COMPGROUPS.NET | Search | Post Question | Groups | Stream | About | Register

### List of number of cycles for each instruction?

• Email
• Follow

Hi,

I am looking for documents that would describe how many clock cycles
each instruction takes on the different x86/x86_64 processors. I can't
seem to find any such documents on AMDs web site or anywhere else.
My CPU is an athlon64 (AMD Athlon(tm) 64 X2 Dual Core Processor 5600+)
but I would also like to have information about others.

Does anyone know of a good reference?

 0

See related articles to this posting

Wolfnoliir wrote:

> I am looking for documents that would describe how many clock cycles
> each instruction takes on the different x86/x86_64 processors. I can't
> seem to find any such documents on AMDs web site or anywhere else.
> My CPU is an athlon64 (AMD Athlon(tm) 64 X2 Dual Core Processor 5600+)
> but I would also like to have information about others.
>
> Does anyone know of a good reference?

The AMD docs are called 'Software Optimization Guide':

The Intel ones 'Optimization Reference Manual':

They have instruction timing lists at the end.

Agner Fog has a document with timing tables as well (manual 4):
http://www.agner.org/optimize/

 0

Wolfnoliir schrieb:
> I am looking for documents that would describe how many clock cycles
> each instruction takes on the different x86/x86_64 processors. I can't
> seem to find any such documents on AMDs web site or anywhere else.

Intel and AMD both have some documentation. Google finds it.

"Intel 64 and IA-32 Architectures Optimization Reference Manual"
<http://www.intel.com/Assets/PDF/manual/248966.pdf>
(Appendix C)

"AMD Athlon Processor x86 Code Optimization Guide"
<http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/22007.pdf>
(for K8, Appendix F)

"Software Optimization Guide for the AMD Hammer Processor"
<http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/40546_GH_Optguide_Ver_3-10_2-12-09.pdf>
(for K10, Appendix C)

> Does anyone know of a good reference?

http://www.agner.org/optimize/instruction_tables.pdf
http://gmplib.org/~tege/x86-timing.pdf

Hendrik vdH

 0

tni wrote:
> Wolfnoliir wrote:
>
>> I am looking for documents that would describe how many clock cycles
>> each instruction takes on the different x86/x86_64 processors. I can't
>> seem to find any such documents on AMDs web site or anywhere else.
>> My CPU is an athlon64 (AMD Athlon(tm) 64 X2 Dual Core Processor 5600+)
>> but I would also like to have information about others.
>>
>> Does anyone know of a good reference?
>
> The AMD docs are called 'Software Optimization Guide':
>
> The Intel ones 'Optimization Reference Manual':
>
> They have instruction timing lists at the end.
>
> Agner Fog has a document with timing tables as well (manual 4):
> http://www.agner.org/optimize/

Thank you!

 0

3 Replies
164 Views

Similar Articles

12/11/2013 5:45:56 PM
[PageSpeed]

Similar Artilces:

Numbered Lists in Pages 09
Tonight I was trying to reformat some text into numbered lists in a Pages 09 document. When I selected Numbered from the list drop-down in the side window, the numbers ended up in a different font than my text. I ended up typing the numbers by hand. It's done now, but I never figured out a way to specify the font of those numbers, or how to change it with the little paragraph button, nor can I think of any reason the numbers shouldn't be the same font as my text. What should I have done? -- "Huan" On 2012-03-23 04:03:01 +0000, Huan said: > Tonight I was trying to ref...

search number inner a list
if I have: var arr= new Array("10","10.8","10,8","40","50","60","70","80","140","150","-5","-7","+8","160","170","180"); and would like to find numbers between 1 and 5 I have 10 10.8 40 140 150 160 170 180 think error becaus it see not number but string; is possible to make a correction at code? and if in the list (in the example I have all numbers; I have also text; is possible to extraxt the interval 1 to 5 with your soluti...

processor clock cycle per instructions
Hai all, Can u pls suggest the method or formula to calculate number of processor clock cycles for each instructions ?It will be greatful to knew this as i have referred the Intel data sheets which includes.I am eager to knew how they r calculating it. regards, fazal If you design the chip, you know how many cycles it takes to do the instruction. If it is a frequently used instruction most chip makers will implement it in hardware with gates, otherwise they may have 'micro-ops' run the instruction with several different instructions running to do that one. On Mon, 3 Dec 2007 21:1...

