Home
last modified time | relevance | path

Searched refs:overhead (Results 1 – 25 of 240) sorted by relevance

12345678910

/Linux-v5.4/tools/testing/selftests/powerpc/pmu/
Dcount_instructions.c30 u64 overhead, bool report) in do_count_loop() argument
45 expected = instructions + overhead; in do_count_loop()
53 printf("Looped for %llu instructions, overhead %llu\n", instructions, overhead); in do_count_loop()
76 u64 current, overhead; in determine_overhead() local
80 overhead = events[0].result.value; in determine_overhead()
85 if (current < overhead) { in determine_overhead()
86 printf("Replacing overhead %llu with %llu\n", overhead, current); in determine_overhead()
87 overhead = current; in determine_overhead()
91 return overhead; in determine_overhead()
97 u64 overhead; in test_body() local
[all …]
/Linux-v5.4/tools/testing/selftests/powerpc/pmu/ebb/
Dinstruction_count_test.c26 uint64_t overhead, bool report) in do_count_loop() argument
46 expected = instructions + overhead; in do_count_loop()
51 printf("Looped for %lu instructions, overhead %lu\n", instructions, overhead); in do_count_loop()
72 uint64_t current, overhead; in determine_overhead() local
76 overhead = event->result.value; in determine_overhead()
81 if (current < overhead) { in determine_overhead()
82 printf("Replacing overhead %lu with %lu\n", overhead, current); in determine_overhead()
83 overhead = current; in determine_overhead()
87 return overhead; in determine_overhead()
112 uint64_t overhead; in instruction_count() local
[all …]
/Linux-v5.4/tools/perf/Documentation/
Dcallchain-overhead-calculation.txt3 The overhead can be shown in two columns as 'Children' and 'Self' when
4 perf collects callchains. The 'self' overhead is simply calculated by
7 'self' overhead values should be 100%.
9 The 'children' overhead is calculated by adding all period values of
10 the child functions so that it can show the total overhead of the
15 It might be confusing that the sum of all the 'children' overhead
17 'self' overhead of its child functions. But with this enabled, users
18 can find which function has the most overhead even if samples are
45 in the usual (self-overhead-only) output of perf report:
64 When the --children option is enabled, the 'self' overhead values of
[all …]
Dperf-config.txt184 'top' means a overhead percentage which is more than 5%.
189 'medium' means a overhead percentage which has more than 0.5%.
192 'normal' means the rest of overhead percentages
346 But if this option is 'false', percent values for overhead are printed i.e.
358 This option control the way to calculate overhead of filtered entries -
368 This is an original overhead and we'll filter out the first 'foo'
369 entry. The value of 'relative' would increase the overhead of 'bar'
371 current overhead (33.33%).
395 flat and folded. This option controls a way to show overhead for each callchain
412 (meaning that it assumes 100% total overhead of 'foo').
[all …]
Dperf-top.txt127 local_weight, abort, in_tx, transaction, overhead, sample, period.
133 overhead, overhead_sys, overhead_us, overhead_children, sample and period.
148 percentage of the overhead column. See --percentage for more info.
152 percentage of the overhead column. See --percentage for more info.
156 percentage of the overhead column. See --percentage for more info.
180 enabled. See the `overhead calculation' section for more details.
197 Do not show entries which have an overhead under that percent.
201 Determine how to display the overhead percentage of filtered entries.
288 Alternatively one can ask for --group and then two overhead columns
346 include::callchain-overhead-calculation.txt[]
Dperf-report.txt47 the overhead column. See --percentage for more info.
57 the overhead column. See --percentage for more info.
62 the overhead column. See --percentage for more info.
99 - overhead: Overhead percentage of sample
109 --time-quantum (default 100ms). Specify with overhead and before it.
136 executed, such as a memory access bottleneck. If a function has high overhead
184 overhead, overhead_sys, overhead_us, overhead_children, sample and period.
