perf defines both __used and __unused variables to use for marking
unused variables. The variable __used is defined to
__attribute__((__unused__)), which contradicts the kernel definition to
__attribute__((__used__)) for new gcc versions. On Android, __used is
also defined in system headers and this leads to warnings like: warning:
'__used__' attribute ignored
__unused is not defined in the kernel and is not a standard definition.
If __unused is included everywhere instead of __used, this leads to
conflicts with glibc headers, since glibc has a variables with this name
in its headers.
The best approach is to use __maybe_unused, the definition used in the
kernel for __attribute__((unused)). In this way there is only one
definition in perf sources (instead of 2 definitions that point to the
same thing: __used and __unused) and it works on both Linux and Android.
This patch simply replaces all instances of __used and __unused with
__maybe_unused.
Signed-off-by: Irina Tirdea <irina.tirdea@intel.com>
Acked-by: Pekka Enberg <penberg@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1347315303-29906-7-git-send-email-irina.tirdea@intel.com
[ committer note: fixed up conflict with a116e05 in builtin-sched.c ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
On some systems (e.g. Android), ALIGN is defined in system headers as
ALIGN(p). The definition of ALIGN used in perf takes 2 parameters:
ALIGN(x,a). This leads to redefinition conflicts.
Redefinition error on Android:
In file included from util/include/linux/list.h:1:0,
from util/callchain.h:5,
from util/hist.h:6,
from util/session.h:4,
from util/build-id.h:4,
from util/annotate.c:11:
util/include/linux/kernel.h:11:0: error: "ALIGN" redefined [-Werror]
bionic/libc/include/sys/param.h:38:0: note: this is the location of
the previous definition
Conflics with system defined ALIGN in Android:
util/event.c: In function 'perf_event__synthesize_comm':
util/event.c:115:32: error: macro "ALIGN" passed 2 arguments, but takes just 1
util/event.c:115:9: error: 'ALIGN' undeclared (first use in this function)
util/event.c:115:9: note: each undeclared identifier is reported only once for
each function it appears in
In order to avoid this redefinition, ALIGN is renamed to PERF_ALIGN.
Signed-off-by: Irina Tirdea <irina.tirdea@intel.com>
Acked-by: Pekka Enberg <penberg@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Irina Tirdea <irina.tirdea@intel.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1347315303-29906-5-git-send-email-irina.tirdea@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Build currently fails:
$ make -C tools/perf O=/tmp/pbuild NO_LIBELF=1
util/symbol.c: In function ‘dso__load’:
util/symbol.c:1128:27: error: ‘struct symsrc’ has no member named ‘dynsym’
CC /tmp/pbuild/util/pager.o
make: *** [/tmp/pbuild/util/symbol.o] Error 1
make: *** Waiting for unfinished jobs....
Moving the dynsym reference to symbol-elf.c reveals that NO_LIBELF requires
NO_LIBUNWIND:
$ make -C tools/perf O=/tmp/pbuild NO_LIBELF=1
LINK /tmp/pbuild/perf
/tmp/pbuild/libperf.a(unwind.o): In function `elf_section_offset':
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:176: undefined reference to `elf_begin'
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:181: undefined reference to `gelf_getehdr'
/tmp/pbuild/libperf.a(unwind.o): In function `elf_section_by_name':
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:157: undefined reference to `elf_nextscn'
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:160: undefined reference to `gelf_getshdr'
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:161: undefined reference to `elf_strptr'
/tmp/pbuild/libperf.a(unwind.o): In function `elf_section_offset':
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:190: undefined reference to `elf_end'
/tmp/pbuild/libperf.a(unwind.o): In function `read_unwind_spec':
/opt/sw/ahern/perf.git/tools/perf/util/unwind.c:190: undefined reference to `elf_end'
collect2: ld returned 1 exit status
make: *** [/tmp/pbuild/perf] Error 1
make: Leaving directory `/opt/sw/ahern/perf.git/tools/perf'
This patch fixes both.
Reviewed-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1345391234-71906-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We keep both a 'runtime' elf image as well as a 'debug' elf image around
and generate symbols by looking at both of these.
