Mathieu Poirier [Wed, 5 Jun 2019 16:16:33 +0000 (10:16 -0600)]
perf cs-etm: Properly set the value of 'old' and 'head' in snapshot mode
This patch adds the necessary intelligence to properly compute the value
of 'old' and 'head' when operating in snapshot mode. That way we can
get the latest information in the AUX buffer and be compatible with the
generic AUX ring buffer mechanic.
Tester notes:
> Leo, have you had the chance to test/review this one? Suzuki?
Sure. I applied this patch on the perf/core branch (with latest
commit 3e4fbf36c1e3 'perf augmented_raw_syscalls: Move reading
filename to the loop') and passed testing with below steps:
# perf record -e cs_etm/@tmc_etr0/ -S -m,64 --per-thread ./sort &
[1] 19097
Bubble sorting array of 30000 elements
# kill -USR2 19097
# kill -USR2 19097
# kill -USR2 19097
[ perf record: Woken up 4 times to write data ]
[ perf record: Captured and wrote 0.753 MB perf.data ]
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190605161633.12245-1-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf data: Fix perf.data documentation for HEADER_CPU_TOPOLOGY
The 'die' info isn't in the same array as core and socket ids, and we
missed the 'dies' string list, that comes right after the 'core' +
'socket' id variable length array, followed by the VLA for the dies.
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: c9cb12c5ba08 ("perf header: Add die information in CPU topology") Link: https://lkml.kernel.org/n/tip-nubi6mxp2n8ofvlx7ph6k3h6@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Kan Liang [Tue, 4 Jun 2019 22:50:42 +0000 (15:50 -0700)]
perf stat: Support per-die aggregation
It is useful to aggregate counts per die. E.g. Uncore becomes die-scope
on Xeon Cascade Lake-AP.
Introduce a new option "--per-die" to support per-die aggregation.
The global id for each core has been changed to socket + die id + core
id. The global id for each die is socket + die id.
Add die information for per-core aggregation. The output of per-core
aggregation will be changed from "S0-C0" to "S0-D0-C0". Any scripts
which rely on the output format of per-core aggregation probably be
broken.
For 'perf stat record/report', there is no die information when
processing the old perf.data. The per-die result will be the same as
per-socket.
Committer notes:
Renamed 'die' variable to 'die_id' to fix the build in some systems:
CC /tmp/build/perf/builtin-script.o
cc1: warnings being treated as errors
builtin-stat.c: In function 'perf_env__get_die':
builtin-stat.c:963: error: declaration of 'die' shadows a global declaration
util/util.h:19: error: shadowed declaration is here
mv: cannot stat `/tmp/build/perf/.builtin-stat.o.tmp': No such file or directory
Signed-off-by: Kan Liang <kan.liang@linux.intel.com> Reviewed-by: Jiri Olsa <jolsa@kernel.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lkml.kernel.org/n/tip-bsnhx7vgsuu6ei307mw60mbj@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Kan Liang [Tue, 4 Jun 2019 22:50:41 +0000 (15:50 -0700)]
perf header: Add die information in CPU topology
With the new CPUID.1F, a new level type of CPU topology, 'die', is
introduced. The 'die' information in CPU topology should be added in
perf header.
To be compatible with old perf.data, the patch checks the section size
before reading the die information. The new info is added at the end of
the cpu_topology section, the old perf tool ignores the extra data. It
never reads data crossing the section boundary.
The new perf tool with the patch can be used on legacy kernel. Add a new
function has_die_topology() to check if die topology information is
supported by kernel. The function only check X86 and CPU 0. Assuming
other CPUs have same topology.
Use similar method for core and socket to support die id and sibling
dies string.
Mathieu Poirier [Fri, 24 May 2019 17:35:08 +0000 (11:35 -0600)]
perf cs-etm: Add support for CPU-wide trace scenarios
Add support for CPU-wide trace scenarios by correlating range packets
with timestamp packets. That way range packets received on different
ETMQ/traceID channels can be processed and synthesized in chronological
order.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-18-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:35:07 +0000 (11:35 -0600)]
perf cs-etm: Add notion of time to decoding code
This patch deals with timestamp packets received from the decoding
library in order to give the front end packet processing loop a handle
on the time instruction conveyed by range packets have been executed at.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-17-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:35:05 +0000 (11:35 -0600)]
perf cs-etm: Add support for multiple traceID queues
When operating in CPU-wide trace mode with a source/sink topology of N:1
packets with multiple traceID will end up in the same cs_etm_queue. In
order to properly decode packets they need to be split in different
queues, i.e one queue per traceID.
As such add support for multiple traceID per cs_etm_queue by adding a
new cs_etm_traceid_queue every time a new traceID is discovered in the
trace stream.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-15-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:35:04 +0000 (11:35 -0600)]
perf cs-etm: Use traceID aware memory callback API
When working with CPU-wide traces different traceID may be found in the
same stream. As such we need to use the decoder callback that provides
the traceID in order to know the thread context being decoded.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-14-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:35:03 +0000 (11:35 -0600)]
perf cs-etm: Move tid/pid to traceid_queue
The tid/pid fields of structure cs_etm_queue are CPU dependent and as
such need to be part of the cs_etm_traceid_queue in order to support
CPU-wide trace scenarios.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-13-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:35:02 +0000 (11:35 -0600)]
perf cs-etm: Move thread to traceid_queue
The thread field of structure cs_etm_queue is CPU dependent and as such
need to be part of the cs_etm_traceid_queue in order to support CPU-wide
trace scenarios.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-12-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:35:00 +0000 (11:35 -0600)]
perf cs-etm: Introduce the concept of trace ID queues
In an ideal world there is one CPU per cs_etm_queue and as such, one
trace ID per cs_etm_queue. In the real world CoreSight topologies allow
multiple CPUs to use the same sink, which translates to multiple trace
IDs per cs_etm_queue.
