Brendan Gregg [Mon, 25 Jul 2016 23:13:35 +0000 (16:13 -0700)]
fix uprobe examples to read correct argument
Brendan Gregg [Mon, 25 Jul 2016 22:02:32 +0000 (15:02 -0700)]
tutorials: end-user, and python developer
Brenden Blanco [Mon, 25 Jul 2016 21:18:54 +0000 (14:18 -0700)]
Merge pull request #627 from brendangregg/master
merge most .c and .py examples
Brendan Gregg [Sun, 24 Jul 2016 23:19:20 +0000 (16:19 -0700)]
return 0 on hello_worlds
Brendan Gregg [Sun, 24 Jul 2016 20:42:47 +0000 (13:42 -0700)]
add nodejs_http_server.py to README list
Brendan Gregg [Sun, 24 Jul 2016 20:40:25 +0000 (13:40 -0700)]
merge task_switch example
Brendan Gregg [Sun, 24 Jul 2016 20:37:20 +0000 (13:37 -0700)]
merge disksnoop example
Brendan Gregg [Sun, 24 Jul 2016 20:34:40 +0000 (13:34 -0700)]
merge bitehist example
Brendan Gregg [Sun, 24 Jul 2016 17:01:46 +0000 (10:01 -0700)]
Merge pull request #615 from chantra/cachetop
[cachetop] top-like cachestat
chantra [Sat, 23 Jul 2016 13:33:11 +0000 (15:33 +0200)]
[cachetop] fix and doc
* pass -fno-color-diagnostics to clang
* remove unicode import (#623)
* add time to cachetop output
* add keybindings to cachetop.8
* add cachetop links to README.md
chantra [Mon, 18 Jul 2016 22:17:45 +0000 (00:17 +0200)]
[cachetop] add example and man page.
make interval a positional parameter.
Emmanuel Bretelle [Thu, 14 Jul 2016 20:04:57 +0000 (13:04 -0700)]
[cachetop] top-like cachestat
Alike cachestat.py but providing cache stats at the process level.
Brendan Gregg [Sat, 23 Jul 2016 00:11:51 +0000 (17:11 -0700)]
USDT Python API and example (#624)
* Python USDT API
Code from @vmg
* Basic USDT example
* retire procstat.py
* improve/fix USDT exceptions
Brendan Gregg [Fri, 22 Jul 2016 01:13:24 +0000 (18:13 -0700)]
Add profile: a CPU profiler (#620)
* Add profile: a CPU profiler
* move Perf to common class
Brendan Gregg [Tue, 19 Jul 2016 03:19:16 +0000 (20:19 -0700)]
Merge pull request #618 from palmtenor/master
Use errno symbols in offcputime.py
Teng Qin [Mon, 18 Jul 2016 20:21:10 +0000 (13:21 -0700)]
Use errno symbol instead of hard-coded numbers in offcputime.py
Brenden Blanco [Mon, 18 Jul 2016 17:20:43 +0000 (10:20 -0700)]
Merge pull request #614 from markdrayton/fix-auto-kprobe
Fix probe detaching and auto-kprobes
Mark Drayton [Fri, 15 Jul 2016 22:55:22 +0000 (23:55 +0100)]
Move open_{kprobes,uprobes,tracepoints} into BPF object
* for #605, this diff moves probe storage from the BPF module to the BPF object,
letting each instantiation clean up its own probes. A module-level counter
for all open probes is provided for the quota check. It also adds a
`cleanup()` function to force cleanup before the `atexit` handler runs.
* for #614, it removes the `len(open_kprobes) == 0` check that prevented more
than one autoload probe working. It fixes the tests that this change breaks by
calling the `cleanup()` function added.
Mark Drayton [Fri, 15 Jul 2016 20:19:07 +0000 (21:19 +0100)]
Store kprobes with string keys, fix num_open_kprobes
Prior to this diff we used inconsistent types for keys in `open_kprobes`. The
results from the regex match (`attach_kprobe(event_re=..)`) and the automatic
`kprobe__` features were passed through `str.decode()`, yielding unicode keys,
but specific matches (i.e. from `attach_kprobe(event=..)`) were stored with
string keys passed down from the caller. Only probes under string keys were
released in `cleanup_kprobes`, leaving attached probes on program exit.
