Home
last modified time | relevance | path

Searched full:throughput (Results 1 – 25 of 260) sorted by relevance

1234567891011

/linux/Documentation/block/
H A Dbfq-iosched.rst14 throughput high).
17 throughput. So, when needed for achieving a lower latency, BFQ builds
18 schedules that may lead to a lower throughput. If your main or only
20 throughput at all times, then do switch off all low-latency heuristics
23 latency and throughput, or on how to maximize throughput.
31 instrumentation, and using the throughput-sync.sh script of the S
50 support is enabled), then the sustainable throughput with BFQ
117 High throughput
120 On hard disks, BFQ achieves up to 30% higher throughput than CFQ, and
121 up to 150% higher throughput than DEADLINE and NOOP, with all the
[all …]
/linux/net/batman-adv/
H A Dbat_v_ogm.c342 …"Sending own OGM2 packet (originator %pM, seqno %u, throughput %u, TTL %d) on interface %s [%pM]\n… in batadv_v_ogm_send_meshif()
344 ntohl(ogm_packet->throughput), ogm_packet->ttl, in batadv_v_ogm_send_meshif()
465 * batadv_v_forward_penalty() - apply a penalty to the throughput metric
470 * @throughput: the current throughput
472 * Apply a penalty on the current throughput metric value based on the
475 * Initially the per hardif hop penalty is applied to the throughput. After
477 * - throughput * 50% if the incoming and outgoing interface are the
478 * same WiFi interface and the throughput is above
480 * - throughput if the outgoing interface is the default
483 * - throughput * node hop penalty otherwise
[all …]
H A Dbat_v_elp.c74 * batadv_v_elp_get_throughput() - get the throughput towards a neighbour
75 * @neigh: the neighbour for which the throughput has to be obtained
76 * @pthroughput: calculated throughput towards the given neighbour in multiples
90 u32 throughput; in batadv_v_elp_get_throughput() local
96 throughput = READ_ONCE(hard_iface->bat_v.throughput_override); in batadv_v_elp_get_throughput()
97 if (throughput != 0) { in batadv_v_elp_get_throughput()
98 *pthroughput = throughput; in batadv_v_elp_get_throughput()
102 /* if this is a wireless device, then ask its throughput through in batadv_v_elp_get_throughput()
134 * the throughput metric to 0. in batadv_v_elp_get_throughput()
147 /* try to estimate the expected throughput based on reported tx in batadv_v_elp_get_throughput()
[all …]
H A Dbat_v.c146 ewma_throughput_init(&hardif_neigh->bat_v.throughput); in batadv_v_hardif_neigh_init()
163 u32 throughput; in batadv_v_neigh_dump_neigh() local
167 throughput = ewma_throughput_read(&hardif_neigh->bat_v.throughput); in batadv_v_neigh_dump_neigh()
168 throughput = throughput * 100; in batadv_v_neigh_dump_neigh()
183 nla_put_u32(msg, BATADV_ATTR_THROUGHPUT, throughput)) in batadv_v_neigh_dump_neigh()
304 u32 throughput; in batadv_v_orig_dump_subentry() local
311 throughput = n_ifinfo->bat_v.throughput * 100; in batadv_v_orig_dump_subentry()
333 nla_put_u32(msg, BATADV_ATTR_THROUGHPUT, throughput) || in batadv_v_orig_dump_subentry()
479 * batadv_v_neigh_cmp() - compare two B.A.T.M.A.N. V neighbours by throughput
486 * is better and 0 if both have equal throughput
[all …]
/linux/net/x25/
H A Dx25_facilities.c103 facilities->throughput = p[1]; in x25_parse_facilities()
210 if (facilities->throughput && (facil_mask & X25_MASK_THROUGHPUT)) { in x25_create_facilities()
212 *p++ = facilities->throughput; in x25_create_facilities()
291 if (theirs.throughput) { in x25_negotiate_facilities()
292 int theirs_in = theirs.throughput & 0x0f; in x25_negotiate_facilities()
293 int theirs_out = theirs.throughput & 0xf0; in x25_negotiate_facilities()
294 int ours_in = ours->throughput & 0x0f; in x25_negotiate_facilities()
295 int ours_out = ours->throughput & 0xf0; in x25_negotiate_facilities()
297 net_dbg_ratelimited("X.25: inbound throughput negotiated\n"); in x25_negotiate_facilities()
298 new->throughput = (new->throughput & 0xf0) | theirs_in; in x25_negotiate_facilities()
[all …]
/linux/tools/testing/selftests/bpf/benchs/
H A Drun_common.sh38 echo -n "throughput: "
39 echo -n "$*" | sed -E "s/.*throughput\s+([0-9]+\.[0-9]+ ± [0-9]+\.[0-9]+\sM\sops\/s).*/\1/"
46 echo -n "hits throughput: "
47 echo -n "$*" | sed -E "s/.* hits throughput\s+([0-9]+\.[0-9]+ ± [0-9]+\.[0-9]+\sM\sops\/s).*/\1/"
50 echo -n ", important_hits throughput: "
51 …echo "$*" | sed -E "s/.*important_hits throughput\s+([0-9]+\.[0-9]+ ± [0-9]+\.[0-9]+\sM\sops\/s).*…
/linux/include/linux/dma/
H A Dk3-psil.h16 * enum udma_tp_level - Channel Throughput Levels
18 * @UDMA_TP_HIGH: High Throughput channel
19 * @UDMA_TP_ULTRAHIGH: Ultra High Throughput channel
45 * @channel_tpl: Desired throughput level for the channel
/linux/tools/testing/selftests/amd-pstate/
H A Dtbench.sh5 # power consumption and throughput etc.when this script trigger tbench
10 # 4) Get power consumption and throughput by amd_pstate_trace.py.
