skills.rest
z88dk
z88dkOFFICIAL
Explore all skills by this author →
Website𝕏@z88dkGitHub

library-am9511

Guides editing and rebuilding the Am9511A APU float library in z88dk.

1.1k|206|Updated Mar 16, 2016
Search Tags:#embedded#assembly#floating point#apu#z88dk#z80#am9511
One-click install
npx skills add https://github.com/z88dk/z88dk --skill library-am9511-z88dk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: library-am9511
Source: https://github.com/z88dk/z88dk/tree/main/.agents/skills/library-am9511
Command: npx skills add https://github.com/z88dk/z88dk --skill library-am9511-z88dk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Working on z88dk's Am9511A (Intel 8231A) hardware APU float library requires deep knowledge of IEEE-754 vs APU float formats, pushf/popf status mapping, and multi-product rebuild steps; mistakes cause Inf/NaN bugs, stack misalignment, or broken library builds. ## Core Features & Use Cases - APU interface reference: Documents push/pop byte order, command byte format, status register error codes, and I/O macros for z80 and 8085 targets. - IEEE specials handling: Explains how am32_fspecial.asm implements Inf/NaN algebra in software since the APU chip has no native Inf/NaN encodings. - Link flags and rebuild workflow: Covers --math-am9511, --am9511, --am9511h flags for classic and newlib, plus safe full-product rebuild commands that preserve ftoa/ftoe in the .lib. - Use Case: When debugging a divide-by-zero returning wrong results on an Am9511A target, use this Skill to trace popf status mapping (DIV0 → ±Inf) and verify the link flags pull in the APU path instead of genmath's sqrt. ## Quick Start Ask the AI to explain how popf maps Am9511A status register error codes to IEEE special values when editing libsrc/math/float/am9511.

Frequently Asked Questions about library-am9511

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I link the Am9511 APU float library in z88dk?▼

Use --math-am9511 or --am9511 for classic z80 targets with the APU at the configured port map. For 8085 use --math-am9511 with -clib=8085; newlib uses --am9511 or -lam9511 with IEEE float mode on the z80 clib only.

How does the Am9511A handle Inf and NaN values?▼

The Am9511A chip has no native Inf or NaN encodings. The library implements IEEE specials in software via am32_fspecial.asm before pushing operands, and popf maps APU status errors to IEEE values: overflow to ±Inf, underflow to 0, and negative sqrt to NaN.

What is the push and pop byte order for the Am9511A?▼

Push operands LSB first with MSB last, and pop results MSB first with LSB last. A wrong byte count or order misaligns the internal byte pointer because the APU stack is a LIFO with wraparound.

Why does my Am9511 library build lose ftoa after rebuilding?▼

Wiping the obj/ directory without re-running the full Makefile drops the cimpl ftoa/ftoe objects from the .lib. Run a full make in libsrc/math/float/am9511 after removing only specific object files, then copy the products into lib/clibs/.

Does z88dk-ticks emulate the Am9511A APU?▼

Yes, z88dk-ticks includes an Am9511 model in src/ticks/am9511.c using default ports 0x42 and 0x43. The model is approximate and not bit-exact silicon, so status-bit accuracy may differ from real hardware.

Am9511A vs math32 software float, which should I use?▼

The Am9511A is a hardware APU with a smaller range (about ±2.7e-20 to 9.2e18) and no Inf/NaN, while math32 is a full software IEEE-754 implementation. Compare both with TIMER benchmarks using --math-am9511 versus the math32 library.

Related Skills

Explore More Software Engineering Skills→
matiaszanollimatiaszanolli
6

audit-nes-hardware

Validate NES APU audio channel and register configurations against hardware documentation.

Community
Advanced
yfyang86yfyang86
1

william-kahan-perspective

Analyze numerical computation problems using IEEE-754 floating-point principles.

Community
Intermediate
openeuler-mirroropeneuler-mirror
3

linux-oom-analyzer

Correlate kernel logs, process memory snapshots, and cgroup metrics to diagnose Linux OOM events.

Official
Advanced
oci-ai-architectsoci-ai-architects
1

frank-method

Plan multi-agent systems using the AGENT Blueprint architecture.

Official
Advanced