Pull perf fixes from Ingo Molnar:
"Two hw-enablement patches, two race fixes, three fixes for regressions
of semantics, plus a number of tooling fixes"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel: Add proper condition to run sched_task callbacks
perf/core: Fix locking for children siblings group read
perf/core: Fix scheduling regression of pinned groups
perf/x86/intel: Fix debug_store reset field for freq events
perf/x86/intel: Add Goldmont Plus CPU PMU support
perf/x86/intel: Enable C-state residency events for Apollo Lake
perf symbols: Accept zero as the kernel base address
Revert "perf/core: Drop kernel samples even though :u is specified"
perf annotate: Fix broken arrow at row 0 connecting jmp instruction to its target
perf evsel: State in the default event name if attr.exclude_kernel is set
perf evsel: Fix attr.exclude_kernel setting for default cycles:p
__GFP_REPEAT was designed to allow retry-but-eventually-fail semantic to
the page allocator. This has been true but only for allocations
requests larger than PAGE_ALLOC_COSTLY_ORDER. It has been always
ignored for smaller sizes. This is a bit unfortunate because there is
no way to express the same semantic for those requests and they are
considered too important to fail so they might end up looping in the
page allocator for ever, similarly to GFP_NOFAIL requests.
Now that the whole tree has been cleaned up and accidental or misled
usage of __GFP_REPEAT flag has been removed for !costly requests we can
give the original flag a better name and more importantly a more useful
semantic. Let's rename it to __GFP_RETRY_MAYFAIL which tells the user
that the allocator would try really hard but there is no promise of a
success. This will work independent of the order and overrides the
default allocator behavior. Page allocator users have several levels of
guarantee vs. cost options (take GFP_KERNEL as an example)
- GFP_KERNEL & ~__GFP_RECLAIM - optimistic allocation without _any_
attempt to free memory at all. The most light weight mode which even
doesn't kick the background reclaim. Should be used carefully because
it might deplete the memory and the next user might hit the more
aggressive reclaim
- GFP_KERNEL & ~__GFP_DIRECT_RECLAIM (or GFP_NOWAIT)- optimistic
allocation without any attempt to free memory from the current
context but can wake kswapd to reclaim memory if the zone is below
the low watermark. Can be used from either atomic contexts or when
the request is a performance optimization and there is another
fallback for a slow path.
- (GFP_KERNEL|__GFP_HIGH) & ~__GFP_DIRECT_RECLAIM (aka GFP_ATOMIC) -
non sleeping allocation with an expensive fallback so it can access
some portion of memory reserves. Usually used from interrupt/bh
context with an expensive slow path fallback.
- GFP_KERNEL - both background and direct reclaim are allowed and the
_default_ page allocator behavior is used. That means that !costly
allocation requests are basically nofail but there is no guarantee of
that behavior so failures have to be checked properly by callers
(e.g. OOM killer victim is allowed to fail currently).
- GFP_KERNEL | __GFP_NORETRY - overrides the default allocator behavior
and all allocation requests fail early rather than cause disruptive
reclaim (one round of reclaim in this implementation). The OOM killer
is not invoked.
- GFP_KERNEL | __GFP_RETRY_MAYFAIL - overrides the default allocator
behavior and all allocation requests try really hard. The request
will fail if the reclaim cannot make any progress. The OOM killer
won't be triggered.
- GFP_KERNEL | __GFP_NOFAIL - overrides the default allocator behavior
and all allocation requests will loop endlessly until they succeed.
This might be really dangerous especially for larger orders.
Existing users of __GFP_REPEAT are changed to __GFP_RETRY_MAYFAIL
because they already had their semantic. No new users are added.
__alloc_pages_slowpath is changed to bail out for __GFP_RETRY_MAYFAIL if
there is no progress and we have already passed the OOM point.
This means that all the reclaim opportunities have been exhausted except
the most disruptive one (the OOM killer) and a user defined fallback
behavior is more sensible than keep retrying in the page allocator.
[akpm@linux-foundation.org: fix arch/sparc/kernel/mdesc.c]
[mhocko@suse.com: semantic fix]
Link: http://lkml.kernel.org/r/20170626123847.GM11534@dhcp22.suse.cz
[mhocko@kernel.org: address other thing spotted by Vlastimil]
Link: http://lkml.kernel.org/r/20170626124233.GN11534@dhcp22.suse.cz
Link: http://lkml.kernel.org/r/20170623085345.11304-3-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Alex Belits <alex.belits@cavium.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: David Daney <david.daney@cavium.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: NeilBrown <neilb@suse.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Which is the case in S/390, where symbols were not being resolved
because machine__get_kernel_start was only setting machine->kernel_start
when the just successfully loaded kernel symtab had its map->start set
to !0, when it was left at (1ULL << 63) assuming a partitioning of the
address space for user/kernel, which is not the case in S/390 nor in
Sparc.