This eliminates the need for the want_symtab/goto restart mechanism
combined with iterating over and reopening the elf images a second time.
Also give dso__synthsize_plt_symbols() the runtime image (which has
dynsyms) instead of the symbol image (which may only have a symtab and
no dynsyms).
Previously if a debug image was found all runtime images were ignored.
This fixes 2 issues:
- Symbol resolution to failure on PowerPC systems with debug symbols
installed, as the debug images lack a '.opd' section which contains
function descriptors.
- On all archs, plt synthesis failed when a debug image was loaded and
that debug image lacks a dynsym section while a runtime image has a
dynsym section.
Assumptions:
- If a .opd section exists, it is contained in the highest priority
image with a dynsym section.
- This generally implies that the debug image lacks a dynsym section
(ie: it is marked as NO_BITS).
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com>
Cc: David Hansen <dave@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Matt Hellsley <matthltc@us.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1344637382-22789-17-git-send-email-cody@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To properly handle platforms with an opd section, both a runtime image
(which contains the opd section but possibly lacks symbols) and a symbol
image (which probably lacks an opd section but has symbols).
The next patch ("perf symbol: use both runtime and debug images")
adjusts the callsite in dso__load() to take advantage of being able to
pass both runtime & debug images.
Assumptions made here:
- The opd section, if it exists in the runtime image, has headers in
both the runtime image and the debug/syms image.
- The index of the opd section (again, only if it exists in the runtime
image) is the same in both the runtime and debug/symbols image.
Both of these are true on RHEL, but it is unclear how accurate they are
in general (on platforms with function descriptors in opd sections).
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com>
Cc: David Hansen <dave@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Matt Hellsley <matthltc@us.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1344637382-22789-16-git-send-email-cody@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Factors opening of certain sections & tracking certain elf info into an
external structure.
The goal here is to keep multiple elfs (and their looked up
sections/indexes) around during the symbol generation process (in
dso__load()).
We need this to properly resolve symbols on PPC due to the use of
function descriptors & the .opd section (ie: symbols which are functions
don't point to their actual location, they point to their function
descriptor in .opd which contains their actual location.
It would be possible to just keep the (Elf *) around, but then we'd end
up with duplicate code for looking up the same sections and checking for
the existence of an important section wouldn't be as clean (and we need
to keep the Elf stuff confined to symtab-elf.c).
Utilized by the later patch
"perf symbols: Use both runtime and debug images"
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com>
Cc: David Hansen <dave@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Matt Hellsley <matthltc@us.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1344637382-22789-12-git-send-email-cody@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
kallsyms__parse() takes a callback that is called on every discovered
symbol. As /proc/kallsyms does not supply symbol sizes, the callback was
simply called with end=start, faking the symbol size to 1.
All of the callbacks (there are 2) used in calls to kallsyms__parse()
are _only_ used as callbacks for kallsyms__parse().
Given that kallsyms__parse() lacks real information about what
end/length should be, don't make up a length in kallsyms__parse().
Instead have the callbacks handle guessing the length.
Also relocate a comment regarding symbol creation to the callback which
does symbol creation (kallsyms__parse() is not in general used to create
symbols).
Signed-off-by: Cody P Schafer <cody@linux.vnet.ibm.com>
Cc: David Hansen <dave@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Matt Hellsley <matthltc@us.ibm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1344637382-22789-3-git-send-email-cody@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Only want to process directories under the guestmnount directory that
have a pid as a name (ie, all digits). Other entries in the guestmount
directory should be ignored. There is already a check that requires the
first character of each entry to be a digit, but atoi is used to convert
the directory name to a pid. For example if guestmount contains a
directory with the name 1foo, atoi converts it to a pid of 1 and a
machine is created with a pid of 1. This is wrong; this directory really
should be ignored. Use strtol to do that.
Signed-off-by: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1343616875-6455-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently we dont care about the file object's endianness. It's possible
we read buildid file object from different architecture than we are
currentlly running on. So we need to care about properly reading such
object's data - handle different endianness properly.
