01 Jun, 2012
4 commits
-
Merge misc patches from Andrew Morton:
- the "misc" tree - stuff from all over the map
- checkpatch updates
- fatfs
- kmod changes
- procfs
- cpumask
- UML
- kexec
- mqueue
- rapidio
- pidns
- some checkpoint-restore feature work. Reluctantly. Most of it
delayed a release. I'm still rather worried that we don't have a
clear roadmap to completion for this work.* emailed from Andrew Morton : (78 patches)
kconfig: update compression algorithm info
c/r: prctl: add ability to set new mm_struct::exe_file
c/r: prctl: extend PR_SET_MM to set up more mm_struct entries
c/r: procfs: add arg_start/end, env_start/end and exit_code members to /proc/$pid/stat
syscalls, x86: add __NR_kcmp syscall
fs, proc: introduce /proc//task//children entry
sysctl: make kernel.ns_last_pid control dependent on CHECKPOINT_RESTORE
aio/vfs: cleanup of rw_copy_check_uvector() and compat_rw_copy_check_uvector()
eventfd: change int to __u64 in eventfd_signal()
fs/nls: add Apple NLS
pidns: make killed children autoreap
pidns: use task_active_pid_ns in do_notify_parent
rapidio/tsi721: add DMA engine support
rapidio: add DMA engine support for RIO data transfers
ipc/mqueue: add rbtree node caching support
tools/selftests: add mq_perf_tests
ipc/mqueue: strengthen checks on mqueue creation
ipc/mqueue: correct mq_attr_ok test
ipc/mqueue: improve performance of send/recv
selftests: add mq_open_tests
... -
While doing the checkpoint-restore in the user space one need to determine
whether various kernel objects (like mm_struct-s of file_struct-s) are
shared between tasks and restore this state.The 2nd step can be solved by using appropriate CLONE_ flags and the
unshare syscall, while there's currently no ways for solving the 1st one.One of the ways for checking whether two tasks share e.g. mm_struct is to
provide some mm_struct ID of a task to its proc file, but showing such
info considered to be not that good for security reasons.Thus after some debates we end up in conclusion that using that named
'comparison' syscall might be the best candidate. So here is it --
__NR_kcmp.It takes up to 5 arguments - the pids of the two tasks (which
characteristics should be compared), the comparison type and (in case of
comparison of files) two file descriptors.Lookups for pids are done in the caller's PID namespace only.
At moment only x86 is supported and tested.
[akpm@linux-foundation.org: fix up selftests, warnings]
[akpm@linux-foundation.org: include errno.h]
[akpm@linux-foundation.org: tweak comment text]
Signed-off-by: Cyrill Gorcunov
Acked-by: "Eric W. Biederman"
Cc: Pavel Emelyanov
Cc: Andrey Vagin
Cc: KOSAKI Motohiro
Cc: Ingo Molnar
Cc: H. Peter Anvin
Cc: Thomas Gleixner
Cc: Glauber Costa
Cc: Andi Kleen
Cc: Tejun Heo
Cc: Matt Helsley
Cc: Pekka Enberg
Cc: Eric Dumazet
Cc: Vasiliy Kulikov
Cc: Alexey Dobriyan
Cc: Valdis.Kletnieks@vt.edu
Cc: Michal Marek
Cc: Frederic Weisbecker
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Add the mq_perf_tests tool I used when creating my mq performance patch.
Also add a local .gitignore to keep the binaries from showing up in git
status output.[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Doug Ledford
Cc: Stephen Rothwell
Cc: Manfred Spraul
Cc: Frederic Weisbecker
Acked-by: KOSAKI Motohiro
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Add a directory to house POSIX message queue subsystem specific tests.
Add first test which checks the operation of mq_open() under various
corner conditions.Signed-off-by: Doug Ledford
Cc: KOSAKI Motohiro
Cc: Doug Ledford
Cc: Joe Korty
Cc: Amerigo Wang
Cc: Serge E. Hallyn
Cc: Jiri Slaby
Cc: Manfred Spraul
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
31 May, 2012
1 commit
-
Pull perf updates from Ingo Molnar.
