linux/tools/perf/builtin-top.c

1379 lines
31 KiB
C
Raw Normal View History

/*
* builtin-top.c
*
* Builtin top command: Display a continuously updated profile of
* any workload, CPU or specific PID.
*
* Copyright (C) 2008, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
*
* Improvements and fixes by:
*
* Arjan van de Ven <arjan@linux.intel.com>
* Yanmin Zhang <yanmin.zhang@intel.com>
* Wu Fengguang <fengguang.wu@intel.com>
* Mike Galbraith <efault@gmx.de>
* Paul Mackerras <paulus@samba.org>
*
* Released under the GPL v2. (and only v2, not any later version)
*/
#include "builtin.h"
#include "perf.h"
#include "util/symbol.h"
#include "util/color.h"
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
#include "util/thread.h"
#include "util/util.h"
#include <linux/rbtree.h>
#include "util/parse-options.h"
#include "util/parse-events.h"
#include "util/debug.h"
#include <assert.h>
#include <fcntl.h>
#include <stdio.h>
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
#include <termios.h>
#include <unistd.h>
#include <errno.h>
#include <time.h>
#include <sched.h>
#include <pthread.h>
#include <sys/syscall.h>
#include <sys/ioctl.h>
#include <sys/poll.h>
#include <sys/prctl.h>
#include <sys/wait.h>
#include <sys/uio.h>
#include <sys/mman.h>
#include <linux/unistd.h>
#include <linux/types.h>
static int fd[MAX_NR_CPUS][MAX_COUNTERS];
static int system_wide = 0;
static int default_interval = 0;
static int count_filter = 5;
static int print_entries;
static int target_pid = -1;
static int inherit = 0;
static int profile_cpu = -1;
static int nr_cpus = 0;
static unsigned int realtime_prio = 0;
static int group = 0;
static unsigned int page_size;
static unsigned int mmap_pages = 16;
static int freq = 1000; /* 1 KHz */
static int delay_secs = 2;
static int zero = 0;
static int dump_symtab = 0;
static bool hide_kernel_symbols = false;
static bool hide_user_symbols = false;
static struct winsize winsize;
static const char *graph_line =
"_____________________________________________________________________"
"_____________________________________________________________________";
static const char *graph_dotted_line =
"---------------------------------------------------------------------"
"---------------------------------------------------------------------"
"---------------------------------------------------------------------";
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
/*
* Source
*/
struct source_line {
u64 eip;
unsigned long count[MAX_COUNTERS];
char *line;
struct source_line *next;
};
static char *sym_filter = NULL;
struct sym_entry *sym_filter_entry = NULL;
static int sym_pcnt_filter = 5;
static int sym_counter = 0;
static int display_weighted = -1;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
/*
* Symbols
*/
struct sym_entry {
struct rb_node rb_node;
struct list_head node;
unsigned long count[MAX_COUNTERS];
unsigned long snap_count;
double weight;
int skip;
u16 name_len;
u8 origin;
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
struct map *map;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
struct source_line *source;
struct source_line *lines;
struct source_line **lines_tail;
pthread_mutex_t source_lock;
};
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
/*
* Source functions
*/
static inline struct symbol *sym_entry__symbol(struct sym_entry *self)
{
return (struct symbol *)(self + 1);
}
static void get_term_dimensions(struct winsize *ws)
{
char *s = getenv("LINES");
if (s != NULL) {
ws->ws_row = atoi(s);
s = getenv("COLUMNS");
if (s != NULL) {
ws->ws_col = atoi(s);
if (ws->ws_row && ws->ws_col)
return;
}
}
#ifdef TIOCGWINSZ
if (ioctl(1, TIOCGWINSZ, ws) == 0 &&
ws->ws_row && ws->ws_col)
return;
#endif
ws->ws_row = 25;
ws->ws_col = 80;
}
static void update_print_entries(struct winsize *ws)
{
print_entries = ws->ws_row;
if (print_entries > 9)
print_entries -= 9;
}
static void sig_winch_handler(int sig __used)
{
get_term_dimensions(&winsize);
update_print_entries(&winsize);
}
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
static void parse_source(struct sym_entry *syme)
{
struct symbol *sym;
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
struct map *map;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
FILE *file;
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
char command[PATH_MAX*2];
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
const char *path;
u64 len;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
if (!syme)
return;
if (syme->lines) {
pthread_mutex_lock(&syme->source_lock);
goto out_assign;
}
sym = sym_entry__symbol(syme);
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
map = syme->map;
path = map->dso->long_name;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
len = sym->end - sym->start;
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
sprintf(command,
"objdump --start-address=0x%016Lx "
"--stop-address=0x%016Lx -dS %s",
map->unmap_ip(map, sym->start),
map->unmap_ip(map, sym->end), path);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
file = popen(command, "r");
if (!file)
return;
pthread_mutex_lock(&syme->source_lock);
syme->lines_tail = &syme->lines;
while (!feof(file)) {
struct source_line *src;
size_t dummy = 0;
char *c;
src = malloc(sizeof(struct source_line));
assert(src != NULL);
memset(src, 0, sizeof(struct source_line));
if (getline(&src->line, &dummy, file) < 0)
break;
if (!src->line)
break;
c = strchr(src->line, '\n');
if (c)
*c = 0;
src->next = NULL;
*syme->lines_tail = src;
syme->lines_tail = &src->next;
if (strlen(src->line)>8 && src->line[8] == ':') {
src->eip = strtoull(src->line, NULL, 16);
src->eip = map->unmap_ip(map, src->eip);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
}
if (strlen(src->line)>8 && src->line[16] == ':') {
src->eip = strtoull(src->line, NULL, 16);
src->eip = map->unmap_ip(map, src->eip);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
}
}
pclose(file);
out_assign:
sym_filter_entry = syme;
pthread_mutex_unlock(&syme->source_lock);
}
static void __zero_source_counters(struct sym_entry *syme)
{
int i;
struct source_line *line;
line = syme->lines;
while (line) {
for (i = 0; i < nr_counters; i++)
line->count[i] = 0;
line = line->next;
}
}
static void record_precise_ip(struct sym_entry *syme, int counter, u64 ip)
{
struct source_line *line;
if (syme != sym_filter_entry)
return;
if (pthread_mutex_trylock(&syme->source_lock))
return;
if (!syme->source)
goto out_unlock;
for (line = syme->lines; line; line = line->next) {
if (line->eip == ip) {
line->count[counter]++;
break;
}
if (line->eip > ip)
break;
}
out_unlock:
pthread_mutex_unlock(&syme->source_lock);
}
static void lookup_sym_source(struct sym_entry *syme)
{
struct symbol *symbol = sym_entry__symbol(syme);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
struct source_line *line;
char pattern[PATH_MAX];
sprintf(pattern, "<%s>:", symbol->name);
pthread_mutex_lock(&syme->source_lock);
for (line = syme->lines; line; line = line->next) {
if (strstr(line->line, pattern)) {
syme->source = line;
break;
}
}
pthread_mutex_unlock(&syme->source_lock);
}
static void show_lines(struct source_line *queue, int count, int total)
{
int i;
struct source_line *line;
line = queue;
for (i = 0; i < count; i++) {
float pcnt = 100.0*(float)line->count[sym_counter]/(float)total;
printf("%8li %4.1f%%\t%s\n", line->count[sym_counter], pcnt, line->line);
line = line->next;
}
}
#define TRACE_COUNT 3
static void show_details(struct sym_entry *syme)
{
struct symbol *symbol;
struct source_line *line;
struct source_line *line_queue = NULL;
int displayed = 0;
int line_queue_count = 0, total = 0, more = 0;
if (!syme)
return;
if (!syme->source)
lookup_sym_source(syme);
if (!syme->source)
return;
symbol = sym_entry__symbol(syme);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
printf("Showing %s for %s\n", event_name(sym_counter), symbol->name);
printf(" Events Pcnt (>=%d%%)\n", sym_pcnt_filter);
pthread_mutex_lock(&syme->source_lock);
line = syme->source;
while (line) {
total += line->count[sym_counter];
line = line->next;
}
line = syme->source;
while (line) {
float pcnt = 0.0;
if (!line_queue_count)
line_queue = line;
line_queue_count++;
if (line->count[sym_counter])
pcnt = 100.0 * line->count[sym_counter] / (float)total;
if (pcnt >= (float)sym_pcnt_filter) {
if (displayed <= print_entries)
show_lines(line_queue, line_queue_count, total);
else more++;
displayed += line_queue_count;
line_queue_count = 0;
line_queue = NULL;
} else if (line_queue_count > TRACE_COUNT) {
line_queue = line_queue->next;
line_queue_count--;
}
line->count[sym_counter] = zero ? 0 : line->count[sym_counter] * 7 / 8;
line = line->next;
}
pthread_mutex_unlock(&syme->source_lock);
if (more)
printf("%d lines not displayed, maybe increase display entries [e]\n", more);
}
/*
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
* Symbols will be added here in event__process_sample and will get out
* after decayed.