Adding:
needs_swap DSO field
dso__swap_init function to initialize DSO's needs_swap
DSO__SWAP to read the data with proper swaps
Together with other endianity patches, this change fixies perf report
discrepancies on origin and target systems as described in test 1 below,
e.g. following perf report diff:
...
0.12% ps [kernel.kallsyms] [k] clear_page
- 0.12% awk bash [.] alloc_word_desc
+ 0.12% awk bash [.] yyparse
0.11% beah-rhts-task libpython2.6.so.1.0 [.] 0x5560e
0.10% perf libc-2.12.so [.] __ctype_toupper_loc
- 0.09% rhts-test-runne bash [.] maybe_make_export_env
+ 0.09% rhts-test-runne bash [.] 0x385a0
0.09% ps [kernel.kallsyms] [k] page_fault
...
Note, running following to test perf endianity handling:
test 1)
- origin system:
# perf record -a -- sleep 10 (any perf record will do)
# perf report > report.origin
# perf archive perf.data
- copy the perf.data, report.origin and perf.data.tar.bz2
to a target system and run:
# tar xjvf perf.data.tar.bz2 -C ~/.debug
# perf report > report.target
# diff -u report.origin report.target
- the diff should produce no output
(besides some white space stuff and possibly different
date/TZ output)
test 1)
- origin system:
# perf record -ag -fo /tmp/perf.data -- sleep 1
- mount origin system root to the target system on /mnt/origin
- target system:
# perf script --symfs /mnt/origin -I -i /mnt/origin/tmp/perf.data \
--kallsyms /mnt/origin/proc/kallsyms
- complete perf.data header is displayed
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1338380624-7443-2-git-send-email-jolsa@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull user-space probe instrumentation from Ingo Molnar:
"The uprobes code originates from SystemTap and has been used for years
in Fedora and RHEL kernels. This version is much rewritten, reviews
from PeterZ, Oleg and myself shaped the end result.
This tree includes uprobes support in 'perf probe' - but SystemTap
(and other tools) can take advantage of user probe points as well.
Sample usage of uprobes via perf, for example to profile malloc()
calls without modifying user-space binaries.
First boot a new kernel with CONFIG_UPROBE_EVENT=y enabled.
If you don't know which function you want to probe you can pick one
from 'perf top' or can get a list all functions that can be probed
within libc (binaries can be specified as well):
$ perf probe -F -x /lib/libc.so.6
To probe libc's malloc():
$ perf probe -x /lib64/libc.so.6 malloc
Added new event:
probe_libc:malloc (on 0x7eac0)
You can now use it in all perf tools, such as:
perf record -e probe_libc:malloc -aR sleep 1
Make use of it to create a call graph (as the flat profile is going to
look very boring):
$ perf record -e probe_libc:malloc -gR make
[ perf record: Woken up 173 times to write data ]
[ perf record: Captured and wrote 44.190 MB perf.data (~1930712
$ perf report | less
32.03% git libc-2.15.so [.] malloc
|
--- malloc
29.49% cc1 libc-2.15.so [.] malloc
|
--- malloc
|
|--0.95%-- 0x208eb1000000000
|
|--0.63%-- htab_traverse_noresize
11.04% as libc-2.15.so [.] malloc
|
--- malloc
|
7.15% ld libc-2.15.so [.] malloc
|
--- malloc
|
5.07% sh libc-2.15.so [.] malloc
|
--- malloc
|
4.99% python-config libc-2.15.so [.] malloc
|
--- malloc
|
4.54% make libc-2.15.so [.] malloc
|
--- malloc
|
|--7.34%-- glob
| |
| |--93.18%-- 0x41588f
| |
| --6.82%-- glob
| 0x41588f
...
Or:
$ perf report -g flat | less
# Overhead Command Shared Object Symbol
# ........ ............. ............. ..........
#
32.03% git libc-2.15.so [.] malloc
27.19%
malloc
29.49% cc1 libc-2.15.so [.] malloc
24.77%
malloc
11.04% as libc-2.15.so [.] malloc
11.02%
malloc
7.15% ld libc-2.15.so [.] malloc
6.57%
malloc
...
