From: Luke Kenneth Casson Leighton Date: Fri, 22 Jul 2022 14:15:34 +0000 (+0100) Subject: shuffle table X-Git-Tag: opf_rfc_ls005_v1~1138 X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=de6c3043a4c8de5073770091672d9e31e8b19e95;p=libreriscv.git shuffle table --- diff --git a/openpower/sv/comparison_table.mdwn b/openpower/sv/comparison_table.mdwn index c64dc9670..0d262cfaa 100644 --- a/openpower/sv/comparison_table.mdwn +++ b/openpower/sv/comparison_table.mdwn @@ -1,31 +1,31 @@ # ISA Comparison Table -| Name | Num
opcodes | Class | Predicate
Masks | Twin
Predication | Explicit
Vector regs | 128-bit | Bigint
capability | LDST
Fault-First | Data-dependent
Fail-first | Predicate-
Result | -|-------------|-------------------|---------------|------------------------|-------------------------|------------------------------|---------|--------------------------|-------------------------|----------------------------------|-------------------------| -| SVP64 | 5(1) | Scalable(2) | yes | yes(3) | no(4) | n/a(5) | yes(6) | yes(7) | yes(8) | yes(9) | -| VSX | 700+ | PackedSIMD | no | no | yes(10) | yes | no | no | no | no | -| NEON | ~250(11) | PredicatedSIMD| yes | no | yes | yes | no | no | no | no | -| SVE2 | ~1000(12) | HSCalable(13) | yes | no | yes | yes | no | yes(7) | no | no | -| AVX-512(14) | ~1000s(15) | PredicatedSIMD| yes | no | yes | yes | no | no | no | no | -| RVV(16) | ~190 | Scalable(17) | yes | no | yes | yes(18) | no | yes | no | no | +| Name | Num
opcodes | Class | Predicate
Masks | Twin
Predication | Explicit
Vector regs | 128-bit | Bigint
capability | LDST
Fault-First | Data-dependent
Fail-first | Predicate-
Result | +|--------------|-------------------|-----------------|------------------------|-------------------------|------------------------------|---------|--------------------------|-------------------------|----------------------------------|-------------------------| +| SVP64 | 5 {1} | Scalable {2} | yes | yes {3} | no {4} | n/a {5} | yes {6} | yes {7} | yes {8} | yes {9} | +| VSX | 700+ | PackedSIMD | no | no | yes {10} | yes | no | no | no | no | +| NEON | ~250 {11} | PredicatedSIMD | yes | no | yes | yes | no | no | no | no | +| SVE2 | ~1000 {12} | HWSCalable {13} | yes | no | yes | yes | no | yes {7} | no | no | +| AVX-512 {14} | ~1000s {15} | PredicatedSIMD | yes | no | yes | yes | no | no | no | no | +| RVV {16} | ~190 | Scalable {17} | yes | no | yes | yes {18}| no | yes | no | no | -* (1): plus EXT001 24-bit prefixing. See [[sv/svp64]] -* (2): A 2-Dimensional Scalable Vector ISA with both Horizontal-First and Vertical-First Modes. See [[sv/vector_isa_comparison]] -* (3): on specific operations. See [[opcode_regs_deduped]] for full list -* (4): SVP64 provides the Vector register concept on top of the **Scalar** GPR, FPR and CR register files. -* (5): SVP64 Vectorises Scalar instructions. If (**optionally**) applied to e.g. VSX Quad-Precision instructions, SVP64 "becomes" 128-bit. -* (6): big-integer add is just `sv.adde`. Mul and divide require addition of two scalar operations -* (7) See [[sv/svp64/appendix]] and [ARM SVE Fault-First](https://alastairreid.github.io/papers/sve-ieee-micro-2017.pdf) -* (8) Based on LD/ST Fail-first, extended to data. See [[sv/svp64/appendix]] -* (9) Turns standard ops into a type of "cmp". See [[sv/svp64/appendix]] -* (10) VSX's Vector Registers are mis-named: they are PackedSIMD. -* (11) difficult to ascertain, see [NEON/VFP](https://developer.arm.com/documentation/den0018/a/NEON-and-VFP-Instruction-Summary/List-of-all-NEON-and-VFP-instructions). +* {1}: plus EXT001 24-bit prefixing. See [[sv/svp64]] +* {2}: A 2-Dimensional Scalable Vector ISA with both Horizontal-First and Vertical-First Modes. See [[sv/vector_isa_comparison]] +* {3}: on specific operations. See [[opcode_regs_deduped]] for full list +* {4}: SVP64 provides the Vector register concept on top of the **Scalar** GPR, FPR and CR register files. +* {5}: SVP64 Vectorises Scalar instructions. It is up to the **implementor** to choose (**optionally**) whether to apply SVP64 to e.g. VSX Quad-Precision (128-bit) instructions. +* {6}: big-integer add is just `sv.adde`. Mul and divide require addition of two scalar operations +* {7} See [[sv/svp64/appendix]] and [ARM SVE Fault-First](https://alastairreid.github.io/papers/sve-ieee-micro-2017.pdf) +* {8} Based on LD/ST Fail-first, extended to data. See [[sv/svp64/appendix]] +* {9} Turns standard ops into a type of "cmp". See [[sv/svp64/appendix]] +* {10} VSX's Vector Registers are mis-named: they are PackedSIMD. +* {11} difficult to ascertain, see [NEON/VFP](https://developer.arm.com/documentation/den0018/a/NEON-and-VFP-Instruction-Summary/List-of-all-NEON-and-VFP-instructions). Critically depends on ARM Scalar instructions -* (12) difficult to exactly ascertain, see ARM Architecture Reference Manual Supplement, DDI 0584. Critically depends on ARM Scalar instructions. -* (13): ARM states that the Scalability is a [Silicon-partner choice](https://developer.arm.com/-/media/Arm%20Developer%20Community/PDF/102340_0001_00_en_introduction-to-sve2.pdf?revision=aae96dd2-5334-4ad3-9a47-393086a20fea). +* {12} difficult to exactly ascertain, see ARM Architecture Reference Manual Supplement, DDI 0584. Critically depends on ARM Scalar instructions. +* {13}: ARM states that the Scalability is a [Silicon-partner choice](https://developer.arm.com/-/media/Arm%20Developer%20Community/PDF/102340_0001_00_en_introduction-to-sve2.pdf?revision=aae96dd2-5334-4ad3-9a47-393086a20fea). this "Scalability independence" is not entirely extended in full to the programmer although ARM requests developers to consider it so, in practice this does not happen. -* (14): [Wikipedia](https://en.wikipedia.org/wiki/AVX-512), [Lifecycle of an instruction set](https://media.handmade-seattle.com/tom-forsyth/) including full slides -* (15): difficult to exactly ascertain, contains subsets. Critically depends on ISA support from earlier x86 ISA subsets (several more thousand instructions). See [SIMD ISA listing](https://www.officedaytime.com/simd512e/) -* (16): [RVV Spec](https://github.com/riscv/riscv-v-spec/blob/master/v-spec.adoc) -* (17): Like the original Cray RVV is a truly scalable Vector ISA (Cray setvl instruction). -* (18): like SVP64 it is up to the hardware implementor to choose whether to support 128-bit elements. +* {14}: [Wikipedia](https://en.wikipedia.org/wiki/AVX-512), [Lifecycle of an instruction set](https://media.handmade-seattle.com/tom-forsyth/) including full slides +* {15}: difficult to exactly ascertain, contains subsets. Critically depends on ISA support from earlier x86 ISA subsets (several more thousand instructions). See [SIMD ISA listing](https://www.officedaytime.com/simd512e/) +* {16}: [RVV Spec](https://github.com/riscv/riscv-v-spec/blob/master/v-spec.adoc) +* {17}: Like the original Cray RVV is a truly scalable Vector ISA (Cray setvl instruction). +* {18}: like SVP64 it is up to the hardware implementor to choose whether to support 128-bit elements.