The Intel PT decoder limits the number of unconditional branches (e.g.
jmps) decoded without consuming any trace packets. Generally, a loop
needs a conditional branch which generates a TNT packet, whereas a "ret"
instruction will generate a TIP or TNT packet. So exceeding the limit is
assumed to be a never-ending loop, which can happen if there has been a
decoding error putting the decoder at the wrong place in the code.
Up until now, the limit of 10000 has been enough but some analytic
purposes have been reported to exceed that.
Increase the limit to 100000, and make it configurable via perf config
intel-pt.max-loops. Also amend the "Never-ending loop" message to
mention the configuration entry.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lore.kernel.org/lkml/20210701175132.3977-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
VM Time Correlation means determining if each TSC packet belongs to a VM
Guest or the Host. When the trace is "in context" that is indicated by
the NR flag in the PIP packet. However, when tracing kernel-only,
userspace only, or using address filters, the trace can be "out of context"
in which case timing packets are produced but not PIP packets.
Nevertheless, it is very unlikely the VM Guest timestamps will be in
the same range as the Host timestamps. Host time ranges are established
by a starting side-band event timestamp, and subsequently by the buffer
timestamp, written when the buffer is copied to the perf.data file.
This patch supports updating the VM Guest timestamp packets, assuming an
unchanging (during perf record) VMX TSC Offset and no VMX TSC scaling.
Furthermore, it is possible to determine what the VMX TSC Offset is,
although not necessarily at the start. The dry-run option lets that
information be determined so that the user can pass it to a subsequent
run. For more detail, refer to the example in the Intel PT documentation
in a subsequent patch.
VM Time Correlation is also performed on the TSC value in PEBs-via-PT
records.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210430070309.17624-12-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
A timestamp should not go backwards. If it does it is assumed that the
7-byte TSC packet value has wrapped. Improve that logic so that it will
not allow the timestamp to go past the buffer timestamp (which is recorded
when the buffer is copied out)
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210430070309.17624-11-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
VM Time Correlation will use time ranges to determine whether a TSC packet
belongs to the Host or Guest. To start, the first non-zero timestamp is
needed. Pass that to the decoder.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210430070309.17624-10-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Even when VMX TSC Offset is not changing (during perf record), different
virtual machines can have different TSC Offsets. There is a Virtual Machine
Control Structure (VMCS) for each virtual CPU, the address of which is
reported to Intel PT in the VMCS packet. We do not know which VMCS belongs
to which virtual machine, so use a tree to keep track of VMCS information.
Then the decoder will be able to use the current VMCS value to look up the
current TSC Offset.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210430070309.17624-9-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Intel PT timestamps are affected by virtualization. While TSC packets can
still be considered to be unique, the TSC values need not be in order any
more. Adjust the algorithm accordingly.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210430070309.17624-8-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The PIP packet NR (non-root) flag indicates whether or not a virtual
machine is being traced (NR=1 => VM). Add support for tracking its value.
In particular note that the PIP packet (outside of PSB+) will be
associated with a TIP packet from which address the NR value takes
effect. At that point, there is a branch from_ip, to_ip with
corresponding from_nr and to_nr.
In the event of VM-Entry failure, there should still PIP and TIP packets
that can be followed in the same way.
Also note that this assumes that a host VMM is not employing VMX controls
that affect Intel PT, e.g. to hide the host from a guest using Intel PT.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210218095801.19576-5-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Retain the PIP packet payload as is, instead of just the CR3, because it
contains also the VMX NR flag which is needed to track VM-Entry.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210218095801.19576-4-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In preparation to support Intel PT decoding of virtual machine traces, add
vmlaunch and vmresume as branch instructions.
Note, sample flags will show "VMentry" even if the VM-Entry fails.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210218095801.19576-3-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The code assumed every CYC-eligible packet has a CYC packet, which is not
the case when CYC thresholds are used. Fix by checking if a CYC packet is
actually present in that case.
Fixes: 5b1dc0fd1d ("perf intel-pt: Add support for samples to contain IPC ratio")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210205175350.23817-4-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The code assumed a change in cycle count means accurate IPC. That is not
correct, for example when sampling both branches and instructions, or at
a FUP packet (which is not CYC-eligible) address. Fix by using an explicit
flag to indicate when IPC can be sampled.
