Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Cache behavior

Avatar for Chen Chen
November 03, 2022
Science
0
84

Cache behavior

Avatar for Chen

Chen

November 03, 2022
Tweet

More Decks by Chen

See All by Chen
Example ROP in MIPS
Avatar for Chen tiffany_04192
0
61
defense
Avatar for Chen tiffany_04192
0
74
Dijkstra’s Algorithm
Avatar for Chen tiffany_04192
0
120
Floyd-Warshall
Avatar for Chen tiffany_04192
0
91
Topological sort DFS
Avatar for Chen tiffany_04192
0
92
BFS
Avatar for Chen tiffany_04192
0
130
DFS
Avatar for Chen tiffany_04192
0
120
Cache miss paper
Avatar for Chen tiffany_04192
0
110
Greedy
Avatar for Chen tiffany_04192
0
110

Other Decks in Science

See All in Science
【RSJ2025】PAMIQ Core: リアルタイム継続学習のための⾮同期推論・学習フレームワーク
Avatar for Geson Anko gesonanko
0
670
データマイニング - グラフデータと経路
Avatar for Y. Yamamoto trycycle
PRO
1
300
なぜ21は素因数分解されないのか？ - Shorのアルゴリズムの現在と壁
Avatar for Daiki Murata daimurat
0
310
データベース10: 拡張実体関連モデル
Avatar for Y. Yamamoto trycycle
PRO
0
1.1k
データマイニング - グラフ埋め込み入門
Avatar for Y. Yamamoto trycycle
PRO
1
180
白金鉱業Vol.21【初学者向け発表枠】身近な例から学ぶ数理最適化の基礎 / Learning the Basics of Mathematical Optimization Through Everyday Examples
Avatar for BrainPad brainpadpr
1
650
AI(人工知能)の過去・現在・未来 —AIは人間を超えるのか—
Avatar for Y-h. Taguchi tagtag
PRO
0
150
データから見る勝敗の法則 / The principle of victory discovered by science (open lecture in NSSU)
Avatar for konakalab konakalab
1
290
データベース12: 正規化(2/2) - データ従属性に基づく正規化
Avatar for Y. Yamamoto trycycle
PRO
0
1.1k
academist Prize 4期生 研究トーク延長戦！「美は世界を救う」っていうけど、どうやって？
Avatar for Jimpei jimpe_hitsuwari
0
480
AkarengaLT vol.41
Avatar for hashimoto-kei hashimoto_kei
1
110
People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text
Avatar for hajime kiyama rudorudo11
0
200

Featured

See All Featured
Facilitating Awesome Meetings
Avatar for Lara Hogan lara
57
6.8k
Conquering PDFs: document understanding beyond plain text
Avatar for Ines Montani inesmontani
PRO
4
2.4k
The Illustrated Guide to Node.js - THAT Conference 2024
Avatar for David Neal reverentgeek
1
280
Max Prin - Stacking Signals: How International SEO Comes Together (And Falls Apart)
Avatar for Tech SEO Connect techseoconnect
PRO
0
100
Testing 201, or: Great Expectations
Avatar for Joseph Mastey jmmastey
46
8.1k
ラッコキーワード サービス紹介資料
Avatar for ラッコ株式会社 rakko
1
2.5M
Fireside Chat
Avatar for paigeccino paigeccino
41
3.8k
Designing for Performance
Avatar for Lara Hogan lara
611
70k
Fashionably flexible responsive web design (full day workshop)
Avatar for Andy Clarke malarkey
408
66k
Exploring the relationship between traditional SERPs and Gen AI search
Avatar for Ray Grieselhuber raygrieselhuber
PRO
2
3.7k
JAMstack: Web Apps at Ludicrous Speed - All Things Open 2022
Avatar for David Neal reverentgeek
1
380
WENDY [Excerpt]
Avatar for Tessa Abrams tessaabrams
9
36k

Transcript

  1. 23 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 0 000 00 00 00 00 6 0 000 00 00 00 00 7 0 000 00 00 00 00 Cache starts empty (valid = 0), 8 blocks (or sets) in cache, 4B blocks, 16-bit address

  2. 24 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 0 000 00 00 00 00 6 0 000 00 00 00 00 7 0 000 00 00 00 00 Access address 0x1234 = 0001 0010 0011 0100 Offset = 0 Index = 5 Tag = 145 Valid bit == 0 à miss

  3. 26 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 145 0F 1E 39 EC 6 0 000 00 00 00 00 7 0 000 00 00 00 00 Access address 0x1234 = 0001 0010 0011 0100 lb: 00 00 00 EC lh: 00 00 39 EC lw: 0F 1E 39 EC Offset = 2 Tag = 145 Index = 5 Valid && tag match à hit

  4. 27 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 145 0F 1E 39 EC 6 0 000 00 00 00 00 7 0 000 00 00 00 00 Valid && tag match à hit lb: 00 00 00 1E lh: 00 00 0F 1E lw: (unaligned) Access address 0x1236 = 0001 0010 0011 0110 Offset = 2 Index = 5 Tag = 145

  5. 28 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 145 0F 1E 39 EC 6 0 000 00 00 00 00 7 0 000 00 00 00 00 Access address 0x1238 = 0001 0010 0011 1000 Offset = 0 Index = 6 Tag = 145 Not valid à miss

  6. 29 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 145 0F 1E 39 EC 6 0 000 00 00 00 00 7 0 000 00 00 00 00 Access address 0x1238 = 0001 0010 0011 1000 1 145 Make request to next level... Wait for it…

  7. 30 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 145 0F 1E 39 EC 6 1 145 3C 99 11 12 7 0 000 00 00 00 00 Access address 0x2234 = 0010 0010 0011 0100 Offset = 0 Index = 5 Tag = 273 Valid, but tags don’t match à miss

  8. 31 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 145 0F 1E 39 EC 6 1 145 3C 99 11 12 7 0 000 00 00 00 00 Access address 0x2234 = 0010 0010 0011 0100 Make request to next level... Wait for it… 273

  9. 32 Cache Behavior Example Set # Valid Tag Data 0

    0 000 00 00 00 00 1 0 000 00 00 00 00 2 0 000 00 00 00 00 3 0 000 00 00 00 00 4 0 000 00 00 00 00 5 1 273 01 CF D0 87 6 1 145 3C 99 11 12 7 0 000 00 00 00 00 Access address 0x2234 = 0010 0010 0011 0100 Note that now, 0x1234’s data is gone Replaced by 0x2234’s data