*/
static LIST_HEAD(active_symbols);
static pthread_mutex_t active_symbols_lock = PTHREAD_MUTEX_INITIALIZER;
/*
* Ordering weight: count-1 * count-2 * ... / count-n
*/
static double sym_weight(const struct sym_entry *sym)
{
double weight = sym->snap_count;
int counter;
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
if (!display_weighted)
return weight;
for (counter = 1; counter < nr_counters-1; counter++)
weight *= sym->count[counter];
weight /= (sym->count[counter] + 1);
return weight;
}
static long samples;
static long userspace_samples;
static const char CONSOLE_CLEAR[] = "";
static void __list_insert_active_sym(struct sym_entry *syme)
{
list_add(&syme->node, &active_symbols);
}
static void list_remove_active_sym(struct sym_entry *syme)
{
pthread_mutex_lock(&active_symbols_lock);
list_del_init(&syme->node);
pthread_mutex_unlock(&active_symbols_lock);
}
static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se)
{
struct rb_node **p = &tree->rb_node;
struct rb_node *parent = NULL;
struct sym_entry *iter;
while (*p != NULL) {
parent = *p;
iter = rb_entry(parent, struct sym_entry, rb_node);
if (se->weight > iter->weight)
p = &(*p)->rb_left;
else
p = &(*p)->rb_right;
}
rb_link_node(&se->rb_node, parent, p);
rb_insert_color(&se->rb_node, tree);
}
static void print_sym_table(void)
{
int printed = 0, j;
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
int counter, snap = !display_weighted ? sym_counter : 0;
float samples_per_sec = samples/delay_secs;
float ksamples_per_sec = (samples-userspace_samples)/delay_secs;
float sum_ksamples = 0.0;
struct sym_entry *syme, *n;
struct rb_root tmp = RB_ROOT;
struct rb_node *nd;
int sym_width = 0, dso_width = 0;
const int win_width = winsize.ws_col - 1;
struct dso *unique_dso = NULL, *first_dso = NULL;
samples = userspace_samples = 0;
/* Sort the active symbols */
pthread_mutex_lock(&active_symbols_lock);
syme = list_entry(active_symbols.next, struct sym_entry, node);
pthread_mutex_unlock(&active_symbols_lock);
list_for_each_entry_safe_from(syme, n, &active_symbols, node) {
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
syme->snap_count = syme->count[snap];
if (syme->snap_count != 0) {
if ((hide_user_symbols &&
syme->origin == PERF_RECORD_MISC_USER) ||
(hide_kernel_symbols &&
syme->origin == PERF_RECORD_MISC_KERNEL)) {
list_remove_active_sym(syme);
continue;
}
syme->weight = sym_weight(syme);
rb_insert_active_sym(&tmp, syme);
sum_ksamples += syme->snap_count;
perf top: Reduce display overhead Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-25 15:57:56 +08:00
for (j = 0; j < nr_counters; j++)
syme->count[j] = zero ? 0 : syme->count[j] * 7 / 8;
} else
list_remove_active_sym(syme);
perf top: Reduce display overhead Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-25 15:57:56 +08:00
}
puts(CONSOLE_CLEAR);
printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
printf( " PerfTop:%8.0f irqs/sec kernel:%4.1f%% [",
samples_per_sec,
100.0 - (100.0*((samples_per_sec-ksamples_per_sec)/samples_per_sec)));
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
if (nr_counters == 1 || !display_weighted) {
printf("%Ld", (u64)attrs[0].sample_period);
if (freq)
printf("Hz ");
else
printf(" ");
}
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
if (!display_weighted)
printf("%s", event_name(sym_counter));
else for (counter = 0; counter < nr_counters; counter++) {
if (counter)
printf("/");
printf("%s", event_name(counter));
}
printf( "], ");
if (target_pid != -1)
printf(" (target_pid: %d", target_pid);
else
printf(" (all");
if (profile_cpu != -1)
printf(", cpu: %d)\n", profile_cpu);
else {
if (target_pid != -1)
printf(")\n");
else
printf(", %d CPUs)\n", nr_cpus);
}
printf("%-*.*s\n", win_width, win_width, graph_dotted_line);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
if (sym_filter_entry) {
show_details(sym_filter_entry);
return;
}
/*
* Find the longest symbol name that will be displayed
*/
for (nd = rb_first(&tmp); nd; nd = rb_next(nd)) {
syme = rb_entry(nd, struct sym_entry, rb_node);
if (++printed > print_entries ||
(int)syme->snap_count < count_filter)
continue;
if (first_dso == NULL)
unique_dso = first_dso = syme->map->dso;
else if (syme->map->dso != first_dso)
unique_dso = NULL;
if (syme->map->dso->long_name_len > dso_width)
dso_width = syme->map->dso->long_name_len;
if (syme->name_len > sym_width)
sym_width = syme->name_len;
}
printed = 0;
if (unique_dso)
printf("DSO: %s\n", unique_dso->long_name);
else {
int max_dso_width = winsize.ws_col - sym_width - 29;
if (dso_width > max_dso_width)
dso_width = max_dso_width;
putchar('\n');
}
if (nr_counters == 1)
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf(" samples pcnt");
else
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf(" weight samples pcnt");
if (verbose)
printf(" RIP ");
printf(" %-*.*s", sym_width, sym_width, "function");
if (!unique_dso)
printf(" DSO");
putchar('\n');
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf(" %s _______ _____",
nr_counters == 1 ? " " : "______");
if (verbose)
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf(" ________________");
printf(" %-*.*s", sym_width, sym_width, graph_line);
if (!unique_dso)
printf(" %-*.*s", dso_width, dso_width, graph_line);
puts("\n");
for (nd = rb_first(&tmp); nd; nd = rb_next(nd)) {
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
struct symbol *sym;
double pcnt;
perf top: Reduce display overhead Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-25 15:57:56 +08:00
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
syme = rb_entry(nd, struct sym_entry, rb_node);
sym = sym_entry__symbol(syme);
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
if (++printed > print_entries || (int)syme->snap_count < count_filter)
continue;
perf top: Reduce display overhead Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-25 15:57:56 +08:00
pcnt = 100.0 - (100.0 * ((sum_ksamples - syme->snap_count) /
sum_ksamples));
perf top: Reduce display overhead Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-25 15:57:56 +08:00
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
if (nr_counters == 1 || !display_weighted)
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf("%20.2f ", syme->weight);
perf top: Reduce display overhead Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-05-25 15:57:56 +08:00
else
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf("%9.1f %10ld ", syme->weight, syme->snap_count);
percent_color_fprintf(stdout, "%4.1f%%", pcnt);
if (verbose)
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
printf(" %016llx", sym->start);
printf(" %-*.*s", sym_width, sym_width, sym->name);
if (!unique_dso)
printf(" %-*.*s", dso_width, dso_width,
dso_width >= syme->map->dso->long_name_len ?
