Definitions

discussion

Add the following Definitions to Section 1.3.1 of Book I

Definition of Simple-V:

In its simplest form, the Simple-V Loop/Vector concept is a Prefixing system (similar to the 8086 REP instruction and the Z80 LDIR) that both augments its following Defined Word-instruction (Suffix), and also may repeat that instruction with optional sequential register offsets from those given in the Suffix. Register numbers may also be extended (larger register files). More advanced features add predication, element-width overrides, and Vertical-First Mode.

Definition of SVP64 Prefixing:

SVP64 is a well-defined implementation of the Simple-V Loop/Vector concept, in a 32-bit Prefix format, that exploits the following instruction (the Defined Word-instruction) using it as a "template". It requires 24 bits, some of which are common to all Suffixes, and some Mode bits are specific to the Defined Word-instruction class: Load/Store-Immediate, Load/Store-Indexed, Arithmetic/Logical, Condition Register operations, and Branch-Conditional. Anything not falling into those five categories is termed "Unvectorizable".

Definition of Horizontal-First:

Normal Cray-style Vectorization, designated Horizontal-First, performs element-level operations (often in parallel) before moving in the usual fashion to the next instruction. The term "Horizontal-First" stems from naturally visually listing program instructions vertically, and register file contents horizontally, whereupon it is clear that register-elements are prioritised.

Definition of Vertical-First:

Vertical-First executes one element operation only then moves on to the next instruction, whereupon if that is also an SVP64-Prefixed instruction the exact same element offset is used. Element offsets are then explicitly advanced by calling a special instruction, svstep. The term "Vertical-First" stems from naturally visually listing program instructions vertically and register file contents horizontally, where moving to the next instruction is a clear priority.

Definition of SVP64Single Prefixing:

A 32-bit Prefix in front of a Defined Word-instruction that extends register numbers (allows larger register files), adds single-bit predication, element-width overrides, and optionally adds Saturation to Arithmetic instructions that normally would not have it. SVP64Single is in Draft only and is yet to be defined.

Definition of "Unvectorizable":

Any operation that inherently makes no sense if repeated (through SVP64 Prefixing) is termed "Unvectorizable". Examples include sc or sync which have no registers. mtmsr is also classed as Unvectorizable because there is only one MSR. Also instructions that simply may not be Prefixed (EXT300-EXT363) are also deemed "Unvectorizable".

Unvectorizable instructions are required to be detected as such if Prefixed (either SVP64 or SVP64Single) and an Illegal Instruction Trap raised.

Hardware Architectural Note: Given that a "pre-classification" Decode Phase is required (identifying whether the Suffix - Defined Word-instruction - is Arithmetic/Logical, CR-op, Load/Store or Branch-Conditional), adding "Unvectorizable" detection to this phase is not unreasonable.

New Prefixed Instruction Encoding space

Proposal: Add new Section 1.6.5 to Book I

The following eight new RESERVED areas are defined within Primary Opcode 9 (EXT009)

Key:

• x - a RESERVED encoding. Illegal Instruction Trap must be raised
• n - a future specification-defined value (currently RESERVED)
• !PO9 - any 6-bit binary value except Primary Opcode 9 (0b010001)
• !ZERO - a non-zero future specification-defined value (currently RESERVED)
• DWd - a "Defined Word-instruction" - Book I Section 1.6 (Public v3.1 p11)
• SVRM - a RESERVED encoding
• SVP64Single: a future RESERVED Prefix encoding
• SVP64: a future RESERVED Loop-Prefix encoding
• EXT200-263: a RESERVED encoding for future Scalar instructions (Vectorizable)
• EXT300-363: a RESERVED encoding for future Scalar instructions (Unvectorizable)

Architectural Resource Allocation Note: Similar to ARM's MOVPRFX instruction and the original x86 REP instruction, despite "influence" over the Suffix, the Suffix is entirely independent of the Prefix. Therefore under no circumstances must different Defined Word-instructions (different from the same Un-Prefixed Defined Word-instruction) be allocated within any EXT{zNN} prefixed or unprefixed space for a given value of z of 0, 2 or 3: the results would be catastrophic. Even if Unvectorizable an instruction Defined Word-instruction space must have the exact same Instruction and exact same Instruction Encoding in all spaces being RESERVED (Illegal Instruction Trap if Unvectorizable) or not be allocated at all. This is required as an inviolate hard rule governing Primary Opcode 9 that may not be revoked under any circumstances. A useful way to think of this is that the Prefix Encoding is, like the 8086 REP instruction, an independent 32-bit Defined Word-instruction.

Note a particular consequence of the application of the above paragraph: due to the fact that the Prefix Encodings are independent, by definition two new Sandbox areas "come into being" in an inviolate manner (i.e. they may not be called anything else, nor may they be revoked rescinded removed or recalled), named SVP64:EXT022 and SVP64Single:EXT022. The only way that these two new areas may be revoked is if EXT022 itself is revoked. All and any re-definitions modifications enhancements clarifications that apply to EXT022 also apply to these two new areas because due to the Prefixes being independent Defined Word-instructions the three areas are actually one and the same area, just as all Scalar Defined Word-instructions are.

Notes:

• PO9-PO1 Prefixed-Prefixed (96-bit) instructions are prohibited. EXT1xx is thus inherently Unvectorizable as the complexity at the Decoder of recognising {PO1-word}{PO9-word}{Defined-word-instruction} becomes too great for High Performance Multi-Issue systems.
• There is however no reason why PO9-PO1 (EXT901?) as an entirely new RESERVED 64-bit Encoding should not be permitted as long as it is clearly marked as Unvectorizable.
• PO1-PO9 Prefixed-Prefixed (96-bit) instructions are also prohibited for the same reason: Multi-Issue Decode complexity is too great.
• There is however no reason why PO1-PO9 (EXT109) as an entirely new RESERVED 64-bit Encoding should not be permitted as long as it is clearly marked as Unvectorizable.
• EXT100-163 instructions (PO1-Prefixed) are also prohibited from being double-PO1-prefixed (not twice prefixed)
• Considerable care is needed both on Architectural Resource Allocation as well as instruction design itself. All new Scalar instructions automatically and inherently must be designed taking their Vectorizeable potential into consideration including VSX in future.
• Once an instruction is allocated in an Unvectorizable area it can never be Vectorized without providing an entirely new Encoding.

\newpage{}