Talk:Memory address

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing (Rated C-class, High-importance)
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.
 

uni[edit]

We are doing an piece at uni on address-binding (runtime, compile-time, load-time) is it just me or does wikipedia lack anything on this? Paige Master 20:26, 13 June 2006 (UTC)

There's a brief allusion to address binding in the register allocation article and the relocation (computer science). Is there anything more specific about address binding? --68.0.124.33 (talk) 02:50, 9 September 2008 (UTC)


One exbibyte is 260 bytes. How is 264 bytes 16 exbibytes? —Preceding unsigned comment added by Bookmunkie (talkcontribs) 00:31, 12 September 2007 (UTC)

264 = 24·260 = 16·260. -- BenRG 14:16, 12 September 2007 (UTC)

Please clarify[edit]

Isn't virtual address a synonym of linear address rather than logical address, as mentioned in the book Understanding The Linux Kernel written by Daniel & Macro and hereKefengx (talk) 14:16, 22 December 2008 (UTC)

I would say quite the opposite. Surely a linear address is a physical address?
But I could do with some help editing this.

SimonTrew (talk) 22:25, 24 February 2009 (UTC)


Spelling Clarification[edit]

Why are we using gibi instead of giga and exibi instead of exa. —Preceding unsigned comment added by 67.110.32.253 (talk) 15:28, 31 August 2010 (UTC)

The infix "bi" is a way of describing the binary proximation, so as to avoid confusion concerning what sizes are actually the case. A kibibyte is therefore 1024 (2^10) bytes, rather than 1000 (10^3). Mebi- is 2^20, Gibi- is 2^30 and so on. The reason for this, of course, is that computers work in terms of bits and bytes - a binary numeral system rather than a decimal numeral system - and physical increments are manufactured in the same fashion. - 88.88.108.249 (talk) 09:56, 1 February 2011 (UTC)

Dubious - virtual addresses translated by the processor[edit]

In the final section, 'Virtual memory versus physical memory', this statement is made: >When the program is actually executed, the virtual addresses are translated by the processor into real memory addresses. I'm not sure that's accurate. I take 'processor' here to mean 'CPU'. My understanding is that the Memory management unit, rather than the CPU, translates virtual addresses into physical addresses. The MMU does that using a Translation Lookaside Buffer. — Preceding unsigned comment added by PlaysWithLife (talkcontribs) 18:03, 29 August 2011 (UTC)

What links here[edit]

Over 200 pages link to this page! Ideally, this article will provide something that satisfies them all! That's a tall order, but a worthwhile goal. http://en.wikipedia.org/wiki/Special:WhatLinksHere/Memory_address Wbm1058 (talk) 21:25, 7 January 2012 (UTC)

36-bit mainframe computers which used 18-bit word addressing[edit]

When I examined this example, at least three questions raised. First of all, "18-bit word addressing" is unclear itself, because could refer either to the address bus (addresses are 18-bit words) or to the addressable data unit (18-bit words are addressed). Could someone add a link to specs?

Second, the relationship of this example to bytes and byte addressing is not expressed.

Third, "36-bit word-addressable machine with an 18-bit data bus" reads as a deep nonsense and the subsequent calculation seems flawed. If addresses were 36-bit, then we have had gigabytes of address space, which is obviously not the case. But the text explicitly states that the data bus is 18-bit – what is 36-bit indeed? Second time this "36-bit" appears in elusive context. But if data bus is 18-bit and we have 218 addresses, multiplication gives us 576 KB – exactly the half of 1.125 MiB mentioned. May be data are nevertheless 36-bit? But is this case referring to "18-bit data bus" is paradoxical and worthless.

Incnis Mrsi (talk) 21:05, 12 January 2012 (UTC)

The address bus has 18 bits, and it addresses 36-bit words. I updated the article with calculations. Sorry for delayed response, I'm not closely monitoring this article, I just check in from time to time. This whole article remains original research. Sometime I may try to update with some citations if I can find them. Feel free to ask for further clarification if you still have questions. Thanks Wbm1058 (talk) 17:59, 20 January 2012 (UTC)
My god. The recent version says:
… a 36-bit word-addressable machine with an 18-bit data bus addresses only 218 (262,144) 36-bit locations …
What is the sense and purpose of "18-bit data bus" in this context if "locations" are 36-bit?! Incnis Mrsi (talk) 18:19, 20 January 2012 (UTC)
sorry, I should have proof-read it better. It should have said address bus. Wbm1058 (talk) 18:38, 20 January 2012 (UTC)
it appears that error was introduced with this edit Wbm1058 (talk) 18:47, 20 January 2012 (UTC)
Thank you. Happily these gaps are filled now. Incnis Mrsi (talk) 20:29, 20 January 2012 (UTC)

"Each memory location in a stored-program computer holds a binary number of some sort."[edit]

Well, of some sort covers a lot of ground, but it is not an adequate description for decimal machines such as the IBM 650 where a memory location held 10 bi-quinary coded decimal digits. While individual digits might be "a binary number of some sort", the 10 digit memory location is base 10, not 2; a decimal number irrespective of how the digits were encoded.50.136.247.190 (talk) 02:30, 31 October 2013 (UTC)