syme->map->dso->long_name :
syme->map->dso->short_name);
printf("\n");
}
}
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
static void prompt_integer(int *target, const char *msg)
{
char *buf = malloc(0), *p;
size_t dummy = 0;
int tmp;
fprintf(stdout, "\n%s: ", msg);
if (getline(&buf, &dummy, stdin) < 0)
return;
p = strchr(buf, '\n');
if (p)
*p = 0;
p = buf;
while(*p) {
if (!isdigit(*p))
goto out_free;
p++;
}
tmp = strtoul(buf, NULL, 10);
*target = tmp;
out_free:
free(buf);
}
static void prompt_percent(int *target, const char *msg)
{
int tmp = 0;
prompt_integer(&tmp, msg);
if (tmp >= 0 && tmp <= 100)
*target = tmp;
}
static void prompt_symbol(struct sym_entry **target, const char *msg)
{
char *buf = malloc(0), *p;
struct sym_entry *syme = *target, *n, *found = NULL;
size_t dummy = 0;
/* zero counters of active symbol */
if (syme) {
pthread_mutex_lock(&syme->source_lock);
__zero_source_counters(syme);
*target = NULL;
pthread_mutex_unlock(&syme->source_lock);
}
fprintf(stdout, "\n%s: ", msg);
if (getline(&buf, &dummy, stdin) < 0)
goto out_free;
p = strchr(buf, '\n');
if (p)
*p = 0;
pthread_mutex_lock(&active_symbols_lock);
syme = list_entry(active_symbols.next, struct sym_entry, node);
pthread_mutex_unlock(&active_symbols_lock);
list_for_each_entry_safe_from(syme, n, &active_symbols, node) {
struct symbol *sym = sym_entry__symbol(syme);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
if (!strcmp(buf, sym->name)) {
found = syme;
break;
}
}
if (!found) {
fprintf(stderr, "Sorry, %s is not active.\n", sym_filter);
sleep(1);
return;
} else
parse_source(found);
out_free:
free(buf);
}
static void print_mapped_keys(void)
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
{
char *name = NULL;
if (sym_filter_entry) {
struct symbol *sym = sym_entry__symbol(sym_filter_entry);
name = sym->name;
}
fprintf(stdout, "\nMapped keys:\n");
fprintf(stdout, "\t[d] display refresh delay. \t(%d)\n", delay_secs);
fprintf(stdout, "\t[e] display entries (lines). \t(%d)\n", print_entries);
if (nr_counters > 1)
fprintf(stdout, "\t[E] active event counter. \t(%s)\n", event_name(sym_counter));
fprintf(stdout, "\t[f] profile display filter (count). \t(%d)\n", count_filter);
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
if (vmlinux_name) {
fprintf(stdout, "\t[F] annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter);
fprintf(stdout, "\t[s] annotate symbol. \t(%s)\n", name?: "NULL");
fprintf(stdout, "\t[S] stop annotation.\n");
}
if (nr_counters > 1)
fprintf(stdout, "\t[w] toggle display weighted/count[E]r. \t(%d)\n", display_weighted ? 1 : 0);
fprintf(stdout,
"\t[K] hide kernel_symbols symbols. \t(%s)\n",
hide_kernel_symbols ? "yes" : "no");
fprintf(stdout,
"\t[U] hide user symbols. \t(%s)\n",
hide_user_symbols ? "yes" : "no");
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
fprintf(stdout, "\t[z] toggle sample zeroing. \t(%d)\n", zero ? 1 : 0);
fprintf(stdout, "\t[qQ] quit.\n");
}
static int key_mapped(int c)
{
switch (c) {
case 'd':
case 'e':
case 'f':
case 'z':
case 'q':
case 'Q':
case 'K':
case 'U':
return 1;
case 'E':
case 'w':
return nr_counters > 1 ? 1 : 0;
case 'F':
case 's':
case 'S':
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
return vmlinux_name ? 1 : 0;
default:
break;
}
return 0;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
}
static void handle_keypress(int c)
{
if (!key_mapped(c)) {
struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
struct termios tc, save;
print_mapped_keys();
fprintf(stdout, "\nEnter selection, or unmapped key to continue: ");
fflush(stdout);
tcgetattr(0, &save);
tc = save;
tc.c_lflag &= ~(ICANON | ECHO);
tc.c_cc[VMIN] = 0;
tc.c_cc[VTIME] = 0;
tcsetattr(0, TCSANOW, &tc);
poll(&stdin_poll, 1, -1);
c = getc(stdin);
tcsetattr(0, TCSAFLUSH, &save);
if (!key_mapped(c))
return;
}
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
switch (c) {
case 'd':
prompt_integer(&delay_secs, "Enter display delay");
if (delay_secs < 1)
delay_secs = 1;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
break;
case 'e':
prompt_integer(&print_entries, "Enter display entries (lines)");
if (print_entries == 0) {
sig_winch_handler(SIGWINCH);
signal(SIGWINCH, sig_winch_handler);
} else
signal(SIGWINCH, SIG_DFL);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
break;
case 'E':
if (nr_counters > 1) {
int i;
fprintf(stderr, "\nAvailable events:");
for (i = 0; i < nr_counters; i++)
fprintf(stderr, "\n\t%d %s", i, event_name(i));
prompt_integer(&sym_counter, "Enter details event counter");
if (sym_counter >= nr_counters) {
fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(0));
sym_counter = 0;
sleep(1);
}
} else sym_counter = 0;
break;
case 'f':
prompt_integer(&count_filter, "Enter display event count filter");
break;
case 'F':
prompt_percent(&sym_pcnt_filter, "Enter details display event filter (percent)");
break;
case 'K':
hide_kernel_symbols = !hide_kernel_symbols;
break;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
case 'q':
case 'Q':
printf("exiting.\n");
exit(0);
case 's':
prompt_symbol(&sym_filter_entry, "Enter details symbol");
break;
case 'S':
if (!sym_filter_entry)
break;
else {
struct sym_entry *syme = sym_filter_entry;
pthread_mutex_lock(&syme->source_lock);
sym_filter_entry = NULL;
__zero_source_counters(syme);
pthread_mutex_unlock(&syme->source_lock);
}
break;
case 'U':
hide_user_symbols = !hide_user_symbols;
break;
