Go to file
Brendan Punsky d40d9e687a rexcode/arm64: model SME's ZA tiles and tile slices
A ZA tile printed as the bare number it is encoded as -- `addha #0,
p0/m, p0/m, z0.s` -- and a tile slice printed as one too, where the
syntax is `za0h.b[w12, 0]`: a tile, taken along its rows or columns,
addressed by one of W12..W15 plus an offset. Neither was anything an
assembler would take.

Tiles are a register class now (ZA0..ZA15, viewed at an element size),
so they print through the same path as every other register. A slice is
its own operand kind holding the four things it is made of, rather than
one packed immediate that only the encoder understood.

Getting that right needed the field layout, and the layout is not what
the encoding table implied: the tile number and the offset share the low
nibble, and how it splits follows the element size -- a byte tile has no
tile bits at all and four of offset, while a quadword tile is all tile
and none. Reading a fixed four bits as the tile made every byte slice
come back as tile 4.

LD1Q/ST1Q scale their index by 16, which the mnemonic's last letter
does not spell the way B/H/W/D do; they were left unscaled when the
other 38 forms were fixed. That also wanted a .q element shape, which
nothing had needed before.

SVE/SME2 decode entries against llvm-mc: 559 byte-exact and 0
mismatched, from 211 and 78 at the start of the session.

What is left is mostly one structural limit: SME's outer products take
five operands (`umopa za0.s, p0/m, p1/m, z0.b, z1.b`) and Instruction
holds four. Widening it would fit -- five Operands is 55 bytes of the 64
-- but it reaches through Encoding, the table, both codecs and the
builders, so it wants its own change.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018UmHLRF11EoWwNWCJ7JGaA
2026-08-28 16:58:33 -04:00
2025-12-11 11:11:53 +00:00
2025-10-28 13:26:56 +00:00
2018-12-27 10:51:15 +00:00
2026-08-27 15:06:35 +01:00
2026-08-09 20:23:34 +02:00
2022-11-04 11:40:07 +00:00
2026-08-12 16:17:49 +01:00
2025-09-26 12:05:16 +02:00
2025-03-18 15:39:18 +00:00
2026-05-17 13:18:48 +01:00
2020-11-03 07:40:17 -03:00
2025-04-09 07:46:44 +11:00
2025-03-28 18:38:08 +01:00

Odin logo
The Data-Oriented Language for Sane Software Development.


The Odin Programming Language

Odin is a general-purpose programming language with distinct typing, built for high performance, modern systems, and built-in data-oriented data types. The Odin Programming Language, the C alternative for the joy of programming.

Website: https://odin-lang.org/

package main

import "core:fmt"

main :: proc() {
	program := "+ + * 😃 - /"
	accumulator := 0

	for token in program {
		switch token {
		case '+': accumulator += 1
		case '-': accumulator -= 1
		case '*': accumulator *= 2
		case '/': accumulator /= 2
		case '😃': accumulator *= accumulator
		case: // Ignore everything else
		}
	}

	fmt.printf("The program \"%s\" calculates the value %d\n",
	           program, accumulator)
}

Documentation

Getting Started

Instructions for downloading and installing the Odin compiler and libraries.

Nightly Builds

Get the latest nightly builds of Odin.

Learning Odin

Overview of Odin

An overview of the Odin programming language.

Frequently Asked Questions (FAQ)

Answers to common questions about Odin.

Packages

Documentation for all the official packages part of the core and vendor library collections.

Examples

Examples on how to write idiomatic Odin code. Shows how to accomplish specific tasks in Odin, as well as how to use packages from core and vendor.

Odin Documentation

Documentation for the Odin language itself.

Odin Discord

Get live support and talk with other Odin programmers on the Odin Discord.

Articles

The Odin Blog

The official blog of the Odin programming language, featuring announcements, news, and in-depth articles by the Odin team and guests.

Warnings

  • The Odin compiler is still in development.
Languages
Odin 82.3%
C++ 12.3%
C 4.6%
Python 0.3%
Lua 0.2%
Other 0.2%