![]() |
|
|
#1 |
|
"Vasiliy"
Apr 2017
Ukraine
32·7 Posts |
I have AMD Athlon X4 860K with L1=16*4+96*2=256 KB
But LLR write L1 cache size : 16 KB Could somebody explain this? pre-thanks |
|
|
|
|
|
#2 |
|
"Curtis"
Feb 2005
Riverside, CA
16DE16 Posts |
In your calculation shown, what label does 16 have? What label does 96 have? (For example, instruction cache or data cache)
Why do you multiply by 4 or 2? Finally, now that you've labeled those items, why would a program only note one of the 16s, and none of the 96s? I wager you'll learn a fair bit about what cache does when you read about the answers to these questions. |
|
|
|
|
|
#3 |
|
"Vasiliy"
Apr 2017
Ukraine
32·7 Posts |
L1 D(Data?)=16KB *4
L1 I(intsructions?)=96KB*4 |
|
|
|
|
|
#4 |
|
"Victor de Hollander"
Aug 2011
the Netherlands
117910 Posts |
L1 is not shared between cores, so why multiply it by 4 (cores)?
|
|
|
|
|
|
#5 |
|
"Kieren"
Jul 2011
In My Own Galaxy!
2×3×1,693 Posts |
P95 and CPUZ report a 6700K as L2=256K. My understanding is that this value is 'per core'. With my previous AMD '8 core' CPU, I think P95 reported L2 per '2-ALU + 1 FPU' unit.
|
|
|
|
|
|
#6 |
|
P90 years forever!
Aug 2002
Yeehaw, FL
827910 Posts |
|
|
|
|
|
|
#7 |
|
"Vasiliy"
Apr 2017
Ukraine
6310 Posts |
Cpu-Z write
L1-D=16Kb*4 L1-I=96Kb*2(in previous message i wrote 96Kb*4. Sorry, my mistake. In first message everything is right) L2=2048Kb So L2 per core=2048/4=512Kb Last fiddled with by vasyannyasha on 2017-12-19 at 07:19 Reason: About mistake in previous message |
|
|
|
![]() |
| Thread Tools | |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Where can I find a Reverse and Add program? I can't find any! | Stargate38 | Programming | 18 | 2015-07-10 06:08 |
| recognizing only 4MB of 8MB L2 cache | Dabigley13 | Information & Answers | 4 | 2015-05-08 07:45 |
| 8k fft and L2 cache | pman | Information & Answers | 1 | 2012-05-23 14:00 |
| L2 cache question | Longshot | Hardware | 5 | 2005-10-26 23:55 |
| L3 cache. What could it buy us? | nomadicus | Hardware | 3 | 2003-08-09 22:23 |