I noticed while writing the first test in 'perf regtest' that to
just test the symbol handling routines one needs to create a
perf session, that is a layer centered on a perf.data file,
events, etc, so I untied these layers.
This reduces the complexity for the users as the number of
parameters to most of the symbols and session APIs now was
reduced while not adding more state to all the map instances by
only having data that is needed to split the kernel (kallsyms
and ELF symtab sections) maps and do vmlinux relocation on the
main kernel map.
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: <1265223128-11786-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Probably this wasn't noticed when testing this on my parisc
machine because I must have copied manually to its cache the
vmlinux file used in the x86_64 machine, now that I tried
looking on a x86-32 machine with a fresh cache, kernel symbols
weren't being resolved even with the right kallsyms copy on its
cache, duh.
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: <1264178102-4203-2-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
For now it just has operations to examine a given file, find its
build-id and add or remove it to/from the cache.
Useful, for instance, when adding binaries sent together with a
perf.data file, so that we can add them to the cache and have
the tools find it when resolving symbols.
It'll also manage the size of the cache like 'ccache' does.
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: <1264008525-29025-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Using this option 'perf buildid-list' will process all samples,
marking the DSOs that had some hits to list just them.
This in turn will be used by a new porcelain, 'perf archive',
that will be just a shell script to create a tarball from the
'perf buildid-list --with-hits' output and the files cached by
'perf record' in ~/.debug.
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: <1263519930-22803-4-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We were always looking at the running machine /proc/modules,
even when processing a perf.data file, which only makes sense
when we're doing 'perf record' and 'perf report' on the same
machine, and in close sucession, or if we don't use modules at
all, right Peter? ;-)
Now, at 'perf record' time we read /proc/modules, find the long
path for modules, and put them as PERF_MMAP events, just like we
did to encode the reloc reference symbol for vmlinux. Talking
about that now it is encoded in .pgoff, so that we can use
.{start,len} to store the address boundaries for the kernel so
that when we reconstruct the kmaps tree we can do lookups right
away, without having to fixup the end of the kernel maps like we
did in the past (and now only in perf record).
One more step in the 'perf archive' direction when we'll finally
be able to collect data in one machine and analyse in another.
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: <1263396139-4798-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
DSOs don't have this problem because the kernel emits a
PERF_MMAP for each new executable mapping it performs on
monitored threads.
To fix the kernel case we simulate the same behaviour, by having
'perf record' to synthesize a PERF_MMAP for the kernel, encoded
like this:
[root@doppio ~]# perf record -a -f sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.344 MB perf.data (~15038 samples) ]
[root@doppio ~]# perf report -D | head -10
0xd0 [0x40]: event: 1
.
. ... raw event: size 64 bytes
. 0000: 01 00 00 00 00 00 40 00 00 00 00 00 00 00 00 00 ......@........
. 0010: 00 00 00 81 ff ff ff ff 00 00 00 00 00 00 00 00 ...............
. 0020: 00 00 00 00 00 00 00 00 5b 6b 65 72 6e 65 6c 2e ........ [kernel
. 0030: 6b 61 6c 6c 73 79 6d 73 2e 5f 74 65 78 74 5d 00 kallsyms._text]
. 0xd0
[0x40]: PERF_RECORD_MMAP 0/0: [0xffffffff81000000((nil)) @ (nil)]: [kernel.kallsyms._text]
I.e. we identify such event as having:
.pid = 0
.filename = [kernel.kallsyms.REFNAME]
.start = REFNAME addr in /proc/kallsyms at 'perf record' time
and use now a hardcoded value of '.text' for REFNAME.
Then, later, in 'perf report', if there are any kernel hits and
thus we need to resolve kernel symbols, we search for REFNAME
and if its address changed, relocation happened and we thus must
change the kernel mapping routines to one that uses .pgoff as
the relocation to apply.
This way we use the same mechanism used for the other DSOs and
don't have to do a two pass in all the kernel symbols.
Reported-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
LKML-Reference: <1262717431-1246-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now a cache will be created in a ~/.debug debuginfo like
hierarchy, so that at the end of a 'perf record' session all the
binaries (with build-ids) involved get collected and indexed by
their build-ids, so that perf report can find them.
This is interesting when developing software where you want to
do a 'perf diff' with the previous build and opens avenues for
lots more interesting tools, like a 'perf diff --graph' that
takes more than two binaries into account.
Tunables for collecting just the symtabs can be added if one
doesn't want to have the full binary, but having the full binary
allows things like 'perf rerecord' or other tools that can
re-run the tests by having access to the exact binary in some
perf.data file, so it may well be interesting to keep the full
binary there.
Space consumption is minimised by trying to use hard links, a
'perf cache' tool to manage the space used, a la ccache is
required to purge older entries.