perf_counter tools: Allow perf top top users to switch between weighted and individual counter display Add [w]eighted hotkey. Pressing [w] toggles between displaying weighted total of all counters, and the counter selected via [E]vent select key. ------------------------------------------------------------------------------ PerfTop: 90395 irqs/sec kernel:16.1% [cache-misses/cache-references/instructions], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 1275408.6 10881 - 5.3% - ffffffff81146f70 : copy_page_c 553683.4 43569 - 21.3% - ffffffff81146f20 : clear_page_c 74075.0 6768 - 3.3% - ffffffff81147190 : copy_user_generic_string 40602.9 7538 - 3.7% - ffffffff81284ba2 : _spin_lock 26882.1 965 - 0.5% - ffffffff8109d280 : file_ra_state_init [w] ------------------------------------------------------------------------------ PerfTop: 91221 irqs/sec kernel:14.5% [10000Hz cache-misses], (all, 4 CPUs) ------------------------------------------------------------------------------ weight samples pcnt RIP kernel function ______ _______ _____ ________________ _______________ 47320.00 - 22.3% - ffffffff81146f20 : clear_page_c 14261.00 - 6.7% - ffffffff810992f5 : __rmqueue 11046.00 - 5.2% - ffffffff81146f70 : copy_page_c 7842.00 - 3.7% - ffffffff81284ba2 : _spin_lock 7234.00 - 3.4% - ffffffff810aa1d6 : unmap_vmas Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-24 16:09:50 +08:00
case 'w':
display_weighted = ~display_weighted;
break;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
case 'z':
zero = ~zero;
break;
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
default:
break;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
}
}
static void *display_thread(void *arg __used)
{
struct pollfd stdin_poll = { .fd = 0, .events = POLLIN };
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
struct termios tc, save;
int delay_msecs, c;
tcgetattr(0, &save);
tc = save;
tc.c_lflag &= ~(ICANON | ECHO);
tc.c_cc[VMIN] = 0;
tc.c_cc[VTIME] = 0;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
repeat:
delay_msecs = delay_secs * 1000;
tcsetattr(0, TCSANOW, &tc);
/* trash return*/
getc(stdin);
do {
print_sym_table();
} while (!poll(&stdin_poll, 1, delay_msecs) == 1);
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
c = getc(stdin);
tcsetattr(0, TCSAFLUSH, &save);
handle_keypress(c);
goto repeat;
return NULL;
}
/* Tag samples to be skipped. */
static const char *skip_symbols[] = {
"default_idle",
"cpu_idle",
"enter_idle",
"exit_idle",
"mwait_idle",
"mwait_idle_with_hints",
"poll_idle",
"ppc64_runlatch_off",
"pseries_dedicated_idle_sleep",
NULL
};
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
static int symbol_filter(struct map *map, struct symbol *sym)
{
struct sym_entry *syme;
const char *name = sym->name;
int i;
/*
* ppc64 uses function descriptors and appends a '.' to the
* start of every instruction address. Remove it.
*/
if (name[0] == '.')
name++;
if (!strcmp(name, "_text") ||
!strcmp(name, "_etext") ||
!strcmp(name, "_sinittext") ||
!strncmp("init_module", name, 11) ||
!strncmp("cleanup_module", name, 14) ||
strstr(name, "_text_start") ||
strstr(name, "_text_end"))
return 1;
syme = symbol__priv(sym);
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
syme->map = map;
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
pthread_mutex_init(&syme->source_lock, NULL);
if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter))
sym_filter_entry = syme;
for (i = 0; skip_symbols[i]; i++) {
if (!strcmp(skip_symbols[i], name)) {
syme->skip = 1;
break;
}
}
if (!syme->skip)
syme->name_len = strlen(sym->name);
return 0;
}
static int parse_symbols(void)
{
if (dsos__load_kernel(vmlinux_name, symbol_filter, 1) <= 0)
return -1;
if (dump_symtab)
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
dsos__fprintf(stderr);
return 0;
}
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
static void event__process_sample(const event_t *self, int counter)
{
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
u64 ip = self->ip.ip;
perf tools: Rewrite and improve support for kernel modules Representing modules as struct map entries, backed by a DSO, etc, using /proc/modules to find where the module is loaded. DSOs now can have a short and long name, so that in verbose mode we can show exactly which .ko or vmlinux image was used. As kernel modules now are a DSO separate from the kernel, we can ask for just the hits for a particular set of kernel modules, just like we can do with shared libraries: [root@doppio linux-2.6-tip]# perf report -n --vmlinux /home/acme/git/build/tip-recvmmsg/vmlinux --modules --dsos \[drm\] | head -15 84.58% 13266 Xorg [k] drm_clflush_pages 4.02% 630 Xorg [k] trace_kmalloc.clone.0 3.95% 619 Xorg [k] drm_ioctl 2.07% 324 Xorg [k] drm_addbufs 1.68% 263 Xorg [k] drm_gem_close_ioctl 0.77% 120 Xorg [k] drm_setmaster_ioctl 0.70% 110 Xorg [k] drm_lastclose 0.68% 106 Xorg [k] drm_open 0.54% 85 Xorg [k] drm_mm_search_free [root@doppio linux-2.6-tip]# Specifying --dsos /lib/modules/2.6.31-tip/kernel/drivers/gpu/drm/drm.ko would have the same effect. Allowing specifying just 'drm.ko' is left for another patch. Processing kallsyms so that per kernel module struct map are instantiated was also left for another patch. That will allow removing the module name from each of its symbols. struct symbol was reduced by removing the ->module backpointer and moving it (well now the map) to struct symbol_entry in perf top, that is its only user right now. The total linecount went down by ~500 lines. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Avi Kivity <avi@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-02 14:29:58 +08:00
struct map *map;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
struct sym_entry *syme;
struct symbol *sym;
u8 origin = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
switch (origin) {
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
case PERF_RECORD_MISC_USER: {
struct thread *thread;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (hide_user_symbols)
return;
thread = threads__findnew(self->ip.pid);
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (thread == NULL)
return;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
map = thread__find_map(thread, ip);
if (map != NULL) {
ip = map->map_ip(map, ip);
sym = map__find_symbol(map, ip, symbol_filter);
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (sym == NULL)
return;
userspace_samples++;
break;
}
}
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
/*
* If this is outside of all known maps,
* and is a negative address, try to look it
* up in the kernel dso, as it might be a
* vsyscall or vdso (which executes in user-mode).
