New features:
- Add 'brstackinsn' field in 'perf script' to reuse the x86 instruction
decoder used in the Intel PT code to study hot paths to samples (Andi Kleen)
Kernel:
- Default UPROBES_EVENTS to Y (Alexei Starovoitov)
- Fix check for kretprobe offset within function entry (Naveen N. Rao)
Infrastructure:
- Introduce util func is_sdt_event() (Ravi Bangoria)
- Make perf_event__synthesize_mmap_events() scale on older kernels where
reading /proc/pid/maps is way slower than reading /proc/pid/task/pid/maps (Stephane Eranian)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=Bg5M
-----END PGP SIGNATURE-----
Merge tag 'perf-core-for-mingo-4.12-20170316' 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:
New features:
- Add 'brstackinsn' field in 'perf script' to reuse the x86 instruction
decoder used in the Intel PT code to study hot paths to samples (Andi Kleen)
Kernel changes:
- Default UPROBES_EVENTS to Y (Alexei Starovoitov)
- Fix check for kretprobe offset within function entry (Naveen N. Rao)
Infrastructure changes:
- Introduce util func is_sdt_event() (Ravi Bangoria)
- Make perf_event__synthesize_mmap_events() scale on older kernels where
reading /proc/pid/maps is way slower than reading /proc/pid/task/pid/maps (Stephane Eranian)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Implement printing instruction sequences as hex dump for branch stacks.
This relies on the x86 instruction decoder used by the PT decoder to
find the lengths of instructions to dump them individually.
This is good enough for pattern matching.
This allows to study hot paths for individual samples, together with
branch misprediction and cycle count / IPC information if available (on
Skylake systems).
% perf record -b ...
% perf script -F brstackinsn
...
read_hpet+67:
ffffffff9905b843 insn: 74 ea # PRED
ffffffff9905b82f insn: 85 c9
ffffffff9905b831 insn: 74 12
ffffffff9905b833 insn: f3 90
ffffffff9905b835 insn: 48 8b 0f
ffffffff9905b838 insn: 48 89 ca
ffffffff9905b83b insn: 48 c1 ea 20
ffffffff9905b83f insn: 39 f2
ffffffff9905b841 insn: 89 d0
ffffffff9905b843 insn: 74 ea # PRED
Only works when no special branch filters are specified.
Occasionally the path does not reach up to the sample IP, as the LBRs
may be frozen before executing a final jump. In this case we print a
special message.
The instruction dumper piggy backs on the existing infrastructure from
the IP PT decoder.
An earlier iteration of this patch relied on a disassembler, but this
version only uses the existing instruction decoder.
Committer note:
Added hint about how to get suitable perf.data files for use with
'-F brstackinsm':
$ perf record usleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ]
$
$ perf script -F brstackinsn
Display of branch stack assembler requested, but non all-branch filter set
Hint: run 'perf record -b ...'
$
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Link: http://lkml.kernel.org/r/20170223234634.583-1-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
$ perf test decoder
57: x86 instruction decoder - new instructions : FAILED!
$
Failed to decode 'rel' value (0xfffffffc vs expected 0): 0f 1b 80 78 56 34 12 bndstx %bnd0,0x12345678(%rax)
Failed to decode 'rel' value (0xfffffffc vs expected 0): 0f 1b 85 78 56 34 12 bndstx %bnd0,0x12345678(%rbp)
Failed to decode 'rel' value (0xfffffffc vs expected 0): 0f 1b 84 01 78 56 34 12 bndstx %bnd0,0x12345678(%rcx,%rax,1)
Failed to decode 'rel' value (0xfffffffc vs expected 0): 0f 1b 84 05 78 56 34 12 bndstx %bnd0,0x12345678(%rbp,%rax,1)
Failed to decode 'rel' value (0xfffffffc vs expected 0): 0f 1b 84 08 78 56 34 12 bndstx %bnd0,0x12345678(%rax,%rcx,1)
There is missing initialization. It only affects the test because it is
checking 'rel' even in cases where there is no value.