To deal with this a new cs_etm_traceid_queue structure is introduced to
enclose all the information related to a single trace ID, allowing a
cs_etm_queue to handle traces generated by any number of CPUs.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-10-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:34:58 +0000 (11:34 -0600)]
perf cs-etm: Move packet queue out of decoder structure
The decoder needs to work with more than one traceID queue if we want to
support CPU-wide scenarios with N:1 source/sink topologies. As such
move the packet buffer and related fields out of the decoder structure
and into the cs_etm_queue structure.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-8-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:34:57 +0000 (11:34 -0600)]
perf cs-etm: Refactor error path in cs_etm_decoder__new()
There is no point in having two different error goto statement since the
openCSD API to free a decoder handles NULL pointers. As such function
cs_etm_decoder__free() can be called to deal with all aspect of freeing
decoder memory.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-7-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:34:56 +0000 (11:34 -0600)]
perf cs-etm: Add handling of switch-CPU-wide events
Add handling of SWITCH-CPU-WIDE events in order to add the tid/pid of
the incoming process to the perf tools machine infrastructure. This
information is later retrieved when a contextID packet is found in the
trace stream.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-6-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:34:55 +0000 (11:34 -0600)]
perf cs-etm: Add handling of itrace start events
Add handling of ITRACE events in order to add the tid/pid of the
executing process to the perf tools machine infrastructure. This
information is later retrieved when a contextID packet is found in the
trace stream.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-5-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:34:54 +0000 (11:34 -0600)]
perf cs-etm: Configure SWITCH_EVENTS in CPU-wide mode
Ask the perf core to generate an event when processes are swapped in/out
of context. That way proper action can be taken by the decoding code
when faced with such event.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-4-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Mathieu Poirier [Fri, 24 May 2019 17:34:53 +0000 (11:34 -0600)]
perf cs-etm: Configure timestamp generation in CPU-wide mode
When operating in CPU-wide mode tracers need to generate timestamps in
order to correlate the code being traced on one CPU with what is executed
on other CPUs.
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org> Tested-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/20190524173508.29044-3-mathieu.poirier@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Jiri Olsa [Fri, 17 May 2019 11:33:47 +0000 (13:33 +0200)]
perf evsel: Remove superfluous nthreads system_wide setup in alloc_fd()
It's already setup in the only caller of this method in
perf_evsel__open(), right before calling perf_evsel__alloc_fd(), no need
to do it again.
Also it's better to have it out of the function before we move it to
libperf.
Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-1k8lhyjxfk7o8v4g3r7eyjc9@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
yuzhoujian [Thu, 30 May 2019 13:29:22 +0000 (14:29 +0100)]
perf record: Add support to collect callchains from kernel or user space only
One can just record callchains in the kernel or user space with this new
options.
We can use it together with "--all-kernel" options.
This two options is used just like print_stack(sys) or print_ustack(usr)
for systemtap.
Shown below is the usage of this new option combined with "--all-kernel"
options:
1. Configure all used events to run in kernel space and just collect
kernel callchains.
$ perf record -a -g --all-kernel --kernel-callchains
2. Configure all used events to run in kernel space and just collect
user callchains.
$ perf record -a -g --all-kernel --user-callchains
Committer notes:
Improved documentation to state that asking for kernel callchains really
is asking for excluding user callchains, and vice versa.
Further mentioned that using both won't get both, but nothing, as both
will be excluded.
Signed-off-by: yuzhoujian <yuzhoujian@didichuxing.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Milian Wolff <milian.wolff@kdab.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/1559222962-22891-1-git-send-email-ufo19890607@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf config: Bail out when a handler returns failure for a key-value pair
So perf_config() uses:
int ret = 0;
perf_config_set__for_each_entry(config_set, section, item) {
...
ret = fn();
if (ret < 0)
break;
}
return ret;
Expecting that that break will imediatelly go to function exit to return
that error value (ret).
The problem is that perf_config_set__for_each_entry() expands into two
nested for() loops, one traversing the sections in a config and the
second the items in each of those sections, so we have to change that
'break' to a goto label right before that final 'return ret'.
With that, for instance 'perf trace' now correctly bails out when a
event that is requested to be added via its 'trace.add_events'
~/.perfconfig entry gets rejected by the kernel BPF verifier:
# perf trace ls
event syntax error: '/home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.o'
\___ Kernel verifier blocks program loading
(add -v to see detail)
Run 'perf list' for a list of valid events
Error: wrong config key-value pair trace.add_events=/home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.o
#
While before it would continue and explode later, when trying to find
maps that would have been in place had that augmented_raw_syscalls.o
precompiled BPF proggie been accepted by the, humm, bast... rigorous
kernel BPF verifier 8-)
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Martin KaFai Lau <kafai@fb.com> Cc: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Song Liu <songliubraving@fb.com> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: Yonghong Song <yhs@fb.com> Fixes: 8a0a9c7e9146 ("perf config: Introduce new init() and exit()") Link: https://lkml.kernel.org/n/tip-qvqxfk9d0rn1l7lcntwiezrr@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Leo Yan [Thu, 6 Jun 2019 13:38:59 +0000 (10:38 -0300)]
perf trace: Exit when failing to build eBPF program
On my Juno board with ARM64 CPUs, perf trace command reports the eBPF
program building failure but the command will not exit and continue to
run. If we define an eBPF event in config file, the event will be
parsed with below flow:
Though the low level functions return back error values when detect eBPF
building failure, but parse_events_option() returns 1 for this case and
trace__config() passes 1 to perf_config(); perf_config() doesn't treat
the returned value 1 as failure and it continues to parse other
configurations. Thus the perf command continues to run even without
enabling eBPF event successfully.
This patch changes error handling in trace__config(), when it detects
failure it will return -1 rather than directly pass error value (1);
finally, perf_config() will directly bail out and perf will exit for
this case.
Committer notes:
Simplified the patch to just check directly the return of
parse_events_option() and it it is non-zero, change err from its initial
zero value to -1.
Signed-off-by: Leo Yan <leo.yan@linaro.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Martin KaFai Lau <kafai@fb.com> Cc: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Song Liu <songliubraving@fb.com> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: Yonghong Song <yhs@fb.com> Fixes: ac96287cae08 ("perf trace: Allow specifying a set of events to add in perfconfig") Link: https://lkml.kernel.org/n/tip-x4i63f5kscykfok0hqim3zma@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf trace: Consume the augmented_raw_syscalls payload
To support the SCA_FILENAME beautifier in more than one syscall arg, as
needed for syscalls such as the rename* family, we need to, after
processing one such arg, bump the augmented pointers so that the next
augmented arg don't reuse data for the previous augmented arguments.
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-4e4cmzyjxb3wkonfo1x9a27y@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Jiri Olsa [Fri, 31 May 2019 13:13:21 +0000 (15:13 +0200)]
perf jvmti: Address gcc string overflow warning for strncpy()
We are getting false positive gcc warning when we compile with gcc9 (9.1.1):
CC jvmti/libjvmti.o
In file included from /usr/include/string.h:494,
from jvmti/libjvmti.c:5:
In function ‘strncpy’,
inlined from ‘copy_class_filename.constprop’ at jvmti/libjvmti.c:166:3:
/usr/include/bits/string_fortified.h:106:10: error: ‘__builtin_strncpy’ specified bound depends on the length of the source argument [-Werror=stringop-overflow=]
106 | return __builtin___strncpy_chk (__dest, __src, __len, __bos (__dest));
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
jvmti/libjvmti.c: In function ‘copy_class_filename.constprop’:
jvmti/libjvmti.c:165:26: note: length computed here
165 | size_t file_name_len = strlen(file_name);
| ^~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
As per Arnaldo's suggestion use strlcpy(), which does the same thing and keeps
gcc silent.
