/openbmc/linux/kernel/ |
H A D | Kconfig.preempt | 1 # SPDX-License-Identifier: GPL-2.0-only 37 This option reduces the latency of the kernel by adding more 40 latency of rescheduling, providing faster application reactions, 44 low priority process to voluntarily preempt itself even if it 52 bool "Preemptible Kernel (Low-Latency Desktop)" 56 This option reduces the latency of the kernel by making 59 permitting a low priority process to be preempted involuntarily 67 embedded system with latency requirements in the milliseconds 71 bool "Fully Preemptible Kernel (Real-Time)" 75 This option turns the kernel into a real-time kernel by replacing [all …]
|
/openbmc/linux/Documentation/block/ |
H A D | bfq-iosched.rst | 5 BFQ is a proportional-share I/O scheduler, with some extra 6 low-latency capabilities. In addition to cgroups support (blkio or io 9 - BFQ guarantees a high system and application responsiveness, and a 10 low latency for time-sensitive applications, such as audio or video 12 - BFQ distributes bandwidth, and not just time, among processes or 16 In its default configuration, BFQ privileges latency over 17 throughput. So, when needed for achieving a lower latency, BFQ builds 19 goal, for a given device, is to achieve the maximum-possible 20 throughput at all times, then do switch off all low-latency heuristics 23 latency and throughput, or on how to maximize throughput. [all …]
|
H A D | deadline-iosched.rst | 10 ----------------------- 11 Refer to Documentation/block/switching-sched.rst for information on 12 selecting an io scheduler on a per-device basis. 14 ------------------------------------------------------------------------------ 17 ----------------------- 27 ----------------------- 33 ------------------------------------ 40 This parameter tunes the balance between per-request latency and aggregate 41 throughput. When low latency is the primary concern, smaller is better (where 42 a value of 1 yields first-come first-served behaviour). Increasing fifo_batch [all …]
|
/openbmc/linux/drivers/net/wireless/intel/iwlwifi/mvm/ |
H A D | quota.c | 1 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause 3 * Copyright (C) 2012-2014, 2018, 2021-2022 Intel Corporation 4 * Copyright (C) 2013-2014 Intel Mobile Communications GmbH 5 * Copyright (C) 2016-2017 Intel Deutschland GmbH 8 #include "fw-api.h" 33 if (vif == data->disabled_vif) in iwl_mvm_quota_iterator() 36 if (!mvmvif->deflink.phy_ctxt) in iwl_mvm_quota_iterator() 40 id = mvmvif->deflink.phy_ctxt->id; in iwl_mvm_quota_iterator() 48 switch (vif->type) { in iwl_mvm_quota_iterator() 50 if (vif->cfg.assoc) in iwl_mvm_quota_iterator() [all …]
|
/openbmc/linux/Documentation/devicetree/bindings/arm/msm/ |
H A D | qcom,idle-state.txt | 3 ARM provides idle-state node to define the cpuidle states, as defined in [1]. 4 cpuidle-qcom is the cpuidle driver for Qualcomm SoCs and uses these idle 5 states. Idle states have different enter/exit latency and residency values. 6 The idle states supported by the QCOM SoC are defined as - 26 Retention: Retention is a low power state where the core is clock gated and 31 state. Retention may have a slightly higher latency than Standby. 44 code in the EL for the SoC. On SoCs with write-back L1 cache, the cache has to 50 be flushed, system bus, clocks - lowered, and SoC main XO clock gated and 51 voltages reduced, provided all cpus enter this state. Since the span of low 52 power modes possible at this state is vast, the exit latency and the residency [all …]
|
/openbmc/linux/arch/alpha/lib/ |