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
perf ui browser: Stop using 'self'
perf annotate browser: Read perf config file for settings
perf config: Allow '_' in config file variable names
perf annotate browser: Make feature toggles global
perf annotate browser: The idx_asm field should be used in asm only view
perf tools: Convert critical messages to ui__error()
perf ui: Make --stdio default when TUI is not supported
tools lib traceevent: Silence compiler warning on 32bit build
perf record: Fix branch_stack type in perf_record_opts
perf tools: Reconstruct event with modifiers from perf_event_attr
perf top: Fix counter name fixup when fallbacking to cpu-clock
perf tools: fix thread_map__new_by_pid_str() memory leak in error path
perf tools: Do not use _FORTIFY_SOURCE when DEBUG=1 is specified
tools lib traceevent: Fix signature of create_arg_item()
tools lib traceevent: Use proper function parameter type
tools lib traceevent: Fix freeing arg on process_dynamic_array()
tools lib traceevent: Fix a possibly wrong memory dereference
tools lib traceevent: Fix a possible memory leak
tools lib traceevent: Allow expressions in __print_symbolic() fields
perf evlist: Explicititely initialize input_name
...
30 May, 2012
7 commits
-
Stop using this python/OOP convention, doesn't really helps. Will do
more from time to time till we get it cleaned up in all of /perf.Suggested-by: Thomas Gleixner
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-5dyxyb8o0gf4yndk27kafbd1@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
The defaults are:
[annotate]
hide_src_code = false
use_offset = true
jump_arrows = true
show_nr_jumps = falseCc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-q4egci70rjgxh7bogbbfpcyf@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
For annotate I want to be able to have variables that are the same as
the ones representing feature toggles.Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-7rhhf6m0a72p2wja4tgv1itg@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
So that when navigating to another function from a call site or when
going to another annotation browser thru the main report/top browser the
options (hide source code, jump arrows, jumpy lines, etc) remains the
last ones selected.Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-0h0tah1zj59p01581snjufne@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
When hide_src_view is true we can't use browser_disasm_line->idx, that
takes into account also non asm lines, we must use browser_disasm_line->idx_asm
instead, otherwise we may end up with an index after the number of
entries, oops, fix it.Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-o1szpyjh3z87yi0n6x0cr8uu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Compiling page-type.c with a recent compiler produces many warnings,
mostly related to signed/unsigned comparisons. This patch cleans up most
of them.One remaining warning is about an unused parameter. The file
doesn't define a __unused macro (or the like) yet. This can be addressed
later.Signed-off-by: Ulrich Drepper
Acked-by: KOSAKI Motohiro
Acked-by: Fengguang Wu
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Programs using /proc/kpageflags need to know about the various flags. The
provides them and the comments in the file
indicate that it is supposed to be used by user-level code. But the file
is not installed.Install the headers and mark the unstable flags as out-of-bounds. The
page-type tool is also adjusted to not duplicate the definitionsSigned-off-by: Ulrich Drepper
Acked-by: KOSAKI Motohiro
Acked-by: Fengguang Wu
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
29 May, 2012
2 commits
-
There were places where use ui__warning (or even fprintf) to show
critical messages. This patch converts them to ui__error so that the
front-end code can implement appropriate behavior.Signed-off-by: Namhyung Kim
Cc: Paul Mackerras
Cc: Pekka Enberg
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1338265382-6872-3-git-send-email-namhyung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo -
The commit dc41b9b8f02db ("perf ui: Change fallback policy of
setup_browser") changed default behavior of the function but missed
setting the use_browser variable to 0 accidently. So perf report ends up
doing nothing in such cases. Fix it.Signed-off-by: Namhyung Kim
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1338216802-5675-1-git-send-email-namhyung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo
27 May, 2012
1 commit
-
The gcc complains about casting a pointer to unsigned long long directly:
SUBDIR ../lib/traceevent/
CC FPIC event-parse.o
CC FPIC trace-seq.o
CC FPIC parse-filter.o
/home/namhyung/project/linux/tools/lib/traceevent/parse-filter.c: In function ‘get_value’:
/home/namhyung/project/linux/tools/lib/traceevent/parse-filter.c:1588: warning: cast from pointer to integer of different size
CC FPIC parse-utils.o
BUILD STATIC LIB libtraceevent.aSigned-off-by: Namhyung Kim
Cc: Frederic Weisbecker
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Steven Rostedt
Link: http://lkml.kernel.org/r/1338003691-3141-1-git-send-email-namhyung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo
26 May, 2012
3 commits
-
The attr.branch_sample_type field is defined as u64 by the API. As
such, we need to ensure the variable holding the value of the branch
stack filters is also u64 otherwise we may lose bits in the future.Note also that the bogus definition of the field in perf_record_opts
caused problems on big-endian PPC systems. Thanks to Anshuman Khandual
for tracking the problem on PPC.Reported-by: Anshuman Khandual
Signed-off-by: Stephane Eranian
Cc: Anshuman Khandual
Cc: Ingo Molnar
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20120525211344.GA7729@quad
Signed-off-by: Arnaldo Carvalho de Melo -
The modifiers:
k kernel space
u user space
h hypervisor
G guest
H host
p, pp, ppp precision level (PEBS)that can be suffixed to an event were lost when tools used event_name()
to reconstruct them from the perf_event_attr entries in a perf.data
file.Fix it by following the defaults used for these modifiers in the current
codebase, so:$ perf record -e instructions:u usleep 1 2> /dev/null
$ perf evlist
instructions:u
$ perf record -e cycles:k usleep 1 2> /dev/null
$ perf evlist
cycles:k
$ perf record -e cycles:kh usleep 1 2> /dev/null
$ perf evlist
cycles:kh
$ perf record -e cache-misses:G usleep 1 2> /dev/null
$ perf evlist
cache-misses:G
$ perf record -e cycles:ppk usleep 1 2> /dev/null
$ perf evlist
cycles:kpp
$Also works with 'top', 'report', etc.
More work needed to cover tracepoints and software events while not
dragging lots of baggage to the python binding, this is a minimal fix
for v3.5.Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-4hl5glle0hxlklw4usva1mkt@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
In 40491eaa "perf top: Update event name when falling back to cpu-clock"
we freed counter->name but didn't reset it to NULL, then when setting it
to the result of event_name(), event_name() would use the cached value,
which by now was overwritten and thus we got garbage or a zero lenght
string.Fix it by just freeing and setting counter->name to NULL, this way
event_name() when called afterwards, will find the right counter name
and cache it again.Found while trying 'cycles:pp' on a machine were :pp couldn't be
honoured. Probably the best fallback here is to tell the user that that
level of precision is not available on the PMU and then go removing 'p',
levels of precision till we get to play 'cycles' and if even that fails,
_then_ get to 'cpu-clock'.But that is the matter for another patch, this one just needs to fix the
caching issue, which in the end will show 'cpu-clock' when tools ask for
the event name being used, which clarifies things for the user, that
will see that 'cycles:pp' or whatever not support event is not being
used, some sort of fallback happened.Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-w1neie2dqli89we1bzwkf4id@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
25 May, 2012
3 commits
-
The namelist array (including its content) was not freed if we fail to
realloc a new 'threads' structure.Signed-off-by: Franck Bui-Huu
Cc: David Ahern
Link: http://lkml.kernel.org/r/1337952109-31995-1-git-send-email-fbuihuu@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo -
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
|
--- malloc29.49% cc1 libc-2.15.so [.] malloc
|
--- malloc
|
|--0.95%-- 0x208eb1000000000
|
|--0.63%-- htab_traverse_noresize11.04% 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%
malloc29.49% cc1 libc-2.15.so [.] malloc
24.77%
malloc11.04% as libc-2.15.so [.] malloc
11.02%
malloc7.15% 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
... -
As:
make DEBUG=1 -C tools/perf
disables optimizations and _FORTIFY_SOURCE in recent distros requires
optimizations to be enabled, seen on a Fedora 17 system:[acme@Fedora17 linux]$ make DEBUG=1 O=/home/acme/git/build/perf/ -C
tools/perf install
In file included from /usr/include/sys/types.h:26:0,
from /usr/include/libelf.h:53,
from /usr/include/gelf.h:53,
from /usr/include/elfutils/libdw.h:53,
from :2:
/usr/include/features.h:314:4: error: #warning _FORTIFY_SOURCE requires
compiling with optimization (-O) [-Werror=cppCc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-4ccyiebqju4uatm31ky7725b@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
24 May, 2012
10 commits
-
The @type should be a type of enum event_type not enum filter_arg_type.