how to measure number of cycles in ISE6.3
hi I need to measure "number of clock cycles" or "execution time", after mapping the VHDL code into Vertex II pro. In Modelsim simulation, it took 350 cycles with 20 MHz clock frequency. I hope the performance after the mapping will be same as the performance in simulation. In ISE 6.3 (or EDK 6.3) , how can we measure the amount of clock cycles? Maybe I can use "big" counter along with my design. Are there any efficient way to do so? Thanx "Jack" <JEmoderatz@yahoo.com> writes: > hi > > I need to measure "number of clock cycle...

Counting number of records from a value list
Hi there, Can someone please tell me how to do this. I have fields that are either checkboxes, drop-down value lists, or radio buttons, I want to be able to calculate the number of records that have a certain value from these fields. For example, if I have a value list that has say One, Two, Three in it, I want to see the total number of records that has one, the total with two etc. Also, if I have a checkbox, I want to see how many records has each option. I need to get this info in order to do a report summarizing different things. Can someone please tell me how to do this? Thanks a ...

number of machine instructions
hello, i'm trying to answer a "prove an upper bound on the number of machine instructions required to process M connections on N objects" exercise from an algorithms book. i don't understand a very basic detail of how i should go about answering the exercise. here's a simplified example question and example programme to hopefully get at what i'd like to know: Prove an upper bound on the number of machine instructions required to process N objects using the below programme. You may assume, for example, that any C assignment statement always requires less than c ins...

Re: Number of Instructions Used by Datastep
Paul, An academically interesting question. My guess is that only a handful of folks would have even a clue--the handful that understand the detailed inner workings of the datastep compiler. Without that kind of intimate knowledge I don't think the question you asked can be answered by anything but a wild ass guess. Besides, why resort to arcane theory when it is so easy to determine empirically what you really want to know. Why not pose your question to http://www.cmg.org/national/about-cmg.html? Joe On Fri, Feb 19, 2010 at 11:42 AM, Paul Walker <walker.627@osu.edu> wrote: &...

Listings: \lstinputlisting and line number reference
Hi, I include my source code with \lstinputlisting and place - as written in the manual - the label with (*@\label{lstbeamer01class}@*) into the source file. If I include \ref{lstbeamer01class} into the main file, it only outputs ??. The log file says that "Reference `lstbeamerclass' on page 14 undefined on input line 384." How could I use the line number reference together with \lstinputlisting? Thansk in advance Christian ...

Using listings package to highlight numbers in code
Dear all, I am using the listings package for a very simple code listing. I would like every number (we are talking integers here so no decimal point to worry about) to be highlighted red. I have tried the following settings: \documentclass[10pt,a4paper]{article} \usepackage{listings} \usepackage{color} \begin{document} \begin{lstlisting}[emph={0,1,2,3,4,5,6,7,8,9}, emphstyle=\color{red}] This is a test 123 + 456 = 789 \end{lstlisting} \end{document} but no success. Can anyone offer any suggestions? Many thanks in advance, Paul -- Department of Computing, Imperial College London ht...

Re: Looking for a list of processor instructions 4636
"Jim Carlock" <spamtrap@crayne.org> wrote in message news:CgxPc.28569$DZ.2788978@twister.tampabay.rr.com... > Was that you that sent the email ? Someone from Atlanta sent > an email. It's hard to see the IP addresses in this newsgroup > here. :-) I think I was in Dallas when I sent it, but who knows what IP address it came from... > I've been trying to follow what's being called what at Intel's > site but it's very confusing. I think I'm just going to forget > about the Katmai, Prescott names and such. Well, there are variants within... Re: Looking for a list of processor instructions #2 "Jim Carlock" <spamtrap@crayne.org> wrote in message news:2r9Pc.18523$DZ.2368241@twister.tampabay.rr.com... > Question about the processor names... I'm seeing that Intel CPUs > are labeled as follows: > > P1/P5/Pentium > P2 > P3 > P4 > P6 > > Then their is a switch to: > > Katmai, Williamette and Prescott. I think they got the Prescott > name from George Bush's grandfather who funded Adolph Hitler > and was a traitor during WWII. :-) What a shame. > > The Katmai, Williamette and Prescott processors... are they ever > re...