H A D | ev6-stxncpy.S | 1 /* SPDX-License-Identifier: GPL-2.0 */ 3 * arch/alpha/lib/ev6-stxncpy.S 4 * 21264 version contributed by Rick Gorton <rick.gorton@api-networks.com> 6 * Copy no more than COUNT bytes of the null-terminated string from 29 * Furthermore, v0, a3-a5, t11, and $at are untouched. 34 * ftp.digital.com/pub/Digital/info/semiconductor/literature/dsc-library.html 36 * E - either cluster 37 * U - upper subcluster; U0 - subcluster U0; U1 - subcluster U1 38 * L - lower subcluster; L0 - subcluster L0; L1 - subcluster L1 66 lda t2, -1 # E : build a mask against false zero [all …]
|
H A D | ev6-stxcpy.S | 1 /* SPDX-License-Identifier: GPL-2.0 */ 3 * arch/alpha/lib/ev6-stxcpy.S 4 * 21264 version contributed by Rick Gorton <rick.gorton@alpha-processor.com> 6 * Copy a null-terminated string from SRC to DST. 21 * Furthermore, v0, a3-a5, t11, and t12 are untouched. 26 * ftp.digital.com/pub/Digital/info/semiconductor/literature/dsc-library.html 28 * E - either cluster 29 * U - upper subcluster; U0 - subcluster U0; U1 - subcluster U1 30 * L - lower subcluster; L0 - subcluster L0; L1 - subcluster L1 58 lda t2, -1 # E : build a mask against false zero [all …]
|
/openbmc/linux/block/ |
H A D | Kconfig | 1 # SPDX-License-Identifier: GPL-2.0 18 - block device files will become unusable 19 - some filesystems (such as ext3) will become unavailable. 86 support for ZAC/ZBC/ZNS host-managed and host-aware zoned block 101 See Documentation/admin-guide/cgroup-v1/blkio-controller.rst for more information. 104 bool "Block throttling .low limit interface support (EXPERIMENTAL)" 107 Add .low limit interface for block throttling. The low limit is a best 128 Enable writeback throttling by default for request-based block devices. 131 bool "Enable support for latency based cgroup IO protection" 134 Enabling this option enables the .latency interface for IO throttling. [all …]
|
H A D | blk-throttle.c | 1 // SPDX-License-Identifier: GPL-2.0 14 #include "blk-cgroup-rwstat.h" 15 #include "blk-stat.h" 16 #include "blk-throttle.h" 31 #define DFL_LATENCY_TARGET (-1L) 36 * For HD, very small latency comes from sequential IO. Such IO is helpless to 46 /* We measure latency for request size from <= 4k to >= 1M */ 55 unsigned long latency; /* ns / 1024 */ member 94 return pd_to_blkg(&tg->pd); in tg_to_blkg() 98 * sq_to_tg - return the throl_grp the specified service queue belongs to [all …]
|
H A D | Kconfig.iosched | 1 # SPDX-License-Identifier: GPL-2.0 14 The Kyber I/O scheduler is a low-overhead scheduler suitable for 16 synchronous writes, it will self-tune queue depths to achieve that 23 BFQ I/O scheduler for BLK-MQ. BFQ distributes the bandwidth of 26 also guarantees a low latency to interactive and soft 27 real-time applications. Details in 28 Documentation/block/bfq-iosched.rst 38 (cgroups-v1) or io (cgroups-v2) controller.
|
/openbmc/linux/Documentation/networking/device_drivers/ethernet/intel/ |
H A D | e1000e.rst | 1 .. SPDX-License-Identifier: GPL-2.0+ 8 Copyright(c) 2008-2018 Intel Corporation. 13 - Identifying Your Adapter 14 - Command Line Parameters 15 - Additional Configurations 16 - Support 48 --------------------- 49 :Valid Range: 0,1,3,4,100-100000 53 vector can generate per second. Increasing ITR lowers latency at the cost of 60 but will increase latency as packets are not processed as quickly. [all …]
|
H A D | e1000.rst | 1 .. SPDX-License-Identifier: GPL-2.0+ 8 Copyright(c) 1999 - 2013 Intel Corporation. 13 - Identifying Your Adapter 14 - Command Line Parameters 15 - Speed and Duplex Configuration 16 - Additional Configurations 17 - Support 50 ------- 54 :Valid Range: 0x01-0x0F, 0x20-0x2F 57 This parameter is a bit-mask that specifies the speed and duplex settings [all …]
|
/openbmc/linux/arch/mips/cavium-octeon/ |
