linux/arch/x86/kernel/dumpstack_64.c
Frederic Weisbecker 0406ca6d8e perf_counter: Ignore the nmi call frames in the x86-64 backtraces
About every callchains recorded with perf record are filled up
including the internal perfcounter nmi frame:

 perf_callchain
 perf_counter_overflow
 intel_pmu_handle_irq
 perf_counter_nmi_handler
 notifier_call_chain
 atomic_notifier_call_chain
 notify_die
 do_nmi
 nmi

We want ignore this frame as it's not interesting for
instrumentation. To solve this, we simply ignore every frames
from nmi context.

New example of "perf report -s sym -c" after this patch:

9.59%  [k] search_by_key
             4.88%
                search_by_key
                reiserfs_read_locked_inode
                reiserfs_iget
                reiserfs_lookup
                do_lookup
                __link_path_walk
                path_walk
                do_path_lookup
                user_path_at
                vfs_fstatat
                vfs_lstat
                sys_newlstat
                system_call_fastpath
                __lxstat
                0x406fb1

             3.19%
                search_by_key
                search_by_entry_key
                reiserfs_find_entry
                reiserfs_lookup
                do_lookup
                __link_path_walk
                path_walk
                do_path_lookup
                user_path_at
                vfs_fstatat
                vfs_lstat
                sys_newlstat
                system_call_fastpath
                __lxstat
                0x406fb1
[...]

For now this patch only solves the problem in x86-64.

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Anton Blanchard <anton@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <1246474930-6088-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-01 22:37:23 +02:00