Fixes: 5b1dc0fd1d ("perf intel-pt: Add support for samples to contain IPC ratio")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: linux-kernel@vger.kernel.org
Link: https://lore.kernel.org/r/20210205175350.23817-3-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add missing CYC packet processing when walking through PSB+. This
improves the accuracy of timestamps that follow PSB+, until the next
MTC.
Fixes: 3d49807870 ("perf tools: Add new Intel PT packet definitions")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20210205175350.23817-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
A single q option decodes ip from only FUP/TIP packets. Make it so that
repeating the q option (i.e. qq) decodes only PSB+, getting ip if there
is a FUP packet within PSB+ (i.e. between PSB and PSBEND).
Example:
$ perf record -e intel_pt//u grep -rI pudding drivers
[ perf record: Woken up 52 times to write data ]
[ perf record: Captured and wrote 57.870 MB perf.data ]
$ time perf script --itrace=bi | wc -l
58948289
real 1m23.863s
user 1m23.251s
sys 0m7.452s
$ time perf script --itrace=biq | wc -l
3385694
real 0m4.453s
user 0m4.455s
sys 0m0.328s
$ time perf script --itrace=biqq | wc -l
1883
real 0m0.047s
user 0m0.043s
sys 0m0.009s
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lore.kernel.org/lkml/20200710151104.15137-13-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Use the new itrace 'q' option to add support for a mode of decoding that
ignores TNT, does not walk object code, but gets the ip from FUP and TIP
packets.
Example:
$ perf record -e intel_pt//u grep -rI pudding drivers
[ perf record: Woken up 52 times to write data ]
[ perf record: Captured and wrote 57.870 MB perf.data ]
$ time perf script --itrace=bi | wc -l
58948289
real 1m23.863s
user 1m23.251s
sys 0m7.452s
$ time perf script --itrace=biq | wc -l
3385694
real 0m4.453s
user 0m4.455s
sys 0m0.328s
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lore.kernel.org/lkml/20200710151104.15137-12-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
While walking code towards a FUP ip, the packet state is
INTEL_PT_STATE_FUP or INTEL_PT_STATE_FUP_NO_TIP. That was mishandled
resulting in the state becoming INTEL_PT_STATE_IN_SYNC prematurely. The
result was an occasional lost EXSTOP event.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lore.kernel.org/lkml/20200710151104.15137-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Now that there's a common version of the decoder for all tools, use it
instead of the local copy.
Also use perf's check-headers.sh script to diff the decoder files to
make sure they remain in sync with the kernel version. Objtool has a
similar check.
Committer notes:
Had to keep this all pointing explicitely to x86 headers/files, i.e.
instead of asm/isnn.h we had to use ../include/asm/insn.h when the files
were in differemt dirs, or just replace "<asm/foo.h>" with "foo.h".
This way we continue to be able to process perf.data files with Intel PT
traces in distros other than x86.
Also fixed up the awk script paths to use $(srcdir)/tools/arch instead
or relative directories so that we keep detached tarballs (make help |
grep perf) working.
For now the include lines in these headers are being ignored so as not
to flag false reports of kernel/tools out of sync.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: x86@kernel.org
Link: http://lore.kernel.org/lkml/8a37e615d2880f039505d693d1e068a009358a2b.1567118001.git.jpoimboe@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
intel-pt-insn-decoder.c includes inat.c directly, so it already has an
implicit dependency on inat.c. The Build file dependency is redundant.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: x86@kernel.org
Link: http://lore.kernel.org/lkml/53776d6d29bc9eceb571d52df8fa32250c58a0f3.1567118001.git.jpoimboe@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Remove the last unneeded use of cache.h in a header, we can check where
it is really needed, i.e. we can remove it and be sure that it isn't
being obtained indirectly.
This is an old file, used by now incorrectly in many places, so it was
providing includes needed indirectly, fixup this fallout.
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-3x3l8gihoaeh7714os861ia7@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Eroding a bit more the tools/perf/util/util.h hodpodge header.