*/
if ((long long)ip >= 0)
return;
/* Fall thru */
case PERF_RECORD_MISC_KERNEL:
if (hide_kernel_symbols)
return;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
sym = kernel_maps__find_symbol(ip, &map);
if (sym == NULL)
return;
break;
default:
return;
}
syme = symbol__priv(sym);
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (!syme->skip) {
syme->count[counter]++;
syme->origin = origin;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
record_precise_ip(syme, counter, ip);
pthread_mutex_lock(&active_symbols_lock);
if (list_empty(&syme->node) || !syme->node.next)
__list_insert_active_sym(syme);
pthread_mutex_unlock(&active_symbols_lock);
++samples;
return;
}
}
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
static void event__process_mmap(event_t *self)
{
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
struct thread *thread = threads__findnew(self->mmap.pid);
if (thread != NULL) {
struct map *map = map__new(&self->mmap, NULL, 0);
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (map != NULL)
thread__insert_map(thread, map);
}
}
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
static void event__process_comm(event_t *self)
{
struct thread *thread = threads__findnew(self->comm.pid);
if (thread != NULL)
thread__set_comm(thread, self->comm.comm);
}
static int event__process(event_t *event)
{
switch (event->header.type) {
case PERF_RECORD_COMM:
event__process_comm(event);
break;
case PERF_RECORD_MMAP:
event__process_mmap(event);
break;
default:
break;
}
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
return 0;
}
struct mmap_data {
int counter;
void *base;
int mask;
unsigned int prev;
};
static unsigned int mmap_read_head(struct mmap_data *md)
{
perf: Do the big rename: Performance Counters -> Performance Events Bye-bye Performance Counters, welcome Performance Events! In the past few months the perfcounters subsystem has grown out its initial role of counting hardware events, and has become (and is becoming) a much broader generic event enumeration, reporting, logging, monitoring, analysis facility. Naming its core object 'perf_counter' and naming the subsystem 'perfcounters' has become more and more of a misnomer. With pending code like hw-breakpoints support the 'counter' name is less and less appropriate. All in one, we've decided to rename the subsystem to 'performance events' and to propagate this rename through all fields, variables and API names. (in an ABI compatible fashion) The word 'event' is also a bit shorter than 'counter' - which makes it slightly more convenient to write/handle as well. Thanks goes to Stephane Eranian who first observed this misnomer and suggested a rename. User-space tooling and ABI compatibility is not affected - this patch should be function-invariant. (Also, defconfigs were not touched to keep the size down.) This patch has been generated via the following script: FILES=$(find * -type f | grep -vE 'oprofile|[^K]config') sed -i \ -e 's/PERF_EVENT_/PERF_RECORD_/g' \ -e 's/PERF_COUNTER/PERF_EVENT/g' \ -e 's/perf_counter/perf_event/g' \ -e 's/nb_counters/nb_events/g' \ -e 's/swcounter/swevent/g' \ -e 's/tpcounter_event/tp_event/g' \ $FILES for N in $(find . -name perf_counter.[ch]); do M=$(echo $N | sed 's/perf_counter/perf_event/g') mv $N $M done FILES=$(find . -name perf_event.*) sed -i \ -e 's/COUNTER_MASK/REG_MASK/g' \ -e 's/COUNTER/EVENT/g' \ -e 's/\<event\>/event_id/g' \ -e 's/counter/event/g' \ -e 's/Counter/Event/g' \ $FILES ... to keep it as correct as possible. This script can also be used by anyone who has pending perfcounters patches - it converts a Linux kernel tree over to the new naming. We tried to time this change to the point in time where the amount of pending patches is the smallest: the end of the merge window. Namespace clashes were fixed up in a preparatory patch - and some stylistic fallout will be fixed up in a subsequent patch. ( NOTE: 'counters' are still the proper terminology when we deal with hardware registers - and these sed scripts are a bit over-eager in renaming them. I've undone some of that, but in case there's something left where 'counter' would be better than 'event' we can undo that on an individual basis instead of touching an otherwise nicely automated patch. ) Suggested-by: Stephane Eranian <eranian@google.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Paul Mackerras <paulus@samba.org> Reviewed-by: Arjan van de Ven <arjan@linux.intel.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: David Howells <dhowells@redhat.com> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: <linux-arch@vger.kernel.org> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 18:02:48 +08:00
struct perf_event_mmap_page *pc = md->base;
int head;
head = pc->data_head;
rmb();
return head;
}
static void mmap_read_counter(struct mmap_data *md)
{
unsigned int head = mmap_read_head(md);
unsigned int old = md->prev;
unsigned char *data = md->base + page_size;
int diff;
/*
* If we're further behind than half the buffer, there's a chance
* the writer will bite our tail and mess up the samples under us.
*
* If we somehow ended up ahead of the head, we got messed up.
*
* In either case, truncate and restart at head.
*/
diff = head - old;
if (diff > md->mask / 2 || diff < 0) {
fprintf(stderr, "WARNING: failed to keep up with mmap data.\n");
/*
* head points to a known good entry, start there.
*/
old = head;
}
for (; old != head;) {
event_t *event = (event_t *)&data[old & md->mask];
event_t event_copy;
size_t size = event->header.size;
/*
* Event straddles the mmap boundary -- header should always
* be inside due to u64 alignment of output.