This diff makes all the keys regular strings. I erred on the side of using
regular strings over `str.decode()`ing them because a) this data isn't passed
outside of Python, b) it's more Python 3 compatible (there is no `.decode()` on
a regular string object in Python 3 so such a change would ultimately need
removing again).
I also cleaned up a few other things:
* removed the call to `awk` for getting probable functions
* removed the `isinstance` checks when cleaning uprobes/tracepoints -- we
should only have string keys in these dicts
* made `num_open_kprobes` skip the perf_events buffers. People likely use this
to check that the right number of probes have been placed so counting
perf_events buffers doesn't make sense here
Brendan Gregg [Thu, 14 Jul 2016 17:13:12 +0000 (10:13 -0700)]
Merge pull request #611 from cdown/failed_opens
killsnoop: s/failed opens/failed kill syscalls/
Chris Down [Thu, 14 Jul 2016 08:08:48 +0000 (09:08 +0100)]
Merge branch 'master' into failed_opens
Brendan Gregg [Wed, 13 Jul 2016 19:05:24 +0000 (12:05 -0700)]
Merge pull request #612 from markdrayton/fileslower
fileslower: try probing vfs_write if __vfs_write is missing
Mark Drayton [Wed, 13 Jul 2016 19:00:35 +0000 (12:00 -0700)]
Merge branch 'master' into fileslower
Mark Drayton [Wed, 13 Jul 2016 17:24:56 +0000 (18:24 +0100)]
fileslower: try probing vfs_write if __vfs_write is missing
Brenden Blanco [Wed, 13 Jul 2016 17:56:34 +0000 (10:56 -0700)]
Merge pull request #610 from bobrik/debian-dockerfile
Build debian packages in docker containers
Chris Down [Wed, 13 Jul 2016 14:18:35 +0000 (15:18 +0100)]
killsnoop: s/failed opens/failed kill syscalls/
Ivan Babrou [Tue, 12 Jul 2016 22:02:35 +0000 (23:02 +0100)]
Build debian packages in docker containers
Brendan Gregg [Tue, 12 Jul 2016 01:27:01 +0000 (18:27 -0700)]
use new tracepoint support (#608)
Brenden Blanco [Mon, 11 Jul 2016 22:02:25 +0000 (15:02 -0700)]
Merge pull request #602 from goldshtn/auto-tp
Full tracepoint support in Clang front-end
Trent Schafer [Mon, 11 Jul 2016 22:02:03 +0000 (08:02 +1000)]
Link to Ubuntu Xenial binary section (#604)
Sasha Goldshtein [Sat, 9 Jul 2016 12:17:44 +0000 (05:17 -0700)]
cc: Use manual parsing instead of std::regex
Older versions of GCC don't support std::regex even though they support
most of C++11. To avoid breaking the build on older systems, such as
Ubuntu 14.04, use manual parsing instead of std::regex.
Sasha Goldshtein [Fri, 8 Jul 2016 17:46:53 +0000 (10:46 -0700)]
tests: Test new tracepoint support
Sasha Goldshtein [Tue, 5 Jul 2016 16:34:56 +0000 (09:34 -0700)]
cc: Rewrite probe functions that refer to tracepoint structures
When a probe function refers to a tracepoint arguments structure,
such as `struct tracepoint__irq__irq_handler_entry`, add that structure
on-the-fly using a Clang frontend action that runs before any other
steps take place.
Typically, the user will create tracepoint probe functions using
the TRACEPOINT_PROBE macro, which avoids the need for specifying
the tracepoint category and event twice in the signature of the
probe function.
Sasha Goldshtein [Sun, 3 Jul 2016 08:33:19 +0000 (01:33 -0700)]
bcc: Auto-tracepoints similar to auto-kprobes
When a function in the BPF program starts with "tracepoint__", parse
the rest of the name as a tracepoint category and name and attach the
tracepoint automatically. For example:
```
int tracepoint__sched__sched_switch(...)