98 …grep Throughput $OUTFILE_TBENCH-perf-$1-$2.log | awk '{print $2}' > $OUTFILE_TBENCH-throughput-$1-…
99 tp_sum=$(awk 'BEGIN {sum=0};{sum += $1};END {print sum}' $OUTFILE_TBENCH-throughput-$1-$2.log)
100 printf "Tbench-$1-#$2 throughput(MB/s): $tp_sum\n" | tee -a $OUTFILE_TBENCH.result
106 # Permance is throughput per second, denoted T/t, where T is throught rendered in t seconds.
150 …bench-$1-#" $OUTFILE_TBENCH.result | grep "throughput(MB/s):" | awk '{print $NF}' > $OUTFILE_TBENC…
151 tp_sum=$(awk 'BEGIN {sum=0};{sum += $1};END {print sum}' $OUTFILE_TBENCH-throughput-$1.log)
152 printf "Tbench-$1 total throughput(MB/s): $tp_sum\n" | tee -a $OUTFILE_TBENCH.result
154 …avg_tp=$(awk 'BEGIN {sum=0};{sum += $1};END {print sum/'$LOOP_TIMES'}' $OUTFILE_TBENCH-throughput-…
[all …]
/linux/drivers/md/
H A Ddm-ps-service-time.c9 * Throughput oriented path selector.
126 * <relative_throughput>: The relative throughput value of in st_add_path()
224 * Case 1: Both have same throughput value. Choose less loaded path. in st_compare_load()
230 * Case 2a: Both have same load. Choose higher throughput path. in st_compare_load()
231 * Case 2b: One path has no throughput value. Choose the other one. in st_compare_load()
272 * Case 4: Service time is equal. Choose higher throughput path. in st_compare_load()
362 MODULE_DESCRIPTION(DM_NAME " throughput oriented path selector");
/linux/Documentation/admin-guide/device-mapper/
H A Ddm-service-time.rst11 The performance value is a relative throughput value among all paths
25 The relative throughput value of the path
42 The relative throughput value of the path
79 and sda has an average throughput 1GB/s and sdb has 4GB/s,
/linux/net/core/
H A Dieee8021q_helpers.c156 * [RFC4594], Section 4.8, recommends High-Throughput Data be marked in ietf_dscp_to_ieee8021q_tt()
160 * By default (as described in Section 2.3), High-Throughput Data will in ietf_dscp_to_ieee8021q_tt()
165 * Throughput Data service class within the constrained 4 Access in ietf_dscp_to_ieee8021q_tt()
166 * Category [IEEE.802.11-2016] model. If the High-Throughput Data in ietf_dscp_to_ieee8021q_tt()
177 * is generally RECOMMENDED to map High-Throughput Data to UP 0, thereby in ietf_dscp_to_ieee8021q_tt()
/linux/block/
H A Dbfq-iosched.c28 * to distribute the device throughput among processes as desired,
29 * without any distortion due to throughput fluctuations, or to device
34 * guarantees that each queue receives a fraction of the throughput
37 * processes issuing sequential requests (to boost the throughput),
76 * preserving both a low latency and a high throughput on NCQ-capable,
81 * the maximum-possible throughput at all times, then do switch off
190 * writes to steal I/O throughput to reads.
240 * because it is characterized by limited throughput and apparently
320 * a) unjustly steal throughput to applications that may actually need
323 * in loss of device throughput with most flash-based storage, and may
[all …]
/linux/drivers/gpu/drm/radeon/
H A Dradeon_benchmark.c80 unsigned int throughput = (n * (size >> 10)) / time; in radeon_benchmark_log_results() local
82 " %d to %d in %u ms, throughput: %u Mb/s or %u MB/s\n", in radeon_benchmark_log_results()
84 throughput * 8, throughput); in radeon_benchmark_log_results()
/linux/drivers/net/wireless/morsemicro/mm81x/
H A Dmmrc.c8 * The default packet size in bits used for calculated throughput of a given
56 * It could mean that some rates will have better probability throughput even
107 * throughput, otherwise the history decays in a single cycle.