This fixes following warning:
$ make
COMPILE FPIC parse-events.o
COMPILE FPIC parse-filter.o
/home/namhyung/project/trace-cmd/parse-filter.c: In function ‘create_arg_item’:
/home/namhyung/project/trace-cmd/parse-filter.c:343:9: warning: comparison between ‘enum filter_arg_type’ and ‘enum event_type’ [-Wenum-compare]
/home/namhyung/project/trace-cmd/parse-filter.c:339:2: warning: case value ‘8’ not in enumerated type ‘enum filter_arg_type’ [-Wswitch]
BUILD STATIC LIB libparsevent.a
BUILD STATIC LIB libtracecmd.a
BUILD trace-cmd
/usr/bin/make -C /home/namhyung/project/trace-cmd/Documentation all
make[1]: Nothing to be done for `all'.
Note: to build the gui, type "make gui"Cc: Borislav Petkov
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337740619-27925-20-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim
Signed-off-by: Steven Rostedt
Signed-off-by: Arnaldo Carvalho de Melo -
The param needs to be updated when setting args up so that
the loop in process_defined_func() can see the correct
param->type for the farg.Cc: Borislav Petkov
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337740619-27925-15-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim
Signed-off-by: Steven Rostedt
Signed-off-by: Arnaldo Carvalho de Melo -
The @arg paremeter should not be freed inside of process_XXX(),
because it'd be freed from the caller of process_arg(). We can
free it only after it was reused for local usage.Cc: Borislav Petkov
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337740619-27925-14-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim
Signed-off-by: Steven Rostedt
Signed-off-by: Arnaldo Carvalho de Melo -
If set_op_prio() failed, the token will be freed at out_free,
then arg->op.op would turn out to be a dangle pointer. After
returning EVENT_ERROR from process_op(), free_arg() will be
called and then it will finally see the dangling pointer.Cc: Borislav Petkov
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337740619-27925-13-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim
Signed-off-by: Steven Rostedt
Signed-off-by: Arnaldo Carvalho de Melo -
If event_read_fields failed in the middle, each member of
struct format_field should be freed also.Cc: Borislav Petkov
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337740619-27925-11-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim
Signed-off-by: Steven Rostedt
Signed-off-by: Arnaldo Carvalho de Melo -
The __print_symbolic() function takes a sequence of key-value pairs for
pretty-printing a constant. The new kvm:kvm_exit print fmt uses the
expression:__print_symbolic(..., { 0x040 + 1, "DB excp" }, ...)
Currently only atoms are supported and this print fmt fails to parse.
This patch adds support for expressions instead of just atoms so that
0x040 + 1 is parsed successfully.Cc: Avi Kivity
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337740619-27925-6-git-send-email-namhyung.kim@lge.com
Signed-off-by: Namhyung Kim
Signed-off-by: Stefan Hajnoczi
Signed-off-by: Steven Rostedt
Signed-off-by: Arnaldo Carvalho de Melo -
…inux into perf/urgent
Pull a 'perf evlist' fix from Arnaldo Carvalho de Melo.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-
It was a global variable, so it was initialized, implicitely, to zero by
being placed in the bss.Now it is just a local variable that is then passed to the __cmd_evlist
routine, so it must be explicitely set to NULL.The problem manifested on a Fedora 17 system, using:
gcc version 4.7.0 20120507 (Red Hat 4.7.0-5) (GCC)
But not on several other systems, by luck.