```
As a result, the sched:sched_switch tracepoint is enabled and the function
is attached to that tracepoint.
Brenden Blanco [Fri, 8 Jul 2016 23:21:38 +0000 (16:21 -0700)]
Fix test failure in test_libbcc (#603)
On some systems, was seeing a failure at tests/cc/test_c_api.cc:172 due
to failure to open the /tmp/perf-pid.map file. Looking through the code,
narrowed it down to an invalid use of c_str() on a temporary
std::string. Fix it by storing the string in a variable.
Signed-off-by: Brenden Blanco <bblanco@plumgrid.com>
Mark Drayton [Fri, 8 Jul 2016 02:49:25 +0000 (03:49 +0100)]
ProcSyms: deduplicate symbol names (#598)
Brenden Blanco [Fri, 8 Jul 2016 00:56:03 +0000 (17:56 -0700)]
Fix for C++ api change in LLVM 3.9 (#600)
Upstream, params() was renamed to parameters(). In order to support both
old and new LLVM, use the unchanged param_begin and param_end API.
Signed-off-by: Brenden Blanco <bblanco@plumgrid.com>
Brendan Gregg [Sat, 2 Jul 2016 01:38:30 +0000 (18:38 -0700)]
A tracepoint example (#596)
Mark Drayton [Fri, 1 Jul 2016 23:47:39 +0000 (00:47 +0100)]
ProcSyms: fix off-by-ones, use binary search to resolve addresses (#594)
* libbcc: fix off-by-one errors in resolving adjacent modules/symbols, add test
* libbcc: use binary search in ProcSyms::Module::find_addr()
Sasha Goldshtein [Fri, 1 Jul 2016 23:45:57 +0000 (02:45 +0300)]
bcc: Add ctypes declarations for new functions (#593)
Add ctypes declarations for `bpf_attach_tracepoint` and
`bpf_detach_tracepoint` in libbcc.py. It works anyway, but now
it's a bit safer.
Brenden Blanco [Thu, 30 Jun 2016 22:38:11 +0000 (15:38 -0700)]
Merge pull request #586 from goldshtn/offcpudist
cpudist: Support off-cpu time reports
Brenden Blanco [Thu, 30 Jun 2016 18:59:50 +0000 (11:59 -0700)]
Merge pull request #590 from goldshtn/bcc-tp-support
bcc: Tracepoint support in libbpf and BPF
Sasha Goldshtein [Thu, 30 Jun 2016 18:07:48 +0000 (11:07 -0700)]
bcc: Add test for tracepoint support
The test asserts that we can enable the sched_switch tracepoint and read
some events from it. The test is also marked to require kernel 4.7 or
later, because that's where the BPF support for tracepoints was introduced.
Sasha Goldshtein [Thu, 30 Jun 2016 17:46:27 +0000 (10:46 -0700)]
cpudist: Protect against potentially negative time deltas
It seems from experimentation that the calculated timestamps between
on- and off-CPU switch events can produce incorrect results, with a
later event having a smaller timestamp. Discard events when the
resulting delta time would be negative.
Sasha Goldshtein [Thu, 30 Jun 2016 14:39:27 +0000 (07:39 -0700)]
cpudist: Use `finish_task_switch` kprobe instead of `sched_switch` tracepoint
The `sched_switch` tracepoint approach requires storing the previous
task's tgid in a map and fetching it from there, because it is not
available as a tracepoint argument. Instead, placing a kprobe on the
`finish_task_switch` function allows cleanly fetching the previous
task's pid and tgid from the task_struct.
Sasha Goldshtein [Thu, 30 Jun 2016 06:35:43 +0000 (23:35 -0700)]
cpudist: Fix extraneous filtering of descheduled tasks
When the `-O` switch was provided, cpudist was unnecessarily filtering
out scheduling events arising from a task waking up when the previous
task was not running. On an idle system, this happens a lot, and causes
events to be missed. This is now fixed.