228 - graph: use a graph tree, displaying absolute overhead rates. (default)
257 - percent: display overhead percent (default)
265 See the `overhead calculation' section for more details. Enabled by
[all …]
Dtips.txt32 Order by the overhead of source file name and line number: perf report -s srcline
39 To separate samples by time use perf report --sort time,overhead,sym
/Linux-v5.4/drivers/net/fddi/skfp/
Dess.c101 static int process_bw_alloc(struct s_smc *smc, long int payload, long int overhead);
120 u_long payload, overhead ; in ess_raf_received_pack() local
258 overhead = ((struct smt_p_3210 *)p)->mib_overhead ; in ess_raf_received_pack()
261 payload, overhead); in ess_raf_received_pack()
266 (void)process_bw_alloc(smc,(long)payload,(long)overhead) ; in ess_raf_received_pack()
309 overhead = ((struct smt_p_3210 *)p)->mib_overhead ; in ess_raf_received_pack()
314 payload, overhead); in ess_raf_received_pack()
319 if(!process_bw_alloc(smc,(long)payload,(long)overhead)) in ess_raf_received_pack()
376 static int process_bw_alloc(struct s_smc *smc, long int payload, long int overhead) in process_bw_alloc() argument
427 if (payload > MAX_PAYLOAD || overhead > 5000) { in process_bw_alloc()
[all …]
/Linux-v5.4/drivers/net/wireless/ralink/rt2x00/
Drt2x00crypto.c68 unsigned int overhead = 0; in rt2x00crypto_tx_overhead() local
71 return overhead; in rt2x00crypto_tx_overhead()
78 overhead += key->icv_len; in rt2x00crypto_tx_overhead()
81 overhead += key->iv_len; in rt2x00crypto_tx_overhead()
85 overhead += 8; in rt2x00crypto_tx_overhead()
88 return overhead; in rt2x00crypto_tx_overhead()
/Linux-v5.4/fs/freevxfs/
Dvxfs_lookup.c108 int overhead = VXFS_DIRBLKOV(sbi, dbp); in vxfs_find_entry() local
110 pos += overhead; in vxfs_find_entry()
111 pg_ofs += overhead; in vxfs_find_entry()
259 int overhead = VXFS_DIRBLKOV(sbi, dbp); in vxfs_readdir() local
261 pos += overhead; in vxfs_readdir()
262 pg_ofs += overhead; in vxfs_readdir()
/Linux-v5.4/net/sctp/
Doutput.c60 packet->size = packet->overhead; in sctp_packet_reset()
95 packet->overhead = sctp_mtu_payload(sp, 0, 0); in sctp_packet_config()
96 packet->size = packet->overhead; in sctp_packet_config()
153 packet->overhead = 0; in sctp_packet_init()
428 pkt_size = packet->overhead; in sctp_packet_pack()
435 else if (auth_len + padded + packet->overhead > in sctp_packet_pack()
445 skb_reserve(nskb, packet->overhead + MAX_HEADER); in sctp_packet_pack()
449 pkt_size -= packet->overhead; in sctp_packet_pack()
572 head = alloc_skb((gso ? packet->overhead : packet->size) + in sctp_packet_transmit()
576 skb_reserve(head, packet->overhead + MAX_HEADER); in sctp_packet_transmit()
[all …]
/Linux-v5.4/net/l2tp/
Dl2tp_eth.c221 unsigned int overhead = 0; in l2tp_eth_adjust_mtu() local
227 overhead += sizeof(struct udphdr); in l2tp_eth_adjust_mtu()
246 overhead += session->hdr_len + ETH_HLEN + l3_overhead; in l2tp_eth_adjust_mtu()
248 mtu = l2tp_tunnel_dst_mtu(tunnel) - overhead; in l2tp_eth_adjust_mtu()
250 dev->mtu = ETH_DATA_LEN - overhead; in l2tp_eth_adjust_mtu()
/Linux-v5.4/include/net/sctp/
Dsctp.h577 __u32 overhead = sizeof(struct sctphdr) + extra; in sctp_mtu_payload() local