*/
if ((old & md->mask) + size != ((old + size) & md->mask)) {
unsigned int offset = old;
unsigned int len = min(sizeof(*event), size), cpy;
void *dst = &event_copy;
do {
cpy = min(md->mask + 1 - (offset & md->mask), len);
memcpy(dst, &data[offset & md->mask], cpy);
offset += cpy;
dst += cpy;
len -= cpy;
} while (len);
event = &event_copy;
}
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (event->header.type == PERF_RECORD_SAMPLE)
event__process_sample(event, md->counter);
else
event__process(event);
old += size;
}
md->prev = old;
}
static struct pollfd event_array[MAX_NR_CPUS * MAX_COUNTERS];
static struct mmap_data mmap_array[MAX_NR_CPUS][MAX_COUNTERS];
static void mmap_read(void)
{
int i, counter;
for (i = 0; i < nr_cpus; i++) {
for (counter = 0; counter < nr_counters; counter++)
mmap_read_counter(&mmap_array[i][counter]);
}
}
int nr_poll;
int group_fd;
static void start_counter(int i, int counter)
{
perf: Do the big rename: Performance Counters -> Performance Events Bye-bye Performance Counters, welcome Performance Events! In the past few months the perfcounters subsystem has grown out its initial role of counting hardware events, and has become (and is becoming) a much broader generic event enumeration, reporting, logging, monitoring, analysis facility. Naming its core object 'perf_counter' and naming the subsystem 'perfcounters' has become more and more of a misnomer. With pending code like hw-breakpoints support the 'counter' name is less and less appropriate. All in one, we've decided to rename the subsystem to 'performance events' and to propagate this rename through all fields, variables and API names. (in an ABI compatible fashion) The word 'event' is also a bit shorter than 'counter' - which makes it slightly more convenient to write/handle as well. Thanks goes to Stephane Eranian who first observed this misnomer and suggested a rename. User-space tooling and ABI compatibility is not affected - this patch should be function-invariant. (Also, defconfigs were not touched to keep the size down.) This patch has been generated via the following script: FILES=$(find * -type f | grep -vE 'oprofile|[^K]config') sed -i \ -e 's/PERF_EVENT_/PERF_RECORD_/g' \ -e 's/PERF_COUNTER/PERF_EVENT/g' \ -e 's/perf_counter/perf_event/g' \ -e 's/nb_counters/nb_events/g' \ -e 's/swcounter/swevent/g' \ -e 's/tpcounter_event/tp_event/g' \ $FILES for N in $(find . -name perf_counter.[ch]); do M=$(echo $N | sed 's/perf_counter/perf_event/g') mv $N $M done FILES=$(find . -name perf_event.*) sed -i \ -e 's/COUNTER_MASK/REG_MASK/g' \ -e 's/COUNTER/EVENT/g' \ -e 's/\<event\>/event_id/g' \ -e 's/counter/event/g' \ -e 's/Counter/Event/g' \ $FILES ... to keep it as correct as possible. This script can also be used by anyone who has pending perfcounters patches - it converts a Linux kernel tree over to the new naming. We tried to time this change to the point in time where the amount of pending patches is the smallest: the end of the merge window. Namespace clashes were fixed up in a preparatory patch - and some stylistic fallout will be fixed up in a subsequent patch. ( NOTE: 'counters' are still the proper terminology when we deal with hardware registers - and these sed scripts are a bit over-eager in renaming them. I've undone some of that, but in case there's something left where 'counter' would be better than 'event' we can undo that on an individual basis instead of touching an otherwise nicely automated patch. ) Suggested-by: Stephane Eranian <eranian@google.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Paul Mackerras <paulus@samba.org> Reviewed-by: Arjan van de Ven <arjan@linux.intel.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: David Howells <dhowells@redhat.com> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: <linux-arch@vger.kernel.org> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 18:02:48 +08:00
struct perf_event_attr *attr;
int cpu;
cpu = profile_cpu;
if (target_pid == -1 && profile_cpu == -1)
cpu = i;
attr = attrs + counter;
attr->sample_type = PERF_SAMPLE_IP | PERF_SAMPLE_TID;
if (freq) {
attr->sample_type |= PERF_SAMPLE_PERIOD;
attr->freq = 1;
attr->sample_freq = freq;
}
attr->inherit = (cpu < 0) && inherit;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
attr->mmap = 1;
try_again:
perf: Do the big rename: Performance Counters -> Performance Events Bye-bye Performance Counters, welcome Performance Events! In the past few months the perfcounters subsystem has grown out its initial role of counting hardware events, and has become (and is becoming) a much broader generic event enumeration, reporting, logging, monitoring, analysis facility. Naming its core object 'perf_counter' and naming the subsystem 'perfcounters' has become more and more of a misnomer. With pending code like hw-breakpoints support the 'counter' name is less and less appropriate. All in one, we've decided to rename the subsystem to 'performance events' and to propagate this rename through all fields, variables and API names. (in an ABI compatible fashion) The word 'event' is also a bit shorter than 'counter' - which makes it slightly more convenient to write/handle as well. Thanks goes to Stephane Eranian who first observed this misnomer and suggested a rename. User-space tooling and ABI compatibility is not affected - this patch should be function-invariant. (Also, defconfigs were not touched to keep the size down.) This patch has been generated via the following script: FILES=$(find * -type f | grep -vE 'oprofile|[^K]config') sed -i \ -e 's/PERF_EVENT_/PERF_RECORD_/g' \ -e 's/PERF_COUNTER/PERF_EVENT/g' \ -e 's/perf_counter/perf_event/g' \ -e 's/nb_counters/nb_events/g' \ -e 's/swcounter/swevent/g' \ -e 's/tpcounter_event/tp_event/g' \ $FILES for N in $(find . -name perf_counter.[ch]); do M=$(echo $N | sed 's/perf_counter/perf_event/g') mv $N $M done FILES=$(find . -name perf_event.*) sed -i \ -e 's/COUNTER_MASK/REG_MASK/g' \ -e 's/COUNTER/EVENT/g' \ -e 's/\<event\>/event_id/g' \ -e 's/counter/event/g' \ -e 's/Counter/Event/g' \ $FILES ... to keep it as correct as possible. This script can also be used by anyone who has pending perfcounters patches - it converts a Linux kernel tree over to the new naming. We tried to time this change to the point in time where the amount of pending patches is the smallest: the end of the merge window. Namespace clashes were fixed up in a preparatory patch - and some stylistic fallout will be fixed up in a subsequent patch. ( NOTE: 'counters' are still the proper terminology when we deal with hardware registers - and these sed scripts are a bit over-eager in renaming them. I've undone some of that, but in case there's something left where 'counter' would be better than 'event' we can undo that on an individual basis instead of touching an otherwise nicely automated patch. ) Suggested-by: Stephane Eranian <eranian@google.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Paul Mackerras <paulus@samba.org> Reviewed-by: Arjan van de Ven <arjan@linux.intel.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: David Howells <dhowells@redhat.com> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: <linux-arch@vger.kernel.org> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 18:02:48 +08:00
fd[i][counter] = sys_perf_event_open(attr, target_pid, cpu, group_fd, 0);
if (fd[i][counter] < 0) {
int err = errno;
if (err == EPERM || err == EACCES)
die("No permission - are you root?\n");
/*
* If it's cycles then fall back to hrtimer
* based cpu-clock-tick sw counter, which
* is always available even if no PMU support:
*/
if (attr->type == PERF_TYPE_HARDWARE
&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {
if (verbose)
warning(" ... trying to fall back to cpu-clock-ticks\n");
attr->type = PERF_TYPE_SOFTWARE;
attr->config = PERF_COUNT_SW_CPU_CLOCK;
goto try_again;
}
printf("\n");
error("perfcounter syscall returned with %d (%s)\n",
fd[i][counter], strerror(err));
perf: Do the big rename: Performance Counters -> Performance Events Bye-bye Performance Counters, welcome Performance Events! In the past few months the perfcounters subsystem has grown out its initial role of counting hardware events, and has become (and is becoming) a much broader generic event enumeration, reporting, logging, monitoring, analysis facility. Naming its core object 'perf_counter' and naming the subsystem 'perfcounters' has become more and more of a misnomer. With pending code like hw-breakpoints support the 'counter' name is less and less appropriate. All in one, we've decided to rename the subsystem to 'performance events' and to propagate this rename through all fields, variables and API names. (in an ABI compatible fashion) The word 'event' is also a bit shorter than 'counter' - which makes it slightly more convenient to write/handle as well. Thanks goes to Stephane Eranian who first observed this misnomer and suggested a rename. User-space tooling and ABI compatibility is not affected - this patch should be function-invariant. (Also, defconfigs were not touched to keep the size down.) This patch has been generated via the following script: FILES=$(find * -type f | grep -vE 'oprofile|[^K]config') sed -i \ -e 's/PERF_EVENT_/PERF_RECORD_/g' \ -e 's/PERF_COUNTER/PERF_EVENT/g' \ -e 's/perf_counter/perf_event/g' \ -e 's/nb_counters/nb_events/g' \ -e 's/swcounter/swevent/g' \ -e 's/tpcounter_event/tp_event/g' \ $FILES for N in $(find . -name perf_counter.[ch]); do M=$(echo $N | sed 's/perf_counter/perf_event/g') mv $N $M done FILES=$(find . -name perf_event.*) sed -i \ -e 's/COUNTER_MASK/REG_MASK/g' \ -e 's/COUNTER/EVENT/g' \ -e 's/\<event\>/event_id/g' \ -e 's/counter/event/g' \ -e 's/Counter/Event/g' \ $FILES ... to keep it as correct as possible. This script can also be used by anyone who has pending perfcounters patches - it converts a Linux kernel tree over to the new naming. We tried to time this change to the point in time where the amount of pending patches is the smallest: the end of the merge window. Namespace clashes were fixed up in a preparatory patch - and some stylistic fallout will be fixed up in a subsequent patch. ( NOTE: 'counters' are still the proper terminology when we deal with hardware registers - and these sed scripts are a bit over-eager in renaming them. I've undone some of that, but in case there's something left where 'counter' would be better than 'event' we can undo that on an individual basis instead of touching an otherwise nicely automated patch. ) Suggested-by: Stephane Eranian <eranian@google.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by: Paul Mackerras <paulus@samba.org> Reviewed-by: Arjan van de Ven <arjan@linux.intel.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: David Howells <dhowells@redhat.com> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: <linux-arch@vger.kernel.org> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 18:02:48 +08:00
die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
exit(-1);
}
assert(fd[i][counter] >= 0);
fcntl(fd[i][counter], F_SETFL, O_NONBLOCK);
/*
* First counter acts as the group leader:
*/
if (group && group_fd == -1)
group_fd = fd[i][counter];
event_array[nr_poll].fd = fd[i][counter];
event_array[nr_poll].events = POLLIN;
nr_poll++;
mmap_array[i][counter].counter = counter;
mmap_array[i][counter].prev = 0;
mmap_array[i][counter].mask = mmap_pages*page_size - 1;
mmap_array[i][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
PROT_READ, MAP_SHARED, fd[i][counter], 0);
if (mmap_array[i][counter].base == MAP_FAILED)
die("failed to mmap with %d (%s)\n", errno, strerror(errno));
}
static int __cmd_top(void)
{
pthread_t thread;
int i, counter;
int ret;
perf top: Support userspace symbols too Example: Compiling the kernel with 'make -k 22 allyesconfig' [root@emilia linux-2.6-tip]# perf top -r 90 ------------------------------------------------------------------------------ PerfTop: 3669 irqs/sec kernel:59.9% [1000Hz cycles], (all, 8 CPUs) ------------------------------------------------------------------------------ samples pcnt function DSO _______ _____ ________________________________ ________________ 3062.00 6.5% clear_page_c [kernel] 2233.00 4.8% _int_malloc /lib64/libc-2.5.so 2100.00 4.5% yylex /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 2029.00 4.3% memset /lib64/libc-2.5.so 1224.00 2.6% page_fault [kernel] 1075.00 2.3% __GI_strlen /lib64/libc-2.5.so 863.00 1.8% sub_preempt_count [kernel] 822.00 1.8% __GI_memcpy /lib64/libc-2.5.so 810.00 1.7% __GI_vfprintf /lib64/libc-2.5.so 786.00 1.7% _int_free /lib64/libc-2.5.so 775.00 1.7% __GI_strcmp /lib64/libc-2.5.so 748.00 1.6% _spin_lock [kernel] 699.00 1.5% main /home/acme/git/build/allyesconfig/scripts/basic/fixdep 659.00 1.4% add_preempt_count [kernel] 649.00 1.4% yyparse /home/acme/git/build/allyesconfig/scripts/genksyms/genksyms 645.00 1.4% preempt_trace [kernel] 635.00 1.4% __GI___libc_free /lib64/libc-2.5.so 597.00 1.3% trace_preempt_on [kernel] 551.00 1.2% __GI___libc_malloc /lib64/libc-2.5.so 516.00 1.1% _spin_lock_irqsave [kernel] 481.00 1.0% copy_user_generic_string [kernel] 479.00 1.0% unmap_vmas [kernel] 429.00 0.9% _IO_file_xsputn_internal /lib64/libc-2.5.so 425.00 0.9% __GI_strncpy /lib64/libc-2.5.so 416.00 0.9% get_page_from_freelist [kernel] 414.00 0.9% malloc_consolidate /lib64/libc-2.5.so 406.00 0.9% get_parent_ip [kernel] 362.00 0.8% __rmqueue [kernel] 347.00 0.7% in_lock_functions [kernel] 316.00 0.7% __d_lookup [kernel] [root@emilia linux-2.6-tip]# More polishing is needed to print just DSO basename when not --verbose, etc. Supporting a 'comm' column requires some more reworking of 'perf top' internals as we will need to use something like the hist entries 'perf report' uses and will be done in another patch. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <1256592199-9608-3-git-send-email-acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-10-27 05:23:19 +08:00
if (target_pid != -1)
event__synthesize_thread(target_pid, event__process);
else
event__synthesize_threads(event__process);
for (i = 0; i < nr_cpus; i++) {
group_fd = -1;
for (counter = 0; counter < nr_counters; counter++)
start_counter(i, counter);
}
/* Wait for a minimal set of events before starting the snapshot */
poll(event_array, nr_poll, 100);
mmap_read();
if (pthread_create(&thread, NULL, display_thread, NULL)) {
printf("Could not create display thread.\n");
exit(-1);
}
if (realtime_prio) {
struct sched_param param;
param.sched_priority = realtime_prio;
if (sched_setscheduler(0, SCHED_FIFO, &param)) {
printf("Could not set realtime priority.\n");
exit(-1);
}
}
while (1) {
int hits = samples;
mmap_read();
if (hits == samples)
ret = poll(event_array, nr_poll, 100);
}
return 0;
}
static const char * const top_usage[] = {
"perf top [<options>]",
NULL
};
static const struct option options[] = {
OPT_CALLBACK('e', "event", NULL, "event",
"event selector. use 'perf list' to list available events",
parse_events),
OPT_INTEGER('c', "count", &default_interval,
"event period to sample"),
OPT_INTEGER('p', "pid", &target_pid,
"profile events on existing pid"),
OPT_BOOLEAN('a', "all-cpus", &system_wide,
"system-wide collection from all CPUs"),