Sasha Goldshtein [Wed, 29 Jun 2016 09:18:06 +0000 (02:18 -0700)]
cpudist: Attempt to resolve pid to command
Use `/proc/$PID/comm`, which may fail, for example if the original
process already exited. This may also produce misleading results
if another process got the same pid, but there's no way around this.
Sasha Goldshtein [Wed, 29 Jun 2016 08:48:08 +0000 (01:48 -0700)]
cpudist: Support off-cpu time reports
Add -O switch, which directs cpudist to collect off-CPU time
statistics. Also restructure the code slightly and added examples
as appropriate.
Sasha Goldshtein [Thu, 30 Jun 2016 13:26:28 +0000 (06:26 -0700)]
bcc: Tracepoint support in libbpf and BPF
Introduce tracepoint support in libbpf via new `bpf_attach_tracepoint`
API, which takes the tracepoint category and name (e.g. "sched",
"sched_switch"). Attach the tracing program to the tracepoint's id
and proceed as usual.
Add `attach_tracepoint` API to Python BPF module, which takes the
tracepoint description as a single string (e.g. "sched:sched_switch").
Load the BPF program with bpf_prog_type set to TRACEPOINT and then
call `bpf_attach_tracepoint` to attach it.
Sasha Goldshtein [Thu, 30 Jun 2016 09:16:39 +0000 (12:16 +0300)]
trace: Specifying a pid with a kernel probe now works (#589)
Due to an incorrectly referenced global variable, specifying a pid
to filter with a kernel probe produced an error. This is now fixed,
for example:
```
TIME PID COMM FUNC
23:46:00 29967 bash sched_switch
23:46:01 29967 bash sched_switch
23:46:01 29967 bash sched_switch
^C
```
Sasha Goldshtein [Wed, 29 Jun 2016 00:57:01 +0000 (03:57 +0300)]
cpudist: Summarize task on-CPU time as histograms (#585)
* cpudist: summarize on-CPU time per task as a histogram
This is the initial implementation of the tool itself, which uses
the sched:sched_switch tracepoint to probe task switches. This is
a slightly more robust approach than using a kernel function, such
as finish_task_switch.
When native BCC support for tracepoints is introduced, this tool
can be reimplemented and reliance on the current Tracepoint module
can be removed.
* cpudist: add man page and examples
Brendan Gregg [Mon, 27 Jun 2016 20:11:18 +0000 (13:11 -0700)]
Merge pull request #584 from ygrek/master
tools: fix getting {ext4,btrfs}_file_operations address (ref #583)
ygrek [Mon, 27 Jun 2016 19:54:55 +0000 (12:54 -0700)]
tools: fix getting {ext4,btrfs}_file_operations address (ref #583)
Brendan Gregg [Mon, 27 Jun 2016 19:33:31 +0000 (12:33 -0700)]
Merge pull request #583 from ygrek/master
ext4slower: fix getting kallsyms address
ygrek [Mon, 27 Jun 2016 18:07:47 +0000 (11:07 -0700)]
ext4slower: fix getting kallsyms address
$ grep ext4_file_operations /proc/kallsyms
ffffffffc0331340 r ext4_file_operations [ext4]
Mark Drayton [Sun, 26 Jun 2016 20:14:44 +0000 (21:14 +0100)]
IPv6 support for tcp* tools (#582)
* tcpretrans: support full IPv6 addresses, fix --lossprobe
* tcpaccept: support full IPv6 addresses, fix timestamps
* tcpconnect: support full IPv6 addresses, fix timestamps
* tcpconnlat: support full IPv6 addresses, fix timestamps
Brenden Blanco [Sun, 26 Jun 2016 16:15:30 +0000 (09:15 -0700)]
Merge pull request #581 from goldshtn/docs-fixes
Update installation instructions for Fedora
Sasha Goldshtein [Sun, 26 Jun 2016 13:19:56 +0000 (16:19 +0300)]
Merge branch 'master' into docs-fixes
Sasha Goldshtein [Sun, 26 Jun 2016 13:15:39 +0000 (06:15 -0700)]
Update installation instructions for Fedora
clang 3.7.0 is known to be buggy, and we recommend that people use
3.7.1. Update install instructions to point to 3.7.1.
Compiling BCC on Fedora requires elfutils-libelf-devel. Add this
package to the install instructions for Fedora.