H A D | Kconfig | 1 # SPDX-License-Identifier: GPL-2.0 11 non-CN63XXP1 hardware, so it is recommended to select "n" 41 Lock the low level TLB fast path into L2. 48 Lock the low level exception handler into L2. 55 Lock the low level interrupt handler into L2. 84 tristate "Module to measure interrupt latency using Octeon CIU Timer" 86 This driver is a module to measure interrupt latency using the 90 will be called octeon-ilm
|
/openbmc/linux/Documentation/devicetree/bindings/memory-controllers/ |
H A D | st,stm32-fmc2-ebi-props.yaml | 1 # SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) 3 --- 4 $id: http://devicetree.org/schemas/memory-controllers/st,stm32-fmc2-ebi-props.yaml# 5 $schema: http://devicetree.org/meta-schemas/core.yaml# 10 - Christophe Kerello <christophe.kerello@foss.st.com> 11 - Marek Vasut <marex@denx.de> 14 st,fmc2-ebi-cs-transaction-type: 33 st,fmc2-ebi-cs-cclk-enable: 40 st,fmc2-ebi-cs-mux-enable: 46 st,fmc2-ebi-cs-buswidth: [all …]
|
/openbmc/linux/Documentation/sound/cards/ |
H A D | emu10k1-jack.rst | 2 Low latency, multichannel audio with JACK and the emu10k1/emu10k2 5 This document is a guide to using the emu10k1 based devices with JACK for low 6 latency, multichannel recording functionality. All of my recent work to allow 12 - Lee Revell, 2005.03.30 15 Until recently, emu10k1 users on Linux did not have access to the same low 16 latency, multichannel features offered by the "kX ASIO" feature of their 25 fairly self explanatory - select Duplex, then for capture and playback select 30 /usr/local/bin/jackd -R -dalsa -r48000 -p64 -n2 -D -Chw:0,2 -Phw:0,3 -S 36 sb-live-mixer.rst (or audigy-mixer.rst).
|
/openbmc/linux/include/net/caif/ |
H A D | caif_device.h | 1 /* SPDX-License-Identifier: GPL-2.0-only */ 3 * Copyright (C) ST-Ericsson AB 2010 16 * struct caif_dev_common - data shared between CAIF drivers and stack. 19 * Link Layer to send flow-stop to CAIF Core. 23 * @link_select: Profile of device, either high-bandwidth or 24 * low-latency. This member is set by CAIF Link 26 * is a high bandwidth or low latency device.
|
/openbmc/linux/Documentation/networking/ |
H A D | tcp-thin.rst | 1 .. SPDX-License-Identifier: GPL-2.0 4 Thin-streams and TCP 7 A wide range of Internet-based services that use reliable transport 8 protocols display what we call thin-stream properties. This means 9 that the application sends data with such a low rate that the 11 effective. In time-dependent scenarios (like online games, control 13 on the data delivery latency, packet loss can be devastating for 19 After analysing a large number of time-dependent interactive 22 lifespan. The combination of time-dependency and the fact that the 25 In order to reduce application-layer latency when packets are lost, [all …]
|
/openbmc/linux/Documentation/admin-guide/pm/ |
H A D | intel_idle.rst | 1 .. SPDX-License-Identifier: GPL-2.0 24 Documentation/admin-guide/pm/cpuidle.rst if you have not done that yet.] 28 processor's functional blocks into low-power states. That instruction takes two 38 only way to pass early-configuration-time parameters to it is via the kernel 42 .. _intel-idle-enumeration-of-states: 50 as C-states (in the ACPI terminology) or idle states. The list of meaningful 51 ``MWAIT`` hint values and idle states (i.e. low-power configurations of the 56 subsystem (see :ref:`idle-states-representation` in 57 Documentation/admin-guide/pm/cpuidle.rst), 66 `below <intel-idle-parameters_>`_.] [all …]
|
/openbmc/linux/Documentation/usb/ |
H A D | ehci.rst | 5 27-Dec-2002 8 USB 2.0-capable host controller hardware. The USB 2.0 standard is 11 - "High Speed" 480 Mbit/sec (60 MByte/sec) 12 - "Full Speed" 12 Mbit/sec (1.5 MByte/sec) 13 - "Low Speed" 1.5 Mbit/sec 15 USB 1.1 only addressed full speed and low speed. High speed devices 23 (TT) in the hub, which turns low or full speed transactions into 31 While usb-storage devices have been available since mid-2001 (working 34 appear to be on hold until more systems come with USB 2.0 built-in. 39 other changes to the Linux-USB core APIs, including the hub driver, [all …]
