| /linux-4.4.14/tools/perf/Documentation/ |
| D | perf-sched.txt | 18 of an arbitrary workload.
21 and other scheduling properties of the workload.
23 'perf sched script' to see a detailed trace of the workload that
26 'perf sched replay' to simulate the workload that was recorded
28 that mimic the workload based on the events in the trace. These
30 of the workload as it occurred when it was recorded - and can repeat
34 workload captured via perf sched record. Columns stand for
|
| D | perf-timechart.txt | 6 perf-timechart - Tool to visualize total system behavior during a workload
18 of an arbitrary workload. By default timechart records only scheduler
|
| D | perf-kmem.txt | 18 of an arbitrary workload.
|
| D | examples.txt | 45 'repeat the workload N times' feature of perf stat:
65 Furthermore, these tracepoints can be used to sample the workload as
|
| D | intel-pt.txt | 526 workload only
529 workload).
531 "workload only" mode is selected by not using the other options but providing a
532 command to run (i.e. the workload).
541 In workload-only mode, the workload is traced but with per-cpu buffers.
542 Inheritance is allowed. Note that you can now trace a workload in per-thread
555 information to decode Intel PT in per-cpu mode, and potentially workload-only
556 mode too if the workload creates new processes.
|
| D | perf-kvm.txt | 24 of an arbitrary workload.
27 of an arbitrary workload and save it into a perf data file. We set the
|
| D | perf-script.txt | 23 'perf script' to see a detailed trace of the workload that was
|
| /linux-4.4.14/tools/perf/tests/ |
| D | perf-record.c | 102 err = sched__get_first_possible_cpu(evlist->workload.pid, &cpu_mask); in test__PERF_RECORD()
114 if (sched_setaffinity(evlist->workload.pid, cpu_mask_size, &cpu_mask) < 0) { in test__PERF_RECORD()
195 if ((pid_t)sample.pid != evlist->workload.pid) { in test__PERF_RECORD()
197 name, evlist->workload.pid, sample.pid); in test__PERF_RECORD()
201 if ((pid_t)sample.tid != evlist->workload.pid) { in test__PERF_RECORD()
203 name, evlist->workload.pid, sample.tid); in test__PERF_RECORD()
212 (pid_t)event->comm.pid != evlist->workload.pid) { in test__PERF_RECORD()
|
| /linux-4.4.14/Documentation/vm/ |
| D | idle_page_tracking.txt | 4 accessed by a workload and which are idle. This information can be useful for
5 estimating the workload's working set size, which, in turn, can be taken into
6 account when configuring the workload parameters, setting memory cgroup limits,
7 or deciding where to place the workload within a compute cluster.
41 workload one should:
43 1. Mark all the workload's pages as idle by setting corresponding bits in
45 /proc/pid/pagemap if the workload is represented by a process, or by
46 filtering out alien pages using /proc/kpagecgroup in case the workload is
49 2. Wait until the workload accesses its working set.
|
| D | zswap.txt | 21 impact to the guest workload and guests sharing the I/O subsystem
|
| D | highmem.txt | 160 machine - although more might work for you and your workload, you're pretty
|
| D | cleancache.txt | 201 overcommitted in a virtual workload); and because the hooks are
|
| D | frontswap.txt | 67 When a workload starts swapping, performance falls through the floor.
|
| D | unevictable-lru.txt | 99 defragmentation, workload management and memory hotplug. The linux kernel
|
| /linux-4.4.14/Documentation/filesystems/nfs/ |
| D | knfsd-stats.txt | 52 Depending on the NFS workload patterns and various network stack
56 However this is a more accurate and less workload-dependent measure
72 pool for the NFS workload (the workload is thread-limited), in which
74 performance of the NFS workload.
91 threads configured than can be used by the NFS workload. This is
97 slow; the idle timeout is 60 minutes. Unless the NFS workload
|
| /linux-4.4.14/tools/perf/util/ |
| D | evlist.c | 46 evlist->workload.pid = -1; in perf_evlist__init()
1518 evlist->workload.pid = fork(); in perf_evlist__prepare_workload()
1519 if (evlist->workload.pid < 0) { in perf_evlist__prepare_workload()
1524 if (!evlist->workload.pid) { in perf_evlist__prepare_workload()
1588 thread_map__set_pid(evlist->threads, 0, evlist->workload.pid); in perf_evlist__prepare_workload()
1602 evlist->workload.cork_fd = go_pipe[1]; in perf_evlist__prepare_workload()
1617 if (evlist->workload.cork_fd > 0) { in perf_evlist__start_workload()
1623 ret = write(evlist->workload.cork_fd, &bf, 1); in perf_evlist__start_workload()
1627 close(evlist->workload.cork_fd); in perf_evlist__start_workload()
|
| D | evlist.h | 53 } workload; member
|
| /linux-4.4.14/Documentation/timers/ |
| D | NO_HZ.txt | 49 However, if you are instead running a light workload with long idle
56 In addition, if you are running either a real-time workload or an HPC
57 workload with short iterations, the scheduling-clock interrupts can
58 degrade your applications performance. If this describes your workload,
235 but do not see any change in your workload's behavior. Is this because
236 your workload isn't affected that much by OS jitter, or is it because
247 possible, then you can conclude that your workload is not all that
322 constraining the workload. For example, the only way to eliminate
|
| /linux-4.4.14/Documentation/ |
| D | kernel-per-CPU-kthreads.txt | 29 # run workload
163 1. Run your workload at a real-time priority, which will allow
201 slowly. Of course, you can also run your workload at
203 but if your workload is CPU-bound, this is a bad idea.