Suggested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ben Gainey <ben.gainey@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/20190531131321.GB1281@krava Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf augmented_raw_syscalls: Move reading filename to the loop
Almost there, next step is to copy more than one filename payload.
Probably to read syscall arg structs, etc we'll need just a variation of
this that will decide what to use, if probe_read_str() or plain
probe_read for structs, i.e. fixed size.
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-uf6u0pld6xe4xuo16f04owlz@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf augmented_raw_syscalls: Change helper to consider just the augmented_filename part
So that we can use it for multiple args, baby steps not to step into the
verifier toes.
In the process make sure we handle -EFAULT from bpf_prog_read_str(), as
this really is needed now that we'll handle more than one augmented
argument, i.e. if there is failure, then we have the argument that fails
have:
(size = 0, err = -EFAULT, value = [] )
followed by the next, lets say that worked for a second pathname:
(size = 4, err = 0, value = "/tmp" )
So we can skip the first while telling the user about the problem and
then process the second.
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-deyvqi39um6gp6hux6jovos8@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf augmented_raw_syscalls: Tell which args are filenames and how many bytes to copy
Since we know what args are strings from reading the syscall
descriptions in tracefs and also already mark such args to be beautified
using the syscall_arg__scnprintf_filename() helper, all we need is to
fill in this info in the 'syscalls' BPF map we were using to state which
syscalls the user is interested in, i.e. the syscall filter.
Right now just set that with PATH_MAX and unroll the syscall arg in the
BPF program, as the verifier isn't liking something clang generates when
unrolling the loop.
This also makes the augmented_raw_syscalls.c program support all arches,
since we removed that set of defines with the hard coded syscall
numbers, all should be automatically set for all arches, with the
syscall id mapping done correcly.
Doing baby steps here, i.e. just the first string arg for a syscall is
printed, syscalls with more than one, say, the various rename* syscalls,
need further work, but lets get first something that the BPF verifier
accepts before increasing the complexity
That commented add_events line is needed for developing this
augmented_raw_syscalls.c BPF program, as if we add it via the
'add_events' mechanism so as to shorten the 'perf trace' command lines,
then we end up not setting up the -v option which precludes us having
access to the bpf verifier log :-\
Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Jesper Dangaard Brouer <brouer@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Leo Yan <leo.yan@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Song Liu <songliubraving@fb.com> Cc: Yonghong Song <yhs@fb.com> Link: https://lkml.kernel.org/n/tip-dn863ya0cbsqycxuy0olvbt1@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adrian Hunter [Mon, 20 May 2019 11:37:24 +0000 (14:37 +0300)]
perf scripts python: exported-sql-viewer.py: Add IPC information to the Branch reports
Enhance the "All branches" and "Selected branches" reports to display IPC
information if it is available.
Committer testing:
So, testing this I noticed that it all starts with the left arrow in every
line, that should mean there is some tree there, i.e. look at all those â–¶
symbols:
Reports -> All Branches:
Time CPU Command PID TID Branch Type In Tx Insn Cnt Cyc Cnt IPC Branch
â–¶ 187836112195670 7 simple-retpolin 23003 23003 trace begin No 0 0 0 0 unknown (unknown) -> 7f6f33d4f110
+_start (ld-2.28.so)
â–¶ 187836112195987 7 simple-retpolin 23003 23003 trace end No 0 883 0 7f6f33d4f110 _start (ld-2.28.so) -> 0 unknown
+(unknown)
â–¶ 187836112199189 7 simple-retpolin 23003 23003 trace begin No 0 0 0 0 unknown (unknown) -> 7f6f33d4f110
+_start (ld-2.28.so)
â–¶ 187836112199189 7 simple-retpolin 23003 23003 call No 0 0 0 7f6f33d4f113 _start+0x3 (ld-2.28.so) -> 7f6f33d4ff50
+_dl_start (ld-2.28.so)
â–¶ 187836112199544 7 simple-retpolin 23003 23003 trace end No 17 996 0.02 7f6f33d4ff73 _dl_start+0x23 (ld-2.28.so) -> 0
+unknown (unknown)
â–¶ 187836112200939 7 simple-retpolin 23003 23003 trace begin No 0 0 0 0 unknown (unknown) -> 7f6f33d4ff73
+_dl_start+0x23 (ld-2.28.so)
â–¶ 187836112201229 7 simple-retpolin 23003 23003 trace end No 1 816 0.00 7f6f33d4ff7a _dl_start+0x2a (ld-2.28.so) -> 0
+unknown (unknown)
â–¶ 187836112203500 7 simple-retpolin 23003 23003 trace begin No 0 0 0 0 unknown (unknown) -> 7f6f33d4ff7a
+_dl_start+0x2a (ld-2.28.so)
But if you click on it, that â–¶ disappears and a new click doesn't make
it reappear, looks buggy, minor oddity, reported to Adrian.
Reports -> Selected Branches, then ask for branches in the ld-2.28.so
DSO:
Adrian Hunter [Mon, 20 May 2019 11:37:22 +0000 (14:37 +0300)]
perf scripts python: export-to-sqlite.py: Export IPC information
Export cycle and instruction counts on samples and calls tables.
Committer testing:
First runs some workload collecting intel_pt with the 'cyc' ter just for
userspace:
[root@quaco adrian.hunter]# perf record -o simple-retpoline.perf.data -e intel_pt/cyc/u ./simple-retpoline
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.035 MB simple-retpoline.perf.data ]
[root@quaco adrian.hunter]#
Then use the export-to-sqlite.py script to see if the changes in this
cset don't make it to break and if the changes in the db schema are the
ones expected:
Adrian Hunter [Mon, 20 May 2019 11:37:17 +0000 (14:37 +0300)]
perf intel-pt: Accumulate cycle count from TSC/TMA/MTC packets
When CYC packets are not available, it is still possible to count cycles
using TSC/TMA/MTC timestamps.
As the timestamp increments in TSC ticks, convert to CPU cycles using
the current core-to-bus ratio.
Do not accumulate cycles when control flow packet generation is not
enabled, nor when time has been "lost", typically due to mwait, which is
indicated by a TSC/TMA packet that is not part of PSB+.
Note, the IPC values are low due to page faults at the beginning of
execution. The additional cycles are due to the time to enter the
kernel, not the actual kernel page fault handler.