Reported-by: Ingo Molnar
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-5e8wolcjs3rgd5i6yi995gfh@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Pull trivial ktest spelling fix from Steven Rostedt:
"I promised Jesper that I would push this for 3.5, but forgot to add it
to my queue. It's just a spelling fix, but it should go in regardless
to hide my inability to get words in the English language correct."Becuse gud spealing is impurtunt.
* tag 'ktest-v3.5-spelling' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest:
ktest: Change singular "paranthesis" to plural "parentheses" -
…rnel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
- Leftover AMD PMU driver fix fix from the end of the v3.4
stabilization cycle.- Late tools/perf/ changes that missed the first round:
* endianness fixes
* event parsing improvements
* libtraceevent fixes factored out from trace-cmd
* perl scripting engine fixes related to libtraceevent,
* testcase improvements
* perf inject / pipe mode fixes
* plus a kernel side fix* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86: Update event scheduling constraints for AMD family 15h models* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
Revert "sched, perf: Use a single callback into the scheduler"
perf evlist: Show event attribute details
perf tools: Bump default sample freq to 4 kHz
perf buildid-list: Work better with pipe mode
perf tools: Fix piped mode read code
perf inject: Fix broken perf inject -b
perf tools: rename HEADER_TRACE_INFO to HEADER_TRACING_DATA
perf tools: Add union u64_swap type for swapping u64 data
perf tools: Carry perf_event_attr bitfield throught different endians
perf record: Fix documentation for branch stack sampling
perf target: Add cpu flag to sample_type if target has cpu
perf tools: Always try to build libtraceevent
perf tools: Rename libparsevent to libtraceevent in Makefile
perf script: Rename struct event to struct event_format in perl engine
perf script: Explicitly handle known default print arg type
perf tools: Add hardcoded name term for pmu events
perf tools: Separate 'mem:' event scanner bits
perf tools: Use allocated list for each parsed event
perf tools: Add support for displaying event parser debug info
perf test: Move parse event automated tests to separated object
23 May, 2012
7 commits
-
Acked-by: Randy Dunlap
Signed-off-by: Jesper Juhl
Signed-off-by: Steven Rostedt -
Pull scheduler changes from Ingo Molnar:
"The biggest change is the cleanup/simplification of the load-balancer:
instead of the current practice of architectures twiddling scheduler
internal data structures and providing the scheduler domains in
colorfully inconsistent ways, we now have generic scheduler code in
kernel/sched/core.c:sched_init_numa() that looks at the architecture's
node_distance() parameters and (while not fully trusting it) deducts a
NUMA topology from it.This inevitably changes balancing behavior - hopefully for the better.