So just check if map__load() was successfull and set
machine->kernel_start to zero, fixing kernel symbol resolution on S/390.
Test performed by Thomas:
----
I like this patch. I have done a new build and removed all my debug output to start
from scratch. Without your patch I get this:
# Samples: 4 of event 'cpu-clock'
# Event count (approx.): 1000000
#
# Children Self Command Shared Object Symbol
# ........ ........ ....... ................ ........................
75.00% 0.00% true [unknown] [k] 0x00000000004bedda
|
---0x4bedda
|
|--50.00%--0x42693a
| |
| --25.00%--0x2a72e0
| 0x2af0ca
| 0x3d1003fe4c0
|
--25.00%--0x4272bc
0x26fa84
and with your patch (I just rebuilt the perf tool, nothing else and used the same
perf.data file as input):
# Samples: 4 of event 'cpu-clock'
# Event count (approx.): 1000000
#
# Children Self Command Shared Object Symbol
# ........ ........ ....... .......................... ..................................
75.00% 0.00% true [kernel.vmlinux] [k] pgm_check_handler
|
---pgm_check_handler
do_dat_exception
handle_mm_fault
__handle_mm_fault
filemap_map_pages
|
|--25.00%--rcu_read_lock_held
| rcu_lockdep_current_cpu_online
| 0x3d1003ff4c0
|
--25.00%--lock_release
Looks good to me....
----
Reported-and-Tested-by: Thomas-Mich Richter <tmricht@linux.vnet.ibm.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Zvonko Kosic <zvonko.kosic@de.ibm.com>
Link: http://lkml.kernel.org/n/tip-dk0n1uzmbe0tbthrpfqlx6bz@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When no event is specified perf will use the "cycles" hardware event
with the highest precision available in the processor, and excluding
kernel events for non-root users, so make that clear in the event name
by setting the "u" event modifier, i.e. "cycles:upp".
E.g.:
The default for root:
# perf record usleep 1
# perf evlist -v
cycles:ppp: ..., precise_ip: 3, exclude_kernel: 0, ...
#
And for !root:
$ perf record usleep 1
$ perf evlist -v
cycles:uppp: ... , precise_ip: 3, exclude_kernel: 1, ...
$
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-lf29zcdl422i9knrgde0uwy3@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull perf/urgent fixes from Arnaldo Carvalho de Melo:
User visible changes:
- Fix max attr.precise_ip probing to make perf use the best cycles:p
available in the processor for non root users (Arnaldo Carvalho de Melo)
- Fix processing of MMAP events for 32-bit binaries on 64-bit systems
when unwind support is not fully integrated, fixing DSO and symbol
resolution (Jiri Olsa)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We currently fail the MMAP event processing if we don't have the MMAP
event's specific arch unwind support compiled in.
That's wrong and can lead to unresolved mmaps in report output for 32bit
binaries on 64bit server, like in this example on x86_64 server:
$ cat ex.c
int main(int argc, char **argv)
{
while (1) {}
}
$ gcc -o ex -m32 ex.c
$ perf record ./ex
^C[ perf record: Woken up 2 times to write data ]
[ perf record: Captured and wrote 0.371 MB perf.data (9322 samples) ]
Before:
$ perf report --stdio
SNIP
# Overhead Command Shared Object Symbol
# ........ ....... ................ ......................
#
100.00% ex [unknown] [.] 0x00000000080483de
0.00% ex [unknown] [.] 0x00000000f76dba4f
0.00% ex [unknown] [.] 0x00000000f76e4c11
0.00% ex [unknown] [.] 0x00000000f76daa30
After:
$ perf report --stdio
SNIP
# Overhead Command Shared Object Symbol
# ........ ....... ............. ...............
#
100.00% ex ex [.] main
0.00% ex ld-2.24.so [.] _dl_start
0.00% ex ld-2.24.so [.] do_lookup_x
0.00% ex ld-2.24.so [.] _start
The fix is not to fail, just warn if there's not unwind support compiled
in.
