Re: [hatari-devel] Re: FPU update

[Douglas Little <doug694@xxxxxxxxxxxxxx>]

Hi,

I can check the version of VASM I was using - probably 1.6c or 1.7. It should assemble without hangs though :)

The test really just hashes binary results upon writing final calcs back to ram - to ensure sensitivity to small errors/differences. Lots of calculations are repeated with different kinds of value ranges, including special values such as NaN, denorm values etc. It won't show you the magnitude of the error - but will give clues about which instructions may misbehave.

I can help explain specific case if it helps.

Doug.

On 21 January 2017 at 16:30, Toni Wilen <twilen@xxxxxxxxxx> wrote:

> I also created a simple 6888x conformance/correctness test recently,
> which was probably posted here a while back. Might be interesting to run
> the softfloat impl. against that sometime.

I tried to compile it (without any message/file handling) without
success.. (vasm for example simply hangs, others give unexplained errors)

I succeeded by running the binary version under Amiga EmuTOS :)

Attached is screenshot that shows errors, what do they mean? (Is it
correct or incorrect value or something else?)

Used softfloat mode, I got much more errors with native mode, probably
due to not having long doubles)