# TPU v6e TEC 指令索引

由 libtpu 0.0.49 的 descriptor、encoder、LLVM TPU printer 与 TEC emitter 生成：登记 **500 种形式、965 个槽与形式组合、2336 条展开签名、1054 个槽与助记符组合**。encoder 拒绝的 125 个槽与形式组合不登记。这里的覆盖是编码、解码与重汇编覆盖，没有在设备上执行。

## 如何阅读

`s[0..31]`、`v[0..63]` 表示寄存器编号，`/` 表示可选的拼写，`iN/uN/f32` 是数值类型，其余字符照写。操作数顺序与 LLVM TPU printer 相同，目的在前。数值类型只限定输入位宽，立即数槽和整个指令包的资源是否够用由汇编器检查。写法见[格式参考](tpu_v6e_tec.md)。

每个形式另有具名字段写法：助记符是 descriptor 中的形式名，操作数是按字段顺序排列的具名参数，用于 printer 写法表达不了的编码。表中只对没有 printer 写法的槽与形式组合列出这种写法，涉及以下形式：`atomic_tile_add_set_done`、`atomic_tile_add_set_done_inverted`、`call_relative`、`call_sreg`、`descriptor_based_dma`、`eup_push`、`general_dma`、`indirect_vreg_stream`、`max_index_scan_bf16`、`max_index_scan_u16`、`min_index_scan_bf16`、`pop_rcmf`、`pop_xrf_write_partial_0`、`pop_xrf_write_partial_1`、`pop_xrf_write_partial_2`、`pop_xrf_write_partial_3`、`pop_xrf_write_partial_4`、`push_scmf`、`read_sync_state_done`、`read_sync_state_value`、`segmented_min_index_scan_bf16`、`segmented_min_scan_bf16`、`single_strided_dma`、`tile_spmem_store_indexed_circular_buffer_return_value_add_bf16`、`tile_spmem_store_indexed_circular_buffer_return_value_add_f32`、`tile_spmem_store_indexed_circular_buffer_return_value_add_s16`、`tile_spmem_store_indexed_circular_buffer_return_value_add_s32`、`unpack_interleaved_compressed_lower_s8_to_bf16`、`unpack_interleaved_compressed_lower_s8_to_s16`、`unpack_interleaved_compressed_lower_u8_to_bf16`、`unpack_interleaved_compressed_lower_u8_to_u16`、`unpack_interleaved_compressed_upper_s8_to_bf16`、`unpack_interleaved_compressed_upper_s8_to_s16`、`unpack_interleaved_compressed_upper_u8_to_bf16`、`unpack_interleaved_compressed_upper_u8_to_u16`、`vector_add_high_f32_bf16`、`vector_add_low_f32_bf16`、`vector_clamp_asymmetric_bf16`、`vector_clamp_asymmetric_f32`、`vector_convert_bf16_to_s4`、`vector_convert_bf16_to_s8`、`vector_convert_bf16_to_u4`、`vector_convert_bf16_to_u8`、`vector_convert_f32_to_bf16`、`vector_convert_s4_to_bf16`、`vector_convert_s8_to_bf16`、`vector_convert_u4_to_bf16`、`vector_convert_u8_to_bf16`、`vector_gt_u16`、`vector_gte_u16`、`vector_lane_left_shift_insert_b16`、`vector_lt_u16`、`vector_lte_u16`、`vector_remap_bf16`、`vector_remap_f32`、`yieldable_sync_done`、`yieldable_sync_equal`、`yieldable_sync_equal_or_done`、`yieldable_sync_greater`、`yieldable_sync_greater_or_done`、`yieldable_sync_greater_or_equal`、`yieldable_sync_greater_or_equal_or_done`、`yieldable_sync_less`、`yieldable_sync_less_or_done`、`yieldable_sync_not_done`、`yieldable_sync_not_equal`、`yieldable_sync_not_equal_or_done`。