Reported-by: Michael Lyle <mlyle@lyle.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20170704131131.27508-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
We should set attr.exclude_kernel when probing for attr.precise_ip
level, otherwise !CAP_SYS_ADMIN users will not default to skidless
samples in capable hardware.
The increase in the paranoid level in commit 0161028b7c ("perf/core:
Change the default paranoia level to 2") broke this, fix it by excluding
kernel samples when probing.
Before:
$ perf record usleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.018 MB perf.data (6 samples) ]
$ perf evlist -v
cycles:u: sample_freq: 4000, sample_type: IP|TID|TIME|PERIOD, exclude_kernel: 1
After:
$ perf record usleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ]
$ perf evlist -v
cycles:ppp: sample_freq: 4000, sample_type: IP|TID|TIME|PERIOD, exclude_kernel: 1, precise_ip: 3
^^^^^^^^^^^^^
^^^^^^^^^^^^^
^^^^^^^^^^^^^
$
To further clarify: we always set .exclude_kernel when non !CAP_SYS_ADMIN
users profile, its just on the attr.precise_ip probing that we weren't doing
so, fix it.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: 7f8d1ade1b ("perf tools: By default use the most precise "cycles" hw counter available")
Link: http://lkml.kernel.org/n/tip-t2qttwhbnua62o5gt75cueml@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull perf updates from Ingo Molnar:
"Most of the changes are for tooling, the main changes in this cycle were:
- Improve Intel-PT hardware tracing support, both on the kernel and
on the tooling side: PTWRITE instruction support, power events for
C-state tracing, etc. (Adrian Hunter)
- Add support to measure SMI cost to the x86 architecture, with
tooling support in 'perf stat' (Kan Liang)
- Support function filtering in 'perf ftrace', plus related
improvements (Namhyung Kim)
- Allow adding and removing fields to the default 'perf script'
columns, using + or - as field prefixes to do so (Andi Kleen)
- Allow resolving the DSO name with 'perf script -F brstack{sym,off},dso'
(Mark Santaniello)
- Add perf tooling unwind support for PowerPC (Paolo Bonzini)
- ... and various other improvements as well"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (84 commits)
perf auxtrace: Add CPU filter support
perf intel-pt: Do not use TSC packets for calculating CPU cycles to TSC
perf intel-pt: Update documentation to include new ptwrite and power events
perf intel-pt: Add example script for power events and PTWRITE
perf intel-pt: Synthesize new power and "ptwrite" events
perf intel-pt: Move code in intel_pt_synth_events() to simplify attr setting
perf intel-pt: Factor out intel_pt_set_event_name()
perf intel-pt: Tidy messages into called function intel_pt_synth_event()
perf intel-pt: Tidy Intel PT evsel lookup into separate function
perf intel-pt: Join needlessly wrapped lines
perf intel-pt: Remove unused instructions_sample_period
perf intel-pt: Factor out common code synthesizing event samples
perf script: Add synthesized Intel PT power and ptwrite events
perf/x86/intel: Constify the 'lbr_desc[]' array and make a function static
perf script: Add 'synth' field for synthesized event payloads
perf auxtrace: Add itrace option to output power events
perf auxtrace: Add itrace option to output ptwrite events
tools include: Add byte-swapping macros to kernel.h
perf script: Add 'synth' event type for synthesized events
x86/insn: perf tools: Add new ptwrite instruction
...
CBR (core-to-bus ratio) packets provide an indication of CPU frequency. A
more accurate measure can be made by counting the cycles (given by CYC
packets) in between other timing packets (either MTC or TSC). Using TSC
packets has at least 2 issues: 1) timing might have stopped (e.g. mwait) or
2) TSC packets within PSB+ might slip past CYC packets. For now, simply do
not use TSC packets for calculating CPU cycles to TSC. That leaves the case
where 2 MTC packets are used, otherwise falling back to the CBR value.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1495786658-18063-37-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Synthesize new power and ptwrite events.
Power events report changes to C-state but I have also added support
for the existing CBR (core-to-bus ratio) packet and included that
when outputting power events.
The PTWRITE packet is associated with the new "ptwrite" instruction,
which is essentially just a way to stuff a 32 or 64 bit value into the
PT trace.
More details can be found in the patches that add documentation and in
the Intel SDM.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1498811805-2335-1-git-send-email-adrian.hunter@intel.com
[ Copy the description of such packet from the patchkit cover message ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