With this in place it will be possible also to introduce new
commands, 'perf archive' and 'perf restore' (or some more
suitable and future proof names) to create a cpio/tar file with
the perf data and the files in the cache that _had_ perf hits of
interest.
There are more aspects to polish, like finding the right vmlinux
file to cache, etc, but this is enough for a first step.
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: <1261957026-15580-10-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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 <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: <1260741029-4430-6-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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 <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1260564622-12392-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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 <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: <1260550239-5372-4-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
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: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Eduardo Habkost <ehabkost@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: <1260550239-5372-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now we have a very high level routine for simple tools to
process IP sample events:
int event__preprocess_sample(const event_t *self,
struct addr_location *al,
symbol_filter_t filter)
It receives the event itself and will insert new threads in the
global threads list and resolve the map and symbol, filling all
this info into the new addr_location struct, so that tools like
annotate and report can further process the event by creating
hist_entries in their specific way (with or without callgraphs,
etc).
It in turn uses the new next layer function:
void thread__find_addr_location(struct thread *self, u8 cpumode,
enum map_type type, u64 addr,
struct addr_location *al,
symbol_filter_t filter)
This one will, given a thread (userspace or the kernel kthread
one), will find the given type (MAP__FUNCTION now, MAP__VARIABLE
too in the near future) at the given cpumode, taking vdsos into
account (userspace hit, but kernel symbol) and will fill all
these details in the addr_location given.
Tools that need a more compact API for plain function
resolution, like 'kmem', can use this other one:
struct symbol *thread__find_function(struct thread *self, u64 addr,
symbol_filter_t filter)
So, to resolve a kernel symbol, that is all the 'kmem' tool
needs, its just a matter of calling:
sym = thread__find_function(kthread, addr, NULL);
The 'filter' parameter is needed because we do lazy
parsing/loading of ELF symtabs or /proc/kallsyms.
With this we remove more code duplication all around, which is
always good, huh? :-)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1259346563-12568-12-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Making the routines that were so far specific to the kernel maps
useful for all threads.
This is done by making the kernel maps be contained in a kernel
"thread".
This gets the kernel specific routines closer to the userspace
counterparts, which will help in reducing the boilerplate for
resolving a symbol, as will be demonstrated in the next patches.
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: <1259346563-12568-9-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
So that the kallsyms loading routines are the direct counterpart
of the vmlinux loading ones, i.e. dso__load_kallsyms is the
counterpart of dso__load_vmlinux.
In the process make them also use the symbols rb tree indexed by
map->type, paving the way for supporting other types of symtabs,
such as the next one to be supported: variables.
This also allowed removal of yet another global variable:
kernel_map__functions.
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: <1259346563-12568-7-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
By using an array of rb_roots in struct dso we can, from a
struct map instance to get the right symbol rb_tree more easily.
This way we can have just one symbol lookup method for struct
map instances, map__find_symbol, instead of one per symtab type
(functions, variables).
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: <1259346563-12568-6-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
And also express its configuration toggles via a struct.
Now all one has to do is to call symbol__init(NULL) if the
defaults are OK, or pass a struct symbol_conf pointer with the
desired configuration.
If a tool uses kernel_maps__find_symbol() to look at the kernel
and modules mappings for a symbol but didn't call symbol__init()
first, that will generate a one time warning too, alerting the
subcommand developer that symbol__init() must be called.
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: <1259071517-3242-2-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now that we can check the buildid to see if it really matches,
this can be done safely:
vmlinux
/boot/vmlinux
/boot/vmlinux-<uts.release>
/lib/modules/<uts.release>/build/vmlinux
/usr/lib/debug/lib/modules/%s/vmlinux
More can be added - if you know about distros that put the
vmlinux somewhere else please let us know.
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: <1259001550-8194-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The write_event() function in builtin-record.c writes out all
mmap()'d DSOs including non-ELF files like GNOME resource files
and such.
Therefore, check for ELF_K_ELF in filename__read_build_id()
before attempting to read the ELF header with gelf_getehdr().
Fixes the following error messages when running "perf kmem
record":
penberg@penberg-laptop:~/src/linux/tools/perf$ perf kmem record
^C[ perf record: Woken up 2 times to write data ]
[ perf record: Captured and wrote 0.753 MB perf.data (~32885 samples) ]
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
filename__read_build_id: cannot get elf header.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <1258885784-11709-1-git-send-email-penberg@cs.helsinki.fi>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We better call this routine after both the kernel and modules
are loaded, because as it was if there weren't modules it would not
be called, resulting in kernel_map->end remaining at zero, so no
map would be found and consequently the kernel symtab wouldn't
get loaded, i.e. no kernel symbols would be resolved.
Also this fixes another case, that is when we _have_ modules,
but the last map would have its ->end address not set before we
loaded its symbols, which would never happen because ->end was
not set.
Reported-by: Ingo Molnar <mingo@elte.hu>
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: <1258821086-11521-1-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