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-natazosyn9rwjka25tvcnyi0@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
For convenience, add the core-to-bus ratio (CBR) value to the decoder
state.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190622093248.581-4-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
PSB+ provides status information only so the core-to-bus ratio (CBR) in
PSB+ will not have changed from its previous value. However, cater for
the possibility of a another CBR change that gets caught up in the PSB+
anyway.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190622093248.581-3-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The core-to-bus ratio (CBR) provides the CPU frequency. With branches
enabled, the decoder was outputting CBR changes only when there was a
branch. That loses the correct time of the change if the trace is not in
context (e.g. not tracing kernel space). Change to output the CBR change
immediately.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190622093248.581-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
PEBS data is encoded in Block Item Packets (BIP). Populate a new structure
intel_pt_blk_items with the values and, upon a Block End Packet (BEP),
report them as a new Intel PT sample type INTEL_PT_BLK_ITEMS.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190610072803.10456-4-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add 3 new packets to supports PEBS via PT, namely Block Begin Packet
(BBP), Block Item Packet (BIP) and Block End Packet (BEP). PEBS data is
encoded into multiple BIP packets that come between BBP and BEP. The BEP
packet might be associated with a FUP packet. That is indicated by using
a separate packet type (INTEL_PT_BEP_IP) similar to other packets types
with the _IP suffix.
Refer to the Intel SDM for more information about PEBS via PT:
https://software.intel.com/en-us/articles/intel-sdm
May 2019 version: Vol. 3B 18.5.5.2 PEBS output to Intel® Processor Trace
Decoding of BIP packets conflicts with single-byte TNT packets. Since
BIP packets only occur in the context of a block (i.e. between BBP and
BEP), that context must be recorded and passed to the packet decoder.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190610072803.10456-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf record:
Alexey Budankov:
- Allow mixing --user-regs with --call-graph=dwarf, making sure that
the minimal set of registers for DWARF unwinding is present in the
set of user registers requested to be present in each sample, while
warning the user that this may make callchains unreliable if more
that the minimal set of registers is needed to unwind.
yuzhoujian:
- Add support to collect callchains from kernel or user space only,
IOW allow setting the perf_event_attr.exclude_callchain_{kernel,user}
bits from the command line.
perf trace:
Arnaldo Carvalho de Melo:
- Remove x86_64 specific syscall numbers from the augmented_raw_syscalls
BPF in-kernel collector of augmented raw_syscalls:sys_{enter,exit}
payloads, use instead the syscall numbers obtainer either by the
arch specific syscalltbl generators or from audit-libs.
- Allow 'perf trace' to ask for the number of bytes to collect for
string arguments, for now ask for PATH_MAX, i.e. the whole
pathnames, which ends up being just a way to speficy which syscall
args are pathnames and thus should be read using bpf_probe_read_str().
- Skip unknown syscalls when expanding strace like syscall groups.
This helps using the 'string' group of syscalls to work in arm64,
where some of the syscalls present in x86_64 that deal with
strings, for instance 'access', are deprecated and this should not
be asked for tracing.
Leo Yan:
- Exit when failing to build eBPF program.
perf config:
Arnaldo Carvalho de Melo:
- Bail out when a handler returns failure for a key-value pair. This
helps with cases where processing a key-value pair is not just a
matter of setting some tool specific knob, involving, for instance
building a BPF program to then attach to the list of events 'perf
trace' will use, e.g. augmented_raw_syscalls.c.
perf.data:
Kan Liang:
- Read and store die ID information available in new Intel processors
in CPUID.1F in the CPU topology written in the perf.data header.
perf stat:
Kan Liang:
- Support per-die aggregation.
Documentation:
Arnaldo Carvalho de Melo:
- Update perf.data documentation about the CPU_TOPOLOGY, MEM_TOPOLOGY,
CLOCKID and DIR_FORMAT headers.
Song Liu:
- Add description of headers HEADER_BPF_PROG_INFO and HEADER_BPF_BTF.
Leo Yan:
- Update default value for llvm.clang-bpf-cmd-template in 'man perf-config'.
JVMTI:
Jiri Olsa:
- Address gcc string overflow warning for strncpy()
core:
- Remove superfluous nthreads system_wide setup in perf_evsel__alloc_fd().
Intel PT:
Adrian Hunter:
- Add support for samples to contain IPC ratio, collecting cycles
information from CYC packets, showing the IPC info periodically, 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.
