If we are running the new perf on an old kernel without support for
sample_id_all, we should fall back to the old unordered processing of
events. If we didn't than we would *always* process events without
timestamps out of order, whether or not we hit a reordering race. In
other words, instead of there being a chance of not attributing samples
correctly, we would guarantee that samples would not be attributed.
While processing all events without timestamps before events with
timestamps may seem like an intuitive solution, it falls down as
PERF_RECORD_EXIT events would also be processed before any samples.
Even with a workaround for that case, samples before/after an exec would
not be attributed correctly.
This patch allows commands to indicate whether they need to fall back to
unordered processing, so that commands that do not care about timestamps
on every event will not be affected. If we do fallback, this will print
out a warning if report -D was invoked.
This patch adds the test in perf_session__new so that we only need to
test once per session. Commands that do not use an event_ops (such as
record and top) can simply pass NULL in it's place.
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <1291951882-sup-6069@au1.ibm.com>
Signed-off-by: Ian Munsie <imunsie@au1.ibm.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Since we check at the beginning of the callers, no need to ask if
dump_trace is set multiple times.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The dump code used by perf report -D is scattered all over the place.
Move it to separate functions.
Acked-by: Ian Munsie <imunsie@au1.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20101207124550.625434869@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If the event has no timestamp assigned then the parse code sets it to
~0ULL which causes the ordering code to enqueue it at the end.
Process it right away.
Reported-by: Ian Munsie <imunsie@au1.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20101207124550.528788441@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
event__name[] is missing an entry for PERF_RECORD_FINISHED_ROUND, but we
happily access the array from the dump code.
Make event__name[] static and provide an accessor function, fix up all
callers and add the missing string.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
LKML-Reference: <20101207124550.432593943@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Now that we have timestamps on FORK, EXIT, COMM, MMAP events we can
sort everything in time order. This fixes the following observed
problem:
mmap(file1) -> pagefault() -> munmap(file1)
mmap(file2) -> pagefault() -> munmap(file2)
Resulted in decoding both pagefaults in file2 because the file1 map
was already replaced by the file2 map when the map address was
identical.
With all events sorted we decode both pagefaults correctly.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <alpine.LFD.2.00.1012051220450.2653@localhost6.localdomain6>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
At perf_session__process_event, so that we reduce the number of lines in eache
tool sample processing routine that now receives a sample_data pointer already
parsed.
This will also be useful in the next patch, where we'll allow sample the
identity fields in MMAP, FORK, EXIT, etc, when it will be possible to see (cpu,
timestamp) just after before every event.
Also validate callchains in perf_session__process_event, i.e. as early as
possible, and keep a counter of the number of events discarded due to invalid
callchains, warning the user about it if it happens.
There is an assumption that was kept that all events have the same sample_type,
that will be dealt with in the future, when this preexisting limitation will be
removed.
Tested-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ian Munsie <imunsie@au1.ibm.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <1291318772-30880-4-git-send-email-acme@infradead.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The ordered sample code allocates singular reference objects struct
sample_queue which have 48byte size on 64bit and 20 bytes on 32bit. That's
silly. Allocate ~64k sized chunks and hand them out.
Performance gain: ~ 15%
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.398713983@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When the sample queue is flushed we free the sample reference objects. Though
we need to malloc new objects when we process further. Stop the malloc/free
orgy and cache the already allocated object for resuage. Only allocate when
the cache is empty.
Performance gain: ~ 10%
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.338488630@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Profiling perf with perf revealed that a large part of the processing time is
spent in malloc/memcpy/free in the sample ordering code. That code copies the
data from the mmap into malloc'ed memory. That's silly. We can keep the mmap
and just store the pointer in the queuing data structure. For 64 bit this is
not a problem as we map the whole file anyway. On 32bit we keep 8 maps around
and unmap the oldest before mmaping the next chunk of the file.
Performance gain: 2.95s -> 1.23s (Faktor 2.4)
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.278787719@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
On 64bit we can map the whole file in one go, on 32bit we can at least map
32MB and not map/unmap tiny chunks of the file.
Base the progress bar on 1/16 of the data size.
Preparatory patch to get rid of the malloc/memcpy/free of trace data.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.213687773@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
No need to check twice.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.152886642@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The progress bar is changed when the file offset changes. This happens only
when the next mmap is done. No need to call ui_progress_update() for every
event.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.094836523@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Replace the pseudo C++ self argument with session and give the mmap related
variables a sensible name. shift is a complete misnomer - it took me several
rounds of cursing to figure out that it's not a shift value.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163820.029687218@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
There is no reason to use a struct sample_event pointer in struct sample_queue
and type cast it when flushing the queue.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163819.969462809@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The homebrewn sort algorithm fails to sort in time order. One of the problem
spots is that it fails to deal with equal timestamps correctly.
