| /linux-6.15/tools/perf/Documentation/ |
| H A D | cpu-and-latency-overheads.txt | 50 parallelization histogram with '--sort=latency,parallelism,comm,symbol --hierarchy'
69 at the given parallelism level:
82 To see the normal function-level profile for particular parallelism levels
83 (number of threads actively running on CPUs), you may use '--parallelism'
84 filter. For example, to see the profile only for low parallelism phases
85 of a workload use '--latency --parallelism=1-2' flags.
|
| H A D | tips.txt | 66 For parallelism histogram, try: perf report --hierarchy --sort latency,parallelism,comm,symbol
67 To analyze particular parallelism levels, try: perf report --latency --parallelism=32-64
68 To see how parallelism changes over time, try: perf report -F time,latency,parallelism --time-quant…
|
| H A D | perf-report.txt | 71 --parallelism::
72 Only consider these parallelism levels. Parallelism level is the number
76 is doing during sequential/low-parallelism phases as compared to
77 high-parallelism phases. This option will affect the percentage of
105 - parallelism: number of running threads at the time of sample
465 and/or --parallelism options and Zoom operations on the TUI (thread, dso, etc).
|
| H A D | topdown.txt | 143 as the parallelism and overlap in the CPU program execution will
|
| /linux-6.15/tools/perf/util/ |
| H A D | addr_location.h | 25 int parallelism; member
|
| H A D | addr_location.c | 20 al->parallelism = 1; in addr_location__init()
|
| H A D | event.c | 771 al->parallelism = max(1, min(machine->parallelism, machine__nr_cpus_avail(machine))); in machine__resolve()
772 if (test_bit(al->parallelism, symbol_conf.parallelism_filter)) in machine__resolve()
779 al->latency = sample->period * 1000 / al->parallelism; in machine__resolve()
|
| H A D | machine.h | 58 int parallelism; member
|
| H A D | hist.h | 248 int parallelism; member
|
| H A D | sort.c | 900 return right->parallelism - left->parallelism; in sort__parallelism_cmp()
910 return test_bit(he->parallelism, parallelism_filter); in hist_entry__parallelism_filter()
916 return repsep_snprintf(bf, size, "%*d", width, he->parallelism); in hist_entry__parallelism_snprintf()
2749 MK_SORT_ENTRY_CHK(parallelism) in MK_SORT_ENTRY_CHK()
|
| H A D | hist.c | 758 .parallelism = al->parallelism, in __hists__add_entry()
2248 if (test_bit(he->parallelism, hists->parallelism_filter)) { in hists__filter_entry_by_parallelism()
|
| H A D | machine.c | 99 machine->parallelism = 1; in machine__init()
687 machine->parallelism += out ? -1 : 1; in machine__process_switch_event()
1902 machine->parallelism--; in machine__process_exit_event()
|
| /linux-6.15/drivers/net/ethernet/sfc/ |
| H A D | efx_channels.c | 127 unsigned int parallelism) in efx_allocate_msix_channels() argument
129 unsigned int n_channels = parallelism; in efx_allocate_msix_channels()
232 efx->n_rx_channels = min(efx->n_rx_channels, parallelism); in efx_allocate_msix_channels()
233 efx->n_tx_channels = min(efx->n_tx_channels, parallelism); in efx_allocate_msix_channels()
259 unsigned int parallelism = efx_wanted_parallelism(efx); in efx_probe_interrupts() local
264 extra_channels, parallelism); in efx_probe_interrupts()
|
| /linux-6.15/drivers/net/ethernet/sfc/siena/ |
| H A D | efx_channels.c | 128 unsigned int parallelism) in efx_allocate_msix_channels() argument
130 unsigned int n_channels = parallelism; in efx_allocate_msix_channels()
233 efx->n_rx_channels = min(efx->n_rx_channels, parallelism); in efx_allocate_msix_channels()
234 efx->n_tx_channels = min(efx->n_tx_channels, parallelism); in efx_allocate_msix_channels()
260 unsigned int parallelism = efx_wanted_parallelism(efx); in efx_siena_probe_interrupts() local
265 extra_channels, parallelism); in efx_siena_probe_interrupts()
|
| /linux-6.15/tools/perf/tests/shell/base_report/ |
| H A D | test_basic.sh | 226 $CMD_PERF report --hierarchy --sort latency,parallelism,comm,symbol --parallelism=1,2 --stdio -i $C…
|
| /linux-6.15/Documentation/block/ |
| H A D | blk-mq.rst | 10 benefiting from the parallelism offered by modern storage devices.
29 device to the operating system. In order to take advantage of the parallelism
|
| /linux-6.15/Documentation/arch/arm/ |
| H A D | kernel_mode_neon.rst | 111 parallelism, and generates NEON code from ordinary C source code. This is fully
|
| /linux-6.15/Documentation/admin-guide/device-mapper/ |
| H A D | vdo.rst | 390 threads increase parallelism and can increase throughput. Physical threads
395 Bio submission threads control the parallelism involved in sending I/O to
|
| /linux-6.15/Documentation/core-api/ |
| H A D | padata.rst | 116 true parallelism is achieved by submitting multiple jobs. parallel() runs with
|
| /linux-6.15/Documentation/virt/hyperv/ |
| H A D | vmbus.rst | 58 channels to achieve higher performance and greater parallelism.
234 higher parallelism and performance. These devices include the
|
| /linux-6.15/Documentation/staging/ |
| H A D | crc32.rst | 141 the parallelism. With the classic Sarwate algorithm, each table lookup
|
| /linux-6.15/Documentation/networking/ |
| H A D | kcm.rst | 236 to the same TCP connection. This allows parallelism between in I/O
|
| /linux-6.15/arch/x86/crypto/ |
| H A D | sha512-avx-asm.S | 170 # stitched to take advantage of instruction-level parallelism.
|
| H A D | sha512-ssse3-asm.S | 165 # stitched to take advantage of instruction-level parallelism.
|
| /linux-6.15/Documentation/admin-guide/hw-vuln/ |
| H A D | core-scheduling.rst | 206 parallelism of the untrusted tasks, it would still solve the above issues while
|