E.g.:
# perf record --cpu 1 -m200000 -a -e intel_pt/cyc/u sleep 0.0001
rounding mmap pages size to 1024M (262144 pages)
[ perf record: Woken up 0 times to write data ]
[ perf record: Captured and wrote 2.208 MB perf.data ]
# perf script --insn-trace --xed -F+ipc,-dso,-cpu,-tid
#
<SNIP + add line numbering to make sense of IPC counts e.g.: (18/3)>
1 cc1 63501.650479626: 7f5219ac27bf _int_free+0x3f jnz 0x7f5219ac2af0 IPC: 0.81 (36/44)
2 cc1 63501.650479626: 7f5219ac27c5 _int_free+0x45 cmp $0x1f, %rbp
3 cc1 63501.650479626: 7f5219ac27c9 _int_free+0x49 jbe 0x7f5219ac2b00
4 cc1 63501.650479626: 7f5219ac27cf _int_free+0x4f test $0x8, %al
5 cc1 63501.650479626: 7f5219ac27d1 _int_free+0x51 jnz 0x7f5219ac2b00
6 cc1 63501.650479626: 7f5219ac27d7 _int_free+0x57 movq 0x13c58a(%rip), %rcx
7 cc1 63501.650479626: 7f5219ac27de _int_free+0x5e mov %rdi, %r12
8 cc1 63501.650479626: 7f5219ac27e1 _int_free+0x61 movq %fs:(%rcx), %rax
9 cc1 63501.650479626: 7f5219ac27e5 _int_free+0x65 test %rax, %rax
10 cc1 63501.650479626: 7f5219ac27e8 _int_free+0x68 jz 0x7f5219ac2821
11 cc1 63501.650479626: 7f5219ac27ea _int_free+0x6a leaq -0x11(%rbp), %rdi
12 cc1 63501.650479626: 7f5219ac27ee _int_free+0x6e mov %rdi, %rsi
13 cc1 63501.650479626: 7f5219ac27f1 _int_free+0x71 shr $0x4, %rsi
14 cc1 63501.650479626: 7f5219ac27f5 _int_free+0x75 cmpq %rsi, 0x13caf4(%rip)
15 cc1 63501.650479626: 7f5219ac27fc _int_free+0x7c jbe 0x7f5219ac2821
16 cc1 63501.650479626: 7f5219ac2821 _int_free+0xa1 cmpq 0x13f138(%rip), %rbp
17 cc1 63501.650479626: 7f5219ac2828 _int_free+0xa8 jnbe 0x7f5219ac28d8
18 cc1 63501.650479626: 7f5219ac28d8 _int_free+0x158 testb $0x2, 0x8(%rbx)
19 cc1 63501.650479628: 7f5219ac28dc _int_free+0x15c jnz 0x7f5219ac2ab0 IPC: 6.00 (18/3)
<SNIP>
- Allow using time ranges with Intel PT, i.e. these features, already
present but not optimially usable with Intel PT, should be now:
Select the second 10% time slice:
$ perf script --time 10%/2
Select from 0% to 10% time slice:
$ perf script --time 0%-10%
Select the first and second 10% time slices:
$ perf script --time 10%/1,10%/2
Select from 0% to 10% and 30% to 40% slices:
$ perf script --time 0%-10%,30%-40%
cs-etm (ARM):
Mathieu Poirier:
- Add support for CPU-wide trace scenarios.
s390:
Thomas Richter:
- Fix missing kvm module load for s390.
- Fix OOM error in TUI mode on s390
- Support s390 diag event display when doing analysis on !s390
architectures.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCXP/1xQAKCRCyPKLppCJ+
J9xcAQCwOITAshE7op7HbKUPtkqiMNu+hpNa3skhxEpGHvKO0AEArpBXtuvEP8EU
PZsp+8vcVrlZ+dZutttgvkRz25mScg8=
=kfFb
-----END PGP SIGNATURE-----
Merge tag 'perf-core-for-mingo-5.3-20190611' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
perf record:
Alexey Budankov:
- Allow mixing --user-regs with --call-graph=dwarf, making sure that
the minimal set of registers for DWARF unwinding is present in the
set of user registers requested to be present in each sample, while
warning the user that this may make callchains unreliable if more
that the minimal set of registers is needed to unwind.
yuzhoujian:
- Add support to collect callchains from kernel or user space only,
IOW allow setting the perf_event_attr.exclude_callchain_{kernel,user}
bits from the command line.
perf trace:
Arnaldo Carvalho de Melo:
- Remove x86_64 specific syscall numbers from the augmented_raw_syscalls
BPF in-kernel collector of augmented raw_syscalls:sys_{enter,exit}
payloads, use instead the syscall numbers obtainer either by the
arch specific syscalltbl generators or from audit-libs.