My first gut reaction was to replace the fancy list with an rbtree, but the
performance is 3 times worse.
Rewrite it so it works.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20101130163819.908482530@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Fix it by explaining what can be happening and giving the number of processed
and lost events.
Also holler if unknown events were found, that can be due to processing a
perf.data file collected using a newer tool where newer events got added on
reporting using an older perf tool, that or a bug, so ask for a report to be
made.
Works on both --tui and --stdio.
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
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>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tool developers have to fill in a 'perf_event_ops' method table to
specify how to handle each event, so far the ones that were not
explicitely especified would get a stub that would just discard the
event.
Change that so that tool developers can get the lost event details and
the total number of such events at the end of 'perf report -D' output.
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
CC: Thomas Gleixner <tglx@linutronix.de>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If we receive two PERF_RECORD_EXIT for the same thread, we can end up
reusing session->last_match and trying to remove the thread twice from
the rb_tree, causing a segfault, so invalidade last_match in
perf_session__remove_thread.
Receiving two PERF_RECORD_EXIT for the same thread is a bug, but its a
harmless one if we make the tool more robust, like this patch does.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Which is at perf_session__destroy_kernel_maps, counterpart to the
perf_session__create_kernel_maps where the kmap structure is located, just
after the vmlinux_maps.
Make it also check if the kernel maps were actually created, which may not
be the case if, for instance, perf_session__new can't complete due to
permission problems in, for instance, a 'perf report' case, when a
segfault will take place, that is how this was noticed.
The problem was introduced in d65a458, thus post .35.
This also adds code to release guest machines as them are also created
in perf_session__create_kernel_maps, so should be deleted on this newly
introduced counterpart, perf_session__destroy_kernel_maps.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
So that we reduce the noise when looking for leaks using tools such as
valgrind.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tidy-up patch to remove some code and struct perf_session data members
which are no longer needed due to the previous patch: "perf tools: Don't
abbreviate file paths relative to the cwd".
LKML-Reference: <new-submission>
Signed-off-by: Dave Martin <dave.martin@linaro.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
If we cannot open our data file, print strerror(errno) for a more
comprehensible error message; and only suggest 'perf record' on ENOENT.
In particular, this fixes the nonsensical advice when:
% sudo perf record sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.009 MB perf.data (~381 samples) ]
% perf trace
failed to open file: perf.data (try 'perf record' first)
%
Cc: Ingo Molnar <mingo@elte.hu>
LPU-Reference: <20100612033615.GA24731@hexapodia.org>
Signed-off-by: Andy Isaacson <adi@hexapodia.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Move them to a session->dead_threads list just like we do with maps that
are replaced, because we may have hist_entries pointing to them.
This fixes a bug when inserting maps for a new thread that reused the
TID, mixing maps for two different threads, causing an endless loop.
The code for insering maps should be made more robust but for .35 this
is the minimalistic patch.
Reported-by: Ingo Molnar <mingo@elte.hu>
Cc: David S. Miller <davem@davemloft.net>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The changes made to support host and guest machines in a session, that
started when the 'perf kvm' tool was introduced ended up introducing a
bug where the host_machine was not having its DSOs traversed for
build-id processing.
Fix it by moving some methods to the right classes and considering the
host_machine when processing build-ids.
Reported-by: Tom Zanussi <tzanussi@gmail.com>
Reported-by: Stephane Eranian <eranian@google.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Without the bloated cplus_demangle from binutils, i.e building with:
$ make NO_DEMANGLE=1 O=~acme/git/build/perf -j3 -C tools/perf/ install
Before:
text data bss dec hex filename
471851 29280 4025056 4526187 45106b /home/acme/bin/perf
After:
[acme@doppio linux-2.6-tip]$ size ~/bin/perf
text data bss dec hex filename
446886 29232 4008576 4484694 446e56 /home/acme/bin/perf
So its a 5.3% size reduction in code, but the interesting part is in the git
diff --stat output:
19 files changed, 20 insertions(+), 1909 deletions(-)
If we ever need some of the things we got from git but weren't using, we just
have to go to the git repo and get fresh, uptodate source code bits.
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The events_stats.total field is too generic, rename it to .total_period,
and also add a comment explaining that it is the sum of all the .period
fields in samples, that is needed because we use auto-freq to avoid
sampling artifacts.
Ditto for events_stats.lost, that is the sum of all lost_event.lost
fields, i.e. the number of events the kernel dropped.
Looking at the users, builtin-sched.c can make use of these fields and
stop doing it again.