236 is feasible only if your workload never requires RCU priority
273 timer down to a level that is acceptable for your workload.
|
| D | iostats.txt | 138 typical workload usually contains a lot of successive and adjacent requests,
|
| D | kernel-parameters.txt | 417 a microbenchmark depending on workload and compiler.
|
| /linux-4.4.14/Documentation/cpu-freq/ |
| D | governors.txt | 178 defines the workload frequency sensitivity threshold in which a lower
182 the performance of the workload running on a CPU will change when
183 frequency changes. A workload with sensitivity of 0% (memory/IO-bound)
185 workload with sensitivity of 100% (CPU-bound) will perform better
|
| D | boost.txt | 45 save power here, though this depends on the workload.
|
| D | pcc-cpufreq.txt | 88 computes the required performance for each processor based on server workload.
|
| /linux-4.4.14/Documentation/filesystems/ |
| D | ceph.txt | 28 across storage nodes in large chunks to distribute workload and
36 dynamically redistributes metadata in response to workload changes,
|
| D | ext4.txt | 63 workload parameter can completely change the ranking of which
|
| /linux-4.4.14/Documentation/block/ |
| D | cfq-iosched.txt | 32 So depending on storage and workload, it might be useful to set slice_idle=0.
99 latency. But if sequential workload is higher(e.g. sequential read),
273 and notice how a workload can lose its disk share and suffer due to
|
| /linux-4.4.14/Documentation/x86/ |
| D | tlb.txt | 30 sizes of the flush will vary greatly depending on the workload as
|
| /linux-4.4.14/tools/perf/ |
| D | builtin-record.c | 656 rec->evlist->workload.pid, in __cmd_record()
735 kill(rec->evlist->workload.pid, SIGTERM); in __cmd_record()
|
| D | builtin-stat.c | 299 child_pid = evsel_list->workload.pid; in __run_perf_stat()
|
| /linux-4.4.14/drivers/s390/char/ |
| D | Kconfig | 87 identification. This is commonly used for workload management and
|
| /linux-4.4.14/Documentation/ABI/stable/ |
| D | sysfs-driver-ib_srp | 63 spread the SRP completion workload over multiple CPU's.
|
| /linux-4.4.14/fs/squashfs/ |
| D | Kconfig | 80 If you have a parallel I/O workload and your system has enough memory,
|
| /linux-4.4.14/Documentation/trace/ |
| D | tracepoint-analysis.txt | 144 Any workload can exhibit variances between runs and it can be important
|
| /linux-4.4.14/mm/ |
| D | Kconfig | 237 by the guest workload. Allowing the compaction & migration for memory
544 reads, can also improve workload performance.
|
| /linux-4.4.14/Documentation/thermal/ |
| D | intel_powerclamp.txt | 40 they’re only used opportunistically, based on workload. With the
|
| /linux-4.4.14/net/ |
| D | Kconfig | 298 flow that greatly exceeds average workload.
|
| /linux-4.4.14/Documentation/sysctl/ |
| D | vm.txt | 667 small benefits in tuning this to a different value if your workload is
797 zone_reclaim may be enabled if it's known that the workload is partitioned
|
| D | kernel.txt | 469 guarantee. If the target workload is already bound to NUMA nodes then this
497 workload pattern changes and minimises performance impact due to remote
|
| /linux-4.4.14/Documentation/cgroups/ |
| D | blkio-controller.txt | 353 On a faster hardware CFQ can be slow, especially with sequential workload.
|
| D | unified-hierarchy.txt | 541 well, resulting in much better overall workload performance.
|
| /linux-4.4.14/Documentation/RCU/ |
| D | checklist.txt | 185 workload is running.
|
| D | whatisRCU.txt | 915 e. Is your workload too update-intensive for normal use of
|
| /linux-4.4.14/Documentation/filesystems/cifs/ |
| D | README | 312 workload/scenario before using this option. Currently, local
343 attribute cache timeout to suit the workload needs. Shorter
|
| /linux-4.4.14/arch/mips/ |
| D | Kconfig | 1997 You should only select this option if you have a workload that
|