Home
last modified time | relevance | path

Searched refs:is (Results 1 – 25 of 298) sorted by relevance

12345678910>>...12

/tools/scripts/
A Dutilities.mak12 # This is used as the common way to specify
14 # newlines; the default is a bizarre string.
27 # The only solution is to change each newline into
47 # the default escape is a bizarre string.
63 # the default escape is a bizarre string.
82 # This function produces text that is suitable for
131 # is-absolute
133 # Usage: bool-value = $(call is-absolute,path)
147 # is-executable
149 # Usage: bool-value = $(call is-executable,path)
[all …]
/tools/testing/vsock/
A DREADME17 4. Boot the guest and ensure that the AF_VSOCK transport is enabled.
40 however, is deferred to Kernel Memory Leak Detector. It is recommended to enable
56 'vsock_perf' is a simple tool to measure vsock performance. It works in
58 starts data transmission to the receiver. After data processing is done,
63 # connect to CID 2, port 1234, send 1G of data, tx buf size is 1M
70 A is calculated as "number of bits to send" / "time in tx loop"
73 # listen port 1234, rx buf size is 1M, socket buf size is 1G, SO_RCVLOWAT is 64K
83 A is calculated as "number of received bits" / "time in rx loop".
84 B is time, spent in 'read()' system call(excluding 'poll()')
85 C is number of 'poll()' wake ups with POLLIN bit set.
[all …]
/tools/perf/Documentation/
A Dintel-hybrid.txt5 For some Intel platforms, such as AlderLake, which is hybrid platform and
34 the event is belong to. Same event name but with different pmu can
52 two events are created automatically. One is for atom, the other is for
60 type. The PMU type ID is stored at attr.config[63:32].
62 PMU type ID is retrieved from sysfs.
72 Cache event is similar. The type PERF_TYPE_HW_CACHE is extended to be
73 PMU aware type. The PMU type ID is stored at attr.config[63:32].
111 type 0 is PERF_TYPE_HARDWARE.
125 The first 'cycles' is core event, the second 'cycles' is atom event.
133 scaled value for core cycles is 160,444,092 and the percentage is 0.47%.
[all …]
A Dperf-config.txt317 Default is true. This option is only applied to TUI.
335 Default is 'false'.
355 Default is 'false'.
371 Default is 'false'.
386 Default is 'false'.
490 If this option is not set and report.children or top.children is
505 control the threshold. Default is 0.5 (%). The overhead is calculated
564 This is identical to 'call-graph.record-mode', except it is
586 '--slab' nor '--page' option is used. Default is 'slab'.
598 This is identical to 'call-graph.record-mode', except it is
[all …]
A Dperf-kwork.txt92 Input file name. (default: perf.data unless stdin is a fifo)
108 have the format seconds.microseconds. If start is not given (i.e., time
109 string is ',x.y') then analysis starts at the beginning of the file. If
110 stop time is not given (i.e, time string is 'x.y,') then analysis goes
126 Input file name. (default: perf.data unless stdin is a fifo)
138 have the format seconds.microseconds. If start is not given (i.e., time
140 stop time is not given (i.e, time string is 'x.y,') then analysis goes
156 Input file name. (default: perf.data unless stdin is a fifo)
179 stop time is not given (i.e, time string is 'x.y,') then analysis goes
195 Input file name. (default: perf.data unless stdin is a fifo)
[all …]
A Djitdump-specification.txt21 The format of the file is binary. It is self-describing in terms of endianness and is portable acro…
31 Each entry in the file is timestamped. All timestamps must use the same clock source. The CLOCK_MON…
55 The record header is specified in order as follows:
82 The record header total_size field is inclusive of all components:
89 …is used to uniquely identify each jitted function. The index can be a monotonically increasing 64-…
96 The record type is optional.
117 The record type is optional.
129 * uint32_t discrim : column discriminator, 0 is default
142 The record type is optional.
152 The record type is optional.
[all …]
A Dperf-help.txt19 If the option '--all' or '-a' is given, then all available commands are
22 If a perf command is named, a manual page for that command is brought
23 up. The 'man' program is used by default for this purpose, but this
26 Note that `perf --help ...` is identical to `perf help ...` because the
27 former is internally converted into the latter.
A Dintel-bts.txt9 notable difference is that Intel BTS has no timing information and as a
10 consequence the present implementation is limited to per-thread recording.
13 code is still needed to pair up calls and returns correctly, consequently much
28 option is:
32 Currently Intel BTS is limited to per-thread tracing so the --per-thread option
33 is also needed.
39 The snapshot option is the same as Intel PT (refer Intel PT documentation).
45 The mmap size option is the same as Intel PT (refer Intel PT documentation).
52 This can be further controlled by option --itrace. The --itrace option is
66 When -D is used, Intel BTS packets are displayed.
[all …]
A Dcallchain-overhead-calculation.txt5 wall-clock overhead). The 'self' overhead is simply calculated by
7 This is the value that perf shows traditionally and sum of all the
10 The 'children' overhead is calculated by adding all period values of
