- 15 Dec, 2009 2 commits
-
-
Arnaldo Carvalho de Melo authored
And it is also needed by 'perf diff'. Signed-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Frédéric Weisbecker <fweisbec@gmail.com> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <1260828571-3613-1-git-send-email-acme@infradead.org> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
Arnaldo Carvalho de Melo authored
Signed-off-by:
-
- 14 Dec, 2009 12 commits
-
-
Arnaldo Carvalho de Melo authored
All tools had copies, and perf diff would have to specify a sample_type_check method just for copying it. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
This is really a generic library routine, so declutter builtin-report.c a bit by moving it to the library. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
As we'll need to sort multiple times for multiple perf sessions, so that we can then do a diff. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
All hist entries are in only one of them, so use just one and a temporary rb_root while sorting/collapsing. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
There is still some more work to do to disentangle map creation from DSO loading, but this happens only for the kernel, and for the early adopters of perf diff, where this disentanglement matters most, we'll be testing different kernels, so no problem here. Further clarification: right now we create the kernel maps for the various modules and discontiguous kernel text maps when loading the DSO, we should do it as a two step process, first creating the maps, for multiple mappings with the same DSO store, then doing the dso load just once, for the first hit on one of the maps sharing this DSO backing store. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
So that we can process two perf.data files. We still need to add a O_MMAP mode for perf_session so that we can do all the mmap stuff in it. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
By having the cwd/cwdlen in the perf_session struct and full_paths in perf_event_ops. Now its just a matter of passing the ops. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
No need for all tools to register it and then immediately call perf_session__process_events. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Pass the event_ops to perf_session__process_events instead. Also move the event_ops definition to session.h, starting to move things around to their right place, trimming the many unneeded headers we have. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
They will need it to get the right threads list, etc. Signed-off-by:
-
David Miller authored
Here, tvec->tv_usec is "unsigned int" not "unsigned long". Since the type is different on every platform, it's probably best to just use long printf formats and cast. Signed-off-by:
David S. Miller <davem@davemloft.net> 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: <20091213.235622.53363059.davem@davemloft.net> Signed-off-by:
-
Hitoshi Mitake authored
This patch adds a new "all" pseudo subsystem and an "all" pseudo suite. These are for testing all subsystem and its all suite, or all suite of one subsystem. (This patch also contains a few trivial comment fixes for bench/* and output style fixes. I judged that there are no necessity to make them into individual patch.) Example of use: | % ./perf bench sched all # Test all suites of sched subsystem | # Running sched/messaging benchmark... | # 20 sender and receiver processes per group | # 10 groups == 400 processes run | | Total time: 0.414 [sec] | | # Running sched/pipe benchmark... | # Extecuted 1000000 pipe operations between two tasks | | Total time: 10.999 [sec] | | 10.999317 usecs/op | 90914 ops/sec | | % ./perf bench all # Test all suites of all subsystems | # Running sched/messaging benchmark... | # 20 sender and receiver processes per group | # 10 groups == 400 processes run | | Total time: 0.420 [sec] | | # Running sched/pipe benchmark... | # Extecuted 1000000 pipe operations between two tasks | | Total time: 11.741 [sec] | | 11.741346 usecs/op | 85169 ops/sec | | # Running mem/memcpy benchmark... | # Copying 1MB Bytes from 0x7ff33e920010 to 0x7ff3401ae010 ... | | 808.407437 MB/Sec Signed-off-by:
Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> LKML-Reference: <1260691319-4683-1-git-send-email-mitake@dcl.info.waseda.ac.jp> Signed-off-by:
-
- 12 Dec, 2009 8 commits
-
-
Arnaldo Carvalho de Melo authored
That does all the initialization boilerplate, opening the file, reading the header, checking if it is valid, etc. And that will as well have the threads list, kmap (now) global variable, etc, so that we can handle two (or more) perf.data files describing sessions to compare. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
It is always wired to dso__find_symbol. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Configurable via symbol_conf.sort_by_name, so that the cost of an extra rb_node on all 'struct symbol' instances is not paid by tools that only want to decode addresses. How to use it: symbol_conf.sort_by_name = true; symbol_init(&symbol_conf); struct map *map = map_groups__find_by_name(kmaps, MAP__VARIABLE, "[kernel.kallsyms]"); if (map == NULL) { pr_err("couldn't find map!\n"); kernel_maps__fprintf(stdout); } else { struct symbol *sym = map__find_symbol_by_name(map, sym_filter, NULL); if (sym == NULL) pr_err("couldn't find symbol %s!\n", sym_filter); else pr_info("symbol %s: %#Lx-%#Lx \n", sym_filter, sym->start, sym->end); } Looking over the vmlinux/kallsyms is common enough that I'll add a variable to the upcoming struct perf_session to avoid the need to use map_groups__find_by_name to get the main vmlinux/kallsyms map. The above example looks on the 'variable' symtab, but it is just like that for the functions one. Also the sort operation is done when we first use map__find_symbol_by_name, in a lazy way. Signed-off-by: -
Arnaldo Carvalho de Melo authored
Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Example: { u64 addr = strtoull(sym_filter, NULL, 16); struct map *map = map_groups__find(kmaps, MAP__VARIABLE, addr); if (map == NULL) pr_err("couldn't find map!\n"); else { struct symbol *sym = map__find_symbol(map, addr, NULL); if (sym == NULL) pr_err("couldn't find addr!\n"); else pr_info("addr %#Lx is in %s global var\n", addr, sym->name); } exit(0); } Added just after symbol__init() call in 'perf top', then: { u64 addr = strtoull(sym_filter, NULL, 16); struct map *map = map_groups__find(kmaps, MAP__VARIABLE, addr); if (map == NULL) pr_err("couldn't find map!\n"); else { struct symbol *sym = map__find_symbol(map, addr, NULL); if (sym == NULL) pr_err("couldn't find addr!\n"); else pr_info("addr %#Lx is in %s global var\n", addr, sym->name); } exit(0); } [root@doppio linux-2.6-tip]# grep ' [dD] ' /proc/kallsyms | grep ' sched' ffffffff817827d8 d sched_nr_latency ffffffff81782ce0 d sched_domains_mutex ffffffff8178c070 d schedstr.22423 ffffffff817909a0 d sched_register_mutex ffffffff81823490 d sched_feat_names ffffffff81823558 d scheduler_running ffffffff818235b8 d sched_clock_running ffffffff818235bc D sched_clock_stable ffffffff81824f00 d sched_switch_trace [root@doppio linux-2.6-tip]# perf top -s 0xffffffff817827d9 addr 0xffffffff817827d9 is in sched_nr_latency global var [root@doppio linux-2.6-tip]# perf top -s ffffffff81782ce0 addr 0xffffffff81782ce0 is in sched_domains_mutex global var [root@doppio linux-2.6-tip]# [root@doppio linux-2.6-tip]# perf top -s ffffffff81782ce0 --vmlinux OFF The file OFF cannot be used, trying to use /proc/kallsyms...addr 0xffffffff81782ce0 is in sched_domains_mutex global var [root@doppio linux-2.6-tip]# perf top -s ffffffff818235bc --vmlinux OFF The file OFF cannot be used, trying to use /proc/kallsyms...addr 0xffffffff818235bc is in sched_clock_stable global var [root@doppio linux-2.6-tip]# So it works with both /proc/kallsyms and with ELF symtabs, either the one on the vmlinux explicitely passed via --vmlinux or in one in the vmlinux_path that matches the buildid for the running kernel or the one found in the buildid header section in a perf.data file. Signed-off-by: -
Arnaldo Carvalho de Melo authored
For selecting the right types of symbols in ELF symtabs. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