Adrian Hunter [Mon, 20 May 2019 11:37:13 +0000 (14:37 +0300)]
perf intel-pt: Add support for samples to contain IPC ratio
Copy the incremental instruction count and cycle count onto 'instructions'
and 'branches' samples.
Because Intel PT does not update the cycle count on every branch or
instruction, the incremental values will often be zero.
When there are values, they will be the number of instructions and
number of cycles since the last update, and thus represent the average
IPC since the last IPC value.
Adrian Hunter [Mon, 20 May 2019 11:37:11 +0000 (14:37 +0300)]
perf intel-pt: Accumulate cycle count from CYC packets
In preparation for providing instructions-per-cycle (IPC) information,
accumulate cycle count from CYC packets.
Although CYC packets are optional (requires config term 'cyc' to enable
cycle-accurate mode when recording), the simplest way to count cycles is
with CYC packets.
The first complication is that cycles must be counted only when also
counting instructions.
That means when control flow packet generation is enabled i.e. between
TIP.PGE and TIP.PGD packets.
Also, sampling the cycle count follows the same rules as sampling the
timestamp, that is, not before the instruction to which the decoder is
walking is reached.
In addition, the cycle count is not accurate for any but the first
branch of a TNT packet.
Alexey Budankov [Thu, 30 May 2019 19:03:36 +0000 (22:03 +0300)]
perf record: Allow mixing --user-regs with --call-graph=dwarf
When DWARF stacks were requested and at the same time that the user
specifies a register set using the --user-regs option the full register
context was being captured on samples:
$ perf record -g --call-graph dwarf,1024 --user-regs=IP,SP,BP -- stack_test2.g.O3
I.e. the --user-regs=IP,SP,BP was being ignored, being overridden by the
needs of --call-graph=dwarf.
After applying the change in this patch the sample data contains the
user specified register, but making sure that at least the minimal set
of register needed for DWARF unwinding (DWARF_MINIMAL_REGS) is
requested.
The user is warned that DWARF unwinding may not work if extra registers
end up being needed.
$ perf record -g --call-graph dwarf,1024 --user-regs=BP -- ls
WARNING: The use of --call-graph=dwarf may require all the user registers, specifying a subset with --user-regs may render DWARF unwinding unreliable, so the minimal registers set (IP, SP) is explicitly forced.
arch COPYING Documentation include Kbuild lbuild MAINTAINERS modules.builtin Module.symvers perf.data.old scripts System.map virt
block CREDITS drivers init Kconfig lib Makefile modules.builtin.modinfo net README security tools vmlinux
certs crypto fs ipc kernel LICENSES mm modules.order perf.data samples sound usr vmlinux.o
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.030 MB perf.data (10 samples) ]
188368474305373 0x5e40 [0x470]: PERF_RECORD_SAMPLE(IP, 0x4002): 23839/23839: 0x401236 period: 1260507 addr: 0x7ffd3d85e96c
... FP chain: nr:0
... user regs: mask 0x1c0 ABI 64-bit
.... BP 0x401260
.... SP 0x7ffd3d85cc20
.... IP 0x401236
... ustack: size 1024, offset 0x58
. data_src: 0x5080021
Committer notes:
Detected build failures on arches where PERF_REGS_ is not available,
such as debian:experimental-x-{mips,mips64,mipsel}, fedora 24 and 30 for
ARC uClibc and glibc, reported to Alexey that provided a patch moving
the DWARF_MINIMAL_REGS from evsel.c to util/perf_regs.h, where it is
guarded by an HAVE_PERF_REGS_SUPPORT ifdef.
Committer testing:
# perf record --user-regs=bp,ax -a sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 1.955 MB perf.data (1773 samples) ]
# perf script -F+uregs | grep AX: | head -5
perf 1719 [000] 181.272398: 1 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffef828fb00
perf 1719 [000] 181.272402: 1 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffef828fb00
perf 1719 [000] 181.272403: 8 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffef828fb00
perf 1719 [000] 181.272405: 181 cycles: ffffffffba06a7c6 native_write_msr+0x6 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffef828fb00
perf 1719 [000] 181.272406: 4405 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffef828fb00
# perf record --call-graph=dwarf --user-regs=bp,ax -a sleep 1
WARNING: The use of --call-graph=dwarf may require all the user registers, specifying a subset with --user-regs may render DWARF unwinding unreliable, so the minimal registers set (IP, SP) is explicitly forced.
[ perf record: Woken up 55 times to write data ]
[ perf record: Captured and wrote 24.184 MB perf.data (2841 samples) ]
[root@quaco ~]# perf script --hide-call-graph -F+uregs | grep AX: | head -5
perf 1729 [000] 211.268006: 1 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffc8679abb0 SP:0x7ffc8679ab78 IP:0x7fa75223a0db
perf 1729 [000] 211.268014: 1 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffc8679abb0 SP:0x7ffc8679ab78 IP:0x7fa75223a0db
perf 1729 [000] 211.268017: 5 cycles: ffffffffba06a7c4 native_write_msr+0x4 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffc8679abb0 SP:0x7ffc8679ab78 IP:0x7fa75223a0db
perf 1729 [000] 211.268020: 48 cycles: ffffffffba06a7c6 native_write_msr+0x6 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffc8679abb0 SP:0x7ffc8679ab78 IP:0x7fa75223a0db
perf 1729 [000] 211.268024: 490 cycles: ffffffffba00e471 intel_bts_enable_local+0x21 (/lib/modules/5.2.0-rc1+/build/vmlinux) ABI:2 AX:0xffffffffffffffda BP:0x7ffc8679abb0 SP:0x7ffc8679ab78 IP:0x7fa75223a0db
#
Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/e7fd37b1-af22-0d94-a0dc-5895e803bbfe@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Leo Yan [Thu, 30 May 2019 09:38:01 +0000 (17:38 +0800)]
perf symbols: Remove unused variable 'err'
Variable 'err' is defined but never used in function symsrc__init(),
remove it and directly return -1 at the end of the function.
Signed-off-by: Leo Yan <leo.yan@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20190530093801.20510-1-leo.yan@linaro.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Song Liu [Tue, 21 May 2019 06:44:06 +0000 (23:44 -0700)]
perf data: Add description of header HEADER_BPF_PROG_INFO and HEADER_BPF_BTF
This patch addes description of HEADER_BPF_PROG_INFO and HEADER_BPF_BTF to
perf.data-file-format.txt.