There are various smaller optimizations, cleanups and fixlets as well"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched: Taint kernel with TAINT_WARN after sleep-in-atomic bug
sched: Remove stale power aware scheduling remnants and dysfunctional knobs
sched/debug: Fix printing large integers on 32-bit platforms
sched/fair: Improve the ->group_imb logic
sched/nohz: Fix rq->cpu_load[] calculations
sched/numa: Don't scale the imbalance
sched/fair: Revert sched-domain iteration breakage
sched/x86: Rewrite set_cpu_sibling_map()
sched/numa: Fix the new NUMA topology bits
sched/numa: Rewrite the CONFIG_NUMA sched domain support
sched/fair: Propagate 'struct lb_env' usage into find_busiest_group
sched/fair: Add some serialization to the sched_domain load-balance walk
sched/fair: Let minimally loaded cpu balance the group
sched: Change rq->nr_running to unsigned int
x86/numa: Check for nonsensical topologies on real hw as well
x86/numa: Hard partition cpu topology masks on node boundaries
x86/numa: Allow specifying node_distance() for numa=fake
x86/sched: Make mwait_usable() heed to "idle=" kernel parameters properly
sched: Update documentation and comments
sched_rt: Avoid unnecessary dequeue and enqueue of pushable tasks in set_cpus_allowed_rt() -
Pull perf changes from Ingo Molnar:
"Lots of changes:- (much) improved assembly annotation support in perf report, with
jump visualization, searching, navigation, visual output
improvements and more.- kernel support for AMD IBS PMU hardware features. Notably 'perf
record -e cycles:p' and 'perf top -e cycles:p' should work without
skid now, like PEBS does on the Intel side, because it takes
advantage of IBS transparently.- the libtracevents library: it is the first step towards unifying
tracing tooling and perf, and it also gives a tracing library for
external tools like powertop to rely on.- infrastructure: various improvements and refactoring of the UI
modules and related code- infrastructure: cleanup and simplification of the profiling
targets code (--uid, --pid, --tid, --cpu, --all-cpus, etc.)- tons of robustness fixes all around
- various ftrace updates: speedups, cleanups, robustness
improvements.- typing 'make' in tools/ will now give you a menu of projects to
build and a short help text to explain what each does.- ... and lots of other changes I forgot to list.
The perf record make bzImage + perf report regression you reported
should be fixed."* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (166 commits)
tracing: Remove kernel_lock annotations
tracing: Fix initial buffer_size_kb state
ring-buffer: Merge separate resize loops
perf evsel: Create events initially disabled -- again
perf tools: Split term type into value type and term type
perf hists: Fix callchain ip printf format
perf target: Add uses_mmap field
ftrace: Remove selecting FRAME_POINTER with FUNCTION_TRACER
ftrace/x86: Have x86 ftrace use the ftrace_modify_all_code()
ftrace: Make ftrace_modify_all_code() global for archs to use
ftrace: Return record ip addr for ftrace_location()
ftrace: Consolidate ftrace_location() and ftrace_text_reserved()
ftrace: Speed up search by skipping pages by address
ftrace: Remove extra helper functions
ftrace: Sort all function addresses, not just per page
tracing: change CPU ring buffer state from tracing_cpumask
tracing: Check return value of tracing_dentry_percpu()
ring-buffer: Reset head page before running self test
ring-buffer: Add integrity check at end of iter read
ring-buffer: Make addition of pages in ring buffer atomic
... -
Pull USB 3.5-rc1 changes from Greg Kroah-Hartman:
"Here is the big USB 3.5-rc1 pull request for the 3.5-rc1 merge window.It's touches a lot of different parts of the kernel, all USB drivers,
due to some API cleanups (getting rid of the ancient err() macro) and
some changes that are needed for USB 3.0 power management updates.There are also lots of new drivers, pimarily gadget, but others as
well. We deleted a staging driver, which was nice, and finally
dropped the obsolete usbfs code, which will make Al happy to never
have to touch that again.There were some build errors in the tree that linux-next found a few
days ago, but those were fixed by the most recent changes (all were
due to us not building with CONFIG_PM disabled.)Signed-off-by: Greg Kroah-Hartman "
* tag 'usb-3.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (477 commits)
xhci: Fix DIV_ROUND_UP compile error.
xhci: Fix compile with CONFIG_USB_SUSPEND=n
USB: Fix core compile with CONFIG_USB_SUSPEND=n
brcm80211: Fix compile error for .disable_hub_initiated_lpm.
Revert "USB: EHCI: work around bug in the Philips ISP1562 controller"
MAINTAINERS: Add myself as maintainer to the USB PHY Layer
USB: EHCI: fix command register configuration lost problem
USB: Remove races in devio.c
USB: ehci-platform: remove update_device
USB: Disable hub-initiated LPM for comms devices.
xhci: Add Intel U1/U2 timeout policy.
xhci: Add infrastructure for host-specific LPM policies.
USB: Add macros for interrupt endpoint types.
xhci: Reserve one command for USB3 LPM disable.
xhci: Some Evaluate Context commands must succeed.