The core uprobes design is fairly straightforward: uprobes probe
points register themselves at (inode:offset) addresses of
libraries/binaries, after which all existing (or new) vmas that map
that address will have a software breakpoint injected at that address.
vmas are COW-ed to preserve original content. The probe points are
kept in an rbtree.
If user-space executes the probed inode:offset instruction address
then an event is generated which can be recovered from the regular
perf event channels and mmap-ed ring-buffer.
Multiple probes at the same address are supported, they create a
dynamic callback list of event consumers.
The basic model is further complicated by the XOL speedup: the
original instruction that is probed is copied (in an architecture
specific fashion) and executed out of line when the probe triggers.
The XOL area is a single vma per process, with a fixed number of
entries (which limits probe execution parallelism).
The API: uprobes are installed/removed via
/sys/kernel/debug/tracing/uprobe_events, the API is integrated to
align with the kprobes interface as much as possible, but is separate
to it.
Injecting a probe point is privileged operation, which can be relaxed
by setting perf_paranoid to -1.
You can use multiple probes as well and mix them with kprobes and
regular PMU events or tracepoints, when instrumenting a task."
Fix up trivial conflicts in mm/memory.c due to previous cleanup of
unmap_single_vma().
* 'perf-uprobes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
perf probe: Detect probe target when m/x options are absent
perf probe: Provide perf interface for uprobes
tracing: Fix kconfig warning due to a typo
tracing: Provide trace events interface for uprobes
tracing: Extract out common code for kprobes/uprobes trace events
tracing: Modify is_delete, is_return from int to bool
uprobes/core: Decrement uprobe count before the pages are unmapped
uprobes/core: Make background page replacement logic account for rss_stat counters
uprobes/core: Optimize probe hits with the help of a counter
uprobes/core: Allocate XOL slots for uprobes use
uprobes/core: Handle breakpoint and singlestep exceptions
uprobes/core: Rename bkpt to swbp
uprobes/core: Make order of function parameters consistent across functions
uprobes/core: Make macro names consistent
uprobes: Update copyright notices
uprobes/core: Move insn to arch specific structure
uprobes/core: Remove uprobe_opcode_sz
uprobes/core: Make instruction tables volatile
uprobes: Move to kernel/events/
uprobes/core: Clean up, refactor and improve the code
...
When loading symbols from DSO we check multiple paths of DSO binary
until we succeed to load symbols ('.symtab' section). Once symbols are
read we try to load also plt symbols.
During the reading of plt symbols, the dso file is reopened from
location given by dso->long_name. This could be wrong in case we want
process buildid binaries.
The change is to make the plt symbols being read from the DSO path, that
normal symbols were read from.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1334756818-6631-1-git-send-email-jolsa@redhat.com
[ committer note: moved dso to be the first parameter of that function ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When building on my Debian/mips system, util/util.c fails to build
because commit 1aed267173 (perf kvm: Do
guest-only counting by default) indirectly includes stdio.h before the
feature selection in util.h is done. This prevents _GNU_SOURCE in
util.h from enabling the declaration of getline(), from now second
inclusion of stdio.h, and the build is broken.
There is another breakage in util/evsel.c caused by include ordering,
but I didn't fully track down the commit that caused it.
The root cause of all this is an inconsistent definition of _GNU_SOURCE,
so I move the definition into the Makefile so that it is passed to all
invocations of the compiler and used uniformly for all system header
files. All other #define and #undef of _GNU_SOURCE are removed as they
cause conflicts with the definition passed to the compiler.
All the features.h definitions (_LARGEFILE64_SOURCE _FILE_OFFSET_BITS=64
and _GNU_SOURCE) are needed by the python glue code too, so they are
moved to BASIC_CFLAGS, and the misleading comments about BASIC_CFLAGS
are removed.
This gives me a clean build on x86_64 (fc12) and mips (Debian).
Cc: David Daney <david.daney@cavium.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Joerg Roedel <joerg.roedel@amd.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1326836461-11952-1-git-send-email-ddaney.cavm@gmail.com
Signed-off-by: David Daney <david.daney@cavium.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