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This is one more thing that started global but are more useful per hist
or per session.
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In cbbc79a we introduced support for multiple events by introducing a
new "event_stat_id" struct and then made several perf_session methods
receive a point to it instead of a pointer to perf_session, and kept the
event_stats and hists rb_tree in perf_session.
While working on the new newt based browser, I realised that it would be
better to introduce a new class, "hists" (short for "histograms"),
renaming the "event_stat_id" struct and the perf_session methods that
were really "hists" methods, as they manipulate only struct hists
members, not touching anything in the other perf_session members.
Other optimizations, such as calculating the maximum lenght of a symbol
name present in an hists instance will be possible as we add them,
avoiding a re-traversal just for finding that information.
The rationale for the name "hists" to replace "event_stat_id" is that we
may have multiple sets of hists for the same event_stat id, as, for
instance, the 'perf diff' tool has, so event stat id is not what
characterizes what this struct and the functions that manipulate it do.
Cc: Eric B Munson <ebmunson@us.ibm.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Using machines__create_kernel_maps(..., HOST_KERNEL_ID) it would create
another machine instance for the host machine, and since 1f626bc we have
it out of the machines rb_tree.
Fix it by using machine__create_kernel_maps(&self->host_machine)
directly.
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We have just one host on a given session, and that is the most common
setup right now, so embed a ->host_machine struct machine instance
directly in the perf_session class, check if we're looking for it before
going to the rb_tree.
This also fixes a problem found when we try to process old perf.data
files where we didn't have MMAP events for the kernel and modules and
thus don't create the kernel maps, do it in event__preprocess_sample if
it wasn't already.
Reported-by: Ingo Molnar <mingo@elte.hu>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Zhang, Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Some events, such as the PERF_RECORD_FINISHED_ROUND event consist of
only an event header and no data. In this case, a 0-length payload
will be read, and the 0 return value will be wrongly interpreted as an
'unexpected end of event stream'.
This patch allows for proper handling of data-less events by skipping
0-length reads.
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
LKML-Reference: <1273038527.6383.51.camel@tropicana>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
The current events reordering algorithm is based on a heuristic that
gets broken once we deal with a very fast flow of events.
Indeed the time period based flushing is not suitable anymore
in the following case, assuming we have a flush period of two
seconds.
CPU 0 | CPU 1
|
cnt1 timestamps | cnt1 timestamps
|
0 | 0
1 | 1
2 | 2
3 | 3
[...] | [...]
4 seconds later
If we spend too much time to read the buffers (case of a lot of
events to record in each buffers or when we have a lot of CPU buffers
to read), in the next pass the CPU 0 buffer could contain a slice
of several seconds of events. We'll read them all and notice we've
reached the period to flush. In the above example we flush the first
half of the CPU 0 buffer, then we read the CPU 1 buffer where we
have events that were on the flush slice and then the reordering
fails.
It's simple to reproduce with:
perf lock record perf bench sched messaging
To solve this, we use a new solution that doesn't rely on an
heuristical time slice period anymore but on a deterministic basis
based on how perf record does its job.
perf record saves the buffers through passes. A pass is a tour
on every buffers from every CPUs. This is made in order: for
each CPU we read the buffers of every counters. So the more
buffers we visit, the later will be the timstamps of their events.
When perf record finishes a pass it records a
PERF_RECORD_FINISHED_ROUND pseudo event.
We record the max timestamp t found in the pass n. Assuming these
timestamps are monotonic across cpus, we know that if a buffer
still has events with timestamps below t, they will be all available
and then read in the pass n + 1.
Hence when we start to read the pass n + 2, we can safely flush every
events with timestamps below t.
============ PASS n =================
CPU 0 | CPU 1
|
cnt1 timestamps | cnt2 timestamps
1 | 2
2 | 3
- | 4 <--- max recorded
============ PASS n + 1 ==============
CPU 0 | CPU 1
|
cnt1 timestamps | cnt2 timestamps
3 | 5
4 | 6
5 | 7 <---- max recorded
Flush every events below timestamp 4
============ PASS n + 2 ==============
CPU 0 | CPU 1
|
cnt1 timestamps | cnt2 timestamps
6 | 8
7 | 9
- | 10
Flush every events below timestamp 7
etc...
It also works on perf.data versions that don't have
PERF_RECORD_FINISHED_ROUND pseudo events. The difference is that
the events will be only flushed in the end of the perf.data
processing. It will then consume more memory and scale less with
large perf.data files.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Currently, perf 'live mode' writes build-ids at the end of the
session, which isn't actually useful for processing live mode events.
What would be better would be to have the build-ids sent before any of
the samples that reference them, which can be done by processing the
event stream and retrieving the build-ids on the first hit. Doing
that in perf-record itself, however, is off-limits.