Requested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Song Liu <songliubraving@fb.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Fixes: 606f972b1361 ("perf bpf: Save bpf_prog_info information as headers to perf.data") Link: http://lkml.kernel.org/r/20190521064406.2498925-1-songliubraving@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Jiri Olsa [Sun, 12 May 2019 15:55:10 +0000 (17:55 +0200)]
sysfs: Add sysfs_update_groups function
Adding sysfs_update_groups function to update
multiple groups.
sysfs_update_groups - given a directory kobject, create a bunch of attribute groups
@kobj: The kobject to update the group on
@groups: The attribute groups to update, NULL terminated
This function update a bunch of attribute groups. If an error occurs when
updating a group, all previously updated groups will be removed together
with already existing (not updated) attributes.
Signed-off-by: Jiri Olsa <jolsa@kernel.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/20190512155518.21468-2-jolsa@kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
Gayatri Kammela [Sat, 11 May 2019 00:03:11 +0000 (17:03 -0700)]
perf/x86/intel/uncore: Add new IMC PCI IDs for KabyLake, AmberLake and WhiskeyLake CPUs
AmberLake and WhiskeyLake have same client uncore events as
KabyLake. Thus add the PCI IDs for AmberLake Y processor lines,
for WhiskeyLake U processor lines and for KabyLake, add H
processor line and workstation.
Platform Device ID
================================
AML Y 2 Core 590Ch
KBL H 4 Core 5910h
KBL 4 Core WorkStation 5918h
WHL U 4 Core 3ED0h
WHL U 4 Core 3E34h
WHL U 2 Core 3E35h
AML Y 4 Core 590Dh
Signed-off-by: Gayatri Kammela <gayatri.kammela@intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Charles Prestopine <charles.d.prestopine@intel.com> Cc: Kan Liang <kan.liang@intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/20190511000311.20733-2-gayatri.kammela@intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org>
Song Liu [Tue, 7 May 2019 16:15:45 +0000 (09:15 -0700)]
perf/core: Allow non-privileged uprobe for user processes
Currently, non-privileged user could only use uprobe with
kernel.perf_event_paranoid = -1
However, setting perf_event_paranoid to -1 leaks other users' processes to
non-privileged uprobes.
To introduce proper permission control of uprobes, we are building the
following system:
A daemon with CAP_SYS_ADMIN is in charge to create uprobes via tracefs;
Users asks the daemon to create uprobes;
Then user can attach uprobe only to processes owned by the user.
This patch allows non-privileged user to attach uprobe to processes owned
by the user.
The following example shows how to use uprobe with non-privileged user.
This is based on Brendan's blog post [1]
Linus Torvalds [Sun, 2 Jun 2019 18:10:01 +0000 (11:10 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
"Two fixes: a quirk for KVM guests running on certain AMD CPUs, and a
KASAN related build fix"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/CPU/AMD: Don't force the CPB cap when running under a hypervisor
x86/boot: Provide KASAN compatible aliases for string routines
Linus Torvalds [Sun, 2 Jun 2019 18:08:12 +0000 (11:08 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf fixes from Ingo Molnar:
"On the kernel side there's a bunch of ring-buffer ordering fixes for a
reproducible bug, plus a PEBS constraints regression fix.
Plus tooling fixes"
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
tools headers UAPI: Sync kvm.h headers with the kernel sources
perf record: Fix s390 missing module symbol and warning for non-root users
perf machine: Read also the end of the kernel
perf test vmlinux-kallsyms: Ignore aliases to _etext when searching on kallsyms
perf session: Add missing swap ops for namespace events
perf namespace: Protect reading thread's namespace
tools headers UAPI: Sync drm/drm.h with the kernel
tools headers UAPI: Sync drm/i915_drm.h with the kernel
tools headers UAPI: Sync linux/fs.h with the kernel
tools headers UAPI: Sync linux/sched.h with the kernel
tools arch x86: Sync asm/cpufeatures.h with the with the kernel
tools include UAPI: Update copy of files related to new fspick, fsmount, fsconfig, fsopen, move_mount and open_tree syscalls
perf arm64: Fix mksyscalltbl when system kernel headers are ahead of the kernel
perf data: Fix 'strncat may truncate' build failure with recent gcc
perf/ring-buffer: Use regular variables for nesting
perf/ring-buffer: Always use {READ,WRITE}_ONCE() for rb->user_page data
perf/ring_buffer: Add ordering to rb->nest increment
perf/ring_buffer: Fix exposing a temporarily decreased data_head
perf/x86/intel/ds: Fix EVENT vs. UEVENT PEBS constraints
Linus Torvalds [Sun, 2 Jun 2019 18:06:13 +0000 (11:06 -0700)]
Merge branch 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull EFI fixes from Ingo Molnar:
"Two EFI fixes: a quirk for weird systabs, plus add more robust error
handling in the old 1:1 mapping code"
* 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
efi: Allow the number of EFI configuration tables entries to be zero
efi/x86/Add missing error handling to old_memmap 1:1 mapping code
Linus Torvalds [Sun, 2 Jun 2019 17:22:38 +0000 (10:22 -0700)]
Merge tag 'spdx-5.2-rc3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull SPDX fixes from Greg KH:
"Here are just two small patches, that fix up some found SPDX
identifier issues.
The first patch fixes an error in a previous SPDX fixup patch, that
causes build errors when doing 'make clean' on the tree (the fact that
almost no one noticed it reflects the fact that kernel developers
don't like doing that option very often...)
The second patch fixes up a number of places in the tree where people
mistyped the string "SPDX-License-Identifier". Given that people can
not even type their own name all the time without mistakes, this was
bound to happen, and odds are, we will have to add some type of check
for this to checkpatch.pl to catch this happening in the future.
Both of these have passed testing by 0-day"
* tag 'spdx-5.2-rc3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
treewide: fix typos of SPDX-License-Identifier
crypto: ux500 - fix license comment syntax error
Linus Torvalds [Sun, 2 Jun 2019 17:21:04 +0000 (10:21 -0700)]
Merge tag 'powerpc-5.2-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux
Pull powerpc fixes from Michael Ellerman:
"A minor fix to our IMC PMU code to print a less confusing error
message when the driver can't initialise properly.
A fix for a bug where a user requesting an unsupported branch sampling
filter can corrupt PMU state, preventing the PMU from counting
properly.
And finally a fix for a bug in our support for kexec_file_load(),
which prevented loading a kernel and initramfs. Most versions of kexec
don't yet use kexec_file_load().
