Re: Problems with float in a kext
Re: Problems with float in a kext
- Subject: Re: Problems with float in a kext
- From: Heiko Panther <email@hidden>
- Date: Sun, 08 Dec 2002 23:45:36 +0100
Sean McBride wrote:
>>What's wrong here? I know that kernel code usually uses float emulation.
>>I couldn't find out which steps to take to activate the emulation. I
>>tried -msoft-float for compiling and linking, but that did not help.
>
> If you want your kext to use the FPU you must NOT specify the -msoft-
> float switch.
Right. Well, I don't really want to use the FPU. The emulation would be
good enough. Anyway, I get that error, no matter if I specify -msoft-float.
> Also note that IOLog does not support %f to print floats. :(
Must be that way to keep the kernel small :) I usually multiply by 1000
and cast to int.
Actually I just noticed that I cast to UInt32 in my example. Here's the
results when cast to int.
Dec 6 17:36:52 Hyp mach_kernel: 500, 2147483647, 2147483647, -2147483648
Heiko
_______________________________________________
darwin-kernel mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/darwin-kernel
Do not post admin requests to the list. They will be ignored.