Fix it.
Reported-and-Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/08c6ad07-7994-3e56-b20e-d75727ca7765@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Update to the new file paths, remove them from introductory comments.
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170218113140.8051-1-bp@alien8.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
So set it only for other compilers, allowing us to overcome yet another
build failure due to an inexistent clang -W option:
error: unknown warning option '-Wno-override-init'; did you mean '-Wno-override-module'? [-Werror,-Wunknown-warning-option]
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-oaa1ici3j8nygp4pzl2oobh3@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To address new warnings emmited by gcc 7, e.g.::
CC /tmp/build/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.o
CC /tmp/build/perf/tests/parse-events.o
util/intel-pt-decoder/intel-pt-pkt-decoder.c: In function 'intel_pt_pkt_desc':
util/intel-pt-decoder/intel-pt-pkt-decoder.c:499:6: error: this statement may fall through [-Werror=implicit-fallthrough=]
if (!(packet->count))
^
util/intel-pt-decoder/intel-pt-pkt-decoder.c:501:2: note: here
case INTEL_PT_CYC:
^~~~
CC /tmp/build/perf/util/intel-pt-decoder/intel-pt-decoder.o
cc1: all warnings being treated as errors
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-mf0hw789pu9x855us5l32c83@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Change Intel PT and BTS to pass up the length and the instruction
bytes of the decoded or sampled instruction in the perf sample.
The decoder already knows this information, we just need to pass it
up. Since it is only a couple of movs it is not very expensive.
Handle instruction cache too. Make sure ilen is always initialized.
Used in the next patch.
[Adrian: re-base on top (and adjust for) instruction buffer size tidy-up]
[Adrian: add BTS support and adjust commit message accordingly]
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Link: http://lkml.kernel.org/r/1475847747-30994-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tidy instruction buffer size usage in preparation for copying the
instruction bytes onto samples.
The instruction buffer is presently used for debugging, so rename its
size macro from INTEL_PT_INSN_DBG_BUF_SZ to INTEL_PT_INSN_BUF_SZ, and
use it everywhere.
Note that the maximum instruction size is 15 which is a less efficient size
to copy than 16, which is why a separate buffer size is used.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1475847747-30994-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The MTC packet provides a 8-bit slice of CTC which is related to TSC by
the TMA packet, however the TMA packet only provides the lower 16 bits
of CTC. If mtc_shift > 8 then some of the MTC bits are not in the CTC
provided by the TMA packet. Fix-up the last_mtc calculated from the TMA
packet by copying the missing bits from the current MTC assuming the
least difference between the two, and that the current MTC comes after
last_mtc.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org # v4.3+
Link: http://lkml.kernel.org/r/1475062896-22274-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In cycle-accurate mode, timestamps can be calculated from CYC packets.
The decoder also estimates timestamps based on the number of
instructions since the last timestamp. For that to work in
cycle-accurate mode, the instruction count needs to be reset to zero
when a timestamp is calculated from a CYC packet, but that wasn't
happening, so fix it.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: stable@vger.kernel.org # v4.3+
Link: http://lkml.kernel.org/r/1475062896-22274-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When address filters are used, the decoder must detect the end of a
filter region (or a branch into a tracestop region) by matching Packet
Generation Disabled (TIP.PGD) packets against the object code using the
IP given in the packet. However, due to errata SKL014 "Intel PT TIP.PGD
May Not Have Target IP Payload", that IP may not be present.
Enable the decoder to handle that by adding a new callback function
'pgd_ip()' which indicates whether the IP is not traced, in which case
that is the point where the trace was disabled.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Link: http://lkml.kernel.org/r/1474641528-18776-16-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The June 2015 Intel SDM introduced IP Compression types 4 and 6. Refer
to section 36.4.2.2 Target IP (TIP) Packet - IP Compression.