Brendan Gregg [Thu, 23 Jun 2016 18:05:29 +0000 (11:05 -0700)]
Merge pull request #579 from bobrik/fix-ipv6-print
Do not throw exception on ipv6 packets in tcpretrans
Ivan Babrou [Thu, 23 Jun 2016 17:11:25 +0000 (18:11 +0100)]
Do not throw exception on ipv6 packets in tcpretrans
Brendan Gregg [Wed, 22 Jun 2016 00:12:38 +0000 (17:12 -0700)]
kernel version guide (#577)
* kernel version guide
* update
4ast [Tue, 21 Jun 2016 15:51:11 +0000 (08:51 -0700)]
Merge pull request #576 from markdrayton/perf-pid-map-fixes
Fix nits for /tmp/perf-pid.map support
Mark Drayton [Tue, 21 Jun 2016 12:37:27 +0000 (13:37 +0100)]
Fix nits for /tmp/perf-pid.map support
4ast [Mon, 20 Jun 2016 16:47:30 +0000 (09:47 -0700)]
Merge pull request #573 from markdrayton/pid-map
Add support for reading symbols from /tmp/perf-pid.map
Brendan Gregg [Wed, 15 Jun 2016 18:50:24 +0000 (11:50 -0700)]
Merge pull request #572 from iovisor/ast_dev
fix install instruction to mention llvm >= 3.7.1
Brendan Gregg [Wed, 15 Jun 2016 18:42:05 +0000 (11:42 -0700)]
Merge branch 'master' into ast_dev
Alexei Starovoitov [Wed, 15 Jun 2016 16:39:28 +0000 (12:39 -0400)]
fix install instruction to mention llvm >= 3.7.1
Signed-off-by: Alexei Starovoitov <ast@fb.com>
Mark Drayton [Wed, 15 Jun 2016 10:53:24 +0000 (11:53 +0100)]
Add support for reading symbols from /tmp/perf-pid.map
This adds basic support for /tmp/perf-pid.map. To cope with processes in
containers, it supports:
* mapping from BCC's PID namespace to the target process's PID namespace
using /proc/pid/status
* resolving a target process's root filesystem using /proc/pid/root
4ast [Wed, 15 Jun 2016 05:14:57 +0000 (22:14 -0700)]
Merge pull request #571 from iovisor/fix_568
Check for NULL result from bpf_attach_kprobe
Brenden Blanco [Wed, 15 Jun 2016 03:57:03 +0000 (20:57 -0700)]
Check for NULL result from bpf_attach_kprobe
The check for NULL in the return value from C functions was comparing to
None incorrectly, causing an error check to pass improperly. Simply
check the truthiness of the return instead seems to be more resilient.
Add a test as well.
Fixes: #568
Signed-off-by: Brenden Blanco <bblanco@plumgrid.com>
Brenden Blanco [Mon, 13 Jun 2016 11:59:42 +0000 (04:59 -0700)]
Merge pull request #563 from qmonnet/debug-output
libbpf.c: Prepend BPF syscall error to log_bug when debug is enabled.
Quentin Monnet [Thu, 9 Jun 2016 12:55:29 +0000 (14:55 +0200)]
python: With DEBUG_BPF, print BPF error even if log_buf is not empty.
When BPF syscall fails and DEBUG_FLAG has been provided to a Python
caller script, the BPF error string would be printed only if log_buf had
not been filled by the kernel. This commit removes this requirement,
printing the error string even if log_buf is not empty.
Signed-off-by: Quentin Monnet <quentin.monnet@6wind.com>
Quentin Monnet [Thu, 9 Jun 2016 12:55:29 +0000 (14:55 +0200)]
python: Print BPF syscall error if DEBUG_BPF is on but log_buf is empty.
Commit
759029fea8066b41b54be5447137db95cb1313c4 provided an option to
store the output from BPF syscall into a buffer (and not to print it
systematically to standard output) on program load in libbpf.c.
