31 insertions(+), 23 deletions(-)

lib/tracing/tracer.c +23 ~~-19~~

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -16,6 +16,7 @@
 #include <linux/module.h>
 #include <linux/linkage.h>
 #include <linux/seq_file.h>
+#include <linux/percpu.h>
 #include <linux/debugfs.h>
 #include <linux/kallsyms.h>
 #include <linux/uaccess.h>

@@ -25,6 +26,7 @@
 #include "tracer_interface.h"
 
 static struct mctracer_trace mctracer_trace;
+static DEFINE_PER_CPU(struct mctracer_trace_cpu, mctracer_trace_cpu);
 static int trace_enabled __read_mostly;
 
 static inline notrace void

@@ -36,21 +38,22 @@ mctracer_add_trace_entry(struct mctracer
 	unsigned long idx, idx_next;
 	struct mctracer_entry *entry;
 	struct task_struct *tsk = current;
+	struct mctracer_trace_cpu *data = tr->data[cpu];
 
-	idx = tr->trace_idx[cpu];
+	idx = data->trace_idx;
 	idx_next = idx + 1;
 
 	if (unlikely(idx_next >= tr->entries)) {
-		atomic_inc(&tr->underrun[cpu]);
+		atomic_inc(&data->underrun);
 		idx_next = 0;
 	}
 
-	tr->trace_idx[cpu] = idx_next;
+	data->trace_idx = idx_next;
 
-	if (unlikely(idx_next != 0 && atomic_read(&tr->underrun[cpu])))
-		atomic_inc(&tr->underrun[cpu]);
+	if (unlikely(idx_next != 0 && atomic_read(&data->underrun)))
+		atomic_inc(&data->underrun);
 
-	entry = tr->trace[cpu] + idx * MCTRACER_ENTRY_SIZE;
+	entry = data->trace + idx * MCTRACER_ENTRY_SIZE;
 	entry->idx	 = atomic_inc_return(&tr->cnt);
 	entry->ip	 = ip;
 	entry->parent_ip = parent_ip;

@@ -73,11 +76,11 @@ static notrace void trace_function(const
 
 	cpu = raw_smp_processor_id();
 
-	atomic_inc(&tr->disabled[cpu]);
-	if (likely(atomic_read(&tr->disabled[cpu]) == 1))
+	atomic_inc(&tr->data[cpu]->disabled);
+	if (likely(atomic_read(&tr->data[cpu]->disabled) == 1))
 		mctracer_add_trace_entry(tr, cpu, ip, parent_ip);
 
-	atomic_dec(&tr->disabled[cpu]);
+	atomic_dec(&tr->data[cpu]->disabled);
 
  out:
 	raw_local_irq_restore(flags);

@@ -93,8 +96,8 @@ static notrace void mctracer_reset(struc
 	int cpu;
 
 	for_each_online_cpu(cpu) {
-		tr->trace_idx[cpu] = 0;
-		atomic_set(&tr->underrun[cpu], 0);
+		tr->data[cpu]->trace_idx = 0;
+		atomic_set(&tr->data[cpu]->underrun, 0);
 	}
 }

@@ -115,16 +118,16 @@ static struct mctracer_entry *mctracer_e
 						 unsigned long idx,
 						 int cpu)
 {
-	struct mctracer_entry *array = tr->trace[cpu];
+	struct mctracer_entry *array = tr->data[cpu]->trace;
 	unsigned long underrun;
 
 	if (idx >= tr->entries)
 		return NULL;
 
-	underrun = atomic_read(&tr->underrun[cpu]);
+	underrun = atomic_read(&tr->data[cpu]->underrun);
 	if (underrun)
 		idx = ((underrun - 1) + idx) % tr->entries;
-	else if (idx >= tr->trace_idx[cpu])
+	else if (idx >= tr->data[cpu]->trace_idx)
 		return NULL;
 
 	return &array[idx];

@@ -139,7 +142,7 @@ static void *find_next_entry(struct mctr
 	int i;
 
 	for_each_possible_cpu(i) {
-		if (!tr->trace[i])
+		if (!tr->data[i]->trace)
 			continue;
 		ent = mctracer_entry_idx(tr, iter->next_idx[i], i);
 		if (ent && (!next || next->idx > ent->idx)) {

@@ -461,6 +464,7 @@ static notrace int mctracer_alloc_buffer
 	int i;
 
 	for_each_possible_cpu(i) {
+		mctracer_trace.data[i] = &per_cpu(mctracer_trace_cpu, i);
 		array = (struct mctracer_entry *)
 			  __get_free_pages(GFP_KERNEL, order);
 		if (array == NULL) {

@@ -468,7 +472,7 @@ static notrace int mctracer_alloc_buffer
 			       " %ld bytes for trace buffer!\n", size);
 			goto free_buffers;
 		}
-		mctracer_trace.trace[i] = array;
+		mctracer_trace.data[i]->trace = array;
 	}
 
 	/*

@@ -489,10 +493,10 @@ static notrace int mctracer_alloc_buffer
 
  free_buffers:
 	for (i-- ; i >= 0; i--) {
-		if (mctracer_trace.trace[i]) {
-			free_pages((unsigned long)mctracer_trace.trace[i],
+		if (mctracer_trace.data[i] && mctracer_trace.data[i]->trace) {
+			free_pages((unsigned long)mctracer_trace.data[i]->trace,
 				   order);
-			mctracer_trace.trace[i] = NULL;
+			mctracer_trace.data[i]->trace = NULL;
 		}
 	}
 	return -ENOMEM;

lib/tracing/tracer.h +8 -4

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -12,15 +12,19 @@ struct mctracer_entry {
 	pid_t pid;
 };
 
+struct mctracer_trace_cpu {
+	void *trace;
+	unsigned long trace_idx;
+	atomic_t      disabled;
+	atomic_t      underrun;
+};
+
 struct mctracer_trace {
-	void	      *trace[NR_CPUS];
-	unsigned long trace_idx[NR_CPUS];
 	unsigned long entries;
 	long	      ctrl;
 	unsigned long iter_flags;
 	atomic_t      cnt;
-	atomic_t      disabled[NR_CPUS];
-	atomic_t      underrun[NR_CPUS];
+	struct mctracer_trace_cpu *data[NR_CPUS];
 };
 
 #endif /* _LINUX_MCOUNT_TRACER_H */

Steven Rostedt

author reviewer

2008-01-15 20:59 UTC #13 lore raw

RFC PATCH v3 13/30 handle accurate time keeping over long delays

In reply to Steven Rostedt

Handle accurate time even if there's a long delay between accumulated clock cycles.

Signed-off-by John Stultz <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/vsyscall_64.c |    5 ++-
 include/asm-x86/vgtod.h       |    2 -
 include/linux/clocksource.h   |   58 ++++++++++++++++++++++++++++++++++++++++--
 kernel/time/timekeeping.c     |   35 +++++++++++++------------
 4 files changed, 80 insertions(+), 20 deletions(-)

arch/x86/kernel/vsyscall_64.c +4 -1

Index: linux-compile.git/arch/x86/kernel/vsyscall_64.c
===================================================================
@@ -86,6 +86,7 @@ void update_vsyscall(struct timespec *wa
 	vsyscall_gtod_data.clock.mask = clock->mask;
 	vsyscall_gtod_data.clock.mult = clock->mult;
 	vsyscall_gtod_data.clock.shift = clock->shift;
+	vsyscall_gtod_data.clock.cycle_accumulated = clock->cycle_accumulated;
 	vsyscall_gtod_data.wall_time_sec = wall_time->tv_sec;
 	vsyscall_gtod_data.wall_time_nsec = wall_time->tv_nsec;
 	vsyscall_gtod_data.wall_to_monotonic = wall_to_monotonic;

@@ -121,7 +122,7 @@ static __always_inline long time_syscall
 
 static __always_inline void do_vgettimeofday(struct timeval * tv)
 {
-	cycle_t now, base, mask, cycle_delta;
+	cycle_t now, base, accumulated, mask, cycle_delta;
 	unsigned seq;
 	unsigned long mult, shift, nsec;
 	cycle_t (*vread)(void);

@@ -135,6 +136,7 @@ static __always_inline void do_vgettimeo
 		}
 		now = vread();
 		base = __vsyscall_gtod_data.clock.cycle_last;
+		accumulated  = __vsyscall_gtod_data.clock.cycle_accumulated;
 		mask = __vsyscall_gtod_data.clock.mask;
 		mult = __vsyscall_gtod_data.clock.mult;
 		shift = __vsyscall_gtod_data.clock.shift;

@@ -145,6 +147,7 @@ static __always_inline void do_vgettimeo
 
 	/* calculate interval: */
 	cycle_delta = (now - base) & mask;
+	cycle_delta += accumulated;
 	/* convert to nsecs: */
 	nsec += (cycle_delta * mult) >> shift;

include/asm-x86/vgtod.h +1 -1

Index: linux-compile.git/include/asm-x86/vgtod.h
===================================================================
@@ -15,7 +15,7 @@ struct vsyscall_gtod_data {
 	struct timezone sys_tz;
 	struct { /* extract of a clocksource struct */
 		cycle_t (*vread)(void);
-		cycle_t	cycle_last;
+		cycle_t	cycle_last, cycle_accumulated;
 		cycle_t	mask;
 		u32	mult;
 		u32	shift;

include/linux/clocksource.h +56 -2

95 lines not on this page · raw message

kernel/time/timekeeping.c +19 ~~-16~~

Index: linux-compile.git/kernel/time/timekeeping.c
===================================================================
@@ -66,16 +66,10 @@ static struct clocksource *clock; /* poi
  */
 static inline s64 __get_nsec_offset(void)
 {
-	cycle_t cycle_now, cycle_delta;
+	cycle_t cycle_delta;
 	s64 ns_offset;
 
-	/* read clocksource: */
-	cycle_now = clocksource_read(clock);
-
-	/* calculate the delta since the last update_wall_time: */
-	cycle_delta = (cycle_now - clock->cycle_last) & clock->mask;
-
-	/* convert to nanoseconds: */
+	cycle_delta = clocksource_get_cycles(clock, clocksource_read(clock));
 	ns_offset = cyc2ns(clock, cycle_delta);
 
 	return ns_offset;

@@ -195,7 +189,7 @@ static void change_clocksource(void)
 
 	clock = new;
 	clock->cycle_last = now;
-
+	clock->cycle_accumulated = 0;
 	clock->error = 0;
 	clock->xtime_nsec = 0;
 	clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH);

@@ -205,9 +199,15 @@ static void change_clocksource(void)
 	printk(KERN_INFO "Time: %s clocksource has been installed.\n",
 	       clock->name);
 }
+
+void timekeeping_accumulate(void)
+{
+	clocksource_accumulate(clock, clocksource_read(clock));
+}
 #else
 static inline void change_clocksource(void) { }
 static inline s64 __get_nsec_offset(void) { return 0; }
+void timekeeping_accumulate(void) { }
 #endif
 
 /**

@@ -302,6 +302,7 @@ static int timekeeping_resume(struct sys
 	timespec_add_ns(&xtime, timekeeping_suspend_nsecs);
 	/* re-base the last cycle value */
 	clock->cycle_last = clocksource_read(clock);
+	clock->cycle_accumulated = 0;
 	clock->error = 0;
 	timekeeping_suspended = 0;
 	write_sequnlock_irqrestore(&xtime_lock, flags);

@@ -448,27 +449,29 @@ static void clocksource_adjust(s64 offse
  */
 void update_wall_time(void)
 {
-	cycle_t offset;
+	cycle_t cycle_now, offset;
 
 	/* Make sure we're fully resumed: */
 	if (unlikely(timekeeping_suspended))
 		return;
 
 #ifdef CONFIG_GENERIC_TIME
-	offset = (clocksource_read(clock) - clock->cycle_last) & clock->mask;
+	cycle_now = clocksource_read(clock);
 #else
-	offset = clock->cycle_interval;
+	cycle_now = clock->cycle_last + clock->cycle_interval;
 #endif
+	offset = (cycle_now - clock->cycle_last) & clock->mask;
+	clocksource_accumulate(clock, cycle_now);
+
 	clock->xtime_nsec += (s64)xtime.tv_nsec << clock->shift;
 
 	/* normally this loop will run just once, however in the
 	 * case of lost or late ticks, it will accumulate correctly.
 	 */
-	while (offset >= clock->cycle_interval) {
+	while (clock->cycle_accumulated >= clock->cycle_interval) {
 		/* accumulate one interval */
 		clock->xtime_nsec += clock->xtime_interval;
-		clock->cycle_last += clock->cycle_interval;
-		offset -= clock->cycle_interval;
+		clock->cycle_accumulated -= clock->cycle_interval;
 
 		if (clock->xtime_nsec >= (u64)NSEC_PER_SEC << clock->shift) {
 			clock->xtime_nsec -= (u64)NSEC_PER_SEC << clock->shift;

@@ -482,7 +485,7 @@ void update_wall_time(void)
 	}
 
 	/* correct the clock when NTP error is too big */
-	clocksource_adjust(offset);
+	clocksource_adjust(clock->cycle_accumulated);
 
 	/* store full nanoseconds into xtime */
 	xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift;

Steven Rostedt

author reviewer

2008-01-15 20:53 UTC #14 lore raw

RFC PATCH v3 14/30 ppc clock accumulate fix

In reply to Steven Rostedt

The following is a quick and dirty fix for powerpc so it includes cycle_accumulated in its calculation. It relies on the fact that the powerpc clocksource is a 64bit counter (don't have to worry about multiple overflows), so the subtraction should be safe.

Signed-off-by John Stultz <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/powerpc/kernel/time.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

arch/powerpc/kernel/time.c +2 -1

Index: linux-compile.git/arch/powerpc/kernel/time.c
===================================================================
@@ -773,7 +773,8 @@ void update_vsyscall(struct timespec *wa
 	stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC;
 	do_div(stamp_xsec, 1000000000);
 	stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC;
-	update_gtod(clock->cycle_last, stamp_xsec, t2x);
+	update_gtod(clock->cycle_last-clock->cycle_accumulated,
+		    stamp_xsec, t2x);
 }
 
 void update_vsyscall_tz(void)

Steven Rostedt

author reviewer

2008-01-15 20:54 UTC #15 lore raw

RFC PATCH v3 15/30 Fixup merge between xtime_cache and timkkeeping starvation fix

In reply to Steven Rostedt

Quoted 7 lines

void update_wall_time(void)
{

(quoted diff) +1 -1

-	cycle_t offset;
+	cycle_t cycle_now, offset;
 
 	/* Make sure we're fully resumed: */
 	if (unlikely(timekeeping_suspended))

Review

Quoted 10 lines

(quoted diff) +3 -2

 		return;
 
 #ifdef CONFIG_GENERIC_TIME
-	offset = (clocksource_read(clock) - clock->cycle_last) & clock->mask;
+	cycle_now = clocksource_read(clock);
 #else
-	offset = clock->cycle_interval;
+	cycle_now = clock->cycle_last + clock->cycle_interval;
 #endif
+	offset = (cycle_now - clock->cycle_last) & clock->mask;

It seems this offset addition was to merge against the colliding xtime_cache changes in mainline. However, I don't think its quite right, and might be causing incorrect time() or vtime() results if NO_HZ is enabled.

Quoted 18 lines

(quoted diff) +4 -3

+	clocksource_accumulate(clock, cycle_now);
+
 	clock->xtime_nsec += (s64)xtime.tv_nsec << clock->shift;
 
 	/* normally this loop will run just once, however in the
 	 * case of lost or late ticks, it will accumulate correctly.
 	 */
-	while (offset >= clock->cycle_interval) {
+	while (clock->cycle_accumulated >= clock->cycle_interval) {
 		/* accumulate one interval */
 		clock->xtime_nsec += clock->xtime_interval;
-		clock->cycle_last += clock->cycle_interval;
-		offset -= clock->cycle_interval;
+		clock->cycle_accumulated -= clock->cycle_interval;
 
 		if (clock->xtime_nsec >= (u64)NSEC_PER_SEC << clock->shift) {
 			clock->xtime_nsec -= (u64)NSEC_PER_SEC << clock->shift;

Review void update_wall_time(void) · line 488

Quoted 5 lines

(quoted diff) +1 -1

@@ -482,7 +485,7 @@ void update_wall_time(void)
 	}
 
 	/* correct the clock when NTP error is too big */
-	clocksource_adjust(offset);
+	clocksource_adjust(clock->cycle_accumulated);

I suspect the following is needed, but haven't been able to test it yet.