308 lines
7.2 KiB
C

/*
* Copyright (C) 1991, 1992 Linus Torvalds
* Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
*/
#include <linux/kallsyms.h>
#include <linux/kprobes.h>
#include <linux/uaccess.h>
#include <linux/utsname.h>
#include <linux/hardirq.h>
#include <linux/kdebug.h>
#include <linux/module.h>
#include <linux/ptrace.h>
#include <linux/kexec.h>
#include <linux/bug.h>
#include <linux/nmi.h>
#include <linux/sysfs.h>
#include <asm/stacktrace.h>
#include "dumpstack.h"
static char x86_stack_ids[][8] = {
[DEBUG_STACK - 1] = "#DB",
[NMI_STACK - 1] = "NMI",
[DOUBLEFAULT_STACK - 1] = "#DF",
[STACKFAULT_STACK - 1] = "#SS",
[MCE_STACK - 1] = "#MC",
#if DEBUG_STKSZ > EXCEPTION_STKSZ
[N_EXCEPTION_STACKS ...
N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
#endif
};
int x86_is_stack_id(int id, char *name)
{
return x86_stack_ids[id - 1] == name;
}
static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
unsigned *usedp, char **idp)
{
unsigned k;
/*
* Iterate over all exception stacks, and figure out whether
* 'stack' is in one of them:
*/
for (k = 0; k < N_EXCEPTION_STACKS; k++) {
unsigned long end = per_cpu(orig_ist, cpu).ist[k];
/*
* Is 'stack' above this exception frame's end?
* If yes then skip to the next frame.
*/
if (stack >= end)
continue;
/*
* Is 'stack' above this exception frame's start address?
* If yes then we found the right frame.
*/
if (stack >= end - EXCEPTION_STKSZ) {
/*
* Make sure we only iterate through an exception
* stack once. If it comes up for the second time
* then there's something wrong going on - just
* break out and return NULL:
*/
if (*usedp & (1U << k))
break;
*usedp |= 1U << k;
*idp = x86_stack_ids[k];
return (unsigned long *)end;
}
/*
* If this is a debug stack, and if it has a larger size than
* the usual exception stacks, then 'stack' might still
* be within the lower portion of the debug stack:
*/
#if DEBUG_STKSZ > EXCEPTION_STKSZ
if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) {
unsigned j = N_EXCEPTION_STACKS - 1;
/*
* Black magic. A large debug stack is composed of
* multiple exception stack entries, which we
* iterate through now. Dont look:
*/
do {
++j;
end -= EXCEPTION_STKSZ;
x86_stack_ids[j][4] = '1' +
(j - N_EXCEPTION_STACKS);
} while (stack < end - EXCEPTION_STKSZ);
if (*usedp & (1U << j))
break;
*usedp |= 1U << j;
*idp = x86_stack_ids[j];
return (unsigned long *)end;
}
#endif
}
return NULL;
}
/*
* x86-64 can have up to three kernel stacks:
* process stack
* interrupt stack
* severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
*/
void dump_trace(struct task_struct *task, struct pt_regs *regs,
unsigned long *stack, unsigned long bp,
const struct stacktrace_ops *ops, void *data)
{
const unsigned cpu = get_cpu();
unsigned long *irq_stack_end =
(unsigned long *)per_cpu(irq_stack_ptr, cpu);
unsigned used = 0;
struct thread_info *tinfo;
int graph = 0;
if (!task)
task = current;
if (!stack) {
unsigned long dummy;
stack = &dummy;
if (task && task != current)
stack = (unsigned long *)task->thread.sp;
}
#ifdef CONFIG_FRAME_POINTER
if (!bp) {
if (task == current) {
/* Grab bp right from our regs */
get_bp(bp);
} else {
/* bp is the last reg pushed by switch_to */
bp = *(unsigned long *) task->thread.sp;
}
}
#endif
/*
* Print function call entries in all stacks, starting at the
* current stack address. If the stacks consist of nested
* exceptions
*/
tinfo = task_thread_info(task);
for (;;) {
char *id;
unsigned long *estack_end;
estack_end = in_exception_stack(cpu, (unsigned long)stack,
&used, &id);
if (estack_end) {
if (ops->stack(data, id) < 0)
break;
bp = print_context_stack(tinfo, stack, bp, ops,
data, estack_end, &graph);
ops->stack(data, "<EOE>");
/*
* We link to the next stack via the
* second-to-last pointer (index -2 to end) in the
* exception stack:
*/
stack = (unsigned long *) estack_end[-2];
continue;
}
if (irq_stack_end) {
unsigned long *irq_stack;
irq_stack = irq_stack_end -
(IRQ_STACK_SIZE - 64) / sizeof(*irq_stack);
if (stack >= irq_stack && stack < irq_stack_end) {
if (ops->stack(data, "IRQ") < 0)
break;
bp = print_context_stack(tinfo, stack, bp,
ops, data, irq_stack_end, &graph);
/*
* We link to the next stack (which would be
* the process stack normally) the last
* pointer (index -1 to end) in the IRQ stack:
*/
stack = (unsigned long *) (irq_stack_end[-1]);
irq_stack_end = NULL;
ops->stack(data, "EOI");
continue;
}
}
break;
}
/*
* This handles the process stack:
*/
bp = print_context_stack(tinfo, stack, bp, ops, data, NULL, &graph);
put_cpu();
}
EXPORT_SYMBOL(dump_trace);
void
show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs,
unsigned long *sp, unsigned long bp, char *log_lvl)
{
unsigned long *stack;
int i;
const int cpu = smp_processor_id();
unsigned long *irq_stack_end =
(unsigned long *)(per_cpu(irq_stack_ptr, cpu));
unsigned long *irq_stack =
(unsigned long *)(per_cpu(irq_stack_ptr, cpu) - IRQ_STACK_SIZE);
/*
* debugging aid: "show_stack(NULL, NULL);" prints the
* back trace for this cpu.
*/
if (sp == NULL) {
if (task)
sp = (unsigned long *)task->thread.sp;
else
sp = (unsigned long *)&sp;
}
stack = sp;
for (i = 0; i < kstack_depth_to_print; i++) {
if (stack >= irq_stack && stack <= irq_stack_end) {
if (stack == irq_stack_end) {
stack = (unsigned long *) (irq_stack_end[-1]);
printk(" <EOI> ");
}
} else {
if (((long) stack & (THREAD_SIZE-1)) == 0)
break;
}
if (i && ((i % STACKSLOTS_PER_LINE) == 0))
printk("\n%s", log_lvl);
printk(" %016lx", *stack++);
touch_nmi_watchdog();
}
printk("\n");
show_trace_log_lvl(task, regs, sp, bp, log_lvl);
}
void show_registers(struct pt_regs *regs)
{
int i;
unsigned long sp;
const int cpu = smp_processor_id();
struct task_struct *cur = current;
sp = regs->sp;
printk("CPU %d ", cpu);
__show_regs(regs, 1);
printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
cur->comm, cur->pid, task_thread_info(cur), cur);
/*
* When in-kernel, we also print out the stack and code at the
* time of the fault..
*/
if (!user_mode(regs)) {
unsigned int code_prologue = code_bytes * 43 / 64;
unsigned int code_len = code_bytes;
unsigned char c;
u8 *ip;
printk(KERN_EMERG "Stack:\n");
show_stack_log_lvl(NULL, regs, (unsigned long *)sp,
regs->bp, KERN_EMERG);
printk(KERN_EMERG "Code: ");
ip = (u8 *)regs->ip - code_prologue;
if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) {
/* try starting at IP */
ip = (u8 *)regs->ip;
code_len = code_len - code_prologue + 1;
}
for (i = 0; i < code_len; i++, ip++) {
if (ip < (u8 *)PAGE_OFFSET ||
probe_kernel_address(ip, c)) {
printk(" Bad RIP value.");
break;
}
if (ip == (u8 *)regs->ip)
printk("<%02x> ", c);
else
printk("%02x ", c);
}
}
printk("\n");
}
int is_valid_bugaddr(unsigned long ip)
{
unsigned short ud2;
if (__copy_from_user(&ud2, (const void __user *) ip, sizeof(ud2)))
return 0;
return ud2 == 0x0b0f;
}