OPT_INTEGER('C', "CPU", &profile_cpu,
"CPU to profile on"),
perf: Enable more compiler warnings Related to a shadowed variable bug fix Valdis Kletnieks noticed that perf does not get built with -Wshadow, which could have helped us avoid the bug. So enable -Wshadow and also enable the following warnings on perf builds, in addition to the already enabled -Wall -Wextra -std=gnu99 warnings: -Wcast-align -Wformat=2 -Wshadow -Winit-self -Wpacked -Wredundant-decls -Wstack-protector -Wstrict-aliasing=3 -Wswitch-default -Wswitch-enum -Wno-system-headers -Wundef -Wvolatile-register-var -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wstrict-prototypes -Wdeclaration-after-statement And change/fix the perf code to build cleanly under GCC 4.3.2. The list of warnings enablement is rather arbitrary: it's based on my (quick) reading of the GCC manpages and trying them on perf. I categorized the warnings based on individually enabling them and looking whether they trigger something in the perf build. If i liked those warnings (i.e. if they trigger for something that arguably could be improved) i enabled the warning. If the warnings seemed to come from language laywers spamming the build with tons of nuisance warnings i generally kept them off. Most of the sign conversion related warnings were in this category. (A second patch enabling some of the sign warnings might be welcome - sign bugs can be nasty.) I also kept warnings that seem to make sense from their manpage description and which produced no actual warnings on our code base. These warnings might still be turned off if they end up being a nuisance. I also left out a few warnings that are not supported in older compilers. [ Note that these changes might break the build on older compilers i did not test, or on non-x86 architectures that produce different warnings, so more testing would be welcome. ] Reported-by: Valdis.Kletnieks@vt.edu Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-08-15 18:26:57 +08:00
OPT_STRING('k', "vmlinux", &vmlinux_name, "file", "vmlinux pathname"),
OPT_BOOLEAN('K', "hide_kernel_symbols", &hide_kernel_symbols,
"hide kernel symbols"),
OPT_INTEGER('m', "mmap-pages", &mmap_pages,
"number of mmap data pages"),
OPT_INTEGER('r', "realtime", &realtime_prio,
"collect data with this RT SCHED_FIFO priority"),
OPT_INTEGER('d', "delay", &delay_secs,
"number of seconds to delay between refreshes"),
OPT_BOOLEAN('D', "dump-symtab", &dump_symtab,
"dump the symbol table used for profiling"),
OPT_INTEGER('f', "count-filter", &count_filter,
"only display functions with more events than this"),
OPT_BOOLEAN('g', "group", &group,
"put the counters into a counter group"),
OPT_BOOLEAN('i', "inherit", &inherit,
"child tasks inherit counters"),
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
OPT_STRING('s', "sym-annotate", &sym_filter, "symbol name",
"symbol to annotate - requires -k option"),
OPT_BOOLEAN('z', "zero", &zero,
"zero history across updates"),
OPT_INTEGER('F', "freq", &freq,
"profile at this frequency"),
OPT_INTEGER('E', "entries", &print_entries,
"display this many functions"),
OPT_BOOLEAN('U', "hide_user_symbols", &hide_user_symbols,
"hide user symbols"),
OPT_BOOLEAN('v', "verbose", &verbose,
"be more verbose (show counter open errors, etc)"),
OPT_END()
};
int cmd_top(int argc, const char **argv, const char *prefix __used)
{
int counter;
symbol__init(sizeof(struct sym_entry));
page_size = sysconf(_SC_PAGE_SIZE);
argc = parse_options(argc, argv, options, top_usage, 0);
if (argc)
usage_with_options(top_usage, options);
/* CPU and PID are mutually exclusive */
if (target_pid != -1 && profile_cpu != -1) {
printf("WARNING: PID switch overriding CPU\n");
sleep(1);
profile_cpu = -1;
}
if (!nr_counters)
nr_counters = 1;
if (delay_secs < 1)
delay_secs = 1;
parse_symbols();
perf_counter tools: Fix/resurrect perf top annotation in a simple interactive form perf top used to have annotation support, but it has bitrotted and removed. This patch restores that: it allows the user to select any symbol in kernel space for source level annotation on the fly, switch between event counters and alter display variables. When symbol details are being displayed, stopping annotation reverts to normal. known keys: [d] select display delay. [e] select display entries (lines). [E] select annotation event counter. [f] select normal display count filter. [F] select annotation display count filter (percentage). [qQ] quit. [s] select annotation symbol and start annotation. [S] stop annotation, revert to normal display. [z] toggle event count zeroing. Sample: ------------------------------------------------------------------------------ PerfTop: 16719 irqs/sec kernel:78.7% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing cache-misses for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% /* adjust length to remove Ethernet CRC */ 0 0.0% if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) 0 0.0% length -= 4; 436 5.0% f039: 41 f6 84 24 5c 29 00 testb $0x1,0x295c(%r12) 0 0.0% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 0 0.0% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 811 9.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 0 0.0% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 7226 82.6% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Available events: 0 cache-misses 1 cache-references 2 instructions 3 cycles Enter details event counter: 2 ------------------------------------------------------------------------------ PerfTop: 15035 irqs/sec kernel:79.0% [cache-misses/cache-references/instructions/cycles], (all, 4 CPUs) ------------------------------------------------------------------------------ Showing instructions for e1000_clean_rx_irq Events Pcnt (>=3%) 0 0.0% int *work_done, int work_to_do) 0 0.0% { 175 0.9% eebf: 55 push %rbp 1898 9.8% eec0: 48 89 e5 mov %rsp,%rbp 0 0.0% 0 0.0% i = rx_ring->next_to_clean; 140 0.7% ef0a: 0f b7 41 1a movzwl 0x1a(%rcx),%eax 670 3.4% ef0e: 89 45 ac mov %eax,-0x54(%rbp) 0 0.0% { 0 0.0% memcpy(skb->data + offset, from, len); 91 0.5% f07b: 49 8b b6 e8 00 00 00 mov 0xe8(%r14),%rsi 1153 5.9% f082: 48 8b b8 e8 00 00 00 mov 0xe8(%rax),%rdi 42 0.2% f089: 8b 4d 84 mov -0x7c(%rbp),%ecx 14 0.1% f08c: 48 83 ef 02 sub $0x2,%rdi 0 0.0% f090: 48 83 ee 02 sub $0x2,%rsi 1618 8.3% f094: f3 a4 rep movsb %ds:(%rsi),%es:(%rdi) 0 0.0% 0 0.0% /* return some buffers to hardware, one at a time is too slow */ 0 0.0% if (cleaned_count >= E1000_RX_BUFFER_WRITE) { 867 4.5% f0e7: 83 7d b0 0f cmpl $0xf,-0x50(%rbp) 0 0.0% 0 0.0% while (rx_desc->status & E1000_RXD_STAT_DD) { 37 0.2% f114: 41 f6 47 0c 01 testb $0x1,0xc(%r15) 4047 20.8% f119: 0f 85 24 fe ff ff jne ef43 <e1000_clean_rx_irq+0x84> Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-07-23 02:36:03 +08:00
parse_source(sym_filter_entry);
/*
* User specified count overrides default frequency.
*/
if (default_interval)
freq = 0;
else if (freq) {
default_interval = freq;
} else {
fprintf(stderr, "frequency and count are zero, aborting\n");
exit(EXIT_FAILURE);
}
/*
* Fill in the ones not specifically initialized via -c:
*/
for (counter = 0; counter < nr_counters; counter++) {
if (attrs[counter].sample_period)
continue;
attrs[counter].sample_period = default_interval;
}
nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
assert(nr_cpus <= MAX_NR_CPUS);
assert(nr_cpus >= 0);
if (target_pid != -1 || profile_cpu != -1)
nr_cpus = 1;
get_term_dimensions(&winsize);
if (print_entries == 0) {
update_print_entries(&winsize);
signal(SIGWINCH, sig_winch_handler);
}
return __cmd_top();
}