Fixup merge between xtime_cache and timekeeping starvation fix.

Signed-off-by John Stultz <[email protected]>

Signed-off-by Steven Rostedt <[email protected]>

---
 kernel/time/timekeeping.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

kernel/time/timekeeping.c +2 -3

Index: linux-compile.git/kernel/time/timekeeping.c
===================================================================
@@ -449,7 +449,7 @@ static void clocksource_adjust(s64 offse
  */
 void update_wall_time(void)
 {
-	cycle_t cycle_now, offset;
+	cycle_t cycle_now;
 
 	/* Make sure we're fully resumed: */
 	if (unlikely(timekeeping_suspended))

@@ -460,7 +460,6 @@ void update_wall_time(void)
 #else
 	cycle_now = clock->cycle_last + clock->cycle_interval;
 #endif
-	offset = (cycle_now - clock->cycle_last) & clock->mask;
 	clocksource_accumulate(clock, cycle_now);
 
 	clock->xtime_nsec += (s64)xtime.tv_nsec << clock->shift;

@@ -491,7 +490,7 @@ void update_wall_time(void)
 	xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift;
 	clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift;
 
-	update_xtime_cache(cyc2ns(clock, offset));
+	update_xtime_cache(cyc2ns(clock, clock->cycle_accumulated));
 
 	/* check to see if there is a new clocksource to use */
 	change_clocksource();

Steven Rostedt

author reviewer

2008-01-15 20:53 UTC #16 lore raw

RFC PATCH v3 16/30 time keeping add cycle_raw for actual incrementation

In reply to Steven Rostedt

The get_monotonic_cycles needs to produce a monotonic counter as output.

This patch adds a cycle_raw to produce an accumulative counter. Unfortunately there is already an cycle_accumulate variable, but that is used to avoid clock source overflow and can also be decremented (probably that name should be changed and we should use that for this patch).

Signed-off-by Steven Rostedt <[email protected]>

Acked-by John Stultz <[email protected]>

---
 include/linux/clocksource.h |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

include/linux/clocksource.h +2 -1

Index: linux-compile.git/include/linux/clocksource.h
===================================================================
@@ -87,7 +87,7 @@ struct clocksource {
 	 * more than one cache line.
 	 */
 	struct {
-		cycle_t cycle_last, cycle_accumulated;
+		cycle_t cycle_last, cycle_accumulated, cycle_raw;
 	} ____cacheline_aligned_in_smp;
 
 	u64 xtime_nsec;

@@ -204,6 +204,7 @@ static inline void clocksource_accumulat
 	cycle_t offset = (now - cs->cycle_last) & cs->mask;
 	cs->cycle_last = now;
 	cs->cycle_accumulated += offset;
+	cs->cycle_raw += offset;
 }
 
 /**

Steven Rostedt

author reviewer

2008-01-15 20:51 UTC #17 lore raw

RFC PATCH v3 17/30 initialize the clock source to jiffies clock.

In reply to Steven Rostedt

The latency tracer can call clocksource_read very early in bootup and before the clock source variable has been initialized. This results in a crash at boot up (even before earlyprintk is initialized). Since the clock->read variable points to NULL.

This patch simply initializes the clock to use clocksource_jiffies, so that any early user of clocksource_read will not crash.

Signed-off-by Steven Rostedt <[email protected]>

Acked-by John Stultz <[email protected]>

---
 include/linux/clocksource.h |    3 +++
 kernel/time/timekeeping.c   |    9 +++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)

include/linux/clocksource.h +3 -0

Index: linux-compile.git/include/linux/clocksource.h
===================================================================
@@ -274,6 +274,9 @@ extern struct clocksource* clocksource_g
 extern void clocksource_change_rating(struct clocksource *cs, int rating);
 extern void clocksource_resume(void);
 
+/* used to initialize clock */
+extern struct clocksource clocksource_jiffies;
+
 #ifdef CONFIG_GENERIC_TIME_VSYSCALL
 extern void update_vsyscall(struct timespec *ts, struct clocksource *c);
 extern void update_vsyscall_tz(void);

kernel/time/timekeeping.c +7 -2

Index: linux-compile.git/kernel/time/timekeeping.c
===================================================================
@@ -53,8 +53,13 @@ static inline void update_xtime_cache(u6
 	timespec_add_ns(&xtime_cache, nsec);
 }
 
-static struct clocksource *clock; /* pointer to current clocksource */
-
+/*
+ * pointer to current clocksource
+ *  Just in case we use clocksource_read before we initialize
+ *  the actual clock source. Instead of calling a NULL read pointer
+ *  we return jiffies.
+ */
+static struct clocksource *clock = &clocksource_jiffies;
 
 #ifdef CONFIG_GENERIC_TIME
 /**

Mathieu Desnoyers

2008-01-15 21:14 UTC #18 lore raw

Re: [RFC PATCH 17/30 v3] initialize the clock source to jiffies clock.

Quoted 8 lines · * Steven Rostedt ([email protected]) wrote:

The latency tracer can call clocksource_read very early in bootup and before the clock source variable has been initialized. This results in a crash at boot up (even before earlyprintk is initialized). Since the clock->read variable points to NULL.
This patch simply initializes the clock to use clocksource_jiffies, so that any early user of clocksource_read will not crash.

Hrm, is it sane at all to use the jiffies as a clocksource at early boot? I thought it was updated by the timer interrupt, which is only activated late in the boot process.

Quoted 43 lines

Signed-off-by Steven Rostedt <[email protected]>

Acked-by John Stultz <[email protected]>

---
 include/linux/clocksource.h |    3 +++
 kernel/time/timekeeping.c   |    9 +++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)

Index: linux-compile.git/include/linux/clocksource.h
===================================================================

include/linux/clocksource.h +3 -0

@@ -274,6 +274,9 @@ extern struct clocksource* clocksource_g
 extern void clocksource_change_rating(struct clocksource *cs, int rating);
 extern void clocksource_resume(void);
 
+/* used to initialize clock */
+extern struct clocksource clocksource_jiffies;
+
 #ifdef CONFIG_GENERIC_TIME_VSYSCALL
 extern void update_vsyscall(struct timespec *ts, struct clocksource *c);
 extern void update_vsyscall_tz(void);

Index: linux-compile.git/kernel/time/timekeeping.c
===================================================================

kernel/time/timekeeping.c +7 -2

@@ -53,8 +53,13 @@ static inline void update_xtime_cache(u6
 	timespec_add_ns(&xtime_cache, nsec);
 }
 
-static struct clocksource *clock; /* pointer to current clocksource */
-
+/*
+ * pointer to current clocksource
+ *  Just in case we use clocksource_read before we initialize
+ *  the actual clock source. Instead of calling a NULL read pointer
+ *  we return jiffies.
+ */
+static struct clocksource *clock = &clocksource_jiffies;
 
 #ifdef CONFIG_GENERIC_TIME
 /**

Signature · 1 line

--

Signature · 3 lines

Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

Steven Rostedt

author reviewer

2008-01-15 21:27 UTC #19 lore raw

Re: [RFC PATCH 17/30 v3] initialize the clock source to jiffies clock.

Quoted 8 lines · Mathieu Desnoyers

Quoted 4 lines
This patch simply initializes the clock to use clocksource_jiffies, so that any early user of clocksource_read will not crash.
Hrm, is it sane at all to use the jiffies as a clocksource at early boot? I thought it was updated by the timer interrupt, which is only activated late in the boot process.

It gives us a bogus value, but we know it's always there. This was discovered in the -rt patch where we had a hard hang at early bootup. Seems that the tracer was calling for a clock source before it was initialized, and it called a null pointer.

The original fix was to just make a dummy timer source that returned zero. But using jiffies seemed a better solution.

-- Steve

Steven Rostedt

author reviewer

2008-01-15 20:58 UTC #20 lore raw

RFC PATCH v3 18/30 add get_monotonic_cycles

In reply to Steven Rostedt

The latency tracer needs a way to get an accurate time without grabbing any locks. Locks themselves might call the latency tracer and cause at best a slow down.

This patch adds get_monotonic_cycles that returns cycles from a reliable clock source in a monotonic fashion.

Signed-off-by Steven Rostedt <[email protected]>

---
 include/linux/clocksource.h |    3 ++
 kernel/time/timekeeping.c   |   48 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+)

include/linux/clocksource.h +3 -0

Index: linux-compile.git/include/linux/clocksource.h
===================================================================
@@ -273,6 +273,9 @@ extern int clocksource_register(struct c
 extern struct clocksource* clocksource_get_next(void);
 extern void clocksource_change_rating(struct clocksource *cs, int rating);
 extern void clocksource_resume(void);
+extern cycle_t get_monotonic_cycles(void);
+extern unsigned long cycles_to_usecs(cycle_t cycles);
+extern cycle_t usecs_to_cycles(unsigned long usecs);
 
 /* used to initialize clock */
 extern struct clocksource clocksource_jiffies;

kernel/time/timekeeping.c +48 -0

54 lines not on this page · raw message

Steven Rostedt

author reviewer

2008-01-15 20:59 UTC #21 lore raw

RFC PATCH v3 19/30 add notrace annotations to timing events

In reply to Steven Rostedt

This patch adds notrace annotations to timer functions that will be used by tracing. This helps speed things up and also keeps the ugliness of printing these functions down.

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/apic_32.c     |    2 +-
 arch/x86/kernel/hpet.c        |    2 +-
 arch/x86/kernel/time_32.c     |    2 +-
 arch/x86/kernel/tsc_32.c      |    2 +-
 arch/x86/kernel/tsc_64.c      |    4 ++--
 arch/x86/lib/delay_32.c       |    6 +++---
 drivers/clocksource/acpi_pm.c |    8 ++++----
 7 files changed, 13 insertions(+), 13 deletions(-)

arch/x86/kernel/apic_32.c +1 -1

Index: linux-compile.git/arch/x86/kernel/apic_32.c
===================================================================
@@ -577,7 +577,7 @@ static void local_apic_timer_interrupt(v
  *   interrupt as well. Thus we cannot inline the local irq ... ]
  */
 
-void fastcall smp_apic_timer_interrupt(struct pt_regs *regs)
+notrace fastcall void smp_apic_timer_interrupt(struct pt_regs *regs)
 {
 	struct pt_regs *old_regs = set_irq_regs(regs);

arch/x86/kernel/hpet.c +1 -1

Index: linux-compile.git/arch/x86/kernel/hpet.c
===================================================================
@@ -295,7 +295,7 @@ static int hpet_legacy_next_event(unsign
 /*
  * Clock source related code
  */
-static cycle_t read_hpet(void)
+static notrace cycle_t read_hpet(void)
 {
 	return (cycle_t)hpet_readl(HPET_COUNTER);
 }

arch/x86/kernel/time_32.c +1 -1

Index: linux-compile.git/arch/x86/kernel/time_32.c
===================================================================
@@ -122,7 +122,7 @@ static int set_rtc_mmss(unsigned long no
 
 int timer_ack;
 
-unsigned long profile_pc(struct pt_regs *regs)
+notrace unsigned long profile_pc(struct pt_regs *regs)
 {
 	unsigned long pc = instruction_pointer(regs);

arch/x86/kernel/tsc_32.c +1 -1

Index: linux-compile.git/arch/x86/kernel/tsc_32.c
===================================================================
@@ -269,7 +269,7 @@ core_initcall(cpufreq_tsc);
 
 static unsigned long current_tsc_khz = 0;
 
-static cycle_t read_tsc(void)
+static notrace cycle_t read_tsc(void)
 {
 	cycle_t ret;

arch/x86/kernel/tsc_64.c +2 -2

Index: linux-compile.git/arch/x86/kernel/tsc_64.c
===================================================================
@@ -248,13 +248,13 @@ __setup("notsc", notsc_setup);
 
 
 /* clock source code: */
-static cycle_t read_tsc(void)
+static notrace cycle_t read_tsc(void)
 {
 	cycle_t ret = (cycle_t)get_cycles_sync();
 	return ret;
 }
 
-static cycle_t __vsyscall_fn vread_tsc(void)
+static notrace cycle_t __vsyscall_fn vread_tsc(void)
 {
 	cycle_t ret = (cycle_t)get_cycles_sync();
 	return ret;

arch/x86/lib/delay_32.c +3 -3

Index: linux-compile.git/arch/x86/lib/delay_32.c
===================================================================
@@ -24,7 +24,7 @@
 #endif
 
 /* simple loop based delay: */
-static void delay_loop(unsigned long loops)
+static notrace void delay_loop(unsigned long loops)
 {
 	int d0;

@@ -39,7 +39,7 @@ static void delay_loop(unsigned long loo
 }
 
 /* TSC based delay: */
-static void delay_tsc(unsigned long loops)
+static notrace void delay_tsc(unsigned long loops)
 {
 	unsigned long bclock, now;

@@ -72,7 +72,7 @@ int read_current_timer(unsigned long *ti
 	return -1;
 }
 
-void __delay(unsigned long loops)
+notrace void __delay(unsigned long loops)
 {
 	delay_fn(loops);
 }

drivers/clocksource/acpi_pm.c +4 -4

Index: linux-compile.git/drivers/clocksource/acpi_pm.c
===================================================================
@@ -30,13 +30,13 @@
  */
 u32 pmtmr_ioport __read_mostly;
 
-static inline u32 read_pmtmr(void)
+static inline notrace u32 read_pmtmr(void)
 {
 	/* mask the output to 24 bits */
 	return inl(pmtmr_ioport) & ACPI_PM_MASK;
 }
 
-u32 acpi_pm_read_verified(void)
+notrace u32 acpi_pm_read_verified(void)
 {
 	u32 v1 = 0, v2 = 0, v3 = 0;

@@ -56,12 +56,12 @@ u32 acpi_pm_read_verified(void)
 	return v2;
 }
 
-static cycle_t acpi_pm_read_slow(void)
+static notrace cycle_t acpi_pm_read_slow(void)
 {
 	return (cycle_t)acpi_pm_read_verified();
 }
 
-static cycle_t acpi_pm_read(void)
+static notrace cycle_t acpi_pm_read(void)
 {
 	return (cycle_t)read_pmtmr();
 }

Steven Rostedt

author reviewer

2008-01-15 20:58 UTC #22 lore raw

RFC PATCH v3 20/30 Add timestamps to tracer

In reply to Steven Rostedt

Add timestamps to trace entries.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |   16 ++++++++++++++++
 lib/tracing/tracer.h |    1 +
 2 files changed, 17 insertions(+)

lib/tracing/tracer.c +16 -0

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -19,12 +19,18 @@
 #include <linux/percpu.h>
 #include <linux/debugfs.h>
 #include <linux/kallsyms.h>
+#include <linux/clocksource.h>
 #include <linux/uaccess.h>
 #include <linux/mcount.h>
 
 #include "tracer.h"
 #include "tracer_interface.h"
 
+static inline notrace cycle_t now(void)
+{
+	return get_monotonic_cycles();
+}
+
 static struct mctracer_trace mctracer_trace;
 static DEFINE_PER_CPU(struct mctracer_trace_cpu, mctracer_trace_cpu);
 static int trace_enabled __read_mostly;

@@ -58,6 +64,7 @@ mctracer_add_trace_entry(struct mctracer
 	entry->ip	 = ip;
 	entry->parent_ip = parent_ip;
 	entry->pid	 = tsk->pid;
+	entry->t	 = now();
 	memcpy(entry->comm, tsk->comm, TASK_COMM_LEN);
 }