- Allow 'perf trace' to ask for the number of bytes to collect for
string arguments, for now ask for PATH_MAX, i.e. the whole
pathnames, which ends up being just a way to speficy which syscall
args are pathnames and thus should be read using bpf_probe_read_str().
- Skip unknown syscalls when expanding strace like syscall groups.
This helps using the 'string' group of syscalls to work in arm64,
where some of the syscalls present in x86_64 that deal with
strings, for instance 'access', are deprecated and this should not
be asked for tracing.
Leo Yan:
- Exit when failing to build eBPF program.
perf config:
Arnaldo Carvalho de Melo:
- Bail out when a handler returns failure for a key-value pair. This
helps with cases where processing a key-value pair is not just a
matter of setting some tool specific knob, involving, for instance
building a BPF program to then attach to the list of events 'perf
trace' will use, e.g. augmented_raw_syscalls.c.
perf.data:
Kan Liang:
- Read and store die ID information available in new Intel processors
in CPUID.1F in the CPU topology written in the perf.data header.
perf stat:
Kan Liang:
- Support per-die aggregation.
Documentation:
Arnaldo Carvalho de Melo:
- Update perf.data documentation about the CPU_TOPOLOGY, MEM_TOPOLOGY,
CLOCKID and DIR_FORMAT headers.
Song Liu:
- Add description of headers HEADER_BPF_PROG_INFO and HEADER_BPF_BTF.
Leo Yan:
- Update default value for llvm.clang-bpf-cmd-template in 'man perf-config'.
JVMTI:
Jiri Olsa:
- Address gcc string overflow warning for strncpy()
core:
- Remove superfluous nthreads system_wide setup in perf_evsel__alloc_fd().
Intel PT:
Adrian Hunter:
- Add support for samples to contain IPC ratio, collecting cycles
information from CYC packets, showing the IPC info periodically, 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.
E.g.:
# perf record --cpu 1 -m200000 -a -e intel_pt/cyc/u sleep 0.0001
rounding mmap pages size to 1024M (262144 pages)
[ perf record: Woken up 0 times to write data ]
[ perf record: Captured and wrote 2.208 MB perf.data ]
# perf script --insn-trace --xed -F+ipc,-dso,-cpu,-tid
#
<SNIP + add line numbering to make sense of IPC counts e.g.: (18/3)>
1 cc1 63501.650479626: 7f5219ac27bf _int_free+0x3f jnz 0x7f5219ac2af0 IPC: 0.81 (36/44)
2 cc1 63501.650479626: 7f5219ac27c5 _int_free+0x45 cmp $0x1f, %rbp
3 cc1 63501.650479626: 7f5219ac27c9 _int_free+0x49 jbe 0x7f5219ac2b00
4 cc1 63501.650479626: 7f5219ac27cf _int_free+0x4f test $0x8, %al
5 cc1 63501.650479626: 7f5219ac27d1 _int_free+0x51 jnz 0x7f5219ac2b00
6 cc1 63501.650479626: 7f5219ac27d7 _int_free+0x57 movq 0x13c58a(%rip), %rcx
7 cc1 63501.650479626: 7f5219ac27de _int_free+0x5e mov %rdi, %r12
8 cc1 63501.650479626: 7f5219ac27e1 _int_free+0x61 movq %fs:(%rcx), %rax
9 cc1 63501.650479626: 7f5219ac27e5 _int_free+0x65 test %rax, %rax
10 cc1 63501.650479626: 7f5219ac27e8 _int_free+0x68 jz 0x7f5219ac2821
11 cc1 63501.650479626: 7f5219ac27ea _int_free+0x6a leaq -0x11(%rbp), %rdi
12 cc1 63501.650479626: 7f5219ac27ee _int_free+0x6e mov %rdi, %rsi
13 cc1 63501.650479626: 7f5219ac27f1 _int_free+0x71 shr $0x4, %rsi
14 cc1 63501.650479626: 7f5219ac27f5 _int_free+0x75 cmpq %rsi, 0x13caf4(%rip)
15 cc1 63501.650479626: 7f5219ac27fc _int_free+0x7c jbe 0x7f5219ac2821
16 cc1 63501.650479626: 7f5219ac2821 _int_free+0xa1 cmpq 0x13f138(%rip), %rbp
17 cc1 63501.650479626: 7f5219ac2828 _int_free+0xa8 jnbe 0x7f5219ac28d8
18 cc1 63501.650479626: 7f5219ac28d8 _int_free+0x158 testb $0x2, 0x8(%rbx)
19 cc1 63501.650479628: 7f5219ac28dc _int_free+0x15c jnz 0x7f5219ac2ab0 IPC: 6.00 (18/3)
<SNIP>
- Allow using time ranges with Intel PT, i.e. these features, already
present but not optimially usable with Intel PT, should be now:
Select the second 10% time slice:
$ perf script --time 10%/2
Select from 0% to 10% time slice:
$ perf script --time 0%-10%
Select the first and second 10% time slices:
$ perf script --time 10%/1,10%/2
Select from 0% to 10% and 30% to 40% slices:
$ perf script --time 0%-10%,30%-40%
cs-etm (ARM):
Mathieu Poirier:
- Add support for CPU-wide trace scenarios.
