[B! kernel] hibomaã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

hiboma id:hiboma

kernelã«é–¢ã™ã‚‹hibomaã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (400)

${{author_name}}$

{{{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}}

ãƒˆãƒ©ãƒ–ãƒ«ã‚·ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚°ã‹ã‚‰Linuxã‚«ãƒ¼ãƒãƒ«ã«æ½œã‚Šè¾¼ã‚€ â€ å®Ÿéš›ã®ã‚µãƒ¼ãƒ“ã‚¹é‹ç”¨ã§æ·±ã‚ãŸã‚«ãƒ¼ãƒãƒ«ã¸ã®ç†è§£
æœ¬è¨˜äº‹ã¯ã€TechFeed Experts Night#19 ã€œ é”äººã«èžãã€Linuxã‚«ãƒ¼ãƒãƒ«ã‚³ãƒ¼ãƒ‰ã®æ©ãæ–¹ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³æ›¸ãèµ·ã“ã—è¨˜äº‹ã«ãªã‚Šã¾ã™ã€‚ ã‚¤ãƒ™ãƒ³ãƒˆãƒšãƒ¼ã‚¸ã®ã‚¿ã‚¤ãƒ ãƒ†ãƒ¼ãƒ–ãƒ«ã‹ã‚‰ã€ãã®ä»–ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã«é–¢ã™ã‚‹è¨˜äº‹ã‚‚ãŠèªã¿é ‚ã‘ã¾ã™ã®ã§ã€ä¸€åº¦ã‚¢ã‚¯ã‚»ã‚¹ã—ã¦ã¿ã¦ãã ã•ã„ã€‚ æœ¬ã‚»ãƒƒã‚·ãƒ§ãƒ³ã®ç™»å£‡è€… ã‚»ãƒƒã‚·ãƒ§ãƒ³å‹•ç”» GMOãƒšãƒ‘ãƒœã®ä¼Šè—¤æ´‹ä¹Ÿã§ã™ã€‚2007å¹´ã‹ã‚‰GMOãƒšãƒ‘ãƒœã«ã¦å‹¤å‹™ã—ã¦ãŠã‚Šã€ãƒ›ã‚¹ãƒ†ã‚£ãƒ³ã‚°ã‚µãƒ¼ãƒ“ã‚¹ã‚„æŠ€è¡“åŸºç›¤ãƒãƒ¼ãƒ ã‚’çµŒã¦ç¾åœ¨ã¯ã‚»ã‚ãƒ¥ãƒªãƒ†ã‚£å¯¾ç–å®¤ã«æ‰€å±žã—ã¦ãŠã‚Šã¾ã™ã€‚ã‚ªãƒ³ãƒ©ã‚¤ãƒ³ã§ã¯TwitterãŠã‚ˆã³Githubã®ã‚¢ã‚«ã‚¦ãƒ³ãƒˆã¨ãƒ–ãƒã‚°ã§æ´»å‹•ã—ã¦ã„ã¾ã™ã€‚ä»Šå›žã¯å®Ÿéš›ã®ã‚µãƒ¼ãƒ“ã‚¹ç’°å¢ƒã§ã®Linuxã‚«ãƒ¼ãƒãƒ«ã®ãƒˆãƒ©ãƒ–ãƒ«ã‚·ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚°äº‹ä¾‹ã¨ã¨ã‚‚ã«ã€ã‚«ãƒ¼ãƒãƒ«ã®è©³ç´°èª¿æŸ»ã‚„ã‚³ãƒ¼ãƒ‰ãƒªãƒ¼ãƒ‡ã‚£ãƒ³ã‚°ã®æ–¹æ³•ã‚’ã”ç´¹ä»‹ã—ã¾ã™ã€‚ ç§ã¯Linuxã‚«ãƒ¼ãƒãƒ«ã®ã‚³ãƒŸãƒƒã‚¿ãƒ¼ã‚’ã—ã¦ã„ã‚‹ã‚ã‘ã§ã¯ãªãã€ã¾ãŸLinuxã‚«ãƒ¼ãƒãƒ«ã®é–‹ç™ºæ¥å‹™ã‚‚ã•ã»ã©è¡Œã£ã¦
hiboma 2023/07/26
æ›¸ãèµ·ã“ã—å…¬é–‹ã•ã‚Œã¾ã—ãŸã€œã€‚ãœã²èªã‚“ã§ãã ã•ã„ã€‚é‹å–¶ã®çš†æ§˜ã‚‚æ„Ÿè¬ã§ã™!

Linux

kernel
ãƒªãƒ³ã‚¯
ã‚³ãƒ³ãƒ†ãƒŠã‚„ã‚³ãƒžãƒ³ãƒ‰ãªã©ã®ãƒªã‚½ãƒ¼ã‚¹å˜ä½ã§å¤–éƒ¨ã¸ã®æ„å›³ã—ãªã„é€šä¿¡ã‚’ãƒ–ãƒãƒƒã‚¯ã™ã‚‹
ç‰¹å®šã®å¤–éƒ¨ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã¸ã®é€šä¿¡ã®åˆ¶é™ã«ã¯ãƒ•ã‚¡ã‚¤ã‚¢ã‚¦ã‚©ãƒ¼ãƒ«ãªã©ã‚’åˆ©ç”¨ã™ã‚‹ã“ã¨ãŒå¤šã„ã§ã™ãŒã€ã‚³ãƒ³ãƒ†ãƒŠã‚„å®Ÿè¡Œã•ã‚ŒãŸã‚³ãƒžãƒ³ãƒ‰åãªã©ã‚’ã‚‚ã¨ã«ã€é€šä¿¡ã‚’åˆ¶å¾¡ã—ãŸã„ã¨ã„ã†éœ€è¦ãŒè‡ªåˆ†ã®ä¸ã§ã‚ã‚Šã¾ã—ãŸã€‚ å…·ä½“çš„ã«ã¯ GitHub Self-hosted runner ã®ã‚ˆã†ãª CI / CD ç’°å¢ƒã§ã€ä¾å˜ãƒ‘ãƒƒã‚±ãƒ¼ã‚¸ã«æ‚ªæ„ã‚ã‚‹ã‚³ãƒ¼ãƒ‰ãŒå…¥ã‚Šè¾¼ã‚“ã§ã—ã¾ã†ã‚ˆã†ãªã‚µãƒ—ãƒ©ã‚¤ãƒã‚§ãƒ¼ãƒ³æ”»æ’ƒãªã©ã‚’æ¤œçŸ¥ãƒ»é˜²å¾¡ã—ã€æ„å›³ã›ãšã«ã‚¯ãƒ¬ãƒ‡ãƒ³ã‚·ãƒ£ãƒ«ãªã©ã®ç§˜åŒ¿ã™ã¹ãæƒ…å ±ãŒå¤–éƒ¨ã«æ¼æ´©ã™ã‚‹ã®ã‚’é˜²ãŽãŸã„ã¨æ€ã£ã¦ã„ã¾ã—ãŸã€‚ ã“ã®ã‚ˆã†ãªã‚µãƒ—ãƒ©ã‚¤ãƒã‚§ãƒ¼ãƒ³æ”»æ’ƒã¸ã®å¯¾ç–ã¯æ§˜ã€…ã§ã™ãŒã€å®Ÿè¡Œæ™‚ã«æ‚ªæ„ã®ã‚ã‚‹å‹•ä½œã‚’æ¤œå‡ºã™ã‚‹ã‚‚ã®ã¨ã—ã¦ã€GitLab ãŒ Falco ã‚’ãƒ™ãƒ¼ã‚¹ã¨ã—ãŸ Package Hunter ãªã©ãŒã‚ã‚Šã¾ã™ã€‚ã“ã®ãƒ„ãƒ¼ãƒ«ã¯ä¾å˜ãƒ‘ãƒƒã‚±ãƒ¼ã‚¸ãªã©ã‚’ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«ã™ã‚‹éš›ã«å®Ÿè¡Œã•ã‚Œã‚‹ã‚·ã‚¹ãƒ†ãƒ ã‚³ãƒ¼ãƒ«ãªã©ã‚’ç›£è¦–ã™ã‚‹ã‚‚ã®ã§ã™ã€‚ æ¤œçŸ¥ã™ã‚‹ã ã‘ã§ã‚ã‚Œã° Package Hu
hiboma 2021/09/27
kernel
ãƒªãƒ³ã‚¯
https://tech.pepabo.com/2020/06/26/kernel-dive-tcp_mem/
hiboma 2020/06/26
TCP oom ã‚„ memory pressure, net.ipv4.tcp_mem ã‚’ã“ã“ã¾ã§è©³ç´°ã«æ›¸ã„ãŸæ–‡ç« ã¯ã€ä»–ã«ç„¡ã„ã‹ã¨æ€ã£ã¦ã¾ã™ã€‚æ°—ã«ãªã‚‹ã‹ãŸã”è¦§ãã ã•ã„ã€‚

kernel
ãƒªãƒ³ã‚¯
TCP æŽ¥æ”¶çª—å£
æœ€è¿‘æœ‰ç©ºé˜…è¯»äº†ä¸€äº› TCP çª—å£ç›¸å…³çš„å†…æ ¸ä»£ç ï¼ŒçŸ¥è¯†ç‚¹ç›¸å¯¹æ¯”è¾ƒçç¢Žï¼Œåšä¸€äº›ç®€å•çš„ç¬”è®°ï¼Œæœ¬ç¯‡ä¸»è¦å…³äºŽ TCP æŽ¥æ”¶çª—å£ã€‚ æœ¬ç¯‡ä¸»è¦åŸºäºŽ Linux v5.2.2 å†…æ ¸ï¼Œå¦‚æœ‰ä¸å‡†ç¡®ä¹‹å¤„ï¼Œä¼šæ—¥åŽä¿®è®¢ã€‚ å‰ç½®çŸ¥è¯†åœ¨ TCP ä¼ è¾“ä¸æåˆ°çš„çª—å£ä¸€èˆ¬æœ‰ä¸‰ä¸ªï¼ŒæŽ¥æ”¶çª—å£ã€å‘é€çª—å£å’Œæ‹¥å¡žçª—å£ï¼š æŽ¥æ”¶çª—å£ï¼šä¸€èˆ¬æ˜¯æŽ¥æ”¶ç«¯å‘å‘é€ç«¯é€šå‘Šè‡ªå·±è¿˜å¯ä»¥æŽ¥æ”¶å’Œç¼“å†²å¤šå°‘æ•°æ®ï¼Œ å¯åœ¨ç”¨æˆ·æ€é€šè¿‡è®¾ç½® SO_RCVBUF ä¸€å®šç¨‹åº¦ä¸Šè°ƒæ•´åˆå§‹çš„æŽ¥å—çª—å£å¤§å°ã€‚ æ‹¥å¡žçª—å£ï¼šå¯ä»¥ç†è§£ä¸ºæ‹¥å¡žç®—æ³•å®šä¹‰çš„é€»è¾‘çª—å£ï¼Œæ˜¯æœ‰å‘é€ç«¯çš„æµæŽ§ç®—æ³•ï¼Œä¸€èˆ¬æ˜¯ MSSï¼ˆæœ€å¤§åˆ†æ®µå¤§å°ï¼‰çš„æ•°å€ã€‚åœ¨ä¸€èˆ¬æ„ä¹‰çš„æ‹¥å¡žç®—æ³•ä¸ï¼Œçª—å£åœ¨æ…¢å¼€å§‹é˜¶æ®µå¢žé•¿ï¼Œå½“çª—å£å€¼åˆ°è¾¾ ssthreshï¼ˆæ…¢å¯åŠ¨é˜ˆå€¼ï¼‰æˆ–æ•°æ®ä¸¢å¤±æ—¶ï¼Œä¼šè¿›è¡Œç›¸åº”è°ƒæ•´ã€‚æ€»ä¹‹ï¼Œè¿™æ˜¯ä¸€ä¸ªç”±æ‹¥å¡žç®—æ³•æ‰€æŽ§åˆ¶çš„çª—å£ã€‚ å‘é€çª—å£ï¼šç®—æ˜¯ä¸€ä¸ªç»¼åˆæ€§çš„çª—å£ï¼Œä¸»è¦ç”¨äºŽè°ƒæ•´å¯å‘é€æ•°æ®çš„å¤§å°ï¼Œå’Œå‘é€ç¼“å†²åŒºå‰©ä½™å¤§å°ï¼Œæ‹¥å¡žçª—å£åŠå‘é€ä¸çš„æ•°æ®åŒ…é‡ï¼ˆpac
hiboma 2020/01/10
__tcp_select_window ã«ã¤ã„ã¦

kernel

tcp
ãƒªãƒ³ã‚¯
BPF(bcc)ã®ãƒ¡ãƒ¢
bpf ã‚„ bcc å‘¨ã‚Šã¯ä½¿ã‚ãªã„ã¨å¿˜ã‚Œã¦ã—ã¾ã†ã®ã§ãƒ¡ãƒ¢ã€‚ Hello, World clang ã§ bpf ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚’ä½œã£ã¦ C ã§ãã‚Œã‚’èªã¿è¾¼ã‚€â€¦ã¨ã„ã†æ–¹æ³•ã‚ˆã‚Šã‚‚ https://github.com/iovisor/bcc/ ã‚’ä½¿ã†æ–¹ãŒæ¥½ãªã®ã§ã€bcc ã§ã‚„ã‚Šã¾ã™ã€‚ C ã§ã‚„ã‚‹ã¨ç’°å¢ƒä½œã‚‹ã¨ã“ã‚ã‹ã‚‰å¤§å¤‰ãªã®ã§â€¦ ref : https://blog.raymond.burkholder.net/index.php?/archives/1000-eBPF-Basics.html execve ã‚·ã‚¹ãƒ†ãƒ ã‚³ãƒ¼ãƒ«ãŒå‘¼ã°ã‚ŒãŸã¨ãã« Hello, World! ã¨è¡¨ç¤ºã™ã‚‹ãƒ—ãƒã‚°ãƒ©ãƒ ã‚’ä½œã‚Šã¾ã™ã€‚ from bcc import BPF bpf_text = """ #include <uapi/linux/ptrace.h> #include <linux/sched.h> #include <
hiboma 2019/12/19
ã‚„ã£ãŸãƒ¼

kernel
ãƒªãƒ³ã‚¯
LKML: Roman Gushchin: [PATCH v2] cgroup, blkcg: prevent dirty inodes to pin dying memory cgroups
[PATCH v2] cgroup, blkcg: prevent dirty inodes to pin dying memory cgroups We've noticed that the number of dying cgroups on our production hosts tends to grow with the uptime. This time it's caused by the writeback code. An inode which is getting dirty for the first time is associated with the wb structure (look at __inode_attach_wb()). It can later be switched to another wb under some conditions
hiboma 2019/12/04
https://hiboma.hatenadiary.jp/entry/2019/12/04/150718 ã«é–¢é€£ã—ã¦ã„ã‚‹

kernel
ãƒªãƒ³ã‚¯
net: increase SOMAXCONN to 4096 - kernel/git/torvalds/linux.git - Linux kernel source tree
hiboma 2019/11/29
5.4 - somaxconn ã®ãƒ‡ãƒ•ã‚©ãƒ«ãƒˆå€¤ãŒ 4096 ã«ã€€ ( 20å¹´é–“å¤‰ã‚ã£ã¦ãªã‹ã£ãŸã‚‰ã—ã„w

kernel
ãƒªãƒ³ã‚¯
SSD Advisory â€“ Linux Kernel AF_PACKET Use After Free (packet_sock) â€“ SecuriTeam Blogs
hiboma 2018/10/23
kernel
ãƒªãƒ³ã‚¯
Intent to Ship: Long Task API - Google Groups
hiboma 2018/06/24
"Reason for revert: this breaks function tracing in /sys/kernel/debug/tracing"

kernel
ãƒªãƒ³ã‚¯
GitHub - mihic/linux-intel-undervolt: Guide to linux undervolting for Haswell and never Intel CPUs
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session.
hiboma 2018/06/14
wrmsr 0x150 8000001c00000000 ã®ãƒã‚°

kernel

linux
ãƒªãƒ³ã‚¯
LKML: Greg Kroah-Hartman: [PATCH 4.16 64/81] fsnotify: Fix fsnotify_mark_connector race
[PATCH 4.16 64/81] fsnotify: Fix fsnotify_mark_connector race 4.16-stable review patch. If anyone has any objections, please let me know. ------------------ From: Robert Kolchmeyer <rkolchmeyer@google.com> commit d90a10e2444ba5a351fa695917258ff4c5709fa5 upstream. fsnotify() acquires a reference to a fsnotify_mark_connector through the SRCU-protected pointer to_tell->i_fsnotify_marks. However, it a
hiboma 2018/06/04
kernel
ãƒªãƒ³ã‚¯
dyndbg(dynamic debug)ã®ä½¿ã„æ–¹ - ã¸ã«ã‚ƒãºã‚“ã¦ï¼ æ—¥ã€…å‹‰å¼·ã®ã¾ã¨ã‚
printk()ã¯ãƒ‡ãƒãƒƒã‚°å‡ºåŠ›ç„¡åŠ¹åŒ–ã®éš›ã«ã¯ã€ã‚³ãƒ¡ãƒ³ãƒˆã‚¢ã‚¦ãƒˆã™ã‚‹ãªã‚Šã—ã¦å†ãƒ“ãƒ«ãƒ‰ã™ã‚‹å¿…è¦ãŒã‚ã‚Šã¾ã™ã€‚ ä½•åº¦ã‚‚ãƒ“ãƒ«ãƒ‰ã‚’ç¹°ã‚Šè¿”ã™ã®ã¯é¢å€’ãªã®ã§ã€ãƒ‡ãƒãƒƒã‚°å‡ºåŠ›ã®æœ‰åŠ¹/ç„¡åŠ¹ã‚’å®Ÿè¡Œæ™‚ã«å‹•çš„ã«åˆ‡ã‚Šæ›¿ãˆã‚‰ã‚Œã‚‹ã‚ˆã†ã«ã™ã‚‹æ©Ÿèƒ½ãŒ dyndbg(dynamic debug)ã§ã™ã€‚ ãªãŠä»¥é™ã®èª¬æ˜Žã¯ã€Linux 3.14.51ã§ç¢ºèªã—ãŸã‚‚ã®ã§ã™ã€‚ ã¾ãŸã€debugfsã®ãƒžã‚¦ãƒ³ãƒˆãƒã‚¤ãƒ³ãƒˆã‚’ã§ç•¥è¨˜ã—ã¦ã„ã¾ã™ã€‚ å‚è€ƒ: debugfsã®ãƒžã‚¦ãƒ³ãƒˆæ–¹æ³• (ãƒžã‚¦ãƒ³ãƒˆãƒã‚¤ãƒ³ãƒˆ /sys/kernel/debug ã®å ´åˆ) sudo mount -t debugfs none /sys/kernel/debug ã‚«ãƒ¼ãƒãƒ«ã‚³ãƒ³ãƒ•ã‚£ã‚®ãƒ¥ãƒ¬ãƒ¼ã‚·ãƒ§ãƒ³ã®å¤‰æ›´ dyndbgã‚’ä½¿ç”¨ã™ã‚‹ã«ã¯ã€ã‚«ãƒ¼ãƒãƒ«ã‚³ãƒ³ãƒ•ã‚£ã‚®ãƒ¥ãƒ¬ãƒ¼ã‚·ãƒ§ãƒ³ã®å¤‰æ›´ãŒå¿…è¦ã§ã™ã€‚ ã¾ãšã€ä»¥ä¸‹ã®å ´æ‰€ã«ã‚ã‚‹ã€ŒCONFIG_DYNAMIC_DEBUGã€ã‚’æœ‰åŠ¹åŒ–ã—ã¾ã™ã€‚ Kernel ha
hiboma 2018/05/18
kernel
ãƒªãƒ³ã‚¯
Re: regression in 4.14-rc2 caused by apparmor: add base infastructure for socket mediation
On Thu, Oct 26, 2017 at 8:54 PM, James Morris <james.l.mor...@oracle.com> wrote: > On Thu, 26 Oct 2017, Linus Torvalds wrote: > >> I'm *very* unhappy with the security layer as is > > What are you unhappy with?We had two big _fundamental_ probl ems this merge window: - untested code that clearly didn't do what it claimed it did, and which caused me to not even accept the main pull request - apparmo
hiboma 2017/10/28
æ¿€ã‚ªã‚³ã®ç¶šã

kernel
ãƒªãƒ³ã‚¯
Re: regression in 4.14-rc2 caused by apparmor: add base infastructure for socket mediation
On Tue, Oct 24, 2017 at 1:57 PM, John Johansen <[email protected]> wrote: > > actually a lot of work and testing has been done. A regression was > found, the fix is in testing and it should land soon, but its not the > regression you are having issues with.Stop gthis f*cking idiocy already! As far as the kernel is concerned, a regressions is THE KERNEL NOT GIVING THE SAME END RESULT WITH
hiboma 2017/10/28
æ¿€ã‚ªã‚³ Linus ðŸ”¥

kernel
ãƒªãƒ³ã‚¯
Linux Load Averages: Solving the Mystery
Recent posts: 28 Apr 2023 Â» eBPF Observability Tools Are Not Security Tools 01 Mar 2023 Â» USENIX SREcon APAC 2022: Computing Performance: What's on the Horizon 17 Feb 2023 Â» USENIX SREcon APAC 2023: CFP 02 May 2022 Â» [email protected] 15 Apr 2022 Â» Netflix End of Series 1 09 Apr 2022 Â» TensorFlow Library Performance 19 Mar 2022 Â» Why Don't You Use ... 26 Sep 2021 Â» The Speed of Time 06 Sep 2021 Â»
hiboma 2017/08/09
kernel
ãƒªãƒ³ã‚¯
Linux_4.12 - Linux Kernel Newbies
Linux 4.12 has been released on Sun, 2 July 2017. Summary: This release includes a new BFQ I/O scheduler which provides a much better interactive experience; it also includes preliminary support for Radeon RX Vega graphic cards and support for USB Type-C connectors; improvements to the live kernel patching feature, support for Intel IMSM's Partial Parity Log which allows to close the RAID5 write h
hiboma 2017/07/04
kernel
ãƒªãƒ³ã‚¯
The 2017 Linux Storage, Filesystem, and Memory-Management Summit [LWN.net]
hiboma 2017/04/04
kernel
ãƒªãƒ³ã‚¯
Automatic NUMA Balancing
hiboma 2017/02/06
kvm

numa

kernel
ãƒªãƒ³ã‚¯
GitHub - intel/numatop: NumaTOP is an observation tool for runtime memory locality characterization and analysis of processes and threads running on a NUMA system.
hiboma 2017/02/05
numa

kernel
ãƒªãƒ³ã‚¯
Googleã‚°ãƒ«ãƒ¼ãƒ—
hiboma 2017/02/05
kernel
ãƒªãƒ³ã‚¯
1 2 3 4 5 6 7 8 9 10 æ¬¡ã®ãƒšãƒ¼ã‚¸