17 values exceeds 100% since each of them is already an accumulation of
40 In this case 'foo' is a child of 'bar', and 'bar' is an immediate
41 child of 'main' so 'foo' also is a child of 'main'. In other words,
42 'main' is a parent of 'foo' and 'bar', and 'bar' is a parent of 'foo'.
65 When the --children option is enabled, the 'self' overhead values of
105 Since v3.16 the 'children' overhead is shown by default and the output
106 is sorted by its values. The 'children' overhead is disabled by
A Dperf-intel-pt.txt57 It is important to start small. That is because it is easy to capture vastly
73 --kcore is used, but access to /proc/kcore is restricted e.g.
262 is the same as:
287 perf_event_attr is displayed if the -vv option is used e.g.
389 If decoding is expected to be reliable and the buffer is large
440 that is supported (0 is always supported).
503 is executed.
606 is the same as:
625 sample size is limited by the maximum event size which is 64KiB. It is
655 The snapshot size is displayed if the option -vv is used e.g.
[all …]
A Dperf-probe.txt45 or lines. If a path of module file is passed, perf-probe
87 syntax is same as PROBE SYNTAX, but NO ARGs.
144 Demangle application symbols. --no-demangle is also available
177 …is also possible to specify a probe point by the source line number or lazy matching by using 'SRC…
179 'SDTEVENT' and 'PROVIDER' is the pre-defined event name which is defined by user SDT (Statically De…
191 See EXAMPLES how it is used.
202 On x86 systems %REG is always the short form of the register: for example %AX. %RAX or %EAX is not …
203 "@user" is a special attribute which means the LOCALVAR will be treated as a user-space memory. Thi…
209 Bitfield is another special type, which takes 3 parameters, bit-width, bit-offset, and container-si…
215 Line range is described by following syntax.
[all …]
A Dperf-arm-spe.txt27 5. Interrupt when the buffer is full
32 This is chosen from a sample population, for SPE this is an IMPLEMENTATION DEFINED choice of all
35 sample. This minimum interval is used by the driver if no interval is specified. A pseudo-random
36 perturbation is also added to the sampling interval by default.
42 Sampling ensures there is only one sampled operation is in flight.
53 The record is appended to a memory buffer
55 Interrupt when the buffer is full
75 hardware. Only one sampled operation is in flight at a time.
123 The SPE interrupt must also be described by the firmware. If the module is loaded and KPTI is
135 The sample period is set from the -c option, and because the minimum interval is used by default
[all …]
A Dperf-record.txt61 The default is 1.
140 <file name> is the name of the object file, <start> is the offset to the
144 If no object file is specified then the kernel is assumed, in which case
153 omitted and <start> is a symbol, then the size is calculated to the end
156 If <size> is omitted and <start> is '*', then the start and size will
314 main limitation is that it is only available on new Intel
494 inheritance is automatically disabled. --per-thread is ignored with a warning
509 is off by default. It is possible to select the registers to sample using their
538 In Snapshot Mode trace data is captured only when signal SIGUSR2 is received
698 record is finished.
[all …]
/tools/memory-model/litmus-tests/
A DREADME6 Test of read-read coherence, that is, whether or not two
10 Test of read-write coherence, that is, whether or not a read
20 Test of write-write coherence, that is, whether or not two
42 S is otherwise known as ISA2.
144 The second is forbidden because smp_store_release() is
177 The end of the name is ".litmus".
216 P0()'s WRITE_ONCE() is read by its first READ_ONCE(), which is a
229 The resulting descriptor is "PodRR". Because P0()'s third access is
233 access, and the resulting descriptor is "Fre". P1()'s first access is
235 thus far is thus "Rfi Once PodRR Once Fre Once".
[all …]
A DWRC+pooncerelease+fencermbonceonce+Once.litmus6 * This litmus test is an extension of the message-passing pattern, where
7 * the first write is moved to a separate process. Because it features
9 * specifically, this litmus test is forbidden because smp_store_release()
10 * is A-cumulative in LKMM.
/tools/perf/pmu-events/
A DREADME5 The main program in this directory, is the 'jevents', which is built and
6 executed _BEFORE_ the perf binary itself is built.
23 folder. Matching is based on the "EventName" field.
46 is derived from JSON file name, 'power8.json').
75 After the 'pmu-events.c' is generated, it is compiled and the resulting
76 'pmu-events.o' is added to 'libperf.a' which is then used to build perf.
95 where 'pm_1plus_ppc_cmpl' is a Power8 PMU event.
103 It is required even if such mapping is 1:1.
105 The mapfile.csv format is expected to be:
121 The header line is the first line in the file, which is
[all …]
/tools/testing/kunit/test_data/
A Dtest_is_test_passed-no_tests_run_no_header.log5 Checking if /dev/shm is on tmpfs...OK
39 Using a channel type which is configured out of UML
41 Using a channel type which is configured out of UML
43 Using a channel type which is configured out of UML
45 Using a channel type which is configured out of UML