For selecting the right types of symbols in /proc/kallsyms, will be followed by elf_symbol_type__is_a, for the same purpose on ELF symtabs. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Using a struct thread instance just to hold the kernel space maps (vmlinux + modules) is overkill and confuses people trying to understand the perf symbols abstractions. The kernel maps are really present in all threads, i.e. the kernel is a library, not a separate thread. So introduce the 'map_groups' abstraction and use it for the kernel maps, now in the kmaps global variable. It, in turn, will move, together with the threads list to the perf_file abstraction, so that we can support multiple perf_file instances, needed by perf diff. Brainstormed-with: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by:
-
- 11 Dec, 2009 5 commits
-
-
Jamie Iles authored
Add definitions of rmb() and cpu_relax() and include the ARM unistd.h header. The __kuser_memory_barrier helper in the helper page is used to provide the correct memory barrier depending on the CPU type. [ The rmb() will work on v6 and v7, segfault on v5. Dynamic detection to add v5 support will be added later. ] Signed-off-by:
Jamie Iles <jamie.iles@picochip.com> Cc: Russell King <linux@arm.linux.org.uk> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mikael Pettersson <mikpe@it.uu.se> LKML-Reference: <1260534009-5394-1-git-send-email-jamie.iles@picochip.com> Signed-off-by:
-
Frederic Weisbecker authored
While converting modify_user_hw_breakpoint() return value, we forgot to handle the off-case. It's not returning a pointer anymore. This solves the build warning reported by Stephen Rothwell against linux-next. Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> LKML-Reference: <1260529122-6260-1-git-send-regression-fweisbec@gmail.com> Signed-off-by:
-
Jamie Iles authored
For embedded platforms, we want to be able to build the perf tools on a build machine to run on a different arch. This patch allows $CROSS_COMPILE to set the cross compiler. Additionally, if NO_LIBPERL is set, then don't use perl include paths as they will be for the host arch. Signed-off-by:
-
Li Zefan authored
For slab, if CONFIG_KMEMTRACE and CONFIG_DEBUG_SLAB are not set, __do_kmalloc() will not track callers: # ./perf record -f -a -R -e kmem:kmalloc ^C # ./perf trace ... perf-2204 [000] 147.376774: kmalloc: call_site=c0529d2d ... perf-2204 [000] 147.400997: kmalloc: call_site=c0529d2d ... Xorg-1461 [001] 147.405413: kmalloc: call_site=0 ... Xorg-1461 [001] 147.405609: kmalloc: call_site=0 ... konsole-1776 [001] 147.405786: kmalloc: call_site=0 ... Signed-off-by:Li Zefan <lizf@cn.fujitsu.com> Reviewed-by:
Pekka Enberg <penberg@cs.helsinki.fi> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: linux-mm@kvack.org <linux-mm@kvack.org> Cc: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro> LKML-Reference: <4B21F8AE.6020804@cn.fujitsu.com> Signed-off-by:
-
Li Zefan authored
Define kmem_trace_alloc_{,node}_notrace() if CONFIG_TRACING is enabled, otherwise perf-kmem will show wrong stats ifndef CONFIG_KMEM_TRACE, because a kmalloc() memory allocation may be traced by both trace_kmalloc() and trace_kmem_cache_alloc(). Signed-off-by:
-
- 10 Dec, 2009 11 commits
-
-
Cyrill Gorcunov authored
Ralf Hildebrandt reported this boot warning: | Running a vanilla 2.6.32 as Xen DomU, I'm getting: | | [ 0.000999] CPU: Physical Processor ID: 0 | [ 0.000999] CPU: Processor Core ID: 1 | [ 0.000999] Performance Events: AMD PMU driver. | [ 0.000999] ------------[ cut here ]------------ | [ 0.000999] WARNING: at arch/x86/kernel/apic/apic.c:249 native_apic_write_dummy So we need to check if APIC functionality is available, and not just in the P6 driver but elsewhere as well. Reported-by:
Ralf Hildebrandt <Ralf.Hildebrandt@charite.de> Signed-off-by:
Cyrill Gorcunov <gorcunov@openvz.org> 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: <20091210165634.GF5086@lenovo> Signed-off-by:
-
Xiao Guangrong authored
Signed-off-by:
Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <4B20BAA6.7010609@cn.fujitsu.com> Signed-off-by:
-
Ingo Molnar authored
Fix: builtin-kmem.c: In function 'parse_caller_opt': builtin-kmem.c:690: error: unused parameter 'arg' builtin-kmem.c: In function 'parse_alloc_opt': builtin-kmem.c:697: error: unused parameter 'arg' Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Pekka Enberg <penberg@cs.helsinki.fi> LKML-Reference: <4B20A195.8030106@cn.fujitsu.com> Signed-off-by: -
Arnaldo Carvalho de Melo authored