s390:
Thomas Richter:
- Fix missing kvm module load for s390.
- Fix OOM error in TUI mode on s390
- Support s390 diag event display when doing analysis on !s390
architectures.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Intel PT decoding is done in time order. In order to support efficient time
interval filtering, add a facility to "fast forward" towards a particular
timestamp. That involves finding the right buffer, stepping to that buffer,
and then stepping forward PSBs. Because decoding must begin at a PSB,
"fast forward" stops at the last PSB that has a timestamp before the target
timestamp.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190604130017.31207-9-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When the decoder gets the next trace buffer, some state is reset if the
buffer is not consecutive to the previous buffer. Add a parameter
'reposition' so that can be done also to support a "fast forward"
facility.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190604130017.31207-8-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Factor out intel_pt_reposition() so it can be reused.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190604130017.31207-7-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Factor out intel_pt_8b_tsc() so it can be reused.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190604130017.31207-6-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a callback function to enable the decoder to lookahead at subsequent
trace buffers. This will be used to implement a "fast forward" facility
which will be needed to support efficient time interval filtering.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190604130017.31207-5-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Based on 1 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms and conditions of the gnu general public license
version 2 as published by the free software foundation this program
is distributed in the hope it will be useful but without any
warranty without even the implied warranty of merchantability or
fitness for a particular purpose see the gnu general public license
for more details
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-only
has been chosen to replace the boilerplate/reference in 263 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Alexios Zavras <alexios.zavras@intel.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190529141901.208660670@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
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+.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190520113728.14389-12-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To make it easier to add new code for different TIP cases, separate each
case.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190520113728.14389-11-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In preparation for using MTC packets to count cycles, record whether
decoding is between a PSB and PSBEND packets.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190520113728.14389-10-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
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.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190520113728.14389-6-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To eliminate some duplication and make the code more understandable,
factor out intel_pt_update_sample_time.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190520113728.14389-5-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Based on 1 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms of the gnu general public license as published by
the free software foundation either version 2 of the license or at
your option any later version this program is distributed in the
hope that it will be useful but without any warranty without even
the implied warranty of merchantability or fitness for a particular
purpose see the gnu general public license for more details you
should have received a copy of the gnu general public license along
with this program if not write to the free software foundation inc
59 temple place suite 330 boston ma 02111 1307 usa
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-or-later
has been chosen to replace the boilerplate/reference in 1334 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Richard Fontana <rfontana@redhat.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190527070033.113240726@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The sample timestamp is updated to ensure that the timestamp represents
the time of the sample and not a branch that the decoder is still
walking towards. The sample timestamp is updated when the decoder
returns, but the decoder does not return for non-taken branches. Update
the sample timestamp then also.
Note that commit 3f04d98e97 ("perf intel-pt: Improve sample
timestamp") was also a stable fix and appears, for example, in v4.4
stable tree as commit a4ebb58fd124 ("perf intel-pt: Improve sample
timestamp").
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org # v4.4+
Fixes: 3f04d98e97 ("perf intel-pt: Improve sample timestamp")
Link: http://lkml.kernel.org/r/20190510124143.27054-4-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The decoder uses its current timestamp in samples. Usually that is a
timestamp that has already passed, but in some cases it is a timestamp
for a branch that the decoder is walking towards, and consequently
hasn't reached.