@@ -250,6 +257,15 @@ static int s_show(struct seq_file *m, vo
 	if (iter->ent == NULL) {
 		seq_printf(m, "mctracer:\n");
 	} else {
+		unsigned long long t;
+		unsigned long usec_rem;
+		unsigned long secs;
+
+		t = cycles_to_usecs(iter->ent->t);
+		usec_rem = do_div(t, 1000000ULL);
+		secs = (unsigned long)t;
+
+		seq_printf(m, "[%5lu.%06lu] ", secs, usec_rem);
 		seq_printf(m, "CPU %d: ", iter->cpu);
 		seq_printf(m, "%s:%d ", iter->ent->comm, iter->ent->pid);
 		seq_print_ip_sym(m, iter->ent->ip, sym_only);

lib/tracing/tracer.h +1 -0

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -5,6 +5,7 @@
 #include <linux/sched.h>
 
 struct mctracer_entry {
+	unsigned long long t;
 	unsigned long idx;
 	unsigned long ip;
 	unsigned long parent_ip;

Steven Rostedt

author reviewer

2008-01-15 20:52 UTC #23 lore raw

RFC PATCH v3 21/30 Sort trace by timestamp

In reply to Steven Rostedt

Now that each entry has a reliable timestamp, we can use the timestamp as the source of sorting the trace and remove the atomic increment.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |    3 +--
 lib/tracing/tracer.h |    2 --
 2 files changed, 1 insertion(+), 4 deletions(-)

lib/tracing/tracer.c +1 -2

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -60,7 +60,6 @@ mctracer_add_trace_entry(struct mctracer
 		atomic_inc(&data->underrun);
 
 	entry = data->trace + idx * MCTRACER_ENTRY_SIZE;
-	entry->idx	 = atomic_inc_return(&tr->cnt);
 	entry->ip	 = ip;
 	entry->parent_ip = parent_ip;
 	entry->pid	 = tsk->pid;

@@ -152,7 +151,7 @@ static void *find_next_entry(struct mctr
 		if (!tr->data[i]->trace)
 			continue;
 		ent = mctracer_entry_idx(tr, iter->next_idx[i], i);
-		if (ent && (!next || next->idx > ent->idx)) {
+		if (ent && (!next || next->t > ent->t)) {
 			next = ent;
 			next_i = i;
 		}

lib/tracing/tracer.h +0 -2

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -6,7 +6,6 @@
 
 struct mctracer_entry {
 	unsigned long long t;
-	unsigned long idx;
 	unsigned long ip;
 	unsigned long parent_ip;
 	char comm[TASK_COMM_LEN];

@@ -24,7 +23,6 @@ struct mctracer_trace {
 	unsigned long entries;
 	long	      ctrl;
 	unsigned long iter_flags;
-	atomic_t      cnt;
 	struct mctracer_trace_cpu *data[NR_CPUS];
 };

Steven Rostedt

author reviewer

2008-01-15 20:50 UTC #24 lore raw

RFC PATCH v3 22/30 speed up the output of the tracer

In reply to Steven Rostedt

The current method of printing out the trace is on every read, do a linear search for the next entry to print. This patch remembers the next entry to look at in the iterator, and if the next read is sequential, it can start reading from the next location.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |   28 +++++++++++++++++++---------
 1 file changed, 19 insertions(+), 9 deletions(-)

lib/tracing/tracer.c +19 -9

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -115,6 +115,7 @@ enum trace_iterator {
 struct mctracer_iterator {
 	struct mctracer_trace *tr;
 	struct mctracer_entry *ent;
+	loff_t pos;
 	unsigned long next_idx[NR_CPUS];
 	int cpu;
 	int idx;

@@ -186,6 +187,8 @@ static void *s_next(struct seq_file *m, 
 	while (ent && iter->idx < i)
 		ent = find_next_entry(iter);
 
+	iter->pos = *pos;
+
 	return ent;
 }

@@ -196,19 +199,25 @@ static void *s_start(struct seq_file *m,
 	loff_t l = 0;
 	int i;
 
-	iter->ent = NULL;
-	iter->cpu = 0;
-	iter->idx = -1;
-
-	for (i = 0; i < NR_CPUS; i++)
-		iter->next_idx[i] = 0;
-
 	/* stop the trace while dumping */
 	if (iter->tr->ctrl)
 		trace_enabled = 0;
 
-	for (p = iter; p && l < *pos; p = s_next(m, p, &l))
-		;
+	if (*pos != iter->pos) {
+		iter->ent = NULL;
+		iter->cpu = 0;
+		iter->idx = -1;
+
+		for (i = 0; i < NR_CPUS; i++)
+			iter->next_idx[i] = 0;
+
+		for (p = iter; p && l < *pos; p = s_next(m, p, &l))
+			;
+
+	} else {
+		l = *pos;
+		p = s_next(m, p, &l);
+	}
 
 	return p;
 }

@@ -296,6 +305,7 @@ static int mctrace_open(struct inode *in
 		return -ENOMEM;
 
 	iter->tr = &mctracer_trace;
+	iter->pos = -1;
 
 	/* TODO stop tracer */
 	ret = seq_open(file, &mctrace_seq_ops);

Steven Rostedt

author reviewer

2008-01-15 21:00 UTC #25 lore raw

RFC PATCH v3 23/30 Add latency_trace format tor tracer

In reply to Steven Rostedt

This patch adds a latency_trace file with the format used by RT in which others have created tools to disect. This file adds some useful recording for tracing, but still does not add actual latency tracing.

Format like (default):

"echo noverbose > /debugfs/tracing/iter_ctrl"

preemption latency trace v1.1.5 on 2.6.24-rc7-tst
--------------------------------------------------------------------
latency: 0 us, #419428/4361791, CPU#1 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4)

    -----------------
    | task: -0 (uid:0 nice:0 policy:0 rt_prio:0)
    -----------------

                 _------=> CPU#            
                / _-----=> irqs-off        
               | / _----=> need-resched    
               || / _---=> hardirq/softirq 
               ||| / _--=> preempt-depth   
               |||| /                      
               |||||     delay             
   cmd     pid ||||| time  |   caller      
      \   /    |||||   \   |   /           
 swapper-0     0d.h. 1595128us+: set_normalized_timespec+0x8/0x2d <c043841d> (ktime_get_ts+0x4a/0x4e <c04499d4>)
 swapper-0     0d.h. 1595131us+: _spin_lock+0x8/0x18 <c0630690> (hrtimer_interrupt+0x6e/0x1b0 <c0449c56>)

Or with verbose turned on:

"echo verbose > /debugfs/tracing/iter_ctrl"

preemption latency trace v1.1.5 on 2.6.24-rc7-tst
--------------------------------------------------------------------
latency: 0 us, #419428/4361791, CPU#1 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4)

    -----------------
    | task: -0 (uid:0 nice:0 policy:0 rt_prio:0)
    -----------------

         swapper     0 0 9 00000000 00000000 [f3675f41] 1595.128ms (+0.003ms): set_normalized_timespec+0x8/0x2d <c043841d> (ktime_get_ts+0x4a/0x4e <c04499d4>)
         swapper     0 0 9 00000000 00000001 [f3675f45] 1595.131ms (+0.003ms): _spin_lock+0x8/0x18 <c0630690> (hrtimer_interrupt+0x6e/0x1b0 <c0449c56>)
         swapper     0 0 9 00000000 00000002 [f3675f48] 1595.135ms (+0.003ms): _spin_lock+0x8/0x18 <c0630690> (hrtimer_interrupt+0x6e/0x1b0 <c0449c56>)

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/tracer.c |  451 +++++++++++++++++++++++++++++++++++++++++++++------
 lib/tracing/tracer.h |   13 +
 2 files changed, 411 insertions(+), 53 deletions(-)

lib/tracing/tracer.c +398 ~~-53~~

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -20,7 +20,9 @@
 #include <linux/debugfs.h>
 #include <linux/kallsyms.h>
 #include <linux/clocksource.h>
+#include <linux/utsrelease.h>
 #include <linux/uaccess.h>
+#include <linux/hardirq.h>
 #include <linux/mcount.h>
 
 #include "tracer.h"

@@ -35,16 +37,25 @@ static struct mctracer_trace mctracer_tr
 static DEFINE_PER_CPU(struct mctracer_trace_cpu, mctracer_trace_cpu);
 static int trace_enabled __read_mostly;
 
+enum trace_flag_type {
+	TRACE_FLAG_IRQS_OFF		= 0x01,
+	TRACE_FLAG_NEED_RESCHED		= 0x02,
+	TRACE_FLAG_HARDIRQ		= 0x04,
+	TRACE_FLAG_SOFTIRQ		= 0x08,
+};
+
 static inline notrace void
 mctracer_add_trace_entry(struct mctracer_trace *tr,
 			 int cpu,
 			 const unsigned long ip,
-			 const unsigned long parent_ip)
+			 const unsigned long parent_ip,
+			 unsigned long flags)
 {
 	unsigned long idx, idx_next;
 	struct mctracer_entry *entry;
 	struct task_struct *tsk = current;
 	struct mctracer_trace_cpu *data = tr->data[cpu];
+	unsigned long pc;
 
 	idx = data->trace_idx;
 	idx_next = idx + 1;

@@ -59,11 +70,18 @@ mctracer_add_trace_entry(struct mctracer
 	if (unlikely(idx_next != 0 && atomic_read(&data->underrun)))
 		atomic_inc(&data->underrun);
 
+	pc = preempt_count();
+
 	entry = data->trace + idx * MCTRACER_ENTRY_SIZE;
+	entry->preempt_count = pc & 0xff;
 	entry->ip	 = ip;
 	entry->parent_ip = parent_ip;
 	entry->pid	 = tsk->pid;
 	entry->t	 = now();
+	entry->flags = (irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) |
+		((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
+		((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
+		(need_resched() ? TRACE_FLAG_NEED_RESCHED : 0);
 	memcpy(entry->comm, tsk->comm, TASK_COMM_LEN);
 }

@@ -84,7 +102,7 @@ static notrace void trace_function(const
 
 	atomic_inc(&tr->data[cpu]->disabled);
 	if (likely(atomic_read(&tr->data[cpu]->disabled) == 1))
-		mctracer_add_trace_entry(tr, cpu, ip, parent_ip);
+		mctracer_add_trace_entry(tr, cpu, ip, parent_ip, flags);
 
 	atomic_dec(&tr->data[cpu]->disabled);

@@ -101,6 +119,11 @@ static notrace void mctracer_reset(struc
 {
 	int cpu;
 
+	tr->time_start = now();
+	tr->saved_latency = 0;
+	tr->critical_start = 0;
+	tr->critical_end = 0;
+
 	for_each_online_cpu(cpu) {
 		tr->data[cpu]->trace_idx = 0;
 		atomic_set(&tr->data[cpu]->underrun, 0);

@@ -110,11 +133,24 @@ static notrace void mctracer_reset(struc
 #ifdef CONFIG_DEBUG_FS
 enum trace_iterator {
 	TRACE_ITER_SYM_ONLY	= 1,
+	TRACE_ITER_VERBOSE	= 2,
+};
+
+/* These must match the bit postions above */
+static const char *trace_options[] = {
+	"symonly",
+	"verbose",
+	NULL
+};
+
+enum trace_file_type {
+	TRACE_FILE_LAT_FMT	= 1,
 };
 
 struct mctracer_iterator {
 	struct mctracer_trace *tr;
 	struct mctracer_entry *ent;
+	unsigned long iter_flags;
 	loff_t pos;
 	unsigned long next_idx[NR_CPUS];
 	int cpu;

@@ -140,37 +176,53 @@ static struct mctracer_entry *mctracer_e
 	return &array[idx];
 }
 
-static void *find_next_entry(struct mctracer_iterator *iter)
+static struct notrace mctracer_entry *
+find_next_entry(struct mctracer_iterator *iter, int *ent_cpu)
 {
 	struct mctracer_trace *tr = iter->tr;
-	struct mctracer_entry *ent;
-	struct mctracer_entry *next = NULL;
-	int next_i = -1;
-	int i;
+	struct mctracer_entry *ent, *next = NULL;
+	int next_cpu = -1;
+	int cpu;
 
-	for_each_possible_cpu(i) {
-		if (!tr->data[i]->trace)
+	for_each_possible_cpu(cpu) {
+		if (!tr->data[cpu]->trace)
 			continue;
-		ent = mctracer_entry_idx(tr, iter->next_idx[i], i);
+		ent = mctracer_entry_idx(tr, iter->next_idx[cpu], cpu);
 		if (ent && (!next || next->t > ent->t)) {
 			next = ent;
-			next_i = i;
+			next_cpu = cpu;
 		}
 	}
+
+	if (ent_cpu)
+		*ent_cpu = next_cpu;
+
+	return next;
+}
+
+static void *find_next_entry_inc(struct mctracer_iterator *iter)
+{
+	struct mctracer_entry *next;
+	int next_cpu = -1;
+
+	next = find_next_entry(iter, &next_cpu);
+
 	if (next) {
-		iter->next_idx[next_i]++;
+		iter->next_idx[next_cpu]++;
 		iter->idx++;
 	}
 	iter->ent = next;
-	iter->cpu = next_i;
+	iter->cpu = next_cpu;
 
 	return next ? iter : NULL;
 }
 
-static void *s_next(struct seq_file *m, void *v, loff_t *pos)
+static void notrace *
+s_next(struct seq_file *m, void *v, loff_t *pos)
 {
 	struct mctracer_iterator *iter = m->private;
 	void *ent;
+	void *last_ent = iter->ent;
 	int i = (int)*pos;
 
 	(*pos)++;

@@ -180,15 +232,18 @@ static void *s_next(struct seq_file *m, 
 		return NULL;
 
 	if (iter->idx < 0)
-		ent = find_next_entry(iter);
+		ent = find_next_entry_inc(iter);
 	else
 		ent = iter;
 
 	while (ent && iter->idx < i)
-		ent = find_next_entry(iter);
+		ent = find_next_entry_inc(iter);
 
 	iter->pos = *pos;
 
+	if (last_ent && !ent)
+		seq_puts(m, "\n\nvim:ft=help\n");
+
 	return ent;
 }

@@ -249,40 +304,239 @@ static void seq_print_symbol(struct seq_
 #endif
 
 static void notrace seq_print_ip_sym(struct seq_file *m,
-				     unsigned long ip,
-				     int sym_only)
+				     unsigned long ip, int sym_only)
 {
+	if (!ip) {
+		seq_printf(m, "0");
+		return;
+	}
+
 	seq_print_symbol(m, "%s", ip);
 	if (!sym_only)
 		seq_printf(m, " <" IP_FMT ">", ip);
 }
 