454 * precision as the throughput is already multiplied by 1000 in in calculate_throughput()
599 * Besides better throughput, also consider this rate better if in generate_table_priority()
635 /* Drop one rate, as the best throughput is zero */ in generate_table_priority()
732 * packet size and theoretical throughput.
987 * Give the best rate at least 2 attempts to keep peak throughput in mmrc_get_rates()
1100 u32 throughput; in mmrc_update() local
1189 throughput = calculate_throughput(tb, i); in mmrc_update()
[all …]
/linux/drivers/net/wireless/intel/iwlegacy/
H A D4965-rs.c139 * The following tables contain the expected throughput metrics for all rates
379 * Static function to get the expected throughput from an il_scale_tbl_info
411 /* Get expected throughput */ in il4965_rs_collect_tx_data()
459 /* Calculate average throughput, if we have enough history. */ in il4965_rs_collect_tx_data()
974 * Set frame tx success limits according to legacy vs. high-throughput,
1002 * Find correct throughput table for given mode of modulation
1046 * Find starting rate for new "search" high-throughput mode of modulation.
1048 * above the current measured throughput of "active" mode, to give new mode
1054 * to decrease to match "active" throughput. When moving from MIMO to SISO,
1068 /* expected "search" throughput */ in il4965_rs_get_best_rate()
[all …]
/linux/Documentation/ABI/testing/
H A Dsysfs-driver-xdata17 The user can read the current PCIe link throughput generated
42 The user can read the current PCIe link throughput generated
/linux/kernel/
H A DKconfig.preempt29 throughput. It will still provide good latencies most of the
49 at the cost of slightly lower throughput.
71 system is under load, at the cost of slightly lower throughput
/linux/tools/testing/selftests/net/ovpn/
H A Dtest-close-socket.sh66 ovpn_cmd_ok "run iperf throughput flow" \
82 ovpn_run_stage "run iperf throughput" ovpn_run_iperf
/linux/drivers/interconnect/imx/
H A Dimx.h24 * High throughput priority level in Regulator mode
29 * Low throughput priority level in Regulator mode
/linux/Documentation/arch/arm/omap/
H A Domap_pm.rst7 throughput constraints to the kernel power management code.
20 latency and throughput, rather than units which are specific to OMAP
46 4. Set the minimum bus throughput needed by a device::
/linux/Documentation/networking/device_drivers/ethernet/intel/
H A De1000.rst148 for bulk throughput traffic.
172 the overall throughput, we recommend that you load the driver as
215 extra latency to frame reception and can end up decreasing the throughput
240 along with RxIntDelay, may improve traffic throughput in specific network
297 along with TxIntDelay, may improve traffic throughput in specific
413 Degradation in throughput performance may be observed in some Jumbo frames
H A De1000e.rst54 increased CPU utilization, though it may help throughput in some circumstances.
84 However, this is generally not suitable for bulk throughput traffic due
123 latency to frame reception and can end up decreasing the throughput of TCP
142 RxIntDelay, may improve traffic throughput in specific network conditions.
164 may improve traffic throughput in specific network conditions.
/linux/Documentation/misc-devices/
H A Ddw-xdata-pcie.rst40 Get link throughput in MB/s::
56 Get link throughput in MB/s::
/linux/drivers/net/wireless/intel/iwlwifi/dvm/
H A Drs.c150 * The following tables contain the expected throughput metrics for all rates
411 * Static function to get the expected throughput from an iwl_scale_tbl_info
441 /* Get expected throughput */ in rs_collect_tx_data()
489 /* Calculate average throughput, if we have enough history. */ in rs_collect_tx_data()
1035 * Set frame tx success limits according to legacy vs. high-throughput,
1062 * Find correct throughput table for given mode of modulation
1109 * Find starting rate for new "search" high-throughput mode of modulation.
1111 * above the current measured throughput of "active" mode, to give new mode
1117 * to decrease to match "active" throughput. When moving from MIMO to SISO,
1130 /* expected "search" throughput */ in rs_get_best_rate()
[all …]
H A Drs.h111 /* uCode API values for OFDM high-throughput (HT) bit rates */
250 LQ_SISO, /* high-throughput types */
280 s32 average_tpt; /* success ratio * expected throughput */
298 const u16 *expected_tpt; /* throughput metrics; expected_tpt_G, etc. */

1234567891011