一、概念
随机数都是由随机数生成器(Random Number Generator)生成的。
1.真随机数 TRUE Random Number
真正的随机数是使用物理现象产生的:比如掷钱币、骰子、转轮、使用电子元件的噪音、核裂变等等,这样的随机数发生器叫做物…
因为在开发过程中遇到了OOM 所以想要信息理解一下这个OOM 到底是真么回事,然后就找到了这篇文章,为了自己学习,所以翻译一下,以备后面查看: Out Of Memory Management内存不足管理检查可用内存确定OOM状态选择过程杀死…
关注了就能看到更多这么棒的文章哦~Some 5.15 development statisticsBy Jonathan CorbetNovember 1, 2021https://lwn.net/Articles/874283/5.15 内核在 10 月 31 日正式发布了,code name 也就顺利成章地起名为 "Trick or Treat" 了。当时&am…
关注了就能看到更多这么棒的文章哦~ The imminent stable-version apocalypse By Jonathan Corbet February 5, 2021 DeepL assisted translation https://lwn.net/Articles/845120/ 人们经常说,stable-kernel release 应该要是稳定的(stable…
Stomped CRC Counters
Stomped CRC counters help in finding the location of bit errors in a network that uses cut-through switches. More precisely, these counters help in finding where bit errors do not exist. Stomped CRC 计数器有助于在使用直通式交换机的网络…
Switch Buffer Management
Recall that during network congestion, a TCP sender relies on the following events for reducing its transmission rate. 回想一下,在网络拥塞期间,TCP 发送端依靠以下事件来降低传输速率。
1. When the sender detects packet loss becaus…
Comparison with Lossless Networks
As Chapter 2 explains, Fibre Channel defines a maximum frame size of 2148 bytes. Users do not have to change the MTU value, and hence, it is never a consideration. 如第 2 章所述,光纤通道定义的最大帧大小为 2148 字节。用户…
TCP Flow Monitoring versus I/O Flow Monitoring
TCP flow monitoring shouldn’t be confused with I/O flow monitoring because of the following reasons: TCP 流量监控不应与 I/O 流量监控混淆,原因如下:
1. TCP belongs to the transport layer (layer 4) of the OS…
Pause Threshold for Long Distance Links长途链路的暂停阈值
This section uses the following basic concepts: 本节使用以下基本概念: Bit Time (BT): It is the time taken to transmit one bit. It is the reciprocal of the bit rate. For example, BT of a 10 GbE po…
Congestion Due to Over-utilization of the Target Link
The same problem may happen on a target/storage link when multiple hosts initiate large-size write I/O operations to the same target. 当多台主机对同一目标启动大容量写入 I/O 操作时,目标/存储链路上可能会…
代码: git clone https://github.com/chasinglulu/linux.git -b X4412-Kernel 设备树文件
linux-4.16.0/arch/arm/boot/dts/exynos4412-x4412.dts
/** X4412 board device tree source** Device tree source file for 9tripods X4412 board which is based on* Sa…
关注了就能看到更多这么棒的文章哦~ Lockless patterns: relaxed access and partial memory barriers February 26, 2021 This article was contributed by Paolo Bonzini DeepL assisted translation https://lwn.net/Articles/846700/ 本系列中的第一篇文章LWN&a…
Location of Ingress No-Drop Queues入口无损队列的位置
Ingress queues for no-drop traffic are maintained by all the ports in a lossless Ethernet network. For the sake of simplicity, Figure 7-1 shows ingress no-drop queue(s) only at one location, but in real…
5.0 Priority Flow Control - Planning and Guidelines优先流量控制 - 规划与指导 This section covers planning, considerations, and general configuration guidelines for enabling PFC on a network. 本节介绍在网络上启用 PFC 的规划、注意事项和一般配置指南。
5.1 S…
一开始看这个函数,还以为要做,网卡设备的down操作。 后来一看代码,原来这个函数是在net/sched/sch_generic.c 文件里,而且不是要做网络接口的down操作。操作的结果是:deactivate transmissions on several devices&…
How iSCSI and NVMe/TCP Exchange Data
The following are the high-level steps for the end devices to exchange data using iSCSI and NVMe/TCP protocols over a TCP/IP network (Figure 8-1). 以下是终端设备通过 TCP/IP 网络使用 iSCSI 和 NVMe/TCP 协议交换数据的高级…
目录:1.寄存器:2.寄存器类型:(1)GPFSEL:GPIO Function Select Registers(功能选择寄存器)(2)GPSET:GPIO Pin Output Set Registers(输…
6.0 Priority Flow Control – Verification带优先级的流量控制 - 验证 6.1 Priority Counters优先级计数器
Priority flow control counters for each interface are available in ethtool. They measure the number of Xon and Xoff (transmit on and off) frames sent and…
NVMe/TCP
NVMe/TCP carries NVMe commands over TCP transport (Figure 1-8) for accessing remote block storage via a lossy or sometimes lossless network. It is an OSI layer 5 (sessions layer) protocol. NVMe/TCP 通过 TCP 传输(图 1-8)传输 NVMe 命令,以便通过有…
1. 内核(kernel)文件系统支持中文配置
内核源码路径下输入:
make menuconfig选择进入File system(文件系统配置选项) DOS/FAT/NT Filesystems —> [] Network File Systems —> -- Native language support —>
deadline
源码基于5.10
0. 私有数据
struct deadline_data {/** run time data*//** requests (deadline_rq s) are present on both sort_list and fifo_list*/struct rb_root sort_list[2];struct list_head fifo_list[2];/** next in sort order. read, write or both ar…
6.0 Configure and Test NVMe over Fabrics Host(s) to Connect to SPDK Target配置和测试 NVMe over Fabrics 主机以连接 SPDK 目标机 The SPDK NVMe-oF target system is spec compliant, which allows for the use of either an SPDK host or Linux Kernel host to co…
[postgresdb223 home]$ pg_ctl start waiting for server to start....2023-08-17 18:51:47.852 CST [1281811] FATAL: could not create shared memory segment: 设备上没有空间 2023-08-17 18:51:47.852 CST [1281811] DETAIL: Failed system call was shmget(key161594131…
Queue Depth Monitoring and Microburst Detection
Queue depth monitoring and microburst detection capture the events that may cause congestion at a lower granularity but are unnoticed by other means due to long polling intervals. 队列深度监控和微爆检测可捕捉…
在kernel代码的任意位置直接调用dump_stack方法即可。dump_stack已经包含在内核符号表中,并在include/linux/kernel.h中被声明。获取栈信息的原理是使用ebp逐层回溯。dump_stack函数在arch/x86/kernel/traps_32.c中定义(kernel 2.6.24)如下:
/** The ar…
4.0 Configure and Test RDMA (All Servers) 配置并测试 RDMA(所有服务器)
4.1 Install Intel LAN Driver, RDMA Driver, and Related Dependencies安装英特尔 LAN 驱动程序、RDMA 驱动程序和相关依赖软件
Download the latest Linux driver package from Intel f…
一、参考
Message logging with printk — The Linux Kernel documentation
如何获得正确的printk格式占位符 — The Linux Kernel documentation
使用printk记录消息 — The Linux Kernel documentation
printk 内核打印 – 人人都懂物联网 (getiot.tech)
内核printk原理…
Location of Ingress No-Drop Queues入口无损队列的位置
Ingress queues for no-drop traffic are maintained by all the ports in a lossless Ethernet network. For the sake of simplicity, Figure 7-1 shows ingress no-drop queue(s) only at one location, but in real…
This chapter covers the following topics. 本章涵盖以下主题 Ethernet flow control. 以太网流量控制 Priority-based Flow Control at layer 2 and layer 3 of the OSI model. OSI 模型第 2 层和第 3 层基于优先级的流量控制。 Congestion spreading in lossless Ethernet n…
2.6.28版本起,内核代码的drivers下增加了一个staging目录,这个目录也是用来存放驱动程序,只是这里的驱动和上层目录的不同,加载的时候内核log会打印如下的语句: MODULE_NAME: module is from the staging directory, t…
1.0 内容简介 This will introduce Ethernet flow control on Intel Ethernet 800 Series Network Adapters with RDMA driver - iRDMA, with a focus on best practices for Linux RDMA traffic. 我们将介绍英特尔以太网 800 系列网络适配器上的以太网流量控制与 RDMA 驱动程序…
用树莓派自身进行编译,首先需要安装树莓派下缺少的库: sudo apt-get install git bc 安装git是为了可以用git获取源码,源码获取方式和上面一样。
配置
cd linux KERNELkernel7 make bcm2709_defconfig
编译与更新
make -j4 zImage mo…