This patch introduces perf-inject, which does the same job while
leaving perf-record untouched. Normal mode perf still records the
build-ids at the end of the session as it should, but for live mode,
perf-inject can be injected in between the record and report steps
e.g.:
perf record -o - ./hackbench 10 | perf inject -v -b | perf report -v -i -
perf-inject reads a perf-record event stream and repipes it to stdout.
At any point the processing code can inject other events into the
event stream - in this case build-ids (-b option) are read and
injected as needed into the event stream.
Build-ids are just the first user of perf-inject - potentially
anything that needs userspace processing to augment the trace stream
with additional information could make use of this facility.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
LKML-Reference: <1272696080-16435-3-git-send-email-tzanussi@gmail.com>
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Those functions operated on members now grouped in 'struct machine', so
move those methods to this new class.
The changes made to 'perf probe' shows that using this abstraction
inserting probes on guests almost got supported for free.
Cc: Avi Kivity <avi@redhat.com>
Cc: Frédéric Weisbecker <fweisbec@gmail.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Zhang, Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
struct kernel_info and kerninfo__ are too vague, what they really
describe are machines, virtual ones or hosts.
There are more changes to introduce helpers to shorten function calls
and to make more clear what is really being done, but I left that for
subsequent patches.
Cc: Avi Kivity <avi@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>
Cc: Zhang, Yanmin <yanmin_zhang@linux.intel.com>
LKML-Reference: <new-submission>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The sample events recorded by perf record are not time ordered
because we have one buffer per cpu for each event (even demultiplexed
per task/per cpu for task bound events). But when we read trace events
we want them to be ordered by time because many state machines are
involved.
There are currently two ways perf tools deal with that:
- use -M to multiplex every buffers (perf sched, perf kmem)
But this creates a lot of contention in SMP machines on
record time.
- use a post-processing time reordering (perf timechart, perf lock)
The reordering used by timechart is simple but doesn't scale well
with huge flow of events, in terms of performance and memory use
(unusable with perf lock for example).
Perf lock has its own samples reordering that flushes its memory
use in a regular basis and that uses a sorting based on the
previous event queued (a new event to be queued is close to the
previous one most of the time).
This patch proposes to export perf lock's samples reordering facility
to the session layer that reads the events. So if a tool wants to
get ordered sample events, it needs to set its
struct perf_event_ops::ordered_samples to true and that's it.
This prepares tracing based perf tools to get rid of the need to
use buffers multiplexing (-M) or to implement their own
reordering.
Also lower the flush period to 2 as it's sufficient already.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
Bypasses the build_id perf header code and replaces it with a
synthesized event and processing function that accomplishes the
same thing, used when reading/writing perf data to/from a pipe.
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: fweisbec@gmail.com
Cc: rostedt@goodmis.org
Cc: k-keiichi@bx.jp.nec.com
Cc: acme@ghostprotocols.net
LKML-Reference: <1270184365-8281-9-git-send-email-tzanussi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Bypasses the tracing_data perf header code and replaces it with
a synthesized event and processing function that accomplishes
the same thing, used when reading/writing perf data to/from a
pipe.
The tracing data is pretty large, and this patch doesn't attempt
to break it down into component events. The tracing_data event
itself doesn't actually contain the tracing data, rather it
arranges for the event processing code to skip over it after
it's read, using the skip return value added to the event
processing loop in a previous patch.
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: fweisbec@gmail.com
Cc: rostedt@goodmis.org
Cc: k-keiichi@bx.jp.nec.com
Cc: acme@ghostprotocols.net
LKML-Reference: <1270184365-8281-8-git-send-email-tzanussi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Bypasses the event type perf header code and replaces it with a
synthesized event and processing function that accomplishes the
same thing, used when reading/writing perf data to/from a pipe.
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: fweisbec@gmail.com
Cc: rostedt@goodmis.org
Cc: k-keiichi@bx.jp.nec.com
Cc: acme@ghostprotocols.net
LKML-Reference: <1270184365-8281-7-git-send-email-tzanussi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Bypasses the attr perf header code and replaces it with a
synthesized event and processing function that accomplishes the
same thing, used when reading/writing perf data to/from a pipe.
Making the attrs into events allows them to be streamed over a
pipe along with the rest of the header data (in later patches).
It also paves the way to allowing events to be added and removed
from perf sessions dynamically.
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Cc: fweisbec@gmail.com
Cc: rostedt@goodmis.org
Cc: k-keiichi@bx.jp.nec.com
Cc: acme@ghostprotocols.net
LKML-Reference: <1270184365-8281-6-git-send-email-tzanussi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>