Existing Intel PT packet decoder did not support type 4, and got type 6
wrong. Because type 3 and type 4 have the same number of bytes, the
packet 'count' has been changed from being the number of ip bytes to
being the type code. That allows the Intel PT decoder to correctly
decide whether to sign-extend or use the last ip. However that also
meant the code had to be adjusted in a number of places.
Currently hardware is not using the new compression types, so this fix
has no effect on existing hardware.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1469005206-3049-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
1/ Replace pcommit with ADR / directed-flushing:
The pcommit instruction, which has not shipped on any product, is
deprecated. Instead, the requirement is that platforms implement either
ADR, or provide one or more flush addresses per nvdimm. ADR
(Asynchronous DRAM Refresh) flushes data in posted write buffers to the
memory controller on a power-fail event. Flush addresses are defined in
ACPI 6.x as an NVDIMM Firmware Interface Table (NFIT) sub-structure:
"Flush Hint Address Structure". A flush hint is an mmio address that
when written and fenced assures that all previous posted writes
targeting a given dimm have been flushed to media.
2/ On-demand ARS (address range scrub):
Linux uses the results of the ACPI ARS commands to track bad blocks
in pmem devices. When latent errors are detected we re-scrub the media
to refresh the bad block list, userspace can also request a re-scrub at
any time.
3/ Support for the Microsoft DSM (device specific method) command format.
4/ Support for EDK2/OVMF virtual disk device memory ranges.
5/ Various fixes and cleanups across the subsystem.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=xCBG
-----END PGP SIGNATURE-----
Merge tag 'libnvdimm-for-4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm updates from Dan Williams:
- Replace pcommit with ADR / directed-flushing.
The pcommit instruction, which has not shipped on any product, is
deprecated. Instead, the requirement is that platforms implement
either ADR, or provide one or more flush addresses per nvdimm.
ADR (Asynchronous DRAM Refresh) flushes data in posted write buffers
to the memory controller on a power-fail event.
Flush addresses are defined in ACPI 6.x as an NVDIMM Firmware
Interface Table (NFIT) sub-structure: "Flush Hint Address Structure".
A flush hint is an mmio address that when written and fenced assures
that all previous posted writes targeting a given dimm have been
flushed to media.
- On-demand ARS (address range scrub).
Linux uses the results of the ACPI ARS commands to track bad blocks
in pmem devices. When latent errors are detected we re-scrub the
media to refresh the bad block list, userspace can also request a
re-scrub at any time.
- Support for the Microsoft DSM (device specific method) command
format.
- Support for EDK2/OVMF virtual disk device memory ranges.
- Various fixes and cleanups across the subsystem.
* tag 'libnvdimm-for-4.8' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (41 commits)
libnvdimm-btt: Delete an unnecessary check before the function call "__nd_device_register"
nfit: do an ARS scrub on hitting a latent media error
nfit: move to nfit/ sub-directory
nfit, libnvdimm: allow an ARS scrub to be triggered on demand
libnvdimm: register nvdimm_bus devices with an nd_bus driver
pmem: clarify a debug print in pmem_clear_poison
x86/insn: remove pcommit
Revert "KVM: x86: add pcommit support"
nfit, tools/testing/nvdimm/: unify shutdown paths
libnvdimm: move ->module to struct nvdimm_bus_descriptor
nfit: cleanup acpi_nfit_init calling convention
nfit: fix _FIT evaluation memory leak + use after free
tools/testing/nvdimm: add manufacturing_{date|location} dimm properties
tools/testing/nvdimm: add virtual ramdisk range
acpi, nfit: treat virtual ramdisk SPA as pmem region
pmem: kill __pmem address space
pmem: kill wmb_pmem()
libnvdimm, pmem: use nvdimm_flush() for namespace I/O writes
fs/dax: remove wmb_pmem()
libnvdimm, pmem: flush posted-write queues on shutdown
...
The pcommit instruction is being deprecated in favor of either ADR
(asynchronous DRAM refresh: flush-on-power-fail) at the platform level, or
posted-write-queue flush addresses as defined by the ACPI 6.x NFIT (NVDIMM
Firmware Interface Table).