+static void notrace print_help_header(struct seq_file *m)
+{
+	seq_puts(m, "                 _------=> CPU#            \n");
+	seq_puts(m, "                / _-----=> irqs-off        \n");
+	seq_puts(m, "               | / _----=> need-resched    \n");
+	seq_puts(m, "               || / _---=> hardirq/softirq \n");
+	seq_puts(m, "               ||| / _--=> preempt-depth   \n");
+	seq_puts(m, "               |||| /                      \n");
+	seq_puts(m, "               |||||     delay             \n");
+	seq_puts(m, "   cmd     pid ||||| time  |   caller      \n");
+	seq_puts(m, "      \\   /    |||||   \\   |   /           \n");
+}
+
+static void notrace print_trace_header(struct seq_file *m,
+				       struct mctracer_iterator *iter)
+{
+	struct mctracer_trace *tr = iter->tr;
+	unsigned long underruns = 0;
+	unsigned long underrun;
+	unsigned long entries   = 0;
+	int sym_only = !!(tr->iter_flags & TRACE_ITER_SYM_ONLY);
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		if (tr->data[cpu]->trace) {
+			underrun = atomic_read(&tr->data[cpu]->underrun);
+			if (underrun) {
+				underruns += underrun;
+				entries += tr->entries;
+			} else
+				entries += tr->data[cpu]->trace_idx;
+		}
+	}
+
+	seq_printf(m, "preemption latency trace v1.1.5 on %s\n",
+		   UTS_RELEASE);
+	seq_puts(m, "-----------------------------------"
+		 "---------------------------------\n");
+	seq_printf(m, " latency: %lu us, #%lu/%lu, CPU#%d |"
+		   " (M:%s VP:%d, KP:%d, SP:%d HP:%d",
+		   cycles_to_usecs(tr->saved_latency),
+		   entries,
+		   (entries + underruns),
+		   smp_processor_id(),
+#if defined(CONFIG_PREEMPT_NONE)
+		   "server",
+#elif defined(CONFIG_PREEMPT_VOLUNTARY)
+		   "desktop",
+#elif defined(CONFIG_PREEMPT_DESKTOP)
+		   "preempt",
+#else
+		   "rt",
+#endif
+		   /* These are reserved for later use */
+		   0, 0, 0, 0);
+#ifdef CONFIG_SMP
+	seq_printf(m, " #P:%d)\n", num_online_cpus());
+#else
+	seq_puts(m, ")\n");
+#endif
+	seq_puts(m, "    -----------------\n");
+	seq_printf(m, "    | task: %.16s-%d "
+		   "(uid:%d nice:%ld policy:%ld rt_prio:%ld)\n",
+		   tr->comm, tr->pid, tr->uid, tr->nice,
+		   tr->policy, tr->rt_priority);
+	seq_puts(m, "    -----------------\n");
+
+	if (tr->critical_start) {
+		seq_puts(m, " => started at: ");
+		seq_print_ip_sym(m, tr->critical_start, sym_only);
+		seq_puts(m, "\n => ended at:   ");
+		seq_print_ip_sym(m, tr->critical_end, sym_only);
+		seq_puts(m, "\n");
+	}
+
+	seq_puts(m, "\n");
+}
+
+
+static void notrace
+lat_print_generic(struct seq_file *m, struct mctracer_entry *entry, int cpu)
+{
+	int hardirq, softirq;
+
+	seq_printf(m, "%8.8s-%-5d ", entry->comm, entry->pid);
+	seq_printf(m, "%d", cpu);
+	seq_printf(m, "%c%c",
+		   (entry->flags & TRACE_FLAG_IRQS_OFF) ? 'd' : '.',
+		   ((entry->flags & TRACE_FLAG_NEED_RESCHED) ? 'N' : '.'));
+
+	hardirq = entry->flags & TRACE_FLAG_HARDIRQ;
+	softirq = entry->flags & TRACE_FLAG_SOFTIRQ;
+	if (hardirq && softirq)
+		seq_putc(m, 'H');
+	else {
+		if (hardirq)
+			seq_putc(m, 'h');
+		else {
+			if (softirq)
+				seq_putc(m, 's');
+			else
+				seq_putc(m, '.');
+		}
+	}
+
+	if (entry->preempt_count)
+		seq_printf(m, "%lx", entry->preempt_count);
+	else
+		seq_puts(m, ".");
+}
+
+unsigned long preempt_mark_thresh = 100;
+
+static void notrace
+lat_print_timestamp(struct seq_file *m, unsigned long long abs_usecs,
+		    unsigned long rel_usecs)
+{
+	seq_printf(m, " %4lldus", abs_usecs);
+	if (rel_usecs > preempt_mark_thresh)
+		seq_puts(m, "!: ");
+	else if (rel_usecs > 1)
+		seq_puts(m, "+: ");
+	else
+		seq_puts(m, " : ");
+}
+
+static void notrace
+print_lat_fmt(struct seq_file *m, struct mctracer_iterator *iter,
+	      unsigned int trace_idx, int cpu)
+{
+	struct mctracer_entry *entry = iter->ent;
+	struct mctracer_entry *next_entry = find_next_entry(iter, NULL);
+	unsigned long abs_usecs;
+	unsigned long rel_usecs;
+	int sym_only = !!(iter->tr->iter_flags & TRACE_ITER_SYM_ONLY);
+	int verbose = !!(iter->tr->iter_flags & TRACE_ITER_VERBOSE);
+
+	if (!next_entry)
+		next_entry = entry;
+	rel_usecs = cycles_to_usecs(next_entry->t - entry->t);
+	abs_usecs = cycles_to_usecs(entry->t - iter->tr->time_start);
+
+	if (verbose) {
+		seq_printf(m, "%16s %5d %d %ld %08lx %08x [%08lx]"
+			   " %ld.%03ldms (+%ld.%03ldms): ",
+			   entry->comm,
+			   entry->pid, cpu, entry->flags,
+			   entry->preempt_count, trace_idx,
+			   cycles_to_usecs(entry->t),
+			   abs_usecs/1000,
+			   abs_usecs % 1000, rel_usecs/1000, rel_usecs % 1000);
+	} else {
+		lat_print_generic(m, entry, cpu);
+		lat_print_timestamp(m, abs_usecs, rel_usecs);
+	}
+	seq_print_ip_sym(m, entry->ip, sym_only);
+	seq_puts(m, " (");
+	seq_print_ip_sym(m, entry->parent_ip, sym_only);
+	seq_puts(m, ")\n");
+}
+
+static void notrace print_trace_fmt(struct seq_file *m,
+				    struct mctracer_iterator *iter)
+{
+	unsigned long usec_rem;
+	unsigned long secs;
+	int sym_only = !!(iter->tr->iter_flags & TRACE_ITER_SYM_ONLY);
+	unsigned long long t;
+
+	t = cycles_to_usecs(iter->ent->t);
+	usec_rem = do_div(t, 1000000ULL);
+	secs = (unsigned long)t;
+
+	seq_printf(m, "[%5lu.%06lu] ", secs, usec_rem);
+	seq_printf(m, "CPU %d: ", iter->cpu);
+	seq_printf(m, "%s:%d ", iter->ent->comm,
+		   iter->ent->pid);
+	seq_print_ip_sym(m, iter->ent->ip, sym_only);
+	if (iter->ent->parent_ip) {
+		seq_printf(m, " <-- ");
+		seq_print_ip_sym(m, iter->ent->parent_ip,
+				 sym_only);
+	}
+	seq_printf(m, "\n");
+}
+
+static int trace_empty(struct mctracer_iterator *iter)
+{
+	struct mctracer_trace_cpu *data;
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		data = iter->tr->data[cpu];
+
+		if (data->trace &&
+		    (data->trace_idx ||
+		     atomic_read(&data->underrun)))
+			return 0;
+	}
+	return 1;
+}
+
 static int s_show(struct seq_file *m, void *v)
 {
 	struct mctracer_iterator *iter = v;
-	int sym_only = !!(iter->tr->iter_flags & TRACE_ITER_SYM_ONLY);
 
 	if (iter->ent == NULL) {
-		seq_printf(m, "mctracer:\n");
+		if (iter->iter_flags & TRACE_FILE_LAT_FMT) {
+			/* print nothing if the buffers are empty */
+			if (trace_empty(iter))
+				return 0;
+			print_trace_header(m, iter);
+			if (!(iter->tr->iter_flags & TRACE_ITER_VERBOSE))
+				print_help_header(m);
+		} else
+			seq_printf(m, "mctracer:\n");
 	} else {
-		unsigned long long t;
-		unsigned long usec_rem;
-		unsigned long secs;
-
-		t = cycles_to_usecs(iter->ent->t);
-		usec_rem = do_div(t, 1000000ULL);
-		secs = (unsigned long)t;
-
-		seq_printf(m, "[%5lu.%06lu] ", secs, usec_rem);
-		seq_printf(m, "CPU %d: ", iter->cpu);
-		seq_printf(m, "%s:%d ", iter->ent->comm, iter->ent->pid);
-		seq_print_ip_sym(m, iter->ent->ip, sym_only);
-		if (iter->ent->parent_ip) {
-			seq_printf(m, " <-- ");
-			seq_print_ip_sym(m, iter->ent->parent_ip,
-					 sym_only);
-		}
-		seq_printf(m, "\n");
+		if (iter->iter_flags & TRACE_FILE_LAT_FMT)
+			print_lat_fmt(m, iter, iter->idx, iter->cpu);
+		else
+			print_trace_fmt(m, iter);
 	}
 
 	return 0;

@@ -295,25 +549,52 @@ static struct seq_operations mctrace_seq
 	.show = s_show,
 };
 
-static int mctrace_open(struct inode *inode, struct file *file)
+static struct mctracer_iterator *
+__mctrace_open(struct inode *inode, struct file *file, int *ret)
 {
 	struct mctracer_iterator *iter;
-	int ret;
 
 	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
-	if (!iter)
-		return -ENOMEM;
+	if (!iter) {
+		*ret = -ENOMEM;
+		goto out;
+	}
 
 	iter->tr = &mctracer_trace;
 	iter->pos = -1;
 
 	/* TODO stop tracer */
-	ret = seq_open(file, &mctrace_seq_ops);
-	if (!ret) {
+	*ret = seq_open(file, &mctrace_seq_ops);
+	if (!*ret) {
 		struct seq_file *m = file->private_data;
 		m->private = iter;
-	} else
+	} else {
 		kfree(iter);
+		iter = NULL;
+	}
+
+ out:
+	return iter;
+}
+
+static int mctrace_open(struct inode *inode, struct file *file)
+{
+	int ret;
+
+	__mctrace_open(inode, file, &ret);
+
+	return ret;
+}
+
+static int mctrace_lt_open(struct inode *inode, struct file *file)
+{
+	struct mctracer_iterator *iter;
+	int ret;
+
+	iter = __mctrace_open(inode, file, &ret);
+
+	if (!ret)
+		iter->iter_flags |= TRACE_FILE_LAT_FMT;
 
 	return ret;
 }

@@ -335,6 +616,13 @@ static struct file_operations mctrace_fo
 	.release = mctrace_release,
 };
 
+static struct file_operations mctrace_lt_fops = {
+	.open = mctrace_lt_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = mctrace_release,
+};
+
 static int mctracer_open_generic(struct inode *inode, struct file *filp)
 {
 	filp->private_data = inode->i_private;

@@ -375,6 +663,17 @@ static ssize_t mctracer_ctrl_write(struc
 	/* When starting a new trace, reset the buffers */
 	if (val)
 		mctracer_reset(tr);
+	else {
+		/* pretty meaningless for now */
+		tr->time_end = now();
+		tr->saved_latency = tr->time_end - tr->time_start;
+		memcpy(tr->comm, current->comm, TASK_COMM_LEN);
+		tr->pid = current->pid;
+		tr->uid = current->uid;
+		tr->nice = current->static_prio - 20 - MAX_RT_PRIO;
+		tr->policy = current->policy;
+		tr->rt_priority = current->rt_priority;
+	}
 
 	if (tr->ctrl ^ val) {
 		if (val)

@@ -399,15 +698,38 @@ static ssize_t mctracer_iter_ctrl_read(s
 				       size_t cnt, loff_t *ppos)
 {
 	struct mctracer_trace *tr = filp->private_data;
-	char buf[64];
+	char *buf;
 	int r = 0;
+	int i;
+	int len = 0;
 
-	if (tr->iter_flags & TRACE_ITER_SYM_ONLY)
-		r = sprintf(buf, "%s", "symonly ");
-	r += sprintf(buf+r, "\n");
+	/* calulate max size */
+	for (i = 0; trace_options[i]; i++) {
+		len += strlen(trace_options[i]);
+		len += 3; /* "no" and space */
+	}
 
-	return simple_read_from_buffer(ubuf, cnt, ppos,
-				       buf, r);
+	/* +2 for \n and \0 */
+	buf = kmalloc(len + 2, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	for (i = 0; trace_options[i]; i++) {
+		if (tr->iter_flags & (1 << i))
+			r += sprintf(buf + r, "%s ", trace_options[i]);
+		else
+			r += sprintf(buf + r, "no%s ", trace_options[i]);
+	}
+
+	r += sprintf(buf + r, "\n");
+	WARN_ON(r >= len + 2);
+
+	r = simple_read_from_buffer(ubuf, cnt, ppos,
+				    buf, r);
+
+	kfree(buf);
+
+	return r;
 }
 
 static ssize_t mctracer_iter_ctrl_write(struct file *filp,

@@ -416,6 +738,9 @@ static ssize_t mctracer_iter_ctrl_write(
 {
 	struct mctracer_trace *tr = filp->private_data;
 	char buf[64];
+	char *cmp = buf;
+	int neg = 0;
+	int i;
 
 	if (cnt > 63)
 		cnt = 63;

@@ -425,8 +750,22 @@ static ssize_t mctracer_iter_ctrl_write(
 
 	buf[cnt] = 0;
 
-	if (strncmp(buf, "symonly", 7) == 0)
-		tr->iter_flags |= TRACE_ITER_SYM_ONLY;
+	if (strncmp(buf, "no", 2) == 0) {
+		neg = 1;
+		cmp += 2;
+	}
+
+	for (i = 0; trace_options[i]; i++) {
+		int len = strlen(trace_options[i]);
+
+		if (strncmp(cmp, trace_options[i], len) == 0) {
+			if (neg)
+				tr->iter_flags &= ~(1 << i);
+			else
+				tr->iter_flags |= (1 << i);
+			break;
+		}
+	}
 
 	filp->f_pos += cnt;

@@ -459,6 +798,12 @@ static void mctrace_init_debugfs(void)
 				    &mctracer_trace, &mctracer_iter_fops);
 	if (!entry)
 		pr_warning("Could not create debugfs 'iter_ctrl' entry\n");
+
+	entry = debugfs_create_file("function_trace", 0444, d_mctracer,
+				    &mctracer_trace, &mctrace_lt_fops);
+	if (!entry)
+		pr_warning("Could not create debugfs 'function_trace' entry\n");
+
 	entry = debugfs_create_file("trace", 0444, d_mctracer,
 				    &mctracer_trace, &mctrace_fops);
 	if (!entry)

lib/tracing/tracer.h +13 -0

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -8,6 +8,8 @@ struct mctracer_entry {
 	unsigned long long t;
 	unsigned long ip;
 	unsigned long parent_ip;
+	unsigned long preempt_count;
+	unsigned long flags;
 	char comm[TASK_COMM_LEN];
 	pid_t pid;
 };

@@ -23,6 +25,17 @@ struct mctracer_trace {
 	unsigned long entries;
 	long	      ctrl;
 	unsigned long iter_flags;
+	char comm[TASK_COMM_LEN];
+	pid_t	      pid;
+	uid_t	      uid;
+	unsigned long nice;
+	unsigned long policy;
+	unsigned long rt_priority;
+	unsigned long saved_latency;
+	unsigned long critical_start;
+	unsigned long critical_end;
+	unsigned long long time_start;
+	unsigned long long time_end;
 	struct mctracer_trace_cpu *data[NR_CPUS];
 };

Steven Rostedt

author reviewer

2008-01-15 21:00 UTC #26 lore raw

RFC PATCH v3 24/30 Split out specific tracing functions

In reply to Steven Rostedt

Several different types of tracing needs to use the same core functions. This patch separates the core functions from more specific ones to allow for future tracing methods.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/Makefile                   |    1 
 lib/tracing/Kconfig            |    7 
 lib/tracing/Makefile           |    3 
 lib/tracing/trace_function.c   |  180 +++++++++++++
 lib/tracing/tracer.c           |  537 +++++++++++++++++------------------------
 lib/tracing/tracer.h           |   91 +++++-
 lib/tracing/tracer_interface.h |   14 -
 7 files changed, 485 insertions(+), 348 deletions(-)

lib/Makefile +1 -0

Index: linux-compile.git/lib/Makefile
===================================================================
@@ -67,6 +67,7 @@ obj-$(CONFIG_SWIOTLB) += swiotlb.o
 obj-$(CONFIG_FAULT_INJECTION) += fault-inject.o
 
 obj-$(CONFIG_MCOUNT) += tracing/
+obj-$(CONFIG_TRACING) += tracing/
 
 lib-$(CONFIG_GENERIC_BUG) += bug.o

lib/tracing/Kconfig +6 -1

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -9,11 +9,16 @@ config MCOUNT
 	bool
 	select FRAME_POINTER
 
