This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: "make check" problems on MIPS
- From: Hartvig Ekner <hartvige at mips dot com>
- To: schwab at suse dot de (Andreas Schwab)
- Cc: hartvige at mips dot com (Hartvig Ekner), libc-alpha at sources dot redhat dot com
- Date: Tue, 9 Apr 2002 22:30:33 +0200 (CEST)
- Subject: Re: "make check" problems on MIPS
Thanks for the pointer - the reason turned out to be very much related:
the "reverse" definition of QNaN and SNaN on MIPS. Due to this, the
casin() function returned SNaN instead of the correct QNaN, which caused
the unexpected invalid operation exception in cacos(), when doing the
final subtractions.
"make check" is now passing on MIPS in its entirety (not only math
tests) on my system (2.4.18 with modified FPU emulator, various patches
to glibc, toolchain from HJL's RedHat 7.1/MIPS miniport, with SDE 5.0
beta used for the math functions compiled wrongly by gcc in the
miniport).
/Hartvig
Andreas Schwab writes:
>
> Hartvig Ekner <hartvige@mips.com> writes:
>
> |> 2) Why does this pass on x86?
>
> I don't know about x86, but the m68k fpu does not raise the invalid
> operation exception when operating on (non-signaling) NANs.
>
> Andreas.
>
> --
> Andreas Schwab, SuSE Labs, schwab@suse.de
> SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
> Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
> "And now for something completely different."
>
--
_ _ _____ ____ Hartvig Ekner Mailto:hartvige@mips.com
|\ /| | |____)(____ Direct: +45 4486 5503
| \/ | | | _____) MIPS Denmark Switch: +45 4486 5555
T E C H N O L O G I E S http://www.mips.com Fax...: +45 4486 5556