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Xiao Guangrong <guangrong.xiao@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Acked-by: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Add support for Intel's AVX-512 instructions to perf tools instruction
decoder used by Intel PT. The kernel's instruction decoder was updated in
a previous patch.
AVX-512 instructions are documented in Intel Architecture Instruction Set
Extensions Programming Reference (February 2016).
AVX-512 instructions are identified by a EVEX prefix which, for the purpose
of instruction decoding, can be treated as though it were a 4-byte VEX
prefix.
Existing instructions which can now accept an EVEX prefix need not be
further annotated in the op code map (x86-opcode-map.txt). In the case of
new instructions, the op code map is updated accordingly.
Also add associated Mask Instructions that are used to manipulate mask
registers used in AVX-512 instructions.
A representative set of instructions is added to the perf tools new
instructions test in a subsequent patch.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: X86 ML <x86@kernel.org>
Link: http://lkml.kernel.org/r/1469003437-32706-4-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
vcvtph2ps does not have an immediate operand, so remove the erroneous
'Ib' from its opcode map entry. Add vcvtph2ps to the perf tools new
instructions test to verify it.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: X86 ML <x86@kernel.org>
Link: http://lkml.kernel.org/r/1469003437-32706-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
That doesn't have -I to match lines.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-7nz9hnbk7a9p91ou927ye5yh@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The check for the maximum code is off-by-one; the current comparison of
a code that is INTEL_PT_ERR_MAX will cause the strlcpy to perform an out
of bounds array access on the intel_pt_err_msgs array.
Fix this with a >= comparison.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1461524203-10224-1-git-send-email-colin.king@canonical.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When running objtool on a ppc64le host to analyze x86 binaries, it
reports a lot of false warnings like:
ipc/compat_mq.o: warning: objtool: compat_SyS_mq_open()+0x91: can't find jump dest instruction at .text+0x3a5
The warnings are caused by the x86 instruction decoder setting the wrong
value for the jump instruction's immediate field because it assumes that
"char == signed char", which isn't true for all architectures. When
converting char to int, gcc sign-extends on x86 but doesn't sign-extend
on ppc64le.
According to the gcc man page, that's a feature, not a bug:
> Each kind of machine has a default for what "char" should be. It is
> either like "unsigned char" by default or like "signed char" by
> default.
>
> Ideally, a portable program should always use "signed char" or
> "unsigned char" when it depends on the signedness of an object.
Conform to the "standards" by changing the "char" casts to "signed
char". This results in no actual changes to the object code on x86.
Note: the x86 decoder now lives in three different locations in the
kernel tree, which are all kept in sync via makefile checks and
warnings: in-kernel, perf, and objtool. This fixes all three locations.
Eventually we should probably try to at least converge the two separate
"tools" locations into a single shared location.
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/9dd4161719b20e6def9564646d68bfbe498c549f.1456962210.git.jpoimboe@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Logging is only used for debugging. Use macros to save calling into the
functions only to return immediately when logging is not enabled.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1443186956-18718-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
TSC packets contain only 7 bytes of TSC. The 8th byte is assumed to
change so infrequently that its value can be inferred. However the
logic must cater for a 7 byte wraparound, which it does by adding 1 to
the top byte.
The existing code was doing that with a while loop even though the
addition should only need to be done once. That logic won't work (will
loop forever) if TSC wraps around at the 8th byte. Theoretically that
would take at least 10 years, unless something else went wrong.
And what else could go wrong. Well, if the chunks of trace data are
processed out of order, it will make it look like the 7-byte TSC has
gone backwards (i.e. wrapped). If that happens 256 times then stuck in
the while loop it will be.