The intel_pt_sample_time() function decides which is which, but was not
handling TNT packets exactly correctly.
In the case of TNT, the timestamp applies to the first branch, so the
decoder must first walk to that branch.
That means intel_pt_sample_time() should return true for TNT, and this
patch makes that change. However, if the first branch is a non-taken
branch (i.e. a 'N'), then intel_pt_sample_time() needs to return false
for subsequent taken branches in the same TNT packet.
To handle that, introduce a new state INTEL_PT_STATE_TNT_CONT to
distinguish the cases.
Note that commit 3f04d98e97 ("perf intel-pt: Improve sample
timestamp") was also a stable fix and appears, for example, in v4.4
stable tree as commit a4ebb58fd124 ("perf intel-pt: Improve sample
timestamp").
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org # v4.4+
Fixes: 3f04d98e97 ("perf intel-pt: Improve sample timestamp")
Link: http://lkml.kernel.org/r/20190510124143.27054-3-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The timestamp used to determine if an instruction sample is made, is an
estimate based on the number of instructions since the last known
timestamp. A consequence is that it might go backwards, which results in
extra samples. Change it so that a sample is only made when the
timestamp goes forwards.
Note this does not affect a sampling period of 0 or sampling periods
specified as a count of instructions.
Example:
Before:
$ perf script --itrace=i10us
ls 13812 [003] 2167315.222583: 3270 instructions:u: 7fac71e2e494 __GI___tunables_init+0xf4 (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 30902 instructions:u: 7fac71e2da0f _dl_cache_libcmp+0x2f (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 10 instructions:u: 7fac71e2d9ff _dl_cache_libcmp+0x1f (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 8 instructions:u: 7fac71e2d9ea _dl_cache_libcmp+0xa (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 14 instructions:u: 7fac71e2d9ea _dl_cache_libcmp+0xa (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 6 instructions:u: 7fac71e2d9ff _dl_cache_libcmp+0x1f (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 14 instructions:u: 7fac71e2d9ff _dl_cache_libcmp+0x1f (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 4 instructions:u: 7fac71e2dab2 _dl_cache_libcmp+0xd2 (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222728: 16423 instructions:u: 7fac71e2477a _dl_map_object_deps+0x1ba (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222734: 12731 instructions:u: 7fac71e27938 _dl_name_match_p+0x68 (/lib/x86_64-linux-gnu/ld-2.28.so)
...
After:
$ perf script --itrace=i10us
ls 13812 [003] 2167315.222583: 3270 instructions:u: 7fac71e2e494 __GI___tunables_init+0xf4 (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222667: 30902 instructions:u: 7fac71e2da0f _dl_cache_libcmp+0x2f (/lib/x86_64-linux-gnu/ld-2.28.so)
ls 13812 [003] 2167315.222728: 16479 instructions:u: 7fac71e2477a _dl_map_object_deps+0x1ba (/lib/x86_64-linux-gnu/ld-2.28.so)
...
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org
Fixes: f4aa081949 ("perf tools: Add Intel PT decoder")
Link: http://lkml.kernel.org/r/20190510124143.27054-2-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
A TSC packet can slip past MTC packets so that the timestamp appears to
go backwards. One estimate is that can be up to about 40 CPU cycles,
which is certainly less than 0x1000 TSC ticks, but accept slippage an
order of magnitude more to be on the safe side.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org
Fixes: 79b58424b8 ("perf tools: Add Intel PT support for decoding MTC packets")
Link: http://lkml.kernel.org/r/20190325135135.18348-1-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Rename build libperf to perf, because it's used to build perf.
The libperf build object name will be used for libperf library.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20190213123246.4015-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Data is copied when the trace is stopped, so packets are never split
between buffers except when processing if the buffer cannot fit in the
address space which can only happen on 32-bit systems. Change the logic
to reflect that.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20190206103947.15750-5-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
CYC packet timestamp calculation depends upon CBR which was being
cleared upon overflow (OVF). That can cause errors due to failing to
synchronize with sideband events. Even if a CBR change has been lost,
the old CBR is still a better estimate than zero. So remove the clearing
of CBR.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org
Link: http://lkml.kernel.org/r/20190206103947.15750-4-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adrian Hunter