 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
.NET Forum / Languages / Managed C++ / June 2006Section attribute "K"?
The docs say: K Cacheable Marks the section as not cacheable and specify /SECTION:.text,!K and DUMPBIN will reveal section characteristics, including "Not Cached." There seems to be a contradiction. Does "K" mark the section as cacheable or not cacheable?  Signature- Vince > The docs say: > [quoted text clipped - 7 lines] > There seems to be a contradiction. Does "K" mark the section as cacheable > or not cacheable? I think the '!' negates the attribute. So K would be Cacheable and !K would be NOT Cacheable. SignatureJeff Partch [VC++ MVP] >> The docs say: >> [quoted text clipped - 10 lines] >I think the '!' negates the attribute. So K would be Cacheable and !K would >be NOT Cacheable. That's what seems to be the case. So the table in the docs indicating: K ... Cacheable ... Marks the section as not cacheable is wrong, eh? Signature- Vince >>> The docs say: >>> [quoted text clipped - 18 lines] > > is wrong, eh? Oh, I'm a bonehead -- I see what you're saying now. :) Whatever its trying to mean, that 'not' has been part of the documentation since at least Oct 2001. I think they try to make it make sense with the... "K [is] peculiar in that the section flag... that correspond to [it] [is] in the negative sense. If you specify [K] on the .text section (/SECTION:.text,K), there will be no difference ... when you run DUMPBIN [because] it was already implicitly cached. To remove [it] specify ... !K..." ...but it still sounds like an explanation of a snake swallowing its tail. SignatureJeff Partch [VC++ MVP] >> That's what seems to be the case. So the table in the docs indicating: >> [quoted text clipped - 5 lines] >to mean, that 'not' has been part of the documentation since at least Oct >2001. It's what you currently see online at MSDN: MSDN Library > Development Tools and Languages > Visual Studio > Visual C++ > Building a C/C++ Program > C/C++ Building Reference > Linking > Linker Options > /SECTION (Specify Section Attributes) Signature- Vince >>> That's what seems to be the case. So the table in the docs indicating: >>> [quoted text clipped - 7 lines] > > It's what you currently see online at MSDN: I know. What I mean is it has said the same thing for ~5 years. While its not impossible that its a typo or a mistake that's gone unnoticed for half-a-decade and several linker versions, I suspect it is trying to say something subtle with intent. The fact that all that gobbledygook about K being peculiar has been added since then and now, indicates to me that somebody has already said 'Huh?', and that the doc owner approved the 'not' and then tried to make its intended meaning more clear. I'm not sure s/he succeeded, because it still sounds backwards to me too. Of course, I'm only a fry-cook and a self-confessed bonehead. :) SignatureJeff Partch [VC++ MVP] Since caching is an esoteric kernel mode thing, maybe the K attribute actually is not handled by the loader, or handled in some not straightforward way. Can the OP tell why he needs noncached memory (outside of kernel & device drivers context)? Regards, --PA Free MagazinesGet these publications absolutely FREE for up to 12 months. There are no hidden fees and no obligation. Simply choose a title, complete the application form and submit it. Read more ...
|
Reply to this Message
|
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.