This is the mail archive of the cygwin@sourceware.cygnus.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Uname -m and arch


On Thu, 29 May 1997, Geoffrey Noer wrote:

> > Under Linux, I would get i586. Is this as it should be? If so, what is
> > the meaning of 6395286?
> Either you have an Intel i6395286 chip in your machine or you're running
> into a cygwin.dll bug.  I have a hunch as to which is more likely.  :-)

Indeed.  Unless Intel's CPU design group has been -really- busy... :)

I noticed this problem under b17 as well, and reported it, but I suspect
it got lost in all the rush.  My solution for the moment was to write my
own 'arch' (for autoconf) and ignore the uname problem since it didn't
-break- anything. 

> What's happening is that uname() isn't setting the processor level
> correctly.  uname() gets some of its info using the SYSTEM_INFO struct.
> This structure has various members many of which aren't supported under
> both Windows 95 and NT.  :-(  I've fixed the development sources so uname
> will behave better under Windows 95 in future releases.

Perhaps a way to set this if you have a specific need, and default to i486
or something otherwise?  Unless you have a way to determine the chipset
accurately. :)  Only reason I suggest a way to set it is if you have
something which depends on 386, 486, or 586 compiler optimizations...

-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]