Thanks to: Anju T Sudhakar, Dave Young, Madhavan Srinivasan, Ravi
Bangoria, Thiago Jung Bauermann"
* tag 'powerpc-5.2-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
powerpc/kexec: Fix loading of kernel + initramfs with kexec_file_load()
powerpc/perf: Fix MMCRA corruption by bhrb_filter
powerpc/powernv: Return for invalid IMC domain
Linus Torvalds [Sun, 2 Jun 2019 17:19:39 +0000 (10:19 -0700)]
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fixes from Paolo Bonzini:
"Fixes for PPC and s390"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: PPC: Book3S HV: Restore SPRG3 in kvmhv_p9_guest_entry()
KVM: PPC: Book3S HV: Fix lockdep warning when entering guest on POWER9
KVM: PPC: Book3S HV: XIVE: Fix page offset when clearing ESB pages
KVM: PPC: Book3S HV: XIVE: Take the srcu read lock when accessing memslots
KVM: PPC: Book3S HV: XIVE: Do not clear IRQ data of passthrough interrupts
KVM: PPC: Book3S HV: XIVE: Introduce a new mutex for the XIVE device
KVM: PPC: Book3S HV: XIVE: Fix the enforced limit on the vCPU identifier
KVM: PPC: Book3S HV: XIVE: Do not test the EQ flag validity when resetting
KVM: PPC: Book3S HV: XIVE: Clear file mapping when device is released
KVM: PPC: Book3S HV: Don't take kvm->lock around kvm_for_each_vcpu
KVM: PPC: Book3S: Use new mutex to synchronize access to rtas token list
KVM: PPC: Book3S HV: Use new mutex to synchronize MMU setup
KVM: PPC: Book3S HV: Avoid touching arch.mmu_ready in XIVE release functions
KVM: s390: Do not report unusabled IDs via KVM_CAP_MAX_VCPU_ID
kvm: fix compile on s390 part 2
Linus Torvalds [Sun, 2 Jun 2019 17:16:09 +0000 (10:16 -0700)]
Merge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/evalenti/linux-soc-thermal
Pull thermal SoC fix from Eduardo Valentin:
"A single revert, detected to cause issues on the tsens driver"
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/evalenti/linux-soc-thermal:
Revert "drivers: thermal: tsens: Add new operation to check if a sensor is enabled"
Linus Torvalds [Sun, 2 Jun 2019 17:14:25 +0000 (10:14 -0700)]
Merge tag 'led-fixes-for-5.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds
Pull LED fix from Jacek Anaszewski:
"Fix for a recent change in LED core, that didn't take into account the
possibility of calling led_blink_setup() from atomic context"
* tag 'led-fixes-for-5.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds:
leds: avoid flush_work in atomic context
Linus Torvalds [Sun, 2 Jun 2019 16:26:34 +0000 (09:26 -0700)]
Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI fixes from James Bottomley:
"Six minor fixes to device drivers and one to the multipath alua
handler.
The most extensive fix is the zfcp port remove prevention one, but
it's impact is only s390"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: libsas: delete sas port if expander discover failed
scsi: libsas: only clear phy->in_shutdown after shutdown event done
scsi: scsi_dh_alua: Fix possible null-ptr-deref
scsi: smartpqi: properly set both the DMA mask and the coherent DMA mask
scsi: zfcp: fix to prevent port_remove with pure auto scan LUNs (only sdevs)
scsi: zfcp: fix missing zfcp_port reference put on -EBUSY from port_remove
scsi: libcxgbi: add a check for NULL pointer in cxgbi_check_route()
Linus Torvalds [Sun, 2 Jun 2019 15:51:30 +0000 (08:51 -0700)]
Merge branch 'akpm' (patches from Andrew)
Merge misc fixes from Andrew Morton:
"Various fixes and followups"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
mm, compaction: make sure we isolate a valid PFN
include/linux/generic-radix-tree.h: fix kerneldoc comment
kernel/signal.c: trace_signal_deliver when signal_group_exit
drivers/iommu/intel-iommu.c: fix variable 'iommu' set but not used
spdxcheck.py: fix directory structures
kasan: initialize tag to 0xff in __kasan_kmalloc
z3fold: fix sheduling while atomic
scripts/gdb: fix invocation when CONFIG_COMMON_CLK is not set
mm/gup: continue VM_FAULT_RETRY processing even for pre-faults
ocfs2: fix error path kobject memory leak
memcg: make it work on sparse non-0-node systems
mm, memcg: consider subtrees in memory.events
prctl_set_mm: downgrade mmap_sem to read lock
prctl_set_mm: refactor checks from validate_prctl_map
kernel/fork.c: make max_threads symbol static
arch/arm/boot/compressed/decompress.c: fix build error due to lz4 changes
arch/parisc/configs/c8000_defconfig: remove obsoleted CONFIG_DEBUG_SLAB_LEAK
mm/vmalloc.c: fix typo in comment
lib/sort.c: fix kernel-doc notation warnings
mm: fix Documentation/vm/hmm.rst Sphinx warnings
When we have holes in a normal memory zone, we could endup having
cached_migrate_pfns which may not necessarily be valid, under heavy memory
pressure with swapping enabled ( via __reset_isolation_suitable(),
triggered by kswapd).
Later if we fail to find a page via fast_isolate_freepages(), we may end
up using the migrate_pfn we started the search with, as valid page. This
could lead to accessing NULL pointer derefernces like below, due to an
invalid mem_section pointer.
Unable to handle kernel NULL pointer dereference at virtual address 0000000000000008 [47/1825]
Mem abort info:
ESR = 0x96000004
Exception class = DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
Data abort info:
ISV = 0, ISS = 0x00000004
CM = 0, WnR = 0
user pgtable: 4k pages, 48-bit VAs, pgdp = 0000000082f94ae9
[0000000000000008] pgd=0000000000000000
Internal error: Oops: 96000004 [#1] SMP
...
CPU: 10 PID: 6080 Comm: qemu-system-aar Not tainted 510-rc1+ #6
Hardware name: AmpereComputing(R) OSPREY EV-883832-X3-0001/OSPREY, BIOS 4819 09/25/2018
pstate: 60000005 (nZCv daif -PAN -UAO)
pc : set_pfnblock_flags_mask+0x58/0xe8
lr : compaction_alloc+0x300/0x950
[...]