-config MCOUNT_TRACER
+config TRACING
+        bool
+	depends on DEBUG_KERNEL
+
+config FUNCTION_TRACER
 	bool "Profiler instrumentation based tracer"
 	depends on DEBUG_KERNEL && HAVE_MCOUNT
 	default n
 	select MCOUNT
+	select TRACING
 	help
 	  Use profiler instrumentation, adding -pg to CFLAGS. This will
 	  insert a call to an architecture specific __mcount routine,

lib/tracing/Makefile +2 -1

Index: linux-compile.git/lib/tracing/Makefile
===================================================================
@@ -1,5 +1,6 @@
 obj-$(CONFIG_MCOUNT) += libmcount.o
 
-obj-$(CONFIG_MCOUNT_TRACER) += tracer.o
+obj-$(CONFIG_TRACING) += tracer.o
+obj-$(CONFIG_FUNCTION_TRACER) += trace_function.o
 
 libmcount-y := mcount.o

lib/tracing/trace_function.c new file +180 -0

180 lines not on this page · raw message

lib/tracing/tracer.c +224 ~~-313~~

824 lines not on this page · raw message

lib/tracing/tracer.h +72 ~~-19~~

112 lines not on this page · raw message

lib/tracing/tracer_interface.h deleted +0 ~~-14~~

Index: linux-compile.git/lib/tracing/tracer_interface.h
===================================================================
@@ -1,14 +0,0 @@
-#ifndef _LINUX_MCTRACER_INTERFACE_H
-#define _LINUX_MCTRACER_INTERFACE_H
-
-#include "tracer.h"
-
-/*
- * Will be at least sizeof(struct mctracer_entry), but callers can request more
- * space for private stuff, such as a timestamp, preempt_count, etc.
- */
-#define MCTRACER_ENTRY_SIZE sizeof(struct mctracer_entry)
-
-#define MCTRACER_NR_ENTRIES (65536UL)
-
-#endif /* _LINUX_MCTRACER_INTERFACE_H */

Steven Rostedt

author reviewer

2008-01-15 20:52 UTC #27 lore raw

RFC PATCH v3 25/30 Trace irq disabled critical timings

In reply to Steven Rostedt

This patch adds latency tracing for critical timings. In /debugfs/tracing/ three files are added:

preempt_max_latency
holds the max latency thus far (in usecs)
(default to large number so one must start latency tracing)

preempt_thresh
threshold (in usecs) to always print out if irqs off is detected to be longer than stated here. If irq_thresh is non-zero, then max_irq_latency is ignored.

preempt_trace
Trace of where the latecy was detected.

preempt_fn_trace_ctrl
0 - don't use mcount
1 - use mcount to trace

Here's an example of a trace with preempt_fn_trace_ctrl == 0

=======
preemption latency trace v1.1.5 on 2.6.24-rc7
--------------------------------------------------------------------
latency: 100 us, #3/3, CPU#1 | (M:rt VP:0, KP:0, SP:0 HP:0 #P:2)

    -----------------
    | task: swapper-0 (uid:0 nice:0 policy:0 rt_prio:0)
    -----------------

=> started at: _spin_lock_irqsave+0x2a/0xb7
=> ended at: _spin_unlock_irqrestore+0x32/0x5f

                 _------=> CPU#            
                / _-----=> irqs-off        
               | / _----=> need-resched    
               || / _---=> hardirq/softirq 
               ||| / _--=> preempt-depth   
               |||| /                      
               |||||     delay             
   cmd     pid ||||| time  |   caller      
      \   /    |||||   \   |   /           
 swapper-0     1d.s3    0us+: _spin_lock_irqsave+0x2a/0xb7 (e1000_update_stats+0x47/0x64c [e1000])
 swapper-0     1d.s3  100us : _spin_unlock_irqrestore+0x32/0x5f (e1000_update_stats+0x641/0x64c [e1000])
 swapper-0     1d.s3  100us : trace_hardirqs_on_caller+0x75/0x89 (_spin_unlock_irqrestore+0x32/0x5f)

vim:ft=help
=======

And this is a trace with irqsoff_fn_trace_ctrl == 1

=======
preemption latency trace v1.1.5 on 2.6.24-rc7
--------------------------------------------------------------------
latency: 102 us, #12/12, CPU#1 | (M:rt VP:0, KP:0, SP:0 HP:0 #P:2)

    -----------------
    | task: swapper-0 (uid:0 nice:0 policy:0 rt_prio:0)
    -----------------

=> started at: _spin_lock_irqsave+0x2a/0xb7
=> ended at: _spin_unlock_irqrestore+0x32/0x5f

                 _------=> CPU#            
                / _-----=> irqs-off        
               | / _----=> need-resched    
               || / _---=> hardirq/softirq 
               ||| / _--=> preempt-depth   
               |||| /                      
               |||||     delay             
   cmd     pid ||||| time  |   caller      
      \   /    |||||   \   |   /           
 swapper-0     1dNs3    0us+: _spin_lock_irqsave+0x2a/0xb7 (e1000_update_stats+0x47/0x64c [e1000])
 swapper-0     1dNs3   46us : e1000_read_phy_reg+0x16/0x225 [e1000] (e1000_update_stats+0x5e2/0x64c [e1000])
 swapper-0     1dNs3   46us : e1000_swfw_sync_acquire+0x10/0x99 [e1000] (e1000_read_phy_reg+0x49/0x225 [e1000])
 swapper-0     1dNs3   46us : e1000_get_hw_eeprom_semaphore+0x12/0xa6 [e1000] (e1000_swfw_sync_acquire+0x36/0x99 [e1000])
 swapper-0     1dNs3   47us : __const_udelay+0x9/0x47 (e1000_read_phy_reg+0x116/0x225 [e1000])
 swapper-0     1dNs3   47us+: __delay+0x9/0x50 (__const_udelay+0x45/0x47)
 swapper-0     1dNs3   97us : preempt_schedule+0xc/0x84 (__delay+0x4e/0x50)
 swapper-0     1dNs3   98us : e1000_swfw_sync_release+0xc/0x55 [e1000] (e1000_read_phy_reg+0x211/0x225 [e1000])
 swapper-0     1dNs3   99us+: e1000_put_hw_eeprom_semaphore+0x9/0x35 [e1000] (e1000_swfw_sync_release+0x50/0x55 [e1000])
 swapper-0     1dNs3  101us : _spin_unlock_irqrestore+0xe/0x5f (e1000_update_stats+0x641/0x64c [e1000])
 swapper-0     1dNs3  102us : _spin_unlock_irqrestore+0x32/0x5f (e1000_update_stats+0x641/0x64c [e1000])
 swapper-0     1dNs3  102us : trace_hardirqs_on_caller+0x75/0x89 (_spin_unlock_irqrestore+0x32/0x5f)

vim:ft=help
=======

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/process_64.c  |    3 
 arch/x86/lib/thunk_64.S       |   18 +
 include/asm-x86/irqflags_32.h |    4 
 include/asm-x86/irqflags_64.h |    4 
 include/linux/irqflags.h      |   37 ++
 include/linux/mcount.h        |   31 ++
 kernel/fork.c                 |    2 
 kernel/lockdep.c              |   25 +
 lib/tracing/Kconfig           |   18 +
 lib/tracing/Makefile          |    1 
 lib/tracing/trace_irqsoff.c   |  558 ++++++++++++++++++++++++++++++++++++++++++
 11 files changed, 680 insertions(+), 21 deletions(-)

arch/x86/kernel/process_64.c +3 -0

Index: linux-compile.git/arch/x86/kernel/process_64.c
===================================================================
@@ -233,7 +233,10 @@ void cpu_idle (void)
 			 */
 			local_irq_disable();
 			enter_idle();
+			/* Don't trace irqs off for idle */
+			stop_critical_timings();
 			idle();
+			start_critical_timings();
 			/* In many cases the interrupt that ended idle
 			   has already called exit_idle. But some idle
 			   loops can be woken up without interrupt. */

arch/x86/lib/thunk_64.S +16 -2

Index: linux-compile.git/arch/x86/lib/thunk_64.S
===================================================================
@@ -47,8 +47,22 @@
 	thunk __up_wakeup,__up
 
 #ifdef CONFIG_TRACE_IRQFLAGS
-	thunk trace_hardirqs_on_thunk,trace_hardirqs_on
-	thunk trace_hardirqs_off_thunk,trace_hardirqs_off
+	/* put return address in rdi (arg1) */
+	.macro thunk_ra name,func
+	.globl \name
+\name:
+	CFI_STARTPROC
+	SAVE_ARGS
+	/* SAVE_ARGS pushs 9 elements */
+	/* the next element would be the rip */
+	movq 9*8(%rsp), %rdi
+	call \func
+	jmp  restore
+	CFI_ENDPROC
+	.endm
+
+	thunk_ra trace_hardirqs_on_thunk,trace_hardirqs_on_caller
+	thunk_ra trace_hardirqs_off_thunk,trace_hardirqs_off_caller
 #endif
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC

include/asm-x86/irqflags_32.h +2 -2

Index: linux-compile.git/include/asm-x86/irqflags_32.h
===================================================================
@@ -139,9 +139,9 @@ static inline int raw_irqs_disabled(void
 static inline void trace_hardirqs_fixup_flags(unsigned long flags)
 {
 	if (raw_irqs_disabled_flags(flags))
-		trace_hardirqs_off();
+		__trace_hardirqs_off();
 	else
-		trace_hardirqs_on();
+		__trace_hardirqs_on();
 }
 
 static inline void trace_hardirqs_fixup(void)

include/asm-x86/irqflags_64.h +2 -2

Index: linux-compile.git/include/asm-x86/irqflags_64.h
===================================================================
@@ -120,9 +120,9 @@ static inline int raw_irqs_disabled(void
 static inline void trace_hardirqs_fixup_flags(unsigned long flags)
 {
 	if (raw_irqs_disabled_flags(flags))
-		trace_hardirqs_off();
+		__trace_hardirqs_off();
 	else
-		trace_hardirqs_on();
+		__trace_hardirqs_on();
 }
 
 static inline void trace_hardirqs_fixup(void)

include/linux/irqflags.h +29 -8

Index: linux-compile.git/include/linux/irqflags.h
===================================================================
@@ -12,10 +12,21 @@
 #define _LINUX_TRACE_IRQFLAGS_H
 
 #ifdef CONFIG_TRACE_IRQFLAGS
-  extern void trace_hardirqs_on(void);
-  extern void trace_hardirqs_off(void);
+# include <linux/mcount.h>
+  extern void trace_hardirqs_on_caller(unsigned long ip);
+  extern void trace_hardirqs_off_caller(unsigned long ip);
   extern void trace_softirqs_on(unsigned long ip);
   extern void trace_softirqs_off(unsigned long ip);
+  extern void trace_hardirqs_on(void);
+  extern void trace_hardirqs_off(void);
+  static inline void notrace __trace_hardirqs_on(void)
+  {
+	trace_hardirqs_on_caller(CALLER_ADDR0);
+  }
+  static inline void notrace __trace_hardirqs_off(void)
+  {
+	trace_hardirqs_off_caller(CALLER_ADDR0);
+  }
 # define trace_hardirq_context(p)	((p)->hardirq_context)
 # define trace_softirq_context(p)	((p)->softirq_context)
 # define trace_hardirqs_enabled(p)	((p)->hardirqs_enabled)

@@ -28,6 +39,8 @@
 #else
 # define trace_hardirqs_on()		do { } while (0)
 # define trace_hardirqs_off()		do { } while (0)
+# define __trace_hardirqs_on()		do { } while (0)
+# define __trace_hardirqs_off()		do { } while (0)
 # define trace_softirqs_on(ip)		do { } while (0)
 # define trace_softirqs_off(ip)		do { } while (0)
 # define trace_hardirq_context(p)	0

@@ -41,24 +54,32 @@
 # define INIT_TRACE_IRQFLAGS
 #endif
 
+#ifdef CONFIG_CRITICAL_IRQSOFF_TIMING
+ extern void stop_critical_timings(void);
+ extern void start_critical_timings(void);
+#else
+# define stop_critical_timings() do { } while (0)
+# define start_critical_timings() do { } while (0)
+#endif
+
 #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
 
 #include <asm/irqflags.h>
 
 #define local_irq_enable() \
-	do { trace_hardirqs_on(); raw_local_irq_enable(); } while (0)
+	do { __trace_hardirqs_on(); raw_local_irq_enable(); } while (0)
 #define local_irq_disable() \
-	do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0)
+	do { raw_local_irq_disable(); __trace_hardirqs_off(); } while (0)
 #define local_irq_save(flags) \
-	do { raw_local_irq_save(flags); trace_hardirqs_off(); } while (0)
+	do { raw_local_irq_save(flags); __trace_hardirqs_off(); } while (0)
 
 #define local_irq_restore(flags)				\
 	do {							\
 		if (raw_irqs_disabled_flags(flags)) {		\
 			raw_local_irq_restore(flags);		\
-			trace_hardirqs_off();			\
+			__trace_hardirqs_off();			\
 		} else {					\
-			trace_hardirqs_on();			\
+			__trace_hardirqs_on();			\
 			raw_local_irq_restore(flags);		\
 		}						\
 	} while (0)

@@ -76,7 +97,7 @@
 #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
 #define safe_halt()						\
 	do {							\
-		trace_hardirqs_on();				\
+		__trace_hardirqs_on();				\
 		raw_safe_halt();				\
 	} while (0)

include/linux/mcount.h +27 -4

Index: linux-compile.git/include/linux/mcount.h
===================================================================
@@ -6,10 +6,6 @@ extern int mcount_enabled;
 
 #include <linux/linkage.h>
 
-#define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
-#define CALLER_ADDR1 ((unsigned long)__builtin_return_address(1))
-#define CALLER_ADDR2 ((unsigned long)__builtin_return_address(2))
-
 typedef void (*mcount_func_t)(unsigned long ip, unsigned long parent_ip);
 
 struct mcount_ops {

@@ -35,4 +31,31 @@ extern void mcount(void);
 # define unregister_mcount_function(ops) do { } while (0)
 # define clear_mcount_function(ops) do { } while (0)
 #endif /* CONFIG_MCOUNT */
+
+
+#ifdef CONFIG_FRAME_POINTER
+/* TODO: need to fix this for ARM */
+# define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
+# define CALLER_ADDR1 ((unsigned long)__builtin_return_address(1))
+# define CALLER_ADDR2 ((unsigned long)__builtin_return_address(2))
+# define CALLER_ADDR3 ((unsigned long)__builtin_return_address(3))
+# define CALLER_ADDR4 ((unsigned long)__builtin_return_address(4))
+# define CALLER_ADDR5 ((unsigned long)__builtin_return_address(5))
+#else
+# define CALLER_ADDR0 ((unsigned long)__builtin_return_address(0))
+# define CALLER_ADDR1 0UL
+# define CALLER_ADDR2 0UL
+# define CALLER_ADDR3 0UL
+# define CALLER_ADDR4 0UL
+# define CALLER_ADDR5 0UL
+#endif
+
+#ifdef CONFIG_CRITICAL_IRQSOFF_TIMING
+  extern void notrace time_hardirqs_on(unsigned long a0, unsigned long a1);
+  extern void notrace time_hardirqs_off(unsigned long a0, unsigned long a1);
+#else
+# define time_hardirqs_on(a0, a1)		do { } while (0)
+# define time_hardirqs_off(a0, a1)		do { } while (0)
+#endif
+
 #endif /* _LINUX_MCOUNT_H */

kernel/fork.c +1 -1

Index: linux-compile.git/kernel/fork.c
===================================================================
@@ -1010,7 +1010,7 @@ static struct task_struct *copy_process(
 
 	rt_mutex_init_task(p);
 