## 按槽查找

[s0](#s0) · [s1](#s1) · [dma](#dma) · [misc](#misc) · [va0](#va0) · [va1](#va1) · [va2](#va2) · [vld](#vld) · [vst](#vst) · [stream](#stream) · [vr](#vr) · [vx](#vx)

## s0

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `call_relative` | `dest=(u5)` |
| `call_sreg` | `dest=(u5), x=(u5)` |
| `pop_rcmf` | `dest=(u5)` |
| `por` | `p[0..15], p[0..31], p[0..31]`<br>`p[0..15], @first_id_occurrence, p[0..31]`<br>`p[0..15], p[0..31], @first_id_occurrence` |
| `sadd.ov.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sadd.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sand.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sbr.abs` | `label / u20` |
| `sbr.abs.clr` | `u20` |
| `sbr.ind` | `s[0..31]` |
| `sbr.rel` | `label / i20` |
| `sc.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `scall.abs` | `lr, label / u20` |
| `sceil.f32` | `s[0..31], s[0..31]` |
| `sclz.u32` | `s[0..31], s[0..31]` |
| `scvt.f32.s32` | `s[0..31], s[0..31]`<br>`s[0..31], f32` |
| `scvt.s32.f32` | `s[0..31], s[0..31]`<br>`s[0..31], f32` |
| `sdelay` | `u11` |
| `sdiv.u32` | `drf, s[0..31], u32`<br>`drf, s[0..31], s[0..31]` |
| `sdivrem.u32` | `drf, s[0..31], u32`<br>`drf, s[0..31], s[0..31]` |
| `seq.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `seq.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sfence` | `无操作数` |
| `sfence.scmf` | `无操作数` |
| `sfence.sel` | `u32`<br>`s[0..31]` |
| `sfence.stream.hbm` | `无操作数` |
| `sfence.stream.spmem` | `无操作数` |
| `sfloor.f32` | `s[0..31], s[0..31]` |
| `sge.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sge.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sge.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sgt.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sgt.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sgt.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `shalt` | `无操作数` |
| `simm.f32` | `s[0..31], f32` |
| `simm.s32` | `s[0..31], u32` |
| `sle.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sle.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sle.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `slt.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `slt.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `slt.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `smax.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], s[0..31], f32` |
| `smax.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `smin.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], s[0..31], f32` |
| `smin.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `smov.u32` | `s[0..31], s[0..31]` |
| `smul.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], f32, s[0..31]` |
| `smul.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `smulhi.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sne.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sne.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sor.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `spop` | `s[0..31], drf`<br>`s[0..31], v2sf` |
| `srdreg.ddr` | `s[0..31]` |
| `srdreg.fsr` | `s[0..31]` |
| `srdreg.gtchi` | `s[0..31]` |
| `srdreg.gtclo` | `s[0..31]` |
| `srdreg.lcchi` | `s[0..31]` |
| `srdreg.lcclo` | `s[0..31]` |
| `srdreg.scid` | `s[0..31]` |
| `srdreg.tag` | `s[0..31]` |
| `srdreg.tbm` | `s[0..31]` |
| `srdreg.tm` | `s[0..31]` |
| `srdreg.yieldreq` | `s[0..31]` |
| `srem.u32` | `drf, s[0..31], u32`<br>`drf, s[0..31], s[0..31]` |
| `ssetifvalue` | `ifvalue, u32`<br>`ifvalue, s[0..31]` |
| `ssetpdepth` | `pdepth, u32`<br>`pdepth, s[0..31]` |
| `ssettag` | `u32`<br>`s[0..31]` |
| `sshla.ov.s32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshll.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshra.s32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshrl.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssub.ov.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssub.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `stileid.u32` | `s[0..31]` |
| `sweird.f32` | `p[0..15], s[0..31]` |
| `sxor.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |

## s1

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `cbreg.add` | `[cbreg:u5], u32`<br>`[cbreg:u5], s[0..31]`<br>`[cbreg:$0x4], u32`<br>`[cbreg:$0x4], s[0..31]` |
| `descriptor_based_dma` | `y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `pop_rcmf` | `dest=(u5)` |
| `por` | `p[0..15], p[0..31], p[0..31]`<br>`p[0..15], @first_id_occurrence, p[0..31]`<br>`p[0..15], p[0..31], @first_id_occurrence` |
| `rdcbreg` | `s[0..31], [cbreg:u5 metadata:u2]` |
| `rddreg` | `s[0..31], [dreg:u6]` |
| `sadd.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], f32, s[0..31]` |
| `sadd.ov.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sadd.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sand.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sc.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sceil.f32` | `s[0..31], s[0..31]` |
| `sclz.u32` | `s[0..31], s[0..31]` |
| `scvt.f32.s32` | `s[0..31], s[0..31]`<br>`s[0..31], f32` |
| `scvt.s32.f32` | `s[0..31], s[0..31]`<br>`s[0..31], f32` |
| `sdelay` | `u11` |
| `seq.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `seq.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sfence` | `无操作数` |
| `sfence.scmf` | `无操作数` |
| `sfence.sel` | `u32`<br>`s[0..31]` |
| `sfence.stream.hbm` | `无操作数` |
| `sfence.stream.spmem` | `无操作数` |
| `sfloor.f32` | `s[0..31], s[0..31]` |
| `sge.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sge.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sge.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sgt.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sgt.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sgt.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `shalt` | `无操作数` |
| `simm.f32` | `s[0..31], f32` |
| `simm.s32` | `s[0..31], u32` |
| `sld` | `s[0..31], [smem:s[0..31]+u32]`<br>`s[0..31], [smem:s[0..31]+s[0..31]]`<br>`s[0..31], [smem:u32]` |
| `sld.cb` | `s[0..31], [smem:u32 cbreg:u5]`<br>`s[0..31], [smem:s[0..31] cbreg:u5]` |
| `sld.cb.upd` | `s[0..31], [smem:u32 cbreg:u5]`<br>`s[0..31], [smem:s[0..31] cbreg:u5]` |
| `sle.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sle.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sle.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `slt.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `slt.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `slt.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `smax.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], s[0..31], f32` |
| `smax.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `smin.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], s[0..31], f32` |
| `smin.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `smov.u32` | `s[0..31], s[0..31]` |
| `sne.f32` | `p[0..15], s[0..31], s[0..31]`<br>`p[0..15], s[0..31], f32` |
| `sne.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sor.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `spop` | `s[0..31], drf`<br>`s[0..31], v2sf` |
| `srdreg.ddr` | `s[0..31]` |
| `srdreg.fsr` | `s[0..31]` |
| `srdreg.gtchi` | `s[0..31]` |
| `srdreg.gtclo` | `s[0..31]` |
| `srdreg.lcchi` | `s[0..31]` |
| `srdreg.lcclo` | `s[0..31]` |
| `srdreg.scid` | `s[0..31]` |
| `srdreg.tag` | `s[0..31]` |
| `srdreg.tbm` | `s[0..31]` |
| `srdreg.tm` | `s[0..31]` |
| `srdreg.yieldreq` | `s[0..31]` |
| `ssetifvalue` | `ifvalue, u32`<br>`ifvalue, s[0..31]` |
| `ssetpdepth` | `pdepth, u32`<br>`pdepth, s[0..31]` |
| `ssettag` | `u32`<br>`s[0..31]` |
| `sshla.ov.s32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshll.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshra.s32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshrl.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sst` | `[smem:u32], s[0..31]`<br>`[smem:s[0..31]], s[0..31]` |
| `sst.cb` | `[smem:u32 cbreg:u5], s[0..31]`<br>`[smem:s[0..31] cbreg:u5], s[0..31]` |
| `sst.cb.upd` | `[smem:u32 cbreg:u5], s[0..31]`<br>`[smem:s[0..31] cbreg:u5], s[0..31]` |
| `ssub.f32` | `s[0..31], s[0..31], s[0