Process qemu-system-aar (pid: 6080, stack limit = 0x0000000095070da5)
Call trace:
set_pfnblock_flags_mask+0x58/0xe8
compaction_alloc+0x300/0x950
migrate_pages+0x1a4/0xbb0
compact_zone+0x750/0xde8
compact_zone_order+0xd8/0x118
try_to_compact_pages+0xb4/0x290
__alloc_pages_direct_compact+0x84/0x1e0
__alloc_pages_nodemask+0x5e0/0xe18
alloc_pages_vma+0x1cc/0x210
do_huge_pmd_anonymous_page+0x108/0x7c8
__handle_mm_fault+0xdd4/0x1190
handle_mm_fault+0x114/0x1c0
__get_user_pages+0x198/0x3c0
get_user_pages_unlocked+0xb4/0x1d8
__gfn_to_pfn_memslot+0x12c/0x3b8
gfn_to_pfn_prot+0x4c/0x60
kvm_handle_guest_abort+0x4b0/0xcd8
handle_exit+0x140/0x1b8
kvm_arch_vcpu_ioctl_run+0x260/0x768
kvm_vcpu_ioctl+0x490/0x898
do_vfs_ioctl+0xc4/0x898
ksys_ioctl+0x8c/0xa0
__arm64_sys_ioctl+0x28/0x38
el0_svc_common+0x74/0x118
el0_svc_handler+0x38/0x78
el0_svc+0x8/0xc
Code: f8607840f100001f8b0114019a801020 (f9400400)
---[ end trace af6a35219325a9b6 ]---
The issue was reported on an arm64 server with 128GB with holes in the
zone (e.g, [32GB@4GB, 96GB@544GB]), with a swap device enabled, while
running 100 KVM guest instances.
This patch fixes the issue by ensuring that the page belongs to a valid
PFN when we fallback to using the lower limit of the scan range upon
failure in fast_isolate_freepages().
Link: http://lkml.kernel.org/r/1558711908-15688-1-git-send-email-suzuki.poulose@arm.com Fixes: 5a811889de10f1eb ("mm, compaction: use free lists to quickly locate a migration target") Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com> Reported-by: Marc Zyngier <marc.zyngier@arm.com> Reviewed-by: Mel Gorman <mgorman@techsingularity.net> Reviewed-by: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Michal Hocko <mhocko@suse.com> Cc: Qian Cai <cai@lca.pw> Cc: Marc Zyngier <marc.zyngier@arm.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Zhenliang Wei [Sat, 1 Jun 2019 05:30:52 +0000 (22:30 -0700)]
kernel/signal.c: trace_signal_deliver when signal_group_exit
In the fixes commit, removing SIGKILL from each thread signal mask and
executing "goto fatal" directly will skip the call to
"trace_signal_deliver". At this point, the delivery tracking of the
SIGKILL signal will be inaccurate.
Therefore, we need to add trace_signal_deliver before "goto fatal" after
executing sigdelset.
Note: SEND_SIG_NOINFO matches the fact that SIGKILL doesn't have any info.
Link: http://lkml.kernel.org/r/20190425025812.91424-1-weizhenliang@huawei.com Fixes: cf43a757fd4944 ("signal: Restore the stop PTRACE_EVENT_EXIT") Signed-off-by: Zhenliang Wei <weizhenliang@huawei.com> Reviewed-by: Christian Brauner <christian@brauner.io> Reviewed-by: Oleg Nesterov <oleg@redhat.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Ivan Delalande <colona@arista.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Deepa Dinamani <deepa.kernel@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Qian Cai [Sat, 1 Jun 2019 05:30:49 +0000 (22:30 -0700)]
drivers/iommu/intel-iommu.c: fix variable 'iommu' set but not used
Commit cf04eee8bf0e ("iommu/vt-d: Include ACPI devices in iommu=pt")
added for_each_active_iommu() in iommu_prepare_static_identity_mapping()
but never used the each element, i.e, "drhd->iommu".
drivers/iommu/intel-iommu.c: In function
'iommu_prepare_static_identity_mapping':
drivers/iommu/intel-iommu.c:3037:22: warning: variable 'iommu' set but
not used [-Wunused-but-set-variable]
struct intel_iommu *iommu;
Fixed the warning by appending a compiler attribute __maybe_unused for it.
Link: http://lkml.kernel.org/r/20190523013314.2732-1-cai@lca.pw Signed-off-by: Qian Cai <cai@lca.pw> Suggested-by: Andrew Morton <akpm@linux-foundation.org> Cc: Joerg Roedel <jroedel@suse.de> Cc: David Woodhouse <dwmw2@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The LICENSE directory has recently changed structure and this makes
spdxcheck fails as per below:
FAIL: "Blob or Tree named 'other' not found"
Traceback (most recent call last):
File "scripts/spdxcheck.py", line 240, in <module>
spdx = read_spdxdata(repo)
File "scripts/spdxcheck.py", line 41, in read_spdxdata
for el in lictree[d].traverse():
[...]
KeyError: "Blob or Tree named 'other' not found"
Fix the script to restore the correctness on checkpatch License checking.
References: 62be257e986d ("LICENSES: Rename other to deprecated")
References: 8ea8814fcdcb ("LICENSES: Clearly mark dual license only licenses") Link: http://lkml.kernel.org/r/20190523084755.56739-1-vincenzo.frascino@arm.com Signed-off-by: Vincenzo Frascino <vincenzo.frascino@arm.com> Cc: Joe Perches <joe@perches.com> Cc: Christoph Hellwig <hch@lst.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Jeremy Cline <jcline@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When building with -Wuninitialized and CONFIG_KASAN_SW_TAGS unset, Clang
warns:
mm/kasan/common.c:484:40: warning: variable 'tag' is uninitialized when
used here [-Wuninitialized]
kasan_unpoison_shadow(set_tag(object, tag), size);
^~~
set_tag ignores tag in this configuration but clang doesn't realize it at
this point in its pipeline, as it points to arch_kasan_set_tag as being
the point where it is used, which will later be expanded to (void
*)(object) without a use of tag. Initialize tag to 0xff, as it removes
this warning and doesn't change the meaning of the code.
Fabiano Rosas [Sat, 1 Jun 2019 05:30:36 +0000 (22:30 -0700)]
scripts/gdb: fix invocation when CONFIG_COMMON_CLK is not set
CLK_GET_RATE_NOCACHE depends on CONFIG_COMMON_CLK. Importing constants.py
when CONFIG_COMMON_CLK is not defined causes:
(gdb) lx-symbols
(...)
File "scripts/gdb/linux/proc.py", line 15, in <module>
from linux import constants
File "scripts/gdb/linux/constants.py", line 2, in <module>
LX_CLK_GET_RATE_NOCACHE = gdb.parse_and_eval("CLK_GET_RATE_NOCACHE")
gdb.error: No symbol "CLK_GET_RATE_NOCACHE" in current context.