USB: Disable USB 3.0 LPM in critical sections.
USB: Add support to enable/disable USB3 link states.
USB: Allow drivers to disable hub-initiated LPM.
USB: Calculate USB 3.0 exit latencies for LPM.
USB: Refactor code to set LPM support flag.
...Conflicts:
arch/arm/mach-exynos/mach-nuri.c
arch/arm/mach-exynos/mach-universal_c210.c
drivers/net/wireless/ath/ath6kl/usb.c -
There was no easy way to see the frequency used, and with the change of
default, we better provide one.[root@sandy linux]# perf evlist -F
cycles: sample_freq=4000
[root@sandy linux]# perf evlist -v
cycles: sample_freq=4000, size: 80, sample_type: 391, read_format: 7, disabled: 1, inherit: 1, mmap: 1, comm: 1, freq: 1, sample_id_all: 1, exclude_guest: 1
[root@sandy linux]#Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-e1p9poez3nwrgycbmwqmhlsu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Quoting Ingo:
"While at it I'd also suggest increasing the default sampling frequency,
from 1000 Hz per CPU to at least 4Khz auto-freq or so - this should work
well all across the board I think. CPUs are getting faster and command/app
run times are getting shorter, 1Khz is a bit low IMO."Requested-by: Ingo Molnar
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Stephane Eranian
Link: http://lkml.kernel.org/n/tip-2jafa6mkrufyekny9ei59lpu@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
In order for perf buildid-list to work with pipe-mode files, it needs to
process buildids and event attr structs.$ perf record -o - noploop 2 | ./perf inject -b | perf buildid-list -i - -H
noploop for 2 seconds
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.084 MB - (~3678 samples) ]
0000000000000000000000000000000000000000 [kernel.kallsyms]
3a0d0629efe74a8da3eeba372cdbd74ad9b8f5d5 /usr/local/bin/noploopThe reason [kernel.kallsyms] shows a 0 build-id comes from the
way buildids are injected in the stream.The buildid for the kernel is provided by a BUILD_ID record. The
[kernel.kallsyms] is provided by a MMAP record. There is no clean and
obvious way to link the two, unfortunately.In regular mode, the kernel buildid is generated from reading the ELF
image or kallsyms and perf knows to associate [kernel.kallsyms] to it.
Later on, when perf processes the [kernel.kallsyms] MMAP record, it will
already have a dso for it.So for now, make sure perf buildid-list shows the buildids for
everything but the kernel image.Signed-off-by: Stephane Eranian
Cc: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337081295-10303-6-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo
22 May, 2012
2 commits
-
In __perf_session__process_pipe_events(), there was a risk we would read
more than what a union perf_event struct can hold. this could happen in
case, perf is reading a file which contains new record types it does not
know about and which are larger than anything it knows about.In general, perf is supposed to skip records it does not understand, but
in pipe mode, those have to be read and ignored. The fixed size header
contains the size of the record, but that size may be larger than union
perf_event, yet it was used as the backing to the read in:union perf_event event;
void *p;size = event->header.size;
p = &event;
p += sizeof(struct perf_event_header);
if (size - sizeof(struct perf_event_header)) {
err = readn(self->fd, p, size - sizeof(struct perf_event_header));We fix this by allocating a buffer based on the size reported in the
header. We reuse the buffer as much as we can. We realloc in case it
becomes too small. In the common case, the performance impact is
negligible.Signed-off-by: Stephane Eranian
Cc: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337081295-10303-3-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
perf inject -b was broken. It would not inject any build_id into the
stream. Furthermore, it would strip samples from the stream.The reason was a missing initialization of the event attribute
structure. The perf_tool.tool.attr() callback was pointing to a simple
repipe. But there was no initialization of the internal data structures
to keep track of events and event ids. That later caused event id
lookups to fail, and sample would get removed.The patch simply adds back the call to perf_event__process_attr() to
initialize the evlist structure and now build_ids are again injected.Signed-off-by: Stephane Eranian
Cc: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1337081295-10303-2-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo