[B! optimization] dannã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

dann id:dann

optimizationã«é–¢ã™ã‚‹dannã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (27)

${{author_name}}$

{{{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}}

ãƒ—ãƒã‚°ãƒ©ãƒ ã‚’é«˜é€ŸåŒ–ã™ã‚‹è©±
9. 9 æœ€é©åŒ–ã«ã¤ã„ã¦ ã€Œç´°ã‹ã„åŠ¹çŽ‡ã®ã“ã¨ã¯å¿˜ã‚Œã¦ã€æ™‚é–“ã® 97% ã«ã¤ã„ã¦è€ƒãˆ ã‚ˆã†ã€‚æ™‚æœŸå°šæ—©ãªæœ€é©åŒ–ã¯è«¸æ‚ªã®æ ¹æºã ã€‚ãã‚Œã§ã‚‚æ®‹ã‚Š 3% ã«ã¤ã„ã¦ã‚‚æ©Ÿä¼šã‚’é€ƒã™ã¹ãã§ã¯ãªã„ã€ - Donald E. Knuth ã€Œãƒ—ãƒã‚°ãƒ©ãƒ æœ€é©åŒ–ã®ç¬¬ä¸€æ³•å‰‡ : æœ€é©åŒ–ã™ã‚‹ãªã€‚ ãƒ—ãƒã‚°ãƒ©ãƒ æœ€é©åŒ–ã®ç¬¬äºŒæ³•å‰‡ ( ä¸Šç´šè€…é™å®š ): ã¾ã ã™ã‚‹ãªã€‚ ã€ - Michael A. Jackson 11. 11 æœ€é©åŒ–ã®å¯¾è±¡ ä¸»ã« Intel ã® Haswell ãƒžã‚¤ã‚¯ãƒã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£ä»¥é™ã‚’å¯¾è±¡ å¤šãã®ãƒ†ã‚¯ãƒ‹ãƒƒã‚¯ã¯ä»–ã®ãƒ—ãƒã‚»ãƒƒã‚µã«ã‚‚å¿œç”¨ã§ãã¾ã™ ãƒ™ãƒ¼ã‚¹ ãƒžã‚¤ã‚¯ãƒã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£ ãƒ—ãƒã‚»ã‚¹ãƒ«ãƒ¼ãƒ« ç™»å ´å¹´ Nehalem Nehalem 45nm 2008 ã€ƒ Westmere 32nm 2010 Sandy Bridge Sandy Bridge 32nm 2011 ã€ƒ Ivy Bridge 22nm 2012 Hasw
dann 2015/03/16
optimization
ãƒªãƒ³ã‚¯
qrintfã®æœ€é©åŒ– - é–‹ç™ºæ—¥è¨˜
H2Oã®å†…éƒ¨ã§ã¯ï¼Œæ•°å€¤ï¼Œæ–‡å—åˆ—ã®ãƒ•ã‚©ãƒ¼ãƒžãƒƒã‚¿ã¨ã—ã¦sprintfãŒç”¨ã„ã‚‰ã‚Œã¦ã„ã¾ã™ï¼Ž æˆ‘ã€…ã¯ï¼Œæœ€è¿‘ã¾ã§sprintfå°‚ç”¨ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰å¤‰æ›å™¨, qrintfã®ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã‚’è¡Œã£ã¦ã„ã¾ã—ãŸï¼Ž æœ¬ç¨¿ã§ã¯qrintfã®æ¦‚è¦ã¨ä»Šå›žé©å¿œã—ãŸå·¥å¤«ï¼Œãã—ã¦ä»Šå¾Œã®èª²é¡Œã«ã¤ã„ã¦ã¾ã¨ã‚ã¦ãŠãã¾ã™ï¼Ž 1. qrintfã¨ã¯ï¼Ÿ qrintfã¨ã¯Cã‚³ãƒ³ãƒ‘ã‚¤ãƒ©ã®ï¼‘ã¤ã§ã‚ã‚‹gccã®ãƒ—ãƒªãƒ—ãƒã‚»ãƒƒã‚µã®ãƒ©ãƒƒãƒ‘ãƒ¼ã§ã‚ã‚Šï¼Œsnprintfã‚’é«˜é€ŸåŒ–ã™ã‚‹ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰å¤‰æ›å™¨ã§ã™ï¼Ž æœ¬ç¨¿åŸ·ç†æ™‚ç‚¹ã§ã®qrintfã®å¤‰æ›å¯¾è±¡ã¯sprintfã¨snprintfã§ã™ï¼Žä»¥ä¸‹æœ¬æ–‡ä¸ã§ã¯ç‰¹ã«ã“ã¨ã‚ã‚Šã®ãªã„é™ã‚Šsprintfã¨è¨˜è¿°ã—ãŸå ´åˆsprintf, snprintfã®ä¸¡æ–¹ã‚’æŒ‡ã—ã¾ã™ï¼Ž qrintfã¯ã‚½ãƒ¼ã‚¹ã‚³ãƒ¼ãƒ‰ã«å‡ºç¾ã™ã‚‹sprintfã§åˆ©ç”¨ã™ã‚‹ãƒ•ã‚©ãƒ¼ãƒžãƒƒãƒˆæ–‡å—åˆ—ã‚’è§£æžã—ï¼Œåž‹ã”ã¨ã«ç”¨æ„ã•ã‚ŒãŸé–¢æ•°å‘¼å‡ºã«ã‚³ãƒ¼ãƒ‰ã‚’æ›¸ãæ›ãˆsprintfã®é«˜é€ŸåŒ–ã‚’è¡Œã„ã¾ã™ï¼Ž
dann 2015/01/02
optimization
ãƒªãƒ³ã‚¯
https://www.arm.com/files/pdf/AT_-_NEON_for_Multimedia_Applications.pdf
dann 2013/07/18
arm

neon

optimization
ãƒªãƒ³ã‚¯
https://elinux.org/images/4/40/Elc2011_anderson_arm.pdf
dann 2013/07/18
arm

optimization

neon
ãƒªãƒ³ã‚¯
Optimizing Code for ARM Cortex-A8 with NEON SIMD - ETH PIXHAWK: MAV Computer Vision
<box green 80% round center>Obeying these guidelines will significantly speed up your code and will perform equally to hand crafted assem bler in most cased. Read the Official ARM NEON Optimization Examples to learn the details why this statement holds.</box> Use the restrict keyword for pointers in function signatures: It signals to the compiler that no other pointer inside the function points at
dann 2013/07/18
arm

optimization
ãƒªãƒ³ã‚¯
Happy Optimization
2. æ˜¨æ—¥ Pathtraq API å…¬é–‹ã—ã¾ã—ãŸ 6000 ä¸‡ URL ã®ã‚¢ã‚¯ã‚»ã‚¹çµ±è¨ˆã‚’ãƒžãƒƒã‚·ãƒ¥ã‚¢ãƒƒãƒ—å¯èƒ½ ã‚ˆã‚ã—ããŠé¡˜ã„ã—ã¾ã™ ã¨ã„ã†ã®ã¯ã€ã•ã¦ãŠãâ€¦ æœ€é©åŒ–ã®è©±ã€‚
dann 2012/09/18
performance

optimization
ãƒªãƒ³ã‚¯
HTML Meta Tag
Practical Optimization: A Gentle Introduction John W. Chinneck Systems and Computer Engineering Carleton University Ottawa, Ontario K1S 5B6 Canada Site has moved to: https://www.optimization101.org/ Redirecting to new site in 5 seconds.
dann 2011/11/28
performance

optimization
ãƒªãƒ³ã‚¯
ã‚³ãƒ¼ãƒ‰ã‚µã‚¤ã‚ºã¨å®Ÿè¡Œé€Ÿåº¦ã®é–¢ä¿‚ã«é–¢ã™ã‚‹è©± - melancholic afternoon
dann 2011/11/05
optimization

intel

simd
ãƒªãƒ³ã‚¯
https://orsj.org/wp-content/or-archives50/pdf/j_mag/Vol.54_J_058.pdf
dann 2011/11/02
optimization
ãƒªãƒ³ã‚¯
apps_SIMD.PDF
dann 2011/11/02
simd

intel

optimization
ãƒªãƒ³ã‚¯
x86/x64æœ€é©åŒ–å‹‰å¼·ä¼š2çµ‚äº†
dann 2011/10/02
x86

optimization

performance

cpu
ãƒªãƒ³ã‚¯
æ¡ä»¶åˆ†å²ã¨cmovã¨maxps
3. è‡ªå·±ç´¹ä»‹ ï®å…‰æˆæ»‹ç”Ÿï¼ˆã‚µã‚¤ãƒœã‚¦ã‚ºãƒ»ãƒ©ãƒœï¼‰ ï®å§‘æ¯ãªæœ€é©åŒ–ãŒå¤§å¥½ã ï®ä»¥å‰ã¯ã‚³ãƒ¼ãƒ‡ãƒƒã‚¯ç³»ã®è¶£å‘³ãƒ»ä»•äº‹ã‚’ã‚„ã£ã¦ã„ãŸ ï®æœ€è¿‘ã¯ä»•äº‹ã®ãŸã‚ã«æ©Ÿæ¢°å¦ç¿’ã‚’å‹‰å¼·ä¸ ï®PRMLã‚ã‚“ã¡ã‚‡ã“å…¬é–‹ä¸ï¼ˆhttps://github.com/herumi/prml/ï¼‰ ï®blog : http://homepage1.nifty.com/herumi/ ï®mail : herumi@nifty.com ï®twitter : @herumi 3 /24 5. ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ« ï®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ© ï®é–¢æ•°çµ„ã¿è¾¼ã¿ã‚¿ã‚¤ãƒ— ï®ãƒ—ãƒã‚°ãƒ©ãƒ è‡ªä½“ã«çµ„ã¿è¾¼ã¾ã‚Œã‚‹ ï®ãã‚Œè‡ªä½“ãŒãƒ—ãƒã‚°ãƒ©ãƒ ã«å½±éŸ¿ã‚’ä¸Žãˆã‚‹ ï®å›žæ•°ã¯æ£ç¢º ï®gprofã‚„DevPartner, Vtune ï®ã‚¿ã‚¤ãƒŸãƒ³ã‚°å‰²ã‚Šè¾¼ã¿ã«ã‚ˆã‚‹é›†è¨ˆã‚¿ã‚¤ãƒ— ï®å‰²ã‚Šè¾¼ã¿ã§ãã®çž¬é–“ã®eipã‚’å–å¾—ã—é›†è¨ˆ ï®ãƒ—ãƒã‚°ãƒ©ãƒ ã«å½±éŸ¿ã‚’ï¼ˆæ®†ã©ï¼‰ä¸Žãˆãªã„ ï®å›žæ•°ã¯ä¸æ£ç¢ºï¼Œæ™‚é–“ã¯æ¦‚ãæ£ç¢º ï®CodeAnalys
dann 2011/08/07
optimization

cpu
ãƒªãƒ³ã‚¯
adv intro
No. 001 ã¾ãšç°¡å˜ãªã¨ã“ã‚ã‹ã‚‰ func( int i ){ if( abs(i) ã“ã‚“ãªã‚½ãƒ¼ã‚¹ã‚’è€ƒãˆã¾ã™. æ™®é€šã«è€ƒãˆã‚‹ã¨ if() ã®ä¸èº«ã¯ i = -5 ã§ã™ã‹ã‚‰, 2å›žæ¡ä»¶åˆ†å²ã™ã‚‹ã“ã¨ã«ãªã‚Šã¾ã™. ã¨ã“ã‚ãŒ if( (unsigned)( i + 5 ) ã¨ã™ã‚‹ã¨1å›žã®åˆ†å²ã§ã™ã¿ã¾ã™. mov eax,i add eax,5 cmp eax,10 ja .else æœ¬ä½“ .else: ãã®ä»– ã¨ã§ãã‚‹ã®ã§ã™. No. 002 å‰å›žã®ã¯ã‚ã¾ã‚Šã«ç°¡å˜ã ã£ãŸã‹ã‚‚çŸ¥ã‚Œã¾ã›ã‚“. ã‚‚ã†å°‘ã—è¤‡é›‘ãªä¾‹ã‚’è€ƒãˆã¦ã¿ã¾ã™. unsigned int ãª x, y, z, wã¨ã„ã†å¤‰æ•°ã«ãŸã„ã—, ãã‚Œã‚‰ãŒå…¨ã¦0ã‹1ã®æ™‚ã®ã¿å‡¦ç†ã—ãŸã„ã¨ã„ã†å ´åˆã§ã™. if( x #ã¡ãªã¿ã«åˆå¾Œã®ã“ã€œã ã®ã‚½ãƒ¼ã‚¹ã®ä¸€éƒ¨ã§ã™. ã•ã¦ã©ã†ã—ã¾ã™ã‹?
dann 2011/08/07
optimization

cpu
ãƒªãƒ³ã‚¯
Using IntelÂ® Hyper-Threading Technology| IntelÂ® Math Kernel Library for macOS
Using Intel.com Search You can easily search the entire Intel.com site in several ways. Brand Name: Core i9 Document Number: 123456 Code Name: Emerald Rapids Special Operators: â€œIce Lakeâ€, Ice AND Lake, Ice OR Lake, Ice* Quick Links You can also try the quick links below to see results for most popular searches. Product Information Support Drivers & Software
dann 2011/07/23
cpi

intel
ãƒªãƒ³ã‚¯
Microsoft PowerPoint - Altix_Optimization.ppt
Altix3000 ã‚·ãƒªãƒ¼ã‚ºã‚·ãƒªãƒ¼ã‚ºã‚·ãƒªãƒ¼ã‚ºã‚·ãƒªãƒ¼ã‚º ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ã¨ã¨ã¨ã¨æœ€é©åŒ–è¬›ç¿’ä¼šæœ€é©åŒ–è¬›ç¿’ä¼šæœ€é©åŒ–è¬›ç¿’ä¼šæœ€é©åŒ–è¬›ç¿’ä¼š æ—¥æœ¬æ—¥æœ¬æ—¥æœ¬æ—¥æœ¬SGIæ ªå¼ä¼šç¤¾æ ªå¼ä¼šç¤¾æ ªå¼ä¼šç¤¾æ ªå¼ä¼šç¤¾ å¹³æˆå¹³æˆå¹³æˆå¹³æˆ17å¹´å¹´å¹´å¹´4æœˆæœˆæœˆæœˆ1æ—¥æ—¥æ—¥æ—¥ PARTPARTPARTPART â…¡â…¡â…¡â…¡ | 2005/4/1 |SGI Altixã‚µãƒ¼ãƒã‚·ã‚¹ãƒ†ãƒ æœ€é©åŒ–ãƒ»ä¸¦åˆ—åŒ–æŠ€æ³•ã€€ã€€ ã€€ã€€ Page 2 Altix3000 ã‚·ãƒªãƒ¼ã‚ºã‚·ãƒªãƒ¼ã‚ºã‚·ãƒªãƒ¼ã‚ºã‚·ãƒªãƒ¼ã‚º ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ã¨ã¨ã¨ã¨æœ€é©åŒ–è¬›ç¿’ä¼šæœ€é©åŒ–è¬›ç¿’ä¼šæœ€é©åŒ–è¬›ç¿’ä¼šæœ€é©åŒ–è¬›ç¿’ä¼š ç¬¬ç¬¬ç¬¬ç¬¬1éƒ¨éƒ¨éƒ¨éƒ¨: ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°æ¦‚è¦æ¦‚è¦æ¦‚è¦æ¦‚è¦ â€¢ã‚·ã‚¹ãƒ†ãƒ ã‚·ã‚¹ãƒ†ãƒ ã‚·ã‚¹ãƒ†ãƒ ã‚·ã‚¹ãƒ†ãƒ æ¦‚è¦æ¦‚è¦æ¦‚è¦æ¦‚è¦ â€¢ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ãƒ—ãƒã‚°ãƒ©ãƒŸãƒ³ã‚°ç’°å¢ƒ
dann 2011/07/11
memory

performance

optimization

cpu
ãƒªãƒ³ã‚¯
HugeDomains.com
dann 2011/07/10
intel

multicore

optimization
ãƒªãƒ³ã‚¯
Memory part 5: What programmers can do [LWN.net]
Benefits for LWN subscribersThe primary benefit from subscribing to LWN is helping to keep us publishing, but, beyond that, subscribers get immediate access to all site content and access to a number of extra site features. Please sign up today! [Editor's note: welcome to part 5 of Ulrich Drepper's "What every programmer should know about memory". This part is the first half of sectionÂ 6, which di
dann 2011/07/10
memory

