[hatari-devel] strange DSP behaviour?

[Douglas Little <doug694@xxxxxxxxxxxxxx>]

I noticed this strange effect on Sunday, probably one for LaurentS :-)

MACR seems to be selectively rounding the accumulator - or not rounding - depending on the weather. Or possibly it is never rounding.

The only external influence on MACR (and other rounding ops) are the global scaling bits and these just affect the constant used to perform rounding. The zeroing of the lower 24bits of the destination accumulator is unconditional - it should always occur. But it didn't happen in the example traced below?

> dr
A: A2: 00  A1: 0055fc  A0: 00ffa6
B: B2: 00  B1: 000fc0  B0: 000000
X: X1: 002000  X0: 080000
Y: Y1: 000000  Y0: 0055fc
R0: ffeb   N0: 0000   M0: ffff
R1: 0000   N1: 29a2   M1: 0002
R2: 0001   N2: 0fc6   M2: ffff
R3: 004e   N3: 0001   M3: ffff
R4: 0000   N4: 006c   M4: ffff
R5: 29a9   N5: eafe   M5: ffff
R6: 151a   N6: ffff   M6: ffff
R7: 49a2   N7: 2400   M7: ffff
LA: 0093   LC: 0001   PC: 008f
SR: 8050  OMR: 02
SP: 05    SSH: 008a  SSL: 8040

> ds
p:008f  4dc7eb         (02 cyc)  macr +x1,y0,b y:(r7)-n7,x1

> dr
A: A2: 00  A1: 0055fc  A0: 00ffa6
B: B2: 00  B1: 000fd5  B0: 7f0000
X: X1: 225100  X0: 080000
Y: Y1: 000000  Y0: 0055fc
R0: ffeb   N0: 0000   M0: ffff
R1: 0000   N1: 29a2   M1: 0002
R2: 0001   N2: 0fc6   M2: ffff
R3: 004e   N3: 0001   M3: ffff
R4: 0000   N4: 006c   M4: ffff
R5: 29a9   N5: eafe   M5: ffff
R6: 151a   N6: ffff   M6: ffff
R7: 25a2   N7: 2400   M7: ffff
LA: 0093   LC: 0001   PC: 0090
SR: 8050  OMR: 02
SP: 05    SSH: 008a  SSL: 8040

I haven't had time to investigate this more closely but this was a direct c&p from the Hatari debugger and seems conclusive - something wrong?

See the DSP user manual, page A-114 for a description of MACR behaviour and the zeroing of LSW after rounding.

http://bitsavers.informatik.uni-stuttgart.de/pdf/motorola/56000/1990_DSP56000_DSP56001_Users_Manual.pdf