-#ifdef CONFIG_TRACE_IRQFLAGS
+#if defined(CONFIG_TRACE_IRQFLAGS) && defined(CONFIG_LOCKDEP)
 	DEBUG_LOCKS_WARN_ON(!p->hardirqs_enabled);
 	DEBUG_LOCKS_WARN_ON(!p->softirqs_enabled);
 #endif

kernel/lockdep.c +23 -2

Index: linux-compile.git/kernel/lockdep.c
===================================================================
@@ -39,6 +39,7 @@
 #include <linux/irqflags.h>
 #include <linux/utsname.h>
 #include <linux/hash.h>
+#include <linux/mcount.h>
 
 #include <asm/sections.h>

@@ -2009,7 +2010,7 @@ void early_boot_irqs_on(void)
 /*
  * Hardirqs will be enabled:
  */
-void trace_hardirqs_on(void)
+void notrace trace_hardirqs_on_caller(unsigned long a0)
 {
 	struct task_struct *curr = current;
 	unsigned long ip;

@@ -2050,14 +2051,27 @@ void trace_hardirqs_on(void)
 	curr->hardirq_enable_ip = ip;
 	curr->hardirq_enable_event = ++curr->irq_events;
 	debug_atomic_inc(&hardirqs_on_events);
+
+#ifdef CONFIG_CRITICAL_IRQSOFF_TIMING
+	time_hardirqs_on(CALLER_ADDR0, a0);
+#endif
 }
+EXPORT_SYMBOL(trace_hardirqs_on_caller);
 
+void notrace trace_hardirqs_on(void) {
+	trace_hardirqs_on_caller(CALLER_ADDR0);
+}
 EXPORT_SYMBOL(trace_hardirqs_on);
 
+void notrace trace_hardirqs_off(void) {
+	trace_hardirqs_off_caller(CALLER_ADDR0);
+}
+EXPORT_SYMBOL(trace_hardirqs_off);
+
 /*
  * Hardirqs were disabled:
  */
-void trace_hardirqs_off(void)
+void notrace trace_hardirqs_off_caller(unsigned long a0)
 {
 	struct task_struct *curr = current;

@@ -2075,10 +2089,17 @@ void trace_hardirqs_off(void)
 		curr->hardirq_disable_ip = _RET_IP_;
 		curr->hardirq_disable_event = ++curr->irq_events;
 		debug_atomic_inc(&hardirqs_off_events);
+#ifdef CONFIG_CRITICAL_IRQSOFF_TIMING
+		time_hardirqs_off(CALLER_ADDR0, a0);
+#endif
 	} else
 		debug_atomic_inc(&redundant_hardirqs_off);
 }
 
+void notrace trace_hardirqs_off(void) {
+	trace_hardirqs_off_caller(CALLER_ADDR0);
+}
+
 EXPORT_SYMBOL(trace_hardirqs_off);
 
 /*

lib/tracing/Kconfig +18 -0

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -24,3 +24,21 @@ config FUNCTION_TRACER
 	  insert a call to an architecture specific __mcount routine,
 	  that the debugging mechanism using this facility will hook by
 	  providing a set of inline routines.
+
+config CRITICAL_IRQSOFF_TIMING
+	bool "Interrupts-off critical section latency timing"
+	default n
+	depends on TRACE_IRQFLAGS_SUPPORT
+	depends on GENERIC_TIME
+	select TRACE_IRQFLAGS
+	select TRACING
+	help
+	  This option measures the time spent in irqs-off critical
+	  sections, with microsecond accuracy.
+
+	  The default measurement method is a maximum search, which is
+	  disabled by default and can be runtime (re-)started
+	  via:
+
+	      echo 0 > /debug/tracing/preempt_max_latency
+

lib/tracing/Makefile +1 -0

Index: linux-compile.git/lib/tracing/Makefile
===================================================================
@@ -2,5 +2,6 @@ obj-$(CONFIG_MCOUNT) += libmcount.o
 
 obj-$(CONFIG_TRACING) += tracer.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace_function.o
+obj-$(CONFIG_CRITICAL_IRQSOFF_TIMING) += trace_irqsoff.o
 
 libmcount-y := mcount.o

lib/tracing/trace_irqsoff.c new file +558 -0

558 lines not on this page · raw message

Steven Rostedt

author reviewer

2008-01-15 20:57 UTC #28 lore raw

RFC PATCH v3 26/30 Add context switch marker to sched.c

In reply to Steven Rostedt

The trace facilities here needs a hook into the context switch events. Since Mathieu Desnoyers has been working on markers for LTTng, I figured I take the marker he has in his patch queue.

The thing is that he only records the prev's pid, next's pid, and prev's state. It would suit me better if it simply passed in the prev and next pointer. Since I don't want to add markers all over the place to get the information that I need. The latency tracer only turns on for short periods of time and already takes up enough memory to store the traces, I don't want to add databases keeping track of changes such as priorities, and such, when I could simply get that information from a context switch.

Yes, that would slow down all context switches a little when tracing is on, but for this, it should be sufficient.

Anyway, I'm not about to start a war on this, so I simply folded and will use Mathieu's proposed marker instead. I'm just worried that in the future we will want more information about the processes in the context switch but will still be stuck with just the pid's and state.

Signed-off-by Steven Rostedt <[email protected]>

Cc Mathieu Desnoyers <[email protected]>

---
 kernel/sched.c |    3 +++
 1 file changed, 3 insertions(+)

kernel/sched.c +3 -0

Index: linux-compile.git/kernel/sched.c
===================================================================
@@ -1933,6 +1933,9 @@ context_switch(struct rq *rq, struct tas
 	struct mm_struct *mm, *oldmm;
 
 	prepare_task_switch(rq, prev, next);
+	trace_mark(kernel_sched_schedule,
+		   "prev_pid %d next_pid %d prev_state %ld",
+		   prev->pid, next->pid, prev->state);
 	mm = next->mm;
 	oldmm = prev->active_mm;
 	/*

Steven Rostedt

author reviewer

2008-01-15 20:55 UTC #29 lore raw

RFC PATCH v3 27/30 Add tracing of context switches

In reply to Steven Rostedt

This patch adds context switch tracing, of the format of:

                 _------=> CPU#            
                / _-----=> irqs-off        
               | / _----=> need-resched    
               || / _---=> hardirq/softirq 
               ||| / _--=> preempt-depth   
               |||| /                      
               |||||     delay             
   cmd     pid ||||| time  |   caller      
      \   /    |||||   \   |   /           
  <idle>-0     1d..3  361us!:  0:140:R --> 5028
    bash-5036  0d..3  552us+:  5036:120:D --> 0
  <idle>-0     0d..3  606us!:  0:140:R --> 5036
    sshd-5028  1d..3  943us!:  5028:120:S --> 0
  <idle>-0     1d..3 1316us+:  0:140:R --> 5028
    bash-5036  0d..3 1388us!:  5036:120:S --> 0
    sshd-5028  1d..3 1772us!:  5028:120:S --> 0

Two files are added to /debugfs/tracing

sched_trace - outputs the above format.

sched_trace_ctrl
0 - turns off context switch tracing.
1 - turns on context switch tracing.

Signed-off-by Steven Rostedt <[email protected]>

Cc Mathieu Desnoyers <[email protected]>

---
 lib/tracing/Kconfig              |   10 +
 lib/tracing/Makefile             |    1 
 lib/tracing/trace_sched_switch.c |  205 +++++++++++++++++++++++++++++++++++++++
 lib/tracing/tracer.c             |   31 +++++
 lib/tracing/tracer.h             |   18 +++
 5 files changed, 264 insertions(+), 1 deletion(-)

lib/tracing/Kconfig +10 -0

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -42,3 +42,13 @@ config CRITICAL_IRQSOFF_TIMING
 
 	      echo 0 > /debug/tracing/preempt_max_latency
 
+config CONTEXT_SWITCH_TRACER
+	bool "Trace process context switches"
+	depends on DEBUG_KERNEL
+	default n
+	select TRACING
+	select MARKERS
+	help
+	  This tracer hooks into the context switch and records
+	  all switching of tasks.
+

lib/tracing/Makefile +1 -0

Index: linux-compile.git/lib/tracing/Makefile
===================================================================
@@ -1,6 +1,7 @@
 obj-$(CONFIG_MCOUNT) += libmcount.o
 
 obj-$(CONFIG_TRACING) += tracer.o
+obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace_function.o
 obj-$(CONFIG_CRITICAL_IRQSOFF_TIMING) += trace_irqsoff.o

lib/tracing/trace_sched_switch.c new file +205 -0

205 lines not on this page · raw message

lib/tracing/tracer.c +31 -0

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================
@@ -31,6 +31,7 @@ enum trace_type
 	__TRACE_FIRST_TYPE = 0,
 
 	TRACE_FN,
+	TRACE_CTX,
 
 	__TRACE_LAST_TYPE
 };

@@ -107,6 +108,24 @@ notrace void tracing_function_trace(stru
 	entry->fn.parent_ip = parent_ip;
 }
 
+notrace void tracing_sched_switch_trace(struct tracing_trace *tr,
+					struct tracing_trace_cpu *data,
+					struct task_struct *prev,
+					int next_pid,
+					unsigned long flags)
+{
+	struct tracing_entry *entry;
+
+	entry = tracing_get_trace_entry(tr, data);
+	tracing_generic_entry_update(entry, flags);
+	entry->type		= TRACE_CTX;
+	entry->ctx.prev_pid	= prev->pid;
+	entry->ctx.prev_prio	= prev->prio;
+	entry->ctx.prev_state	= prev->state;
+	entry->ctx.next_pid	= next_pid;
+	/* would like to save next prio, but we cant :-( */
+}
+
 #ifdef CONFIG_DEBUG_FS
 enum trace_iterator {
 	TRACE_ITER_SYM_ONLY	= 1,

@@ -414,6 +433,8 @@ lat_print_timestamp(struct seq_file *m, 
 		seq_puts(m, " : ");
 }
 
+static const char state_to_char[] = "RSDTtZX";
+
 static void notrace
 print_lat_fmt(struct seq_file *m, struct tracing_iterator *iter,
 	      unsigned int trace_idx, int cpu)

@@ -424,6 +445,7 @@ print_lat_fmt(struct seq_file *m, struct
 	unsigned long rel_usecs;
 	int sym_only = !!(trace_flags & TRACE_ITER_SYM_ONLY);
 	int verbose = !!(trace_flags & TRACE_ITER_VERBOSE);
+	int S;
 
 	if (!next_entry)
 		next_entry = entry;

@@ -450,6 +472,15 @@ print_lat_fmt(struct seq_file *m, struct
 		seq_print_ip_sym(m, entry->fn.parent_ip, sym_only);
 		seq_puts(m, ")\n");
 		break;
+	case TRACE_CTX:
+		S = entry->ctx.prev_state < sizeof(state_to_char) ?
+			state_to_char[entry->ctx.prev_state] : 'X';
+		seq_printf(m, " %d:%d:%c --> %d\n",
+			   entry->ctx.prev_pid,
+			   entry->ctx.prev_prio,
+			   S,
+			   entry->ctx.next_pid);
+		break;
 	}
 }

lib/tracing/tracer.h +17 -1

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -10,6 +10,14 @@ struct tracing_function {
 	unsigned long parent_ip;
 };
 
+struct tracing_sched_switch {
+	unsigned int prev_pid;
+	unsigned char prev_prio;
+	unsigned char prev_state;
+	unsigned int next_pid;
+	unsigned char next_prio;
+};
+
 struct tracing_entry {
 	char type;
 	char cpu;  /* who will want to trace more than 256 CPUS? */

@@ -18,7 +26,10 @@ struct tracing_entry {
 	int pid;
 	cycle_t t;
 	char comm[TASK_COMM_LEN];
-	struct tracing_function fn;
+	union {
+		struct tracing_function fn;
+		struct tracing_sched_switch ctx;
+	};
 };
 
 struct tracing_trace_cpu {

@@ -76,6 +87,11 @@ void tracing_function_trace(struct traci
 			    unsigned long ip,
 			    unsigned long parent_ip,
 			    unsigned long flags);
+void tracing_sched_switch_trace(struct tracing_trace *tr,
+				struct tracing_trace_cpu *data,
+				struct task_struct *prev,
+				int next_pid,
+				unsigned long flags);
 
 extern struct file_operations tracing_fops;
 extern struct file_operations tracing_lt_fops;

Steven Rostedt

author reviewer

2008-01-15 20:56 UTC #30 lore raw

RFC PATCH v3 28/30 Generic command line storage

In reply to Steven Rostedt

Saving the comm of tasks for each trace is very expensive. This patch includes in the context switch hook, a way to store the last 100 command lines of tasks. This table is examined when a trace is to be printed.

Note: The comm may be destroyed if other traces are performed. Later (TBD) patches may simply store this information in the trace itself.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/Kconfig              |    1 
 lib/tracing/trace_function.c     |    2 
 lib/tracing/trace_irqsoff.c      |    3 +
 lib/tracing/trace_sched_switch.c |   14 +++-
 lib/tracing/tracer.c             |  112 +++++++++++++++++++++++++++++++++++++--
 lib/tracing/tracer.h             |    5 +
 6 files changed, 128 insertions(+), 9 deletions(-)

lib/tracing/Kconfig +1 -0

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -19,6 +19,7 @@ config FUNCTION_TRACER
 	default n
 	select MCOUNT
 	select TRACING
+	select CONTEXT_SWITCH_TRACER
 	help
 	  Use profiler instrumentation, adding -pg to CFLAGS. This will
 	  insert a call to an architecture specific __mcount routine,

lib/tracing/trace_function.c +2 -0

Index: linux-compile.git/lib/tracing/trace_function.c
===================================================================
@@ -70,9 +70,11 @@ static void function_trace_ctrl_update(s
 	if (tr->ctrl ^ val) {
 		if (val) {
 			trace_enabled = 1;
+			atomic_inc(&trace_record_cmdline);
 			register_mcount_function(&trace_ops);
 		} else {
 			trace_enabled = 0;
+			atomic_dec(&trace_record_cmdline);
 			unregister_mcount_function(&trace_ops);
 		}
 		tr->ctrl = val;

lib/tracing/trace_irqsoff.c +3 -0

Index: linux-compile.git/lib/tracing/trace_irqsoff.c
===================================================================
@@ -93,6 +93,9 @@ static void update_max_tr(struct tracing
 	save->policy = current->policy;
 	save->rt_priority = current->rt_priority;
 
+	/* record this tasks comm */
+	tracing_record_cmdline(current);
+
 	/* from memcpy above: save->trace = data->trace */
 	data->trace = max_buffer;
 	max_buffer = save->trace;

lib/tracing/trace_sched_switch.c +11 -3

Index: linux-compile.git/lib/tracing/trace_sched_switch.c
===================================================================
@@ -31,7 +31,7 @@ static notrace void sched_switch_callbac
 	va_list ap;
 	int cpu;
 
-	if (!trace_enabled)
+	if (!atomic_read(&trace_record_cmdline))
 		return;
 
 	va_start(ap, format);

@@ -49,6 +49,11 @@ static notrace void sched_switch_callbac
 	/* Ignore prev_state, since we get that from prev itself */
 	va_end(ap);
 
+	tracing_record_cmdline(prev);
+
+	if (!trace_enabled)
+		return;
+
 	raw_local_irq_save(flags);
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];

@@ -82,10 +87,13 @@ static void sched_switch_trace_ctrl_upda
 		sched_switch_reset(tr);
 
 	if (tr->ctrl ^ val) {
-		if (val)
+		if (val) {
+			atomic_inc(&trace_record_cmdline);
 			trace_enabled = 1;
-		else
+		} else {
+			atomic_dec(&trace_record_cmdline);
 			trace_enabled = 0;
+		}
 		tr->ctrl = val;
 	}
 }

lib/tracing/tracer.c +107 -5

168 lines not on this page · raw message

lib/tracing/tracer.h +4 -1

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -25,7 +25,6 @@ struct tracing_entry {
 	char preempt_count; /* assumes PREEMPT_MASK is 8 bits or less */
 	int pid;
 	cycle_t t;
-	char comm[TASK_COMM_LEN];
 	union {
 		struct tracing_function fn;
 		struct tracing_sched_switch ctx;

@@ -92,11 +91,15 @@ void tracing_sched_switch_trace(struct t
 				struct task_struct *prev,
 				int next_pid,
 				unsigned long flags);
+void tracing_record_cmdline(struct task_struct *tsk);
 
 extern struct file_operations tracing_fops;
 extern struct file_operations tracing_lt_fops;
 extern struct file_operations tracing_ctrl_fops;
 
+extern atomic_t trace_record_cmdline;
+extern atomic_t trace_record_cmdline_disabled;
+
 static inline notrace cycle_t now(void)
 {
 	return get_monotonic_cycles();

Mathieu Desnoyers

2008-01-15 21:30 UTC #31 lore raw

Re: [RFC PATCH 28/30 v3] Generic command line storage

Quoted 5 lines · * Steven Rostedt ([email protected]) wrote:

Saving the comm of tasks for each trace is very expensive. This patch includes in the context switch hook, a way to store the last 100 command lines of tasks. This table is examined when a trace is to be printed.