But doing so, it only stores the content of attr.log_buf, while the
error string--resulting from a failed BPF syscall--is no more displayed
when the DEBUG_BPF flag is used in the Python script responsible for
converting and injecting the code.
This commit proposes a fix for this bug by printing the error message
(associated to the return value from the syscall) from the Python
caller, when all the following conditions are met:
- the syscall fails,
- the DEBUG_BPF flag has been provided, and
- log_buf is empty (has not been filled by kernel).
Note: when DEBUG_BPF is not provided, the error string is printed in the
C wrapper in libbpf.c (bpf_prog_load) anyway.
Fixes:
759029fe ("Add option for custom log string to bpf_prog_load")
Signed-off-by: Quentin Monnet <quentin.monnet@6wind.com>
Brendan Gregg [Wed, 8 Jun 2016 18:44:49 +0000 (11:44 -0700)]
Merge pull request #562 from evverx/change-kernel-user-stacks-order
Change kernel/user stacks order
Evgeny Vereshchagin [Tue, 7 Jun 2016 20:33:54 +0000 (06:33 +1000)]
offcputime: add the -d option
Closes #559
Evgeny Vereshchagin [Tue, 7 Jun 2016 20:06:33 +0000 (06:06 +1000)]
offcputime: change user/kernel stack order
see #559
Brendan Gregg [Tue, 7 Jun 2016 15:49:19 +0000 (08:49 -0700)]
Merge pull request #561 from evverx/folded-dont-miss-backtrace-part
offcputime -f: don't miss backtrace bounds
Evgeny Vereshchagin [Tue, 7 Jun 2016 08:00:01 +0000 (18:00 +1000)]
offcputime -f: don't miss backtrace bounds
See #559
Brendan Gregg [Mon, 6 Jun 2016 19:45:02 +0000 (12:45 -0700)]
Merge pull request #560 from evverx/clarify-offcputime
Update manpage for offcputime
Brendan Gregg [Mon, 6 Jun 2016 18:18:54 +0000 (11:18 -0700)]
Merge pull request #554 from evverx/fix-p-option
Fix tools[/old]/offcputime -p PID
Evgeny Vereshchagin [Mon, 6 Jun 2016 09:25:54 +0000 (09:25 +0000)]
Update manpage for offcputime
Closes #558
Evgeny Vereshchagin [Fri, 27 May 2016 06:13:52 +0000 (06:13 +0000)]
Fix tools[/old]/offcputime -p PID
We should extract pid before THREAD_FILTER
4ast [Fri, 3 Jun 2016 01:15:43 +0000 (18:15 -0700)]
Merge pull request #555 from markdrayton/execsnoop
execsnoop: use BPF_PERF_OUTPUT, add PPID
Mark Drayton [Thu, 2 Jun 2016 09:53:34 +0000 (10:53 +0100)]
execsnoop: show PPID in output
Mark Drayton [Thu, 2 Jun 2016 09:53:20 +0000 (10:53 +0100)]
execsnoop: use BPF_PERF_OUTPUT instead of trace pipe
Brenden Blanco [Wed, 1 Jun 2016 01:21:10 +0000 (18:21 -0700)]
Merge pull request #557 from ryanmiao/mypr
set use_errno to True for libbcc.so
Brenden Blanco [Wed, 1 Jun 2016 01:16:06 +0000 (18:16 -0700)]
Merge pull request #544 from abirchall/abirchall_release
Add user space stack traces to offcputime
Jincheng Miao [Sun, 29 May 2016 05:17:03 +0000 (13:17 +0800)]
set use_errno to True for libbcc.so
Currently, errno could not be accessed by libbcc.py, like:
```
binding socket to 'eth0'
cannot create raw socket: Operation not permitted
Traceback (most recent call last):
File "/home/ryan/Software/bcc/examples/networking/http_filter/http-parse-simple.py", line 73, in <module>
BPF.attach_raw_socket(function_http_filter, interface)
File "/usr/lib/python2.7/site-packages/bcc/__init__.py", line 330, in attach_raw_socket
raise Exception("Failed to open raw device %s: %s" % (dev, errstr))
Exception: Failed to open raw device eth0: Success
```
The exception prints the errno as `Success`, instead of `EPERM`.