Fix that by getting rid of the unnecessary while loop.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1443186956-18718-4-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add xsavec, xsaves and xrstors to the op code map and the perf tools new
instructions test. To run the test:
$ tools/perf/perf test "x86 ins"
39: Test x86 instruction decoder - new instructions : Ok
Or to see the details:
$ tools/perf/perf test -v "x86 ins" 2>&1 | grep 'xsave\|xrst'
For information about xsavec, xsaves and xrstors, refer the Intel SDM.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qiaowei Ren <qiaowei.ren@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1441196131-20632-8-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add rdpkru and wrpkru to the op code map and the perf tools new
instructions test. In the case of the test, only the bytes can be
tested at the moment since binutils doesn't support the instructions
yet. To run the test:
$ tools/perf/perf test "x86 ins"
39: Test x86 instruction decoder - new instructions : Ok
Or to see the details:
$ tools/perf/perf test -v "x86 ins" 2>&1 | grep pkru
For information about rdpkru and wrpkru, refer the Intel SDM.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qiaowei Ren <qiaowei.ren@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1441196131-20632-7-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Intel Architecture Instruction Set Extensions Programing Reference (Oct
2014) describes 3 new memory instructions, namely clflushopt, clwb and
pcommit. Add them to the op code map and the perf tools new
instructions test. e.g.
$ tools/perf/perf test "x86 ins"
39: Test x86 instruction decoder - new instructions : Ok
Or to see the details:
$ tools/perf/perf test -v "x86 ins"
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qiaowei Ren <qiaowei.ren@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1441196131-20632-6-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Intel SHA Extensions are explained in the Intel Architecture
Instruction Set Extensions Programing Reference (Oct 2014).
There are 7 new instructions. Add them to the op code map
and the perf tools new instructions test. e.g.
$ tools/perf/perf test "x86 ins"
39: Test x86 instruction decoder - new instructions : Ok
Or to see the details:
$ tools/perf/perf test -v "x86 ins" 2>&1 | grep sha
Committer note:
3 lines of details, for the curious:
$ perf test -v "x86 ins" 2>&1 | grep sha256msg1 | tail -3
Decoded ok: 0f 38 cc 84 08 78 56 34 12 sha256msg1 0x12345678(%rax,%rcx,1),%xmm0
Decoded ok: 0f 38 cc 84 c8 78 56 34 12 sha256msg1 0x12345678(%rax,%rcx,8),%xmm0
Decoded ok: 44 0f 38 cc bc c8 78 56 34 12 sha256msg1 0x12345678(%rax,%rcx,8),%xmm15
$
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qiaowei Ren <qiaowei.ren@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1441196131-20632-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The MPX instructions are presently not described in the SDM
opcode maps, and there are not encoding characters for bnd
registers, address method or operand type. So the kernel
opcode map is using 'Gv' for bnd registers and 'Ev' for
everything else. That is fine because the instruction
decoder does not use that information anyway, except as
an indication that there is a ModR/M byte.
Nevertheless, in some cases the 'Gv' and 'Ev' are the wrong
way around, BNDLDX and BNDSTX have 2 operands not 3, and it
wouldn't hurt to identify the mandatory prefixes.
This has no effect on the decoding of valid instructions,
but the addition of the mandatory prefixes will cause some
invalid instructions to error out that wouldn't have
previously.
Note that perf tools has a copy of the instruction decoder
and provides a test for new instructions which includes MPX
instructions e.g.
$ perf test "x86 ins"
39: Test x86 instruction decoder - new instructions : Ok
Or to see the details:
$ perf test -v "x86 ins"
Commiter notes:
And to see these MPX instructions specifically:
$ perf test -v "x86 ins" 2>&1 | grep bndldx | head -3
Decoded ok: 0f 1a 00 bndldx (%eax),%bnd0
Decoded ok: 0f 1a 05 78 56 34 12 bndldx 0x12345678,%bnd0
Decoded ok: 0f 1a 18 bndldx (%eax),%bnd3
$ perf test -v "x86 ins" 2>&1 | grep bndstx | head -3
Decoded ok: 0f 1b 00 bndstx %bnd0,(%eax)
Decoded ok: 0f 1b 05 78 56 34 12 bndstx %bnd0,0x12345678
Decoded ok: 0f 1b 18 bndstx %bnd3,(%eax)
$
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qiaowei Ren <qiaowei.ren@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1441196131-20632-4-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf tools has a copy of the x86 instruction decoder used by the kernel.
