Subject: Re: soft-float builds
To: Simon Burge <email@example.com>
From: Jason R Thorpe <firstname.lastname@example.org>
Date: 05/30/2001 20:03:03
On Thu, May 31, 2001 at 12:10:38PM +1000, Simon Burge wrote:
> > Yes, esp. for e.g. PowerPC, where it changes the ABI for e.g. stdarg. Do
> > you have kernel FP emulation for FPU-less PowerPCs? If so, then you can
> > still run FPU-using PowerPC code by emulating the register accesses, but
> > that'll be slow...
> We have kernel FP emulation, but yes, it's not quite as fast :) Early
> userland work was done with macppc binaries.
Ah, cool. Y'know, all the kernel FP emulation should be done with the
softfloat lib that's now in libkern (Ross added it for FP completion on
the Alpha port).
> > Actually, are there ELF header bits that indiciate "uses FPU"? I know
> > there are for MIPS...
> Do you have a reference for that? We don't seem to have any flags in
> our source tree for those (that I can see)...
Hm, maybe it's in the MIPS supplemental document -- which I have somewhere
upstairs buried in a lot of other books -- I can look later -- but file(1)
can recognize it, too... it's in the e_flags word, I believe. You should
check the PowerPC processor supplement.
Anyway, in that case, you can just force the correct behavior with the
linker -- "Don't let softfloat objects link against non-softfloat objects".
-- Jason R. Thorpe <email@example.com>