This patch would fix this problem.
Signed-off-by: Jincheng Miao <jincheng.miao@gmail.com>
Brenden Blanco [Fri, 27 May 2016 00:42:13 +0000 (17:42 -0700)]
Merge pull request #552 from evverx/install-tools-old
Install tools/old
Evgeny Vereshchagin [Thu, 26 May 2016 09:11:44 +0000 (09:11 +0000)]
Point people to tools/old
Evgeny Vereshchagin [Thu, 26 May 2016 08:50:50 +0000 (08:50 +0000)]
Install tools/old too
4ast [Tue, 24 May 2016 20:30:39 +0000 (13:30 -0700)]
Merge pull request #551 from vmg/vmg/lua-para
lua: Allow LuaJIT parameterized types in `open_perf_buffer`
Vicent Marti [Tue, 24 May 2016 15:57:34 +0000 (17:57 +0200)]
lua: Allow LuaJIT parameterized types in `open_perf_buffer`
4ast [Tue, 24 May 2016 15:43:51 +0000 (08:43 -0700)]
Merge pull request #550 from markdrayton/trace-fields
Use BPF_PERF_OUTPUT for fileslower and stacksnoop
Mark Drayton [Tue, 24 May 2016 14:01:01 +0000 (07:01 -0700)]
Use BPF_PERF_OUTPUT for fileslower and stacksnoop
Andrew Birchall [Tue, 24 May 2016 08:44:41 +0000 (01:44 -0700)]
Update documentation and man page for offcputime
Andrew Birchall [Thu, 5 May 2016 17:56:40 +0000 (10:56 -0700)]
Add user space stack traces to offcputime
Summary:
* generalize `KernelSymbolCache` into `SymbolCache` which supports user/kernel space symbols
* create `BPF.sym()` for resolving user/kernel space symbols
* `_ksym_cache` => `_sym_caches` and create `BPF._sym_cache()` to leverage the symbol caches
* update `tools/offcputime.py` to print user space stack traces
Test Plan:
```
dev[bcc](abirchall_next): ~/bcc_run_tool.sh offcputime -U -k -f 1
ERROR: Displaying user stacks for kernel threads doesn't make sense.
devbig680[bcc](abirchall_next): ~/bcc_run_tool.sh offcputime -K -f 1 | grep python2 | head -n 1
python2.7;system_call_fastpath;sys_futex;do_futex;futex_wait;futex_wait_queue_me;schedule 19
dev[bcc](abirchall_next): ~/bcc_run_tool.sh offcputime -U -f 1 | grep python2 | head -n 1
python2.7;clone;start_thread;t_bootstrap;PyEval_CallObjectWithKeywords;PyObject_Call;instancemethod_call;PyObject_Call;function_call;PyEval_EvalCodeEx;PyEval_EvalFrameEx;PyObject_Call;function_call;PyEval_EvalCodeEx;PyEval_EvalFrameEx;time_sleep;PyEval_RestoreThread 5
devbig680[bcc](abirchall_next): ~/bcc_run_tool.sh offcputime -f 1 | grep python2 | head -n 1
python2.7;system_call_fastpath;sys_futex;do_futex;futex_wait;futex_wait_queue_me;schedule;[unknown];__libc_start_main;Py_Main;RunModule;PyObject_Call;function_call;PyEval_EvalCodeEx;PyEval_EvalFrameEx;PyEval_EvalFrameEx;PyEval_EvalCodeEx;PyEval_EvalFrameEx;lock_PyThread_acquire_lock;PyEval_RestoreThread 39
```
Brenden Blanco [Mon, 23 May 2016 02:55:06 +0000 (19:55 -0700)]
Merge pull request #535 from MQasimSarfraz/patch-1
Updated monitor.c to take care of host byte order
Brenden Blanco [Fri, 20 May 2016 17:31:30 +0000 (10:31 -0700)]
Merge pull request #546 from iamkafai/master
Fix a few fd leaks