Link: http://lkml.kernel.org/r/20190523195313.24701-1-farosas@linux.ibm.com Fixes: e7e6f462c1be ("scripts/gdb: print cached rate in lx-clk-summary") Signed-off-by: Fabiano Rosas <farosas@linux.ibm.com> Reviewed-by: Stephen Boyd <swboyd@chromium.org> Cc: Jan Kiszka <jan.kiszka@siemens.com> Cc: Kieran Bingham <kbingham@kernel.org> Cc: Leonard Crestez <leonard.crestez@nxp.com> Cc: Jackie Liu <liuyun01@kylinos.cn> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mike Rapoport [Sat, 1 Jun 2019 05:30:33 +0000 (22:30 -0700)]
mm/gup: continue VM_FAULT_RETRY processing even for pre-faults
When get_user_pages*() is called with pages = NULL, the processing of
VM_FAULT_RETRY terminates early without actually retrying to fault-in all
the pages.
If the pages in the requested range belong to a VMA that has userfaultfd
registered, handle_userfault() returns VM_FAULT_RETRY *after* user space
has populated the page, but for the gup pre-fault case there's no actual
retry and the caller will get no pages although they are present.
This issue was uncovered when running post-copy memory restore in CRIU
after d9c9ce34ed5c ("x86/fpu: Fault-in user stack if
copy_fpstate_to_sigframe() fails").
After this change, the copying of FPU state to the sigframe switched from
copy_to_user() variants which caused a real page fault to get_user_pages()
with pages parameter set to NULL.
In post-copy mode of CRIU, the destination memory is managed with
userfaultfd and lack of the retry for pre-fault case in get_user_pages()
causes a crash of the restored process.
Making the pre-fault behavior of get_user_pages() the same as the "normal"
one fixes the issue.
Link: http://lkml.kernel.org/r/1557844195-18882-1-git-send-email-rppt@linux.ibm.com Fixes: d9c9ce34ed5c ("x86/fpu: Fault-in user stack if copy_fpstate_to_sigframe() fails") Signed-off-by: Mike Rapoport <rppt@linux.ibm.com> Tested-by: Andrei Vagin <avagin@gmail.com> [https://travis-ci.org/avagin/linux/builds/533184940] Tested-by: Hugh Dickins <hughd@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: Borislav Petkov <bp@suse.de> Cc: Pavel Machek <pavel@ucw.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If a call to kobject_init_and_add() fails we should call kobject_put()
otherwise we leak memory.
Add call to kobject_put() in the error path of call to
kobject_init_and_add(). Please note, this has the side effect that the
release method is called if kobject_init_and_add() fails.
Link: http://lkml.kernel.org/r/20190513033458.2824-1-tobin@kernel.org Signed-off-by: Tobin C. Harding <tobin@kernel.org> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Changwei Ge <gechangwei@live.cn> Cc: Gang He <ghe@suse.com> Cc: Jun Piao <piaojun@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jiri Slaby [Sat, 1 Jun 2019 05:30:26 +0000 (22:30 -0700)]
memcg: make it work on sparse non-0-node systems
We have a single node system with node 0 disabled:
Scanning NUMA topology in Northbridge 24
Number of physical nodes 2
Skipping disabled node 0
Node 1 MemBase 0000000000000000 Limit 00000000fbff0000
NODE_DATA(1) allocated [mem 0xfbfda000-0xfbfeffff]
This causes crashes in memcg when system boots:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
#PF error: [normal kernel read fault]
...
RIP: 0010:list_lru_add+0x94/0x170
...
Call Trace:
d_lru_add+0x44/0x50
dput.part.34+0xfc/0x110
__fput+0x108/0x230
task_work_run+0x9f/0xc0
exit_to_usermode_loop+0xf5/0x100
It is reproducible as far as 4.12. I did not try older kernels. You have
to have a new enough systemd, e.g. 241 (the reason is unknown -- was not
investigated). Cannot be reproduced with systemd 234.
The system crashes because the size of lru array is never updated in
memcg_update_all_list_lrus and the reads are past the zero-sized array,
causing dereferences of random memory.
The root cause are list_lru_memcg_aware checks in the list_lru code. The
test in list_lru_memcg_aware is broken: it assumes node 0 is always
present, but it is not true on some systems as can be seen above.
So fix this by avoiding checks on node 0. Remember the memcg-awareness by
a bool flag in struct list_lru.
Link: http://lkml.kernel.org/r/20190522091940.3615-1-jslaby@suse.cz Fixes: 60d3fd32a7a9 ("list_lru: introduce per-memcg lists") Signed-off-by: Jiri Slaby <jslaby@suse.cz> Acked-by: Michal Hocko <mhocko@suse.com> Suggested-by: Vladimir Davydov <vdavydov.dev@gmail.com> Acked-by: Vladimir Davydov <vdavydov.dev@gmail.com> Reviewed-by: Shakeel Butt <shakeelb@google.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Chris Down [Sat, 1 Jun 2019 05:30:22 +0000 (22:30 -0700)]
mm, memcg: consider subtrees in memory.events
memory.stat and other files already consider subtrees in their output, and
we should too in order to not present an inconsistent interface.
The current situation is fairly confusing, because people interacting with
cgroups expect hierarchical behaviour in the vein of memory.stat,
cgroup.events, and other files. For example, this causes confusion when
debugging reclaim events under low, as currently these always read "0" at
non-leaf memcg nodes, which frequently causes people to misdiagnose breach
behaviour. The same confusion applies to other counters in this file when
debugging issues.
Aggregation is done at write time instead of at read-time since these
counters aren't hot (unlike memory.stat which is per-page, so it does it
at read time), and it makes sense to bundle this with the file
notifications.
After this patch, events are propagated up the hierarchy:
[root@ktst ~]# cat /sys/fs/cgroup/system.slice/memory.events
low 0
high 0
max 0
oom 0
oom_kill 0
[root@ktst ~]# systemd-run -p MemoryMax=1 true
Running as unit: run-r251162a189fb4562b9dabfdc9b0422f5.service
[root@ktst ~]# cat /sys/fs/cgroup/system.slice/memory.events
low 0
high 0
max 7
oom 1
oom_kill 1
As this is a change in behaviour, this can be reverted to the old
behaviour by mounting with the `memory_localevents' flag set. However, we
use the new behaviour by default as there's a lack of evidence that there
are any current users of memory.events that would find this change
undesirable.
akpm: this is a behaviour change, so Cc:stable. THis is so that
forthcoming distros which use cgroup v2 are more likely to pick up the
revised behaviour.
Link: http://lkml.kernel.org/r/20190208224419.GA24772@chrisdown.name Signed-off-by: Chris Down <chris@chrisdown.name> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Shakeel Butt <shakeelb@google.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Tejun Heo <tj@kernel.org> Cc: Roman Gushchin <guro@fb.com> Cc: Dennis Zhou <dennis@kernel.org> Cc: Suren Baghdasaryan <surenb@google.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / mirror_ubuntu-eoan-kernel.git / log