optimization

cpu
ãƒªãƒ³ã‚¯
åœéƒ¨æ˜Œå¹³ã®ã‚ã¾ã‚Šreblogã—ãªã„tumblr - æœ€é€Ÿã® memset64 ã‚’æ±‚ã‚ã¦ ä»Šå›žã®ãŠé¡Œã¯ char å¹…ã˜ã‚ƒãªãã¦ word å¹…ã®...
void* memset64(void* d, uint64_t i, int n) { asm("cld; rep stosq;" :: "D"(d), "a"(i), "c"(n) : "memory"); return d; } æœ€è¿‘ã® CPU ã¯ã‚¯ã‚½è³¢ã„ã€‚ãã®ãŸã‚ã€ä¸‹æ‰‹ã«æ‰‹ã§ loop unrolling ã™ã‚‹ã‚ˆã‚Šã‚‚ã€é€†ã« CPU ã«ã€Œã“ã“ã¯ãƒ«ãƒ¼ãƒ—ãªã‚“ã ãžãŠãŠãŠï½žã€ã¨ã„ã†ã®ã‚’æ˜Žç¤ºçš„ã«æŒ‡ç¤ºã—ã¦ã‚ã’ãŸã»ã†ãŒ CPU å´ãŒå‹æ‰‹ã‹ã¤ä¸æ°—å‘³ã«æœ€é©ãªæŒ™å‹•ã‚’ç¤ºã—ã‚„ãŒã‚‹ã€‚ãã†ã„ã†æ„å‘³ã§ã¯ç¾ä»£ã¯ã‚ã‚‹ç¨®ã® CISC ãƒ«ãƒã‚µãƒ³ã‚¹ã¨ã‚ˆã¹ã‚‹æ™‚æœŸã§ã¯ã‚ã‚‹ã®ã ã‚ã†( 10 å¹´å¾Œã¯ã‚ã‹ã‚‰ã‚“ãŒ)ã€‚
dann 2011/07/09
gcc

optimization
ãƒªãƒ³ã‚¯
Software optimization resources. C++ and assembly. Windows, Linux, BSD, Mac OS X
See also my blog Contents Optimization manuals Vector class library Object file converter and disassem bler Subroutine library ForwardCom: An open standard instruction set for high performance microprocessors Test programs for measuring clock cycles in C++ and assem bly code Floating point exception tracking and NaN propagation CPUID manipulation program Links Optimization manuals This series of fiv
dann 2011/07/09
performance

c++

cpu

optimization
ãƒªãƒ³ã‚¯
Intel optimization
ï»¿ Intelãƒ—ãƒã‚»ãƒƒã‚µæœ€é©åŒ–ãƒžãƒ‹ãƒ¥ã‚¢ãƒ«ã‚’èªã‚‚ã† ä»Šã¾ã§ã‚ã‚‹ç¨‹åº¦ã¯èªã‚“ã§ãŸã‘ã©, ã€ŽIntel 64 and IA-32 Architectures Optimization Reference Manualã€ã‚’èªã¿ç›´ã—ã¦, æ°—ã¥ã„ãŸã“ã¨ãªã©ã‚’ã¾ã¨ã‚ã¦ã¿ã‚ˆã†ã¨ã„ã†è©¦ã¿. ä»Šæ‰‹å…ƒã«ã‚ã‚‹ã®ã¯248966-024 April 2011ã®ã‚‚ã®. æ—¥æœ¬èªžç‰ˆã‚‚ã‚ã‚Šã¾ã™ãŒ, å¤šå°‘å¤ã„ã®ã§è‹±èªžç‰ˆãŒã‚ˆã„ã§ã—ã‚‡ã†. è¡¨è¨˜ã«ã¤ã„ã¦ ãƒ—ãƒã‚»ãƒƒã‚µã¨æ›¸ã„ãŸã‚Š, CPUã¨æ›¸ã„ãŸã‚Šã—ã¾ã™. Intel64ã‚’x64ã¨æ›¸ã„ãŸã‚Šã—ã¾ã™. ä¾‹ãˆã°3ã‚µã‚¤ã‚¯ãƒ«ã®ã“ã¨ã‚’å€‹äººçš„ãªæ…£ç¿’ã§3clkã¨æ›¸ãã“ã¨ãŒå¤šã„ã§ã™. ã™ã„ã¾ã›ã‚“. é–“é•ã„ãªã©ã”ã–ã„ã¾ã—ãŸã‚‰ãƒ¡ãƒ¼ãƒ«(herumi@nifty.com)ã‹@herumiã«ãŠé¡˜ã„ã—ã¾ã™. 2ç« Intel64/IA-32 CPUã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£ 2.1 Sandy Bridgeæ¦‚è¦ 2.1.
dann 2011/07/09
intel

optimization

cpu
ãƒªãƒ³ã‚¯
1 2 æ¬¡ã®ãƒšãƒ¼ã‚¸