- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I notice that there are differences between win32 and em64t ivf modules. For example, if ctxptr is of type T_CONTEXT,ctxptr%FLOATSAVE%STATUSWORD is valid in win32but not so in em64t. How does one get around this? and are these differences docummented?
Gerry
Link Copied
5 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
It looks to me as if Microsoft defines this structure differently on the two architectures, which I do not find surprising. Anything you're doing in this area is architecture-specific, so you need to conditionalize appropriately.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
So much for single code base propaganda. I takeit that the differences between the ivf win32 and em64t modules haven't been documented.
Gerry
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Eh? I assume Microsoft documents this - it is their data structure.
As for single code base, when you start writing code that is explicit to an architecture, you'll run into issues. A normal Fortran program, properly written, should not see an architecture dependency.
As for single code base, when you start writing code that is explicit to an architecture, you'll run into issues. A normal Fortran program, properly written, should not see an architecture dependency.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
MADsblionel:Eh? I assume Microsoft documents this - it is their data structure.
My query relates to Intel modules providing interfaces to Microsoft API. The IVF modules are a mess compared to the Microsoft header files.
MADsblionel:As for single code base, when you start writing code that is explicit to an architecture, you'll run into issues. A normal Fortran program, properly written, should not see an architecture dependency.
Yes I agree, IVF is not for writing normal Fortran programs for commodity chips, eh!
Gerry
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Gerry,
If you have identified an incompatibility between definitions in the Win32 modules and those provided by MSVC, please elaborate and we'll be glad to take a look. I am "on the road" this week and can't study this issue in depth right now, but from what I can see the MSVC definitions also differ by architecture.
I am somewhat puzzled at your assumption that code designed to read hardware registers can be made to be platform independent. Please enlighten me.
If you have identified an incompatibility between definitions in the Win32 modules and those provided by MSVC, please elaborate and we'll be glad to take a look. I am "on the road" this week and can't study this issue in depth right now, but from what I can see the MSVC definitions also differ by architecture.
I am somewhat puzzled at your assumption that code designed to read hardware registers can be made to be platform independent. Please enlighten me.
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page