Re: Looking for a list of processor instructions 11804
"Ivan Korotkov" <koroNOSPAMtkov2@ztelDOT.ru> wrote in message news:opsbu0ns0675ztla@news.zebratelecom.ru... > So you mean Prescott would be slower than Northwood if clockrate were the > same??? Are there any enhancements (except for clockrate) at all in it? > And what are these 8 pipeline stages? I thought there aren't any exact > numbers of stages in pipeline since they are not so distinct. Depending on the benchmark used, Prescott tends to be slower than Northwood at the same clock speed. It's not as bad, however, as the difference between the P6 and the...

Re: Number of Instructions Used by Datastep #2
Dear SAS-L-ers, Barry Schwarz posted the following: > Once you start talking about processing a data set, you introduce a > whole new set of I/O variables such as throughput, channel > conflicts/collisions, network performance, etc. These are unrelated > to processing speeds and instruction counts but will impact run time, > sometimes dominantly so. > > But getting back to your question, as opposed to your goal, consider > attacking the problem in reverse. Run a data step with no I/O but > numerous iterations and time it. Then double the iterations. Then > t...

Re: Looking for a list of processor instructions 147484
"Bx.Cornwell" writes: <...> > >Intel sure did some strange things... > >also... was the 286 able to run w/ an 8087 on some boards? was there a 187 ? No, 80287 was different from 8087, and yes, there is a dataheet for an 80C187. Steve ...

HP 49G+ Find Maximum Number In A List 12267
Hello, Today I was trying to figure out how to find the maximum number in a list when finally, and by blind luck, I came up with \<< \<< MAX \>> STREAM \>>. This code is a lot simpler than what I was first trying to do. I just wanted to share that with the group. Regards, Mark ...

Embedding: Is it possible to limit number of virtual instructions executed?
Hi experts- I have a potential use case of embedding Python where it must co- operate with the host c/c++ application in a single thread. That is, the c code and python interpreter need to cooperate and time share, or yield to each other. The main loop would need to do something like: check for events for events with c handlers: process in c for events with python handlers: add to queue if have time: python_continue_execution(number of vm instructions) The goal is to be able to set some limit on the amount of time in the interpreter in each loop, and not rely on the python sid...

Compare Directory File Lists down to the Version Number Level
I have come up with a program called Directory Checker 2.80 that helps people insure that they have the right versions of their files in the right directories. Directory Checker is a utility that enables people to compare the contents of two Windows Directories side by side down to the File Version Number level. It is especially helpful when you are having difficulty making sure that the latest Versions of your files are in the correct directories (i.e., OCX's, DLL's EXE's, Word Docs, Crystal Reports, etc.) This program is available for download at "http://bbussell.airweb....

Sas tops fortune list as number 1 employer
http://news.cnet.com/8301-1001_3-10440381-92.html SAS tops the best employer list again with Google at number 4 Sent from my iPhone ...

Why are 2 clocks the minimum clock cycle for the fastest instructions for Intel chip
Hi, Recently I read Intel 486 insruction set and found that the minimum clock cycle for an instructions is 2 clocks. Usually we design in ASIC and FPGA using 1 clock for a register exchange instruction. What is the reason for 2 clocks of minimum clock cycles? Weng Weng Tianxiang wrote: > Recently I read Intel 486 insruction set and found that the minimum > clock cycle for an instructions is 2 clocks. > > Usually we design in ASIC and FPGA using 1 clock for a register > exchange instruction. What is the reason for 2 clocks of minimum clock > cycles? I don't know th...

concatenate the elements in each list of a list of lists
I already asked a similar question, but encounter problems with python... How can I concatenate the elements in each list of a list of lists list_of_listsA = [['klas*', '*', '*'], ['mooi*', '*', '*', '*'], ['arm*', '*', '*(haar)']] wanted result: list_of_listsA = [['klas* * *'] ['mooi* * * *'] ['arm* * *(haar)']] Thanks a lot ! On Jul 23, 5:33=A0pm, antar2 <desoth...@yahoo.com> wrote: > I already asked a similar question, but encounter problems with > python... > How c...