Searched refs:latencies (Results 1 – 25 of 49) sorted by relevance
12
5 The only two tunables for the Kyber scheduler are the target latencies for7 these target latencies.
20 service time for a request. As we focus mainly on read latencies, this is
51 setting is the software limit. Some devices exhibit large latencies when53 smaller discards and potentially help reduce latencies induced by large
10 the service quality. Extreme latencies are caused by TCP's19 streams provoke high latencies when using TCP is unfortunate.33 to experience high retransmission latencies.
151 zfcp_sdev->latencies.write.channel.min = 0xFFFFFFFF; in zfcp_scsi_slave_alloc()152 zfcp_sdev->latencies.write.fabric.min = 0xFFFFFFFF; in zfcp_scsi_slave_alloc()153 zfcp_sdev->latencies.read.channel.min = 0xFFFFFFFF; in zfcp_scsi_slave_alloc()154 zfcp_sdev->latencies.read.fabric.min = 0xFFFFFFFF; in zfcp_scsi_slave_alloc()155 zfcp_sdev->latencies.cmd.channel.min = 0xFFFFFFFF; in zfcp_scsi_slave_alloc()156 zfcp_sdev->latencies.cmd.fabric.min = 0xFFFFFFFF; in zfcp_scsi_slave_alloc()157 spin_lock_init(&zfcp_sdev->latencies.lock); in zfcp_scsi_slave_alloc()
279 struct zfcp_latencies latencies; member
424 struct zfcp_latencies *lat = &zfcp_sdev->latencies; \455 struct zfcp_latencies *lat = &zfcp_sdev->latencies; \
2078 lat = &zfcp_sdev->latencies.read; in zfcp_fsf_req_trace()2083 lat = &zfcp_sdev->latencies.write; in zfcp_fsf_req_trace()2086 lat = &zfcp_sdev->latencies.cmd; in zfcp_fsf_req_trace()2091 spin_lock(&zfcp_sdev->latencies.lock); in zfcp_fsf_req_trace()2095 spin_unlock(&zfcp_sdev->latencies.lock); in zfcp_fsf_req_trace()
9 detect large system latencies induced by the behavior of certain underlying22 you are trying to keep event service latencies down in the microsecond range.57 change to a default of 10 usecs. If any latencies that exceed the threshold is
198 - Calculating latencies between two events
11 throughput. It will still provide good latencies most of the18 latencies.
3350 const u16 *latencies; in pri_wm_latency_show() local3353 latencies = dev_priv->wm.skl_latency; in pri_wm_latency_show()3355 latencies = dev_priv->wm.pri_latency; in pri_wm_latency_show()3357 wm_latency_show(m, latencies); in pri_wm_latency_show()3365 const u16 *latencies; in spr_wm_latency_show() local3368 latencies = dev_priv->wm.skl_latency; in spr_wm_latency_show()3370 latencies = dev_priv->wm.spr_latency; in spr_wm_latency_show()3372 wm_latency_show(m, latencies); in spr_wm_latency_show()3380 const u16 *latencies; in cur_wm_latency_show() local3383 latencies = dev_priv->wm.skl_latency; in cur_wm_latency_show()[all …]
6 perf-sched - Tool to trace/measure scheduler properties (latencies)20 'perf sched latency' to report the per task scheduling latencies
17 multiqueue and other fast devices. Given target latencies for reads and
132 The IO controller will attempt to maintain average IO latencies below
60 - ibm,cpu-idle-state-latencies-ns:62 exit-latencies (in ns) for the idle states in
8 To improve performance and minimize average latencies, many contemporary CPUs
25 determinism and well-bound latencies. Even in the worst-case, PI will
91 Say N here if you need ultimate kernel/user switch latencies
19 histogram of latencies. All these counters may be accessed by sending74 collect histogram of latencies. The
20 channels. With a post 2.6.9 Linux kernel, latencies down to 64 (1.33 ms) or
75 you should take into account the latencies and jitter introduced by240 latencies. But if it is too small slave won't be able to capture clear edge
92 in order to avoid degrading from-idle transition latencies.240 eliminate scheduling-clock interrupt latencies. Here are some254 to degrade your latencies -- and that this degradation can
116 the address space, each with varying latencies. This enables
230 slowdowns and noticeable latencies. Normally Speedstep should be used