Instead of saving the comm at context switch, could we save them when a process exits ? (actually, if we want to do this right, we would also have to trace the changes of "comm" upon exec(), so we deal correctly with processes that do multiple execs).

This way, we would not duplicate the comm of processes still active in the system, and would trigger on a much lower event-rate trace point.

And the idea would be to save only the comm of processes present in your summary (top X processes); why do you save the last 100 ? (if the system has more than 100 active tasks, this code will always be executed, which seems bad)

Also, if we do as I propose, we would have to keep process exit/exec tracing active until the summary is cleared, even though the rest of "high event rate" tracing could be stopped sooner.

Mathieu

Quoted 308 lines

Note: The comm may be destroyed if other traces are performed. Later (TBD) patches may simply store this information in the trace itself.

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/Kconfig              |    1 
 lib/tracing/trace_function.c     |    2 
 lib/tracing/trace_irqsoff.c      |    3 +
 lib/tracing/trace_sched_switch.c |   14 +++-
 lib/tracing/tracer.c             |  112 +++++++++++++++++++++++++++++++++++++--
 lib/tracing/tracer.h             |    5 +
 6 files changed, 128 insertions(+), 9 deletions(-)

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================

lib/tracing/Kconfig +1 -0

@@ -19,6 +19,7 @@ config FUNCTION_TRACER
 	default n
 	select MCOUNT
 	select TRACING
+	select CONTEXT_SWITCH_TRACER
 	help
 	  Use profiler instrumentation, adding -pg to CFLAGS. This will
 	  insert a call to an architecture specific __mcount routine,

Index: linux-compile.git/lib/tracing/trace_function.c
===================================================================

lib/tracing/trace_function.c +2 -0

@@ -70,9 +70,11 @@ static void function_trace_ctrl_update(s
 	if (tr->ctrl ^ val) {
 		if (val) {
 			trace_enabled = 1;
+			atomic_inc(&trace_record_cmdline);
 			register_mcount_function(&trace_ops);
 		} else {
 			trace_enabled = 0;
+			atomic_dec(&trace_record_cmdline);
 			unregister_mcount_function(&trace_ops);
 		}
 		tr->ctrl = val;

Index: linux-compile.git/lib/tracing/trace_irqsoff.c
===================================================================

lib/tracing/trace_irqsoff.c +3 -0

@@ -93,6 +93,9 @@ static void update_max_tr(struct tracing
 	save->policy = current->policy;
 	save->rt_priority = current->rt_priority;
 
+	/* record this tasks comm */
+	tracing_record_cmdline(current);
+
 	/* from memcpy above: save->trace = data->trace */
 	data->trace = max_buffer;
 	max_buffer = save->trace;

Index: linux-compile.git/lib/tracing/trace_sched_switch.c
===================================================================

lib/tracing/trace_sched_switch.c +11 -3

@@ -31,7 +31,7 @@ static notrace void sched_switch_callbac
 	va_list ap;
 	int cpu;
 
-	if (!trace_enabled)
+	if (!atomic_read(&trace_record_cmdline))
 		return;
 
 	va_start(ap, format);

@@ -49,6 +49,11 @@ static notrace void sched_switch_callbac
 	/* Ignore prev_state, since we get that from prev itself */
 	va_end(ap);
 
+	tracing_record_cmdline(prev);
+
+	if (!trace_enabled)
+		return;
+
 	raw_local_irq_save(flags);
 	cpu = raw_smp_processor_id();
 	data = tr->data[cpu];

@@ -82,10 +87,13 @@ static void sched_switch_trace_ctrl_upda
 		sched_switch_reset(tr);
 
 	if (tr->ctrl ^ val) {
-		if (val)
+		if (val) {
+			atomic_inc(&trace_record_cmdline);
 			trace_enabled = 1;
-		else
+		} else {
+			atomic_dec(&trace_record_cmdline);
 			trace_enabled = 0;
+		}
 		tr->ctrl = val;
 	}
 }

Index: linux-compile.git/lib/tracing/tracer.c
===================================================================

lib/tracing/tracer.c +107 -5

@@ -49,6 +49,88 @@ void notrace tracing_reset(struct tracin
 	atomic_set(&data->underrun, 0);
 }
 
+#define SAVED_CMDLINES 128
+static unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1];
+static unsigned map_cmdline_to_pid[SAVED_CMDLINES];
+static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN];
+static int cmdline_idx;
+static DEFINE_SPINLOCK(trace_cmdline_lock);
+atomic_t trace_record_cmdline;
+atomic_t trace_record_cmdline_disabled;
+
+static void trace_init_cmdlines(void)
+{
+	memset(&map_pid_to_cmdline, -1, sizeof(map_pid_to_cmdline));
+	memset(&map_cmdline_to_pid, -1, sizeof(map_cmdline_to_pid));
+	cmdline_idx = 0;
+}
+
+notrace void trace_stop_cmdline_recording(void);
+
+static void notrace trace_save_cmdline(struct task_struct *tsk)
+{
+	unsigned map;
+	unsigned idx;
+
+	if (!tsk->pid || unlikely(tsk->pid > PID_MAX_DEFAULT))
+		return;
+
+	/*
+	 * It's not the end of the world if we don't get
+	 * the lock, but we also don't want to spin
+	 * nor do we want to disable interrupts,
+	 * so if we miss here, then better luck next time.
+	 */
+	if (!spin_trylock(&trace_cmdline_lock))
+		return;
+
+	idx = map_pid_to_cmdline[tsk->pid];
+	if (idx >= SAVED_CMDLINES) {
+		idx = (cmdline_idx + 1) % SAVED_CMDLINES;
+
+		map = map_cmdline_to_pid[idx];
+		if (map <= PID_MAX_DEFAULT)
+			map_pid_to_cmdline[map] = (unsigned)-1;
+
+		map_pid_to_cmdline[tsk->pid] = idx;
+
+		cmdline_idx = idx;
+	}
+
+	memcpy(&saved_cmdlines[idx], tsk->comm, TASK_COMM_LEN);
+
+	spin_unlock(&trace_cmdline_lock);
+}
+
+static notrace char *trace_find_cmdline(int pid)
+{
+	char *cmdline = "<...>";
+	unsigned map;
+
+	if (!pid)
+		return "<idle>";
+
+	if (pid > PID_MAX_DEFAULT)
+		goto out;
+
+	map = map_pid_to_cmdline[pid];
+	if (map >= SAVED_CMDLINES)
+		goto out;
+
+	cmdline = saved_cmdlines[map];
+
+ out:
+	return cmdline;
+}
+
+void tracing_record_cmdline(struct task_struct *tsk)
+{
+	if (atomic_read(&trace_record_cmdline_disabled))
+		return;
+
+	trace_save_cmdline(tsk);
+}
+
 static inline notrace struct tracing_entry *
 tracing_get_trace_entry(struct tracing_trace *tr,
 			struct tracing_trace_cpu *data)

@@ -90,7 +172,6 @@ tracing_generic_entry_update(struct trac
 		((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
 		((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
 		(need_resched() ? TRACE_FLAG_NEED_RESCHED : 0);
-	memcpy(entry->comm, tsk->comm, TASK_COMM_LEN);
 }
 
 notrace void tracing_function_trace(struct tracing_trace *tr,

@@ -242,6 +323,8 @@ static void *s_start(struct seq_file *m,
 	loff_t l = 0;
 	int i;
 
+	atomic_inc(&trace_record_cmdline_disabled);
+
 	/* let the tracer grab locks here if needed */
 	if (iter->tr->start)
 		iter->tr->start(iter);

@@ -269,6 +352,8 @@ static void s_stop(struct seq_file *m, v
 {
 	struct tracing_iterator *iter = m->private;
 
+	atomic_dec(&trace_record_cmdline_disabled);
+
 	/* let the tracer release locks here if needed */
 	if (iter->tr->stop)
 		iter->tr->stop(iter);

@@ -390,8 +475,11 @@ static void notrace
 lat_print_generic(struct seq_file *m, struct tracing_entry *entry, int cpu)
 {
 	int hardirq, softirq;
+	char *comm;
 
-	seq_printf(m, "%8.8s-%-5d ", entry->comm, entry->pid);
+	comm = trace_find_cmdline(entry->pid);
+
+	seq_printf(m, "%8.8s-%-5d ", comm, entry->pid);
 	seq_printf(m, "%d", cpu);
 	seq_printf(m, "%c%c",
 		   (entry->flags & TRACE_FLAG_IRQS_OFF) ? 'd' : '.',

@@ -453,9 +541,12 @@ print_lat_fmt(struct seq_file *m, struct
 	abs_usecs = cycles_to_usecs(entry->t - iter->tr->time_start);
 
 	if (verbose) {
+		char *comm;
+
+		comm = trace_find_cmdline(entry->pid);
 		seq_printf(m, "%16s %5d %d %d %08x %08x [%08lx]"
 			   " %ld.%03ldms (+%ld.%03ldms): ",
-			   entry->comm,
+			   comm,
 			   entry->pid, cpu, entry->flags,
 			   entry->preempt_count, trace_idx,
 			   cycles_to_usecs(entry->t),

@@ -491,6 +582,9 @@ static void notrace print_trace_fmt(stru
 	unsigned long secs;
 	int sym_only = !!(trace_flags & TRACE_ITER_SYM_ONLY);
 	unsigned long long t;
+	char *comm;
+
+	comm = trace_find_cmdline(iter->ent->pid);
 
 	t = cycles_to_usecs(iter->ent->t);
 	usec_rem = do_div(t, 1000000ULL);

@@ -498,7 +592,7 @@ static void notrace print_trace_fmt(stru
 
 	seq_printf(m, "[%5lu.%06lu] ", secs, usec_rem);
 	seq_printf(m, "CPU %d: ", iter->cpu);
-	seq_printf(m, "%s:%d ", iter->ent->comm,
+	seq_printf(m, "%s:%d ", comm,
 		   iter->ent->pid);
 	switch (iter->ent->type) {
 	case TRACE_FN:

@@ -812,6 +906,14 @@ static __init int trace_init_debugfs(voi
 	return 0;
 }
 
-device_initcall(trace_init_debugfs);
+static __init int trace_init(void)
+{
+	trace_init_cmdlines();
+
+	return trace_init_debugfs();
+
+}
+
+device_initcall(trace_init);
 
 #endif /* CONFIG_DEBUG_FS */

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================

lib/tracing/tracer.h +4 -1

@@ -25,7 +25,6 @@ struct tracing_entry {
 	char preempt_count; /* assumes PREEMPT_MASK is 8 bits or less */
 	int pid;
 	cycle_t t;
-	char comm[TASK_COMM_LEN];
 	union {
 		struct tracing_function fn;
 		struct tracing_sched_switch ctx;

@@ -92,11 +91,15 @@ void tracing_sched_switch_trace(struct t
 				struct task_struct *prev,
 				int next_pid,
 				unsigned long flags);
+void tracing_record_cmdline(struct task_struct *tsk);
 
 extern struct file_operations tracing_fops;
 extern struct file_operations tracing_lt_fops;
 extern struct file_operations tracing_ctrl_fops;
 
+extern atomic_t trace_record_cmdline;
+extern atomic_t trace_record_cmdline_disabled;
+
 static inline notrace cycle_t now(void)
 {
 	return get_monotonic_cycles();

Signature · 1 line

--

Signature · 3 lines

Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

Steven Rostedt

author reviewer

2008-01-15 22:15 UTC #32 lore raw

Re: [RFC PATCH 28/30 v3] Generic command line storage

Quoted 12 lines · Mathieu Desnoyers

Quoted 5 lines · * Steven Rostedt ([email protected]) wrote:
Saving the comm of tasks for each trace is very expensive. This patch includes in the context switch hook, a way to store the last 100 command lines of tasks. This table is examined when a trace is to be printed.
Instead of saving the comm at context switch, could we save them when a process exits ? (actually, if we want to do this right, we would also have to trace the changes of "comm" upon exec(), so we deal correctly with processes that do multiple execs).

Quoted 7 lines

This way, we would not duplicate the comm of processes still active in the system, and would trigger on a much lower event-rate trace point.
And the idea would be to save only the comm of processes present in your summary (top X processes); why do you save the last 100 ? (if the system has more than 100 active tasks, this code will always be executed, which seems bad)

100 was just an arbitrary number that seems to be fine in my current tests. But yes, a function trace over a thousand tasks will clean them out. But the recording is only done when a trace is active.

Quoted 4 lines

Also, if we do as I propose, we would have to keep process exit/exec tracing active until the summary is cleared, even though the rest of "high event rate" tracing could be stopped sooner.

One thing I planned on doing was at a low path point (hit max latency to record, etc) we could copy the array into the trace buffer, so the info stays with the trace.

Going with your approach (marker at exit), I could flush the traces that were saved, as well as the comms of the current tasks that are still running. The thing is, this all starts to get a bit intrusive into the rest of the kernel. What is suppose to be a compact latency tracer, starts to become larger, problematic, entity. Now changes in exit could require updates to the tracer. Which is something I want to avoid.

-- Steve

Steven Rostedt

author reviewer

2008-01-15 20:53 UTC #33 lore raw

RFC PATCH v3 29/30 make varaible size buffers for traces

In reply to Steven Rostedt

Each trace now can have the size of its trace entries modified through command line arguments.

trace_fn_entries - function trace entries (default 65536) trace_irq_entries - irq off trace entries (default 512) trace_ctx_entries - schedule switch entries (default 16384)

Signed-off-by Steven Rostedt <[email protected]>

---
 lib/tracing/trace_function.c     |   14 ++++++++++++--
 lib/tracing/trace_irqsoff.c      |   14 ++++++++++++--
 lib/tracing/trace_sched_switch.c |   14 ++++++++++++--
 lib/tracing/tracer.h             |    1 -
 4 files changed, 36 insertions(+), 7 deletions(-)

lib/tracing/trace_function.c +12 -2

Index: linux-compile.git/lib/tracing/trace_function.c
===================================================================
@@ -18,6 +18,16 @@
 static struct tracing_trace function_trace __read_mostly;
 static DEFINE_PER_CPU(struct tracing_trace_cpu, function_trace_cpu);
 static int trace_enabled __read_mostly;
+static unsigned long trace_nr_entries = (65536UL);
+
+static int __init set_nr_entries(char *str)
+{
+	if (!str)
+		return 0;
+	trace_nr_entries = simple_strtoul(str, &str, 0);
+	return 1;
+}
+__setup("trace_fn_entries=", set_nr_entries);
 
 static notrace void function_trace_reset(struct tracing_trace *tr)
 {

@@ -132,7 +142,7 @@ static void function_trace_close(struct 
 
 __init static int function_trace_alloc_buffers(void)
 {
-	const int order = page_order(TRACING_NR_ENTRIES * TRACING_ENTRY_SIZE);
+	const int order = page_order(trace_nr_entries * TRACING_ENTRY_SIZE);
 	const unsigned long size = (1UL << order) << PAGE_SHIFT;
 	struct tracing_entry *array;
 	int i;

@@ -156,7 +166,7 @@ __init static int function_trace_alloc_b
 	function_trace.entries = size / TRACING_ENTRY_SIZE;
 
 	pr_info("function tracer: %ld bytes allocated for %ld",
-		size, TRACING_NR_ENTRIES);
+		size, trace_nr_entries);
 	pr_info(" entries of %ld bytes\n", (long)TRACING_ENTRY_SIZE);
 	pr_info("   actual entries %ld\n", function_trace.entries);

lib/tracing/trace_irqsoff.c +12 -2

Index: linux-compile.git/lib/tracing/trace_irqsoff.c
===================================================================
@@ -25,6 +25,16 @@ static unsigned long preempt_max_latency
 static unsigned long preempt_thresh;
 static __cacheline_aligned_in_smp DEFINE_MUTEX(max_mutex);
 static int trace_enabled __read_mostly;
+static unsigned long trace_nr_entries = (512UL);
+
+static int __init set_nr_entries(char *str)
+{
+	if (!str)
+		return 0;
+	trace_nr_entries = simple_strtoul(str, &str, 0);
+	return 1;
+}
+__setup("trace_irq_entries=", set_nr_entries);
 
 /*
  * max trace is switched with this buffer.