The expectation is that the copy will be kept more-or-less in-synch with
the kernel version. Consequently it is helpful to know if there are
differences.
This patch adds a check into the perf tools build so that a diff is done
on the sources, and a warning is printed if they are different. Note
that the warning is not fatal and the build continues as normal.
The check is done as part of building the instruction decoder, so, like
a compiler warning, it is not seen unless the instruction decoder has to
be re-compiled. e.g.
$ make -C tools/perf >/dev/null
$ echo "/* blah */" >> tools/perf/util/intel-pt-decoder/inat_types.h
$ make -C tools/perf >/dev/null
Warning: Intel PT: x86 instruction decoder differs from kernel
$ make -C tools/perf >/dev/null
$
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qiaowei Ren <qiaowei.ren@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1441196131-20632-2-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
I hit following building error randomly:
...
/bin/sh: /path/to/kernel/buildperf/util/intel-pt-decoder/inat-tables.c: No such file or directory
...
LINK /path/to/kernel/buildperf/plugin_mac80211.so
LINK /path/to/kernel/buildperf/plugin_kmem.so
LINK /path/to/kernel/buildperf/plugin_xen.so
LINK /path/to/kernel/buildperf/plugin_hrtimer.so
In file included from util/intel-pt-decoder/intel-pt-insn-decoder.c:25:0:
util/intel-pt-decoder/inat.c:24:25: fatal error: inat-tables.c: No such file or directory
#include "inat-tables.c"
^
compilation terminated.
make[4]: *** [/path/to/kernel/buildperf/util/intel-pt-decoder/intel-pt-insn-decoder.o] Error 1
make[4]: *** Waiting for unfinished jobs....
LINK /path/to/kernel/buildperf/plugin_function.so
This is caused by tools/perf/util/intel-pt-decoder/Build that, it tries
to generate $(OUTPUT)util/intel-pt-decoder/inat-tables.c atomatically
but forget to ensure the existance of $(OUTPUT)util/intel-pt-decoder
directory.
This patch fixes it by adding $(call rule_mkdir) like other similar rules.
Signed-off-by: Wang Nan <wangnan0@huawei.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1441087005-107540-1-git-send-email-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
It is theoretically possible to process perf.data files created on x86
and that contain Intel PT or Intel BTS data, on any other architecture,
which is why it is possible for there to be build errors on powerpc
caused by pt/bts.
The errors were:
util/intel-pt-decoder/intel-pt-insn-decoder.c: In function ‘intel_pt_insn_decoder’:
util/intel-pt-decoder/intel-pt-insn-decoder.c:138:3: error: switch missing default case [-Werror=switch-default]
switch (insn->immediate.nbytes) {
^
cc1: all warnings being treated as errors
linux-acme.git/tools/perf/perf-obj/libperf.a(libperf-in.o): In function `intel_pt_synth_branch_sample':
sources/linux-acme.git/tools/perf/util/intel-pt.c:871: undefined reference to `tsc_to_perf_time'
linux-acme.git/tools/perf/perf-obj/libperf.a(libperf-in.o): In function `intel_pt_sample':
sources/linux-acme.git/tools/perf/util/intel-pt.c:915: undefined reference to `tsc_to_perf_time'
sources/linux-acme.git/tools/perf/util/intel-pt.c:962: undefined reference to `tsc_to_perf_time'
linux-acme.git/tools/perf/perf-obj/libperf.a(libperf-in.o): In function `intel_pt_process_event':
sources/linux-acme.git/tools/perf/util/intel-pt.c:1454: undefined reference to `perf_time_to_tsc'
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>
Cc: Wang Nan <wangnan0@huawei.com>
Cc: Zefan Li <lizefan@huawei.com>
Cc: pi3orama@163.com
Link: http://lkml.kernel.org/r/1441046384-28663-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
A TRACESTOP packet is produced when an Intel PT trace enters a defined
region of the address space at which point the tracing stops.