47 Using a channel type which is configured out of UML
49 Using a channel type which is configured out of UML
51 Using a channel type which is configured out of UML
53 Using a channel type which is configured out of UML
55 Using a channel type which is configured out of UML
[all …]
A Dtest_output_isolated_correctly.log61 Using a channel type which is configured out of UML
63 Using a channel type which is configured out of UML
65 Using a channel type which is configured out of UML
67 Using a channel type which is configured out of UML
69 Using a channel type which is configured out of UML
71 Using a channel type which is configured out of UML
73 Using a channel type which is configured out of UML
75 Using a channel type which is configured out of UML
77 Using a channel type which is configured out of UML
79 Using a channel type which is configured out of UML
[all …]
/tools/testing/selftests/net/packetdrill/
A Dtcp_zerocopy_epoll_oneshot.pkt4 // This is a test to confirm that EPOLLERR is only fired once for an FD when
5 // EPOLLONESHOT is set.
8 // is correctly fired only once, when EPOLLONESHOT is set. send another packet
9 // with MSG_ZEROCOPY. confirm that EPOLLERR is not fired. Rearm the FD and
10 // confirm that EPOLLERR is correctly set.
49 // rearm the FD and verify the EPOLLERR is fired again.
A Dtcp_zerocopy_epoll_exclusive.pkt4 // EPOLLERR is known to be not edge-triggered unlike EPOLLIN and EPOLLOUT but
5 // it is not level-triggered either. this tests verify that the same behavior is
9 // is correctly fired only once, when EPOLLET is set. send another packet with
10 // MSG_ZEROCOPY. confirm that EPOLLERR is correctly fired again only once.
/tools/memory-model/Documentation/
A Drecipes.txt15 there is only one CPU or only one memory location is accessed, and the
28 called is not defined; the object code is allowed to use either
252 pointers. The general approach is shown below:
288 that is expanded as additional primes are needed.
442 It is tempting to assume that CPU0()'s store to x is globally ordered
466 of z is 2, then the final value of r1 must also be 1, but it really is
468 is that in this version, CPU2() is not part of the release-acquire chain.
518 On CPU0, the store is to @cond and the load is in waitqueue_active().
531 needed in which situations, but this is not the case. There is a pattern
538 relation is called "reads-from" and is usually abbreviated "rf".
[all …]
A Dglossary.txt2 glossaries, it is not intended to be read front to back (except perhaps
18 In this case, because the address of "p->a" on line 3 is computed
31 An example special acquire operation is smp_load_acquire(),
44 there is said to be a coherence link from the second CPU to
48 is a "coherence internal" (coi) link. The term "coherence
49 external" (coe) link is used when it is necessary to exclude
84 are all equal to zero is forbidden by LKMM.
112 external" (fre) link is used when it is necessary to exclude
144 CPU, there is said to be a reads-from link from the second
152 is a "reads-from internal" (rfi) link. The term "reads-from
[all …]
A Dcontrol-dependencies.txt4 A major difficulty with control dependencies is that current compilers
5 do not support them. One purpose of this document is therefore to
11 One such challenge is that control dependencies order only later stores.
13 unless a read memory barrier is provided. Consider the following code:
19 This is not guaranteed to provide any ordering because some types of CPUs
31 However, stores are not speculated. This means that ordering is
49 "a" is always non-zero, it would be well within its rights to optimize
59 It is tempting to try use control dependencies to enforce ordering on
89 conditional is absolutely required, and must be present in the final
146 make sure that MAX is greater than one, perhaps as follows:
[all …]
/tools/tracing/rtla/
A DMakefile.config20 $(info libtraceevent version is too low, it must be at least $(LIBTRACEEVENT_MIN_VERSION))
27 $(info libtraceevent is missing. Please install libtraceevent-dev/libtraceevent-devel)
36 $(info libtracefs version is too low, it must be at least $(LIBTRACEFS_MIN_VERSION))
43 $(info libtracefs is missing. Please install libtracefs-dev/libtracefs-devel)
52 $(info libcpupower is missing, building without --deepest-idle-state support.)
70 $(info libbpf is missing, building without BPF skeleton support.)
79 $(info clang is missing or does not support BPF CO-RE, building without BPF skeleton support.)
88 $(info bpftool is missing or not supporting skeletons, building without BPF skeleton support.)
/tools/perf/
A Ddesign.txt35 When creating a new counter fd, 'perf_event_attr' is:
72 is divided into 3 bit-fields:
83 this is, with the following encoding:
153 counters. A "counting" counter is one that is used for counting the
177 A "sampling" counter is one that is set up to generate an interrupt
226 CPU is in user, kernel and/or hypervisor mode.
243 pid == 0: if the pid parameter is zero, the counter is attached to the
274 (A single counter on its own is created with group_fd = -1 and is
277 A counter group is scheduled onto the CPU as a unit, that is, it will
291 as where the ring-buffer head is.
[all …]

Completed in 34 milliseconds

12345678910>>...12