580 overhead += sp->pf->af->net_header_len; in sctp_mtu_payload()
582 overhead += sizeof(struct ipv6hdr); in sctp_mtu_payload()
584 if (WARN_ON_ONCE(mtu && mtu <= overhead)) in sctp_mtu_payload()
585 mtu = overhead; in sctp_mtu_payload()
587 return mtu ? mtu - overhead : overhead; in sctp_mtu_payload()
/Linux-v5.4/fs/ext4/
Dresize.c77 ext4_grpblk_t overhead; in ext4_group_overhead_blocks() local
78 overhead = ext4_bg_num_gdb(sb, group); in ext4_group_overhead_blocks()
80 overhead += 1 + in ext4_group_overhead_blocks()
82 return overhead; in ext4_group_overhead_blocks()
97 unsigned overhead; in verify_group_input() local
109 overhead = ext4_group_overhead_blocks(sb, group); in verify_group_input()
110 metaend = start + overhead; in verify_group_input()
112 input->blocks_count - 2 - overhead - sbi->s_itb_per_group; in verify_group_input()
264 unsigned overhead; in ext4_alloc_group_tables() local
282 overhead = ext4_group_overhead_blocks(sb, src_group); in ext4_alloc_group_tables()
[all …]
/Linux-v5.4/net/dsa/
Dtag_ksz.c127 .overhead = KSZ8795_INGRESS_TAG_LEN,
202 .overhead = KSZ9477_INGRESS_TAG_LEN,
240 .overhead = KSZ_INGRESS_TAG_LEN,
Dtag_brcm.c179 .overhead = BRCM_TAG_LEN,
207 .overhead = BRCM_TAG_LEN,
/Linux-v5.4/Documentation/driver-api/mmc/
Dmmc-async-req.rst8 How significant is the cache maintenance overhead?
11 pre-fetch makes the cache overhead relatively significant. If the DMA
13 transfer, the DMA preparation overhead would not affect the MMC performance.
75 Pseudocode to handle is_first_req scenario with minimal prepare overhead::
/Linux-v5.4/Documentation/virt/
Dparavirt_ops.rst25 known that the overhead of indirect call isn't very important.
29 are called frequently and are performance critical. The overhead is
/Linux-v5.4/Documentation/usb/
Dohci.rst20 - improved robustness; bugfixes; and less overhead
30 to overhead in IRQ processing. When interrupt transfers are queued, those
/Linux-v5.4/mm/
DKconfig.debug35 By default this option will have a small overhead, e.g. by not
37 architectures. Even bigger overhead comes when the debugging is
85 poison on free. This reduces some of the overhead of the
112 kernel code. However the runtime performance overhead is virtually
/Linux-v5.4/drivers/net/fddi/skfp/h/
Dsmt.h801 struct smt_p_3210 overhead ; /* current SBA overhead */ member
818 struct smt_p_3210 overhead ; /* current SBA overhead */ member
834 struct smt_p_3210 overhead ; /* current SBA overhead */ member
856 struct smt_p_3210 overhead ; /* current SBA overhead */ member
/Linux-v5.4/Documentation/filesystems/ext4/
Dbigalloc.rst11 blocks to reduce both fragmentation and metadata overhead. The
23 also shrinking the amount of file system overhead for metadata.
/Linux-v5.4/fs/fscache/
DKconfig22 The gathering of statistics adds a certain amount of overhead to
39 The generation of this histogram adds a certain amount of overhead to
/Linux-v5.4/Documentation/netlabel/
Dlsm_interface.rst46 sidestep much of this overhead once a mapping has been established. Once the
53 bypassed as well which should result in a significant reduction in overhead.
/Linux-v5.4/net/mac80211/
Drc80211_minstrel_ht.h78 unsigned int overhead; member

12345678910