[B! kernel] krrrrã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

krrrr id:krrrr

kernelã«é–¢ã™ã‚‹krrrrã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (9)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

ã‚½ã‚±ãƒƒãƒˆAPIãŒé…ã™ãŽã‚‹ï¼Ÿæ–°ãŸãªio_uringã‚’è©¦ã™ï¼
æ–°ã—ã„APIãŒä½œã‚‰ã‚Œã‚‹ãŸã³ã«ã€ç§ãŸã¡ã¯ã€å¤ã„APIã‚’ç½®ãæ›ãˆã‚‹ã ã‘ã§é«˜é€ŸåŒ–ã¨ã„ã†å¤¢ã‚’ã¿ã¾ã™ã€‚ä½•åº¦å¤¢ç ´ã‚Œã¦ã‚‚ã€é«˜é€ŸãªAPIãŒè¿½åŠ ã•ã‚ŒãŸã¨èžã‘ã°ã€è©¦ã•ãšã«ã¯ã„ã‚‰ã‚Œã¾ã›ã‚“ã‚ˆãï¼ ä»Šå›žã¯ã€Linuxã‚«ãƒ¼ãƒãƒ«5.1ã§è¿½åŠ ã•ã‚ŒãŸio_uringã‚’ä½¿ã£ã¦ã€Rustã®asyncãƒ©ãƒ³ã‚¿ã‚¤ãƒ ã‚’å®Ÿè£…ã—ã€gRPCã‚µãƒ¼ãƒã®ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ã‚’å®Ÿè¡Œã—ã¦ã¿ã¾ã—ãŸã€‚ io_uringã¨ã¯io_uringã¯ã€ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã¨ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®éžåŒæœŸI/Oã®ãŸã‚ã«é–‹ç™ºã•ã‚Œã¾ã—ãŸã€‚åŒæœŸã‚ˆã‚Šã‚‚éžåŒæœŸã®ã»ã†ãŒãŠã—ã‚ƒã‚Œã€ãã†ã„ã†é›°å›²æ°—ã‚ã‚Šã¾ã™ã‚ˆãï¼ã‚¯ãƒ©ã‚¦ãƒ‰ãƒã‚¤ãƒ†ã‚£ãƒ–ã‚‚ã€éžåŒæœŸã«APIã‚’ä»‹ã—ã¦ã€ãªã‚“ã‹ã‚„ã£ã¦ã‚‹ã‚„ã¤ã§ã™ã‚ˆãã€‚ io_uringã®ã‚¤ãƒ³ã‚¿ãƒ¼ãƒ•ã‚§ã‚¤ã‚¹ã¯ã€é«˜ã„æ€§èƒ½ã‚’ç›®æŒ‡ã—ã€1ï¼‰ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã¨ã‚«ãƒ¼ãƒãƒ«é–“ã§ã®ãƒ¡ãƒ¢ãƒªã‚³ãƒ”ãƒ¼ã‚’é¿ã‘ã‚‹ã€2ï¼‰è¤‡æ•°ã®I/Oè¦æ±‚ã‚’ä¸€åº¦ã«ã‚«ãƒ¼ãƒãƒ«ã«ä¼ãˆã‚‹ã“ã¨ãŒã§ãã‚‹ã€ã¨ã„ã†å·¥å¤«ãŒã•ã‚Œã¦ã„ã¾ã™ã€‚ ä¸‹å›³ã®ã‚ˆã†ã«ã€ã‚¢ãƒ—ãƒªã‚±
krrrr 2020/12/01
kernel

async

linux
ãƒªãƒ³ã‚¯
https://kernel.dk/io_uring.pdf
krrrr 2020/11/17
linux

kernel

async

io_uring
ãƒªãƒ³ã‚¯
io_uring By Example: An Article Series
io_uring is a clever new, high-performance interface for asynchronous I/O for Linux without the drawbacks of the aio set of APIs. In this 3-part article series, we look at how to use io_uring to get the most common programming tasks done under Linux. We write a series of programs of increasing complexity to slowly but steadily introduce the features of io_uring. While we will use liburing for the
krrrr 2020/11/17
kernel

linux

io_uring
ãƒªãƒ³ã‚¯
Escaping a chroot jail/1 | PyTux
Everybody will tell you that a chroot jail (that is, making a process think that a directory is instead the root folder, and not letting it access or modify anything outside of that) is ineffective against a process with root privileges1 (UID 0). Letâ€™s see why. The escape basically works like this: We create a temporary folder (I named mine .42, hidden not to draw too much attention) and we chroot
krrrr 2019/11/27
chroot

jail

kernel
ãƒªãƒ³ã‚¯
ãã£ã¨æ˜Žæ—¥ã‹ã‚‰å½¹ç«‹ã¤eBPFã®ã—ãã¿
ç¬¬15å›žã‚«ãƒ¼ãƒãƒ«/VMæŽ¢æ¤œéšŠã§ã®ç™ºè¡¨. eBPFã®ãƒãƒ¼ãƒ€ã®ä»•çµ„ã¿ã‚„Verifierã®ä»•çµ„ã¿ãªã©ã«ã¤ã„ã¦1ã¤ã®eBPFãƒ—ãƒã‚°ãƒ©ãƒ ã‚’é¡Œæã«æŽ˜ã‚Šä¸‹ã’ã¾ã™.
krrrr 2019/08/23
eBPF

kernel

linux

network
ãƒªãƒ³ã‚¯
Namespaces in operation, part 1: namespaces overview [LWN.net]
Namespaces in operation, part 1: namespaces overview LWN.net needs you!Without subscribers, LWN would simply not exist. Please consider signing up for a subscription and helping to keep LWN publishing The Linux 3.8 merge window saw the acceptance of Eric Biederman's sizeable series of user namespace and related patches. Although there rem ain some details to finishâ€”for example, a number of Linux fi
krrrr 2018/11/12
linux

system

kernel

namespaces
ãƒªãƒ³ã‚¯
Why is the kernel community replacing iptables with BPF?
Apr 17, 2018Why is the kernel community replacing iptables with BPF? Author Note: this is a post by long-time Linux kernel networking developer and creator of the Cilium project, Thomas Graf The Linux kernel community recently announced bpfilter, which will replace the long-standing in-kernel implementation of iptables with high-performance network filtering powered by Linux BPF, all while guarant
krrrr 2018/04/18
kernel

linux

network

BPF

iptables
ãƒªãƒ³ã‚¯
Writing an OS in Rust (Second Edition)
This blog series creates a small operating system in the Rust programming language. Each post is a small tutorial and includes all needed code, so you can follow along if you like. The source code is also available in the corresponding Github repository. Latest post: Async/Await The first step in creating our own operating system kernel is to create a Rust executable that does not link the standar
krrrr 2016/01/04
kernel
ãƒªãƒ³ã‚¯
Linuxã§ãƒãƒ¼ãƒ‰ãƒãƒ©ãƒ³ã‚µã‚„ã‚ãƒ£ãƒƒã‚·ãƒ¥ã‚µãƒ¼ãƒã‚’ãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã•ã›ã‚‹ãŸã‚ã®ã‚«ãƒ¼ãƒãƒ«ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚° - ã‚†ã†ã†ããƒ–ãƒã‚°
æœ¬è¨˜äº‹ã®å…¬é–‹å¾Œã®2016å¹´7æœˆã«ã¯ã¦ãªã«ãŠã‘ã‚‹ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°äº‹ä¾‹ã‚’ç´¹ä»‹ã—ãŸã€‚ ã¯ã¦ãªã«ãŠã‘ã‚‹Linuxãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¹ã‚¿ãƒƒã‚¯ãƒ‘ãƒ•ã‚©ãƒ¼ãƒžãƒ³ã‚¹æ”¹å–„ / Linux network performance improvement at hatena - Speaker Deck HAProxy ã‚„ nginx ãªã©ã®ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ãƒãƒ¼ãƒ‰ãƒãƒ©ãƒ³ã‚µã‚„ãƒªãƒãƒ¼ã‚¹ãƒ—ãƒã‚ã‚·ã€memcached ãªã©ã® KVS ã®ã‚ˆã†ãªé«˜ãƒ‘ã‚±ãƒƒãƒˆãƒ¬ãƒ¼ãƒˆã«ãªã‚Šã‚„ã™ã„ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã«ãŠã„ã¦ã€å˜ä¸€ã® CPU ã‚³ã‚¢ã«è² è·ãŒåã‚Šã€ãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã—ãªã„ã“ã¨ãŒã‚ã‚Šã¾ã™ã€‚ ä»Šå›žã¯ã€ã“ã®ã‚ˆã†ãªãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã«ãŠã„ã¦ CPU è² è·ãŒãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã—ãªã„ç†ç”±ã¨ã€ãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã•ã›ã‚‹ãŸã‚ã® Linux ã‚«ãƒ¼ãƒãƒ«ã®ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¹ã‚¿ãƒƒã‚¯ã®ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°æ‰‹æ³•ã¨ã—ã¦ RFS (Receive Flow Steering) ã‚’
krrrr 2015/03/31
linux

kernel
ãƒªãƒ³ã‚¯
1