As off_t is a long, so breaking things on 32-bit land. Now buildids work on 32-bit land. [root@ana ~]# uname -a Linux ana.ghostprotocols.net 2.6.31.6-162.fc12.i686 #1 SMP Fri Dec 4 01:09:09 EST 2009 i686 i686 i386 GNU/Linux [root@ana ~]# perf buildid-list | tail -5 136ee6792ba2ae57870ecd87369f4ae3194d5b27 /lib/libreadline.so.6.0 d202dcb1ad48d140065783657d37ae3f2d9ab83f /usr/bin/gdb 0a56c0c00dcc2e9e581ae9997f31957c9c4671df /usr/lib/libdwarf.so.0.0 5f9e6ac95241cbb3227608e0ff2a2e0cbbe72439 /home/acme/bin/perf 925d19eccc2ddb1c9d74dd178a011426f1b124a8 /bin/sleep [root@ana ~]# Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Signed-off-by:
-
Li Zefan authored
Before: $ ./perf kmem ... -l, --line <num> show n lines --raw-ip show raw ip instead of symbol After: $ ./perf kmem ... -l, --line <num> show n lines --raw-ip show raw ip instead of symbol Signed-off-by: -
Li Zefan authored
As Ingo suggested, make "perf kmem" show help information. "perf kmem stat [--caller] [--alloc] .." will show memory statistics. Signed-off-by:
-
Mike Frysinger authored
The core ftrace code (trace_clock_local) calls sched_clock() directly, so we don't want to recurisvely trigger the ftrace code. Rather than update every sched_clock() definition, tag the prototype for everyone as notrace. Signed-off-by:
Mike Frysinger <vapier@gentoo.org> Cc: Peter Zijlstra <peterz@infradead.org> LKML-Reference: <1260407223-10900-1-git-send-email-vapier@gentoo.org> Signed-off-by:
-
Frederic Weisbecker authored
When we have a maximum latency reported for a task, we need a convenient way to find the matching location to the raw traces or to perf sched map that shows where the task has been eventually scheduled in. This gives a pointer to retrieve the events that occured during this max latency. Signed-off-by:
-
Julia Lawall authored
Memset should be given the size of the structure, not the size of the pointer. The semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ type T; T *x; expression E; @@ memset(x, E, sizeof( + * x)) // </smpl> Signed-off-by:
Julia Lawall <julia@diku.dk> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <Pine.LNX.4.64.0912092026000.1870@ask.diku.dk> Signed-off-by:
-
Xiao Guangrong authored
fix: [<c0477471>] ? printk+0x1d/0x24 [<c01c98f9>] ? perf_prepare_sample+0x269/0x280 [<c0149231>] warn_slowpath_common+0x71/0xd0 [<c01c98f9>] ? perf_prepare_sample+0x269/0x280 [<c01492aa>] warn_slowpath_null+0x1a/0x20 [<c01c98f9>] perf_prepare_sample+0x269/0x280 [<c016e9f3>] ? cpu_clock+0x53/0x90 [<c01cc368>] __perf_event_overflow+0x2a8/0x300 [<c01ccc3b>] perf_event_overflow+0x1b/0x30 [<c01ccccf>] perf_swevent_hrtimer+0x7f/0x120 This is because 'data.raw' variable not initialize. Signed-off-by:
Peter Zijlstra <peterz@infradead.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <4B208E93.1010801@cn.fujitsu.com> Signed-off-by:
-
- 09 Dec, 2009 2 commits
-
-
Xiao Guangrong authored
In current code, task's execute time is got by reading '/proc/<pid>/sched' file, it's wrong if the task is created by pthread_create(), because every thread task has same pid. This way also has two demerits: 1: 'perf sched replay' can't work if the kernel is not compiled with the 'CONFIG_SCHED_DEBUG' option 2: perf tool should depend on proc file system So, this patch uses PERF_COUNT_SW_TASK_CLOCK to get task's execution time instead of reading /proc file. Changelog v2 -> v3: use PERF_COUNT_SW_TASK_CLOCK instead of rusage() as Ingo's suggestion Reported-by:Torok Edwin <edwintorok@gmail.com> Signed-off-by:
-
Frederic Weisbecker authored
When we define the common event fields in kprobe, we invert the error handling and return immediately in case of success. Then we omit to define specific kprobes fields (ip and nargs), and specific kretprobes fields (func, ret_ip, nargs). And we only define them when we fail to create common fields. The most visible consequence is that we can't create filter for k(ret)probes specific fields. This patch re-invert the success/error handling to fix it. Reported-by:
Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by:
Masami Hiramatsu <mhiramat@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Li Zefan <lizf@cn.fujitsu.com> LKML-Reference: <1260263815-5167-1-git-send-regression-fweisbec@gmail.com> Signed-off-by:
-