|
/openbmc/phosphor-dbus-interfaces/yaml/xyz/openbmc_project/Inventory/Item/ |
H A D | Dimm.interface.yaml | 4 - name: MemoryDataWidth 8 - name: MemorySizeInKB 12 - name: MemoryDeviceLocator 16 - name: MemoryType 20 - name: MemoryTypeDetail 24 - name: MaxMemorySpeedInMhz 28 - name: MemoryAttributes 33 - name: MemoryConfiguredSpeedInMhz 37 - name: ECC 40 Error-Correcting Code. [all …]
|
/openbmc/u-boot/drivers/ddr/fsl/ |
H A D | ddr2_dimm_params.c | 1 // SPDX-License-Identifier: GPL-2.0 18 * --- ----- ------ 26 * 0 low 1GB 1GB 46 * Convert a two-nibble BCD value into a cycle time. 47 * While the spec calls for nano-seconds, picos are returned. 143 * CAS latency given the DRAM clock period. The SPD only 145 * frequency the DIMM runs at, the shorter its CAS latency can. 146 * be. If the DIMM is operating at a sufficiently low frequency, 147 * it may be able to run at a CAS latency shorter than the 148 * shortest SPD-defined CAS latency. [all …]
|
H A D | ddr1_dimm_params.c | 1 // SPDX-License-Identifier: GPL-2.0 19 * --- ----- ------ 27 * 0 low 1GB 1GB 47 * Convert a two-nibble BCD value into a cycle time. 48 * While the spec calls for nano-seconds, picos are returned. 162 * CAS latency given the DRAM clock period. The SPD only 164 * frequency the DIMM runs at, the shorter its CAS latency can be. 165 * If the DIMM is operating at a sufficiently low frequency, 166 * it may be able to run at a CAS latency shorter than the 167 * shortest SPD-defined CAS latency. [all …]
|
/openbmc/linux/include/uapi/linux/caif/ |
H A D | caif_socket.h | 1 /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ 4 * Copyright (C) ST-Ericsson AB 2010 16 * enum caif_link_selector - Physical Link Selection. 17 * @CAIF_LINK_HIGH_BANDW: Physical interface for high-bandwidth 19 * @CAIF_LINK_LOW_LATENCY: Physical interface for low-latency 32 * enum caif_channel_priority - CAIF channel priorities. 35 * @CAIF_PRIO_LOW: Low-priority channel. 55 * enum caif_protocol_type - CAIF Channel type. 78 * enum caif_at_type - AT Service Endpoint 85 * enum caif_debug_type - Content selection for debug connection [all …]
|
/openbmc/linux/drivers/net/ethernet/amazon/ena/ |
H A D | ena_admin_defs.h | 1 /* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ 3 * Copyright 2015-2020 Amazon.com, Inc. or its affiliates. All rights reserved. 59 /* descriptors and headers are in device memory (a.k.a Low Latency 119 * 1 : ctrl_data - control buffer address valid 120 * 2 : ctrl_data_indirect - control buffer address 142 * 7:5 : sq_direction - 0x1 - Tx; 0x2 - Rx 193 * 7:5 : sq_direction - 0x1 - Tx, 0x2 - Rx 199 /* 3:0 : placement_policy - Describing where the SQ 201 * 0x1 - descriptors and headers are in OS memory, 202 * 0x3 - descriptors and headers in device memory [all …]
|
/openbmc/linux/kernel/trace/ |
H A D | trace_hwlat.c | 1 // SPDX-License-Identifier: GPL-2.0 3 * trace_hwlat.c - A simple Hardware Latency detector. 20 * Although certain hardware-inducing latencies are necessary (for example, 22 * and remote management) they can wreak havoc upon any OS-level performance 23 * guarantees toward low-latency, especially when the OS is not even made 27 * sampling the built-in CPU timer, looking for discontiguous readings. 31 * environment requiring any kind of low-latency performance 34 * Copyright (C) 2008-2009 Jon Masters, Red Hat, Inc. <jcm@redhat.com> 35 * Copyright (C) 2013-2016 Steven Rostedt, Red Hat, Inc. <srostedt@redhat.com> 67 static char *thread_mode_str[] = { "none", "round-robin", "per-cpu" }; [all …]
|