This patch just adds decoder support.
Support for specifying TRACESTOP regions is left until later.
For details refer to the June 2015 or later Intel 64 and IA-32
Architectures SDM Chapter 36 Intel Processor Trace.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-25-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
CYC packets provide even finer grain timestamp information than MTC and
TSC packets. A CYC packet contains the number of CPU cycles since the
last CYC packet.
This patch just adds decoder support. The CPU frequency can be related
to TSC using the Maximum Non-Turbo Ratio in combination with the CBR
(core-to-bus ratio) packet. However more accuracy is achieved by simply
interpolating the number of cycles between other timing packets like MTC
or TSC. This patch takes the latter approach.
Support for a default value and validation of values is provided by a
later patch. Also documentation is updated in a separate patch.
For details refer to the June 2015 or later Intel 64 and IA-32
Architectures SDM Chapter 36 Intel Processor Trace.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-23-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
MTC packets provide finer grain timestamp information than TSC packets.
MTC packets record time using the hardware crystal clock (CTC) which is
related to TSC packets using a TMA packet.
This patch just adds decoder support.
Support for a default value and validation of values is provided by a
later patch. Also documentation is updated in a separate patch.
For details refer to the June 2015 or later Intel 64 and IA-32
Architectures SDM Chapter 36 Intel Processor Trace.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-21-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Record additional information in the AUXTRACE_INFO event in preparation
for decoding MTC and CYC packets. Pass the information to the decoder.
The AUXTRACE_INFO record can be extended by using the size to indicate
the presence of new members.
The additional information includes PMU config bit positions and the TSC
to CTC (hardware crystal clock) ratio needed to decode MTC packets.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-20-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
New features have been added to Intel PT which include a number of new
packet definitions.
This patch adds packet definitions for new packets: TMA, MTC, CYC, VMCS,
TRACESTOP and MNT. Also another bit in PIP is defined.
This patch only adds support for the definitions. Later patches add
support for decoding TMA, MTC, CYC and TRACESTOP which is where those
packets are explained.
VMCS and the newly defined bit in PIP are used with virtualization which
is not supported yet. MNT is a maintenance packet which the decoder
should ignore.
For details, refer to the June 2015 or later Intel 64 and IA-32
Architectures SDM Chapter 36 Intel Processor Trace.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-19-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The period on synthesized 'instructions' samples was being set to a
fixed value, whereas the correct value is the number of instructions
since the last sample, which is a value that the decoder can provide.
So do it that way.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-14-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Fix some include paths and add missing inat_types.h.
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: http://lkml.kernel.org/r/55D77696.60102@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add support for decoding an Intel Processor Trace.
Intel PT trace data must be 'decoded' which involves walking the object
code and matching the trace data packets.
The decoder requests a buffer of binary data via a get_trace()
call-back, which it decodes using instruction information which it gets
via another call-back walk_insn().
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-6-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a facility to log Intel Processor Trace decoding. The log is
intended for debugging purposes only.
The log file name is "intel_pt.log" and is opened in the current
directory. The log contains a record of all packets and instructions
decoded and can get very large (10 MB would be a small one).
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-5-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add support for decoding instructions for Intel Processor Trace. The
kernel x86 instruction decoder is copied for this.
This essentially provides intel_pt_get_insn() which takes a binary
buffer, uses the kernel's x86 instruction decoder to get details of the
instruction and then categorizes it for consumption by an Intel PT
decoder.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1439450095-30122-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add support for decoding Intel Processor Trace packets.
This essentially provides intel_pt_get_packet() which takes a buffer of
binary data and returns the decoded packet.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/1437150840-31811-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