@@ -497,7 +507,7 @@ static void notrace irqsoff_trace_close(
 
 __init static int trace_irqsoff_alloc_buffers(void)
 {
-	const int order = page_order(TRACING_NR_ENTRIES * TRACING_ENTRY_SIZE);
+	const int order = page_order(trace_nr_entries * TRACING_ENTRY_SIZE);
 	const unsigned long size = (1UL << order) << PAGE_SHIFT;
 	struct tracing_entry *array;
 	int i;

@@ -535,7 +545,7 @@ __init static int trace_irqsoff_alloc_bu
 	max_tr.stop = irqsoff_stop;
 
 	pr_info("irqs off tracer: %ld bytes allocated for %ld",
-		size, TRACING_NR_ENTRIES);
+		size, trace_nr_entries);
 	pr_info(" entries of %d bytes\n", (int)TRACING_ENTRY_SIZE);
 	pr_info("   actual entries %ld\n", irqsoff_trace.entries);

lib/tracing/tracer.h +0 -1

Index: linux-compile.git/lib/tracing/tracer.h
===================================================================
@@ -76,7 +76,6 @@ struct tracing_iterator {
 };
 
 #define TRACING_ENTRY_SIZE sizeof(struct tracing_entry)
-#define TRACING_NR_ENTRIES (65536UL)
 
 void notrace tracing_reset(struct tracing_trace_cpu *data);
 int tracing_open_generic(struct inode *inode, struct file *filp);

lib/tracing/trace_sched_switch.c +12 -2

Index: linux-compile.git/lib/tracing/trace_sched_switch.c
===================================================================
@@ -18,6 +18,16 @@ static struct tracing_trace sched_switch
 static DEFINE_PER_CPU(struct tracing_trace_cpu, sched_switch_trace_cpu);
 
 static int trace_enabled __read_mostly;
+static unsigned long trace_nr_entries = (16384UL);
+
+static int __init set_nr_entries(char *str)
+{
+	if (!str)
+		return 0;
+	trace_nr_entries = simple_strtoul(str, &str, 0);
+	return 1;
+}
+__setup("trace_ctx_entries=", set_nr_entries);
 
 static notrace void sched_switch_callback(const struct marker *mdata,
 					  void *private_data,

@@ -144,7 +154,7 @@ static void sched_switch_trace_close(str
 
 __init static int sched_switch_trace_alloc_buffers(void)
 {
-	const int order = page_order(TRACING_NR_ENTRIES * TRACING_ENTRY_SIZE);
+	const int order = page_order(trace_nr_entries * TRACING_ENTRY_SIZE);
 	const unsigned long size = (1UL << order) << PAGE_SHIFT;
 	struct tracing_entry *array;
 	int ret;

@@ -170,7 +180,7 @@ __init static int sched_switch_trace_all
 	sched_switch_trace.entries = size / TRACING_ENTRY_SIZE;
 
 	pr_info("sched_switch tracer: %ld bytes allocated for %ld",
-		size, TRACING_NR_ENTRIES);
+		size, trace_nr_entries);
 	pr_info(" entries of %ld bytes\n", (long)TRACING_ENTRY_SIZE);
 	pr_info("   actual entries %ld\n", sched_switch_trace.entries);

Steven Rostedt

author reviewer

2008-01-15 20:59 UTC #34 lore raw

RFC PATCH v3 30/30 trace preempt off critical timings

In reply to Steven Rostedt

Add preempt off timings. A lot of this code is taken from the RT patch latency trace that was written by Ingo Molnar.

Now instead of just tracing irqs off, preemption off can be selected to be recorded.

When this is selected, it shares the same files as irqs off timings. One can either trace preemption off, irqs off, or one or the other off. But once it is configured in the kernel for what to trace, it can't be changed.

Signed-off-by Steven Rostedt <[email protected]>

---
 arch/x86/kernel/process_32.c |    3 +++
 include/linux/irqflags.h     |    3 ++-
 include/linux/mcount.h       |    8 ++++++++
 include/linux/preempt.h      |    2 +-
 kernel/sched.c               |   24 +++++++++++++++++++++++-
 lib/tracing/Kconfig          |   24 ++++++++++++++++++++++++
 lib/tracing/Makefile         |    1 +
 lib/tracing/trace_irqsoff.c  |   40 ++++++++++++++++++++++++++++++----------
 8 files changed, 92 insertions(+), 13 deletions(-)

lib/tracing/Kconfig +24 -0

Index: linux-compile.git/lib/tracing/Kconfig
===================================================================
@@ -43,6 +43,30 @@ config CRITICAL_IRQSOFF_TIMING
 
 	      echo 0 > /debug/tracing/preempt_max_latency
 
+	  (Note that kernel size and overhead increases with this option
+	  enabled. This option and the preempt-off timing option can be
+	  used together or separately.)
+
+config CRITICAL_PREEMPT_TIMING
+	bool "Preemption-off critical section latency timing"
+	default n
+	depends on GENERIC_TIME
+	depends on PREEMPT
+	select TRACING
+	help
+	  This option measures the time spent in preemption off critical
+	  sections, with microsecond accuracy.
+
+	  The default measurement method is a maximum search, which is
+	  disabled by default and can be runtime (re-)started
+	  via:
+
+	      echo 0 > /debug/tracing/preempt_max_latency
+
+	  (Note that kernel size and overhead increases with this option
+	  enabled. This option and the irqs-off timing option can be
+	  used together or separately.)
+
 config CONTEXT_SWITCH_TRACER
 	bool "Trace process context switches"
 	depends on DEBUG_KERNEL

lib/tracing/Makefile +1 -0

Index: linux-compile.git/lib/tracing/Makefile
===================================================================
@@ -4,5 +4,6 @@ obj-$(CONFIG_TRACING) += tracer.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace_function.o
 obj-$(CONFIG_CRITICAL_IRQSOFF_TIMING) += trace_irqsoff.o
+obj-$(CONFIG_CRITICAL_PREEMPT_TIMING) += trace_irqsoff.o
 
 libmcount-y := mcount.o

lib/tracing/trace_irqsoff.c +30 ~~-10~~

Index: linux-compile.git/lib/tracing/trace_irqsoff.c
===================================================================
@@ -36,6 +36,12 @@ static int __init set_nr_entries(char *s
 }
 __setup("trace_irq_entries=", set_nr_entries);
 
+#ifdef CONFIG_CRITICAL_PREEMPT_TIMING
+# define preempt_trace() (preempt_count())
+#else
+# define preempt_trace() (0)
+#endif
+
 /*
  * max trace is switched with this buffer.
  */

@@ -208,7 +214,7 @@ start_critical_timing(unsigned long ip, 
 
 	data->critical_sequence = max_sequence;
 	data->preempt_timestamp = now();
-	data->critical_start = parent_ip;
+	data->critical_start = parent_ip ? : ip;
 	tracing_reset(data);
 
 	local_save_flags(flags);

@@ -232,18 +238,19 @@ stop_critical_timing(unsigned long ip, u
 	atomic_inc(&data->disabled);
 	local_save_flags(flags);
 	tracing_function_trace(tr, data, ip, parent_ip, flags);
-	check_critical_timing(tr, data, parent_ip, cpu);
+	check_critical_timing(tr, data, parent_ip ? : ip, cpu);
 	data->critical_start = 0;
 	atomic_dec(&data->disabled);
 }
 
+/* start and stop critical timings used to for stoppage (in idle) */
 void notrace start_critical_timings(void)
 {
 	unsigned long flags;
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (preempt_trace() || irqs_disabled_flags(flags))
 		start_critical_timing(CALLER_ADDR0, 0);
 }

@@ -253,10 +260,11 @@ void notrace stop_critical_timings(void)
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (preempt_trace() || irqs_disabled_flags(flags))
 		stop_critical_timing(CALLER_ADDR0, 0);
 }
 
+#ifdef CONFIG_CRITICAL_IRQSOFF_TIMING
 #ifdef CONFIG_LOCKDEP
 void notrace time_hardirqs_on(unsigned long a0, unsigned long a1)
 {

@@ -264,7 +272,7 @@ void notrace time_hardirqs_on(unsigned l
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irqs_disabled_flags(flags))
 		stop_critical_timing(a0, a1);
 }

@@ -274,7 +282,7 @@ void notrace time_hardirqs_off(unsigned 
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irqs_disabled_flags(flags))
 		start_critical_timing(a0, a1);
 }

@@ -313,7 +321,7 @@ void notrace trace_hardirqs_on(void)
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irqs_disabled_flags(flags))
 		stop_critical_timing(CALLER_ADDR0, 0);
 }
 EXPORT_SYMBOL(trace_hardirqs_on);

@@ -324,7 +332,7 @@ void notrace trace_hardirqs_off(void)
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irqs_disabled_flags(flags))
 		start_critical_timing(CALLER_ADDR0, 0);
 }
 EXPORT_SYMBOL(trace_hardirqs_off);

@@ -335,7 +343,7 @@ void notrace trace_hardirqs_on_caller(un
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irqs_disabled_flags(flags))
 		stop_critical_timing(CALLER_ADDR0, caller_addr);
 }
 EXPORT_SYMBOL(trace_hardirqs_on_caller);

@@ -346,13 +354,25 @@ void notrace trace_hardirqs_off_caller(u
 
 	local_save_flags(flags);
 
-	if (irqs_disabled_flags(flags))
+	if (!preempt_trace() && irqs_disabled_flags(flags))
 		start_critical_timing(CALLER_ADDR0, caller_addr);
 }
 EXPORT_SYMBOL(trace_hardirqs_off_caller);
 
 #endif /* CONFIG_LOCKDEP */
+#endif /*  CONFIG_CRITICAL_IRQSOFF_TIMING */
 
+#ifdef CONFIG_CRITICAL_PREEMPT_TIMING
+void notrace trace_preempt_on(unsigned long a0, unsigned long a1)
+{
+	stop_critical_timing(a0, a1);
+}
+
+void notrace trace_preempt_off(unsigned long a0, unsigned long a1)
+{
+	start_critical_timing(a0, a1);
+}
+#endif /* CONFIG_CRITICAL_PREEMPT_TIMING */
 
 #ifdef CONFIG_MCOUNT
 static void notrace irqsoff_trace_call(unsigned long ip,

include/linux/preempt.h +1 -1

Index: linux-compile.git/include/linux/preempt.h
===================================================================
@@ -10,7 +10,7 @@
 #include <linux/linkage.h>
 #include <linux/list.h>
 
-#ifdef CONFIG_DEBUG_PREEMPT
+#if defined(CONFIG_DEBUG_PREEMPT) || defined(CONFIG_CRITICAL_PREEMPT_TIMING)
   extern void fastcall add_preempt_count(int val);
   extern void fastcall sub_preempt_count(int val);
 #else

kernel/sched.c +23 -1

Index: linux-compile.git/kernel/sched.c
===================================================================
@@ -63,6 +63,7 @@
 #include <linux/reciprocal_div.h>
 #include <linux/unistd.h>
 #include <linux/pagemap.h>
+#include <linux/mcount.h>
 
 #include <asm/tlb.h>
 #include <asm/irq_regs.h>

@@ -3502,26 +3503,44 @@ void scheduler_tick(void)
 #endif
 }
 
-#if defined(CONFIG_PREEMPT) && defined(CONFIG_DEBUG_PREEMPT)
+#if defined(CONFIG_PREEMPT) && (defined(CONFIG_DEBUG_PREEMPT) || \
+				defined(CONFIG_CRITICAL_PREEMPT_TIMING))
+
+static inline unsigned long get_parent_ip(unsigned long addr)
+{
+	if (in_lock_functions(addr)) {
+		addr = CALLER_ADDR2;
+		if (in_lock_functions(addr))
+			addr = CALLER_ADDR3;
+	}
+	return addr;
+}
 
 void fastcall add_preempt_count(int val)
 {
+#ifdef CONFIG_DEBUG_PREEMPT
 	/*
 	 * Underflow?
 	 */
 	if (DEBUG_LOCKS_WARN_ON((preempt_count() < 0)))
 		return;
+#endif
 	preempt_count() += val;
+#ifdef CONFIG_DEBUG_PREEMPT
 	/*
 	 * Spinlock count overflowing soon?
 	 */
 	DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >=
 				PREEMPT_MASK - 10);
+#endif
+	if (preempt_count() == val)
+		trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
 }
 EXPORT_SYMBOL(add_preempt_count);
 
 void fastcall sub_preempt_count(int val)
 {
+#ifdef CONFIG_DEBUG_PREEMPT
 	/*
 	 * Underflow?
 	 */

@@ -3533,7 +3552,10 @@ void fastcall sub_preempt_count(int val)
 	if (DEBUG_LOCKS_WARN_ON((val < PREEMPT_MASK) &&
 			!(preempt_count() & PREEMPT_MASK)))
 		return;
+#endif
 
+	if (preempt_count() == val)
+		trace_preempt_on(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1));
 	preempt_count() -= val;
 }
 EXPORT_SYMBOL(sub_preempt_count);

arch/x86/kernel/process_32.c +3 -0

Index: linux-compile.git/arch/x86/kernel/process_32.c
===================================================================
@@ -195,7 +195,10 @@ void cpu_idle(void)
 				play_dead();
 
 			__get_cpu_var(irq_stat).idle_timestamp = jiffies;
+			/* Don't trace irqs off for idle */
+			stop_critical_timings();
 			idle();
+			start_critical_timings();
 		}
 		tick_nohz_restart_sched_tick();
 		preempt_enable_no_resched();

include/linux/mcount.h +8 -0

Index: linux-compile.git/include/linux/mcount.h
===================================================================
@@ -58,4 +58,12 @@ extern void mcount(void);
 # define time_hardirqs_off(a0, a1)		do { } while (0)
 #endif
 
+#ifdef CONFIG_CRITICAL_PREEMPT_TIMING
+  extern void notrace trace_preempt_on(unsigned long a0, unsigned long a1);
+  extern void notrace trace_preempt_off(unsigned long a0, unsigned long a1);
+#else
+# define trace_preempt_on(a0, a1)		do { } while (0)
+# define trace_preempt_off(a0, a1)		do { } while (0)
+#endif
+
 #endif /* _LINUX_MCOUNT_H */

include/linux/irqflags.h +2 -1

Index: linux-compile.git/include/linux/irqflags.h
===================================================================
@@ -54,7 +54,8 @@
 # define INIT_TRACE_IRQFLAGS
 #endif
 
-#ifdef CONFIG_CRITICAL_IRQSOFF_TIMING
+#if defined(CONFIG_CRITICAL_IRQSOFF_TIMING) || \
+	defined(CONFIG_CRITICAL_PREEMPT_TIMING)
  extern void stop_critical_timings(void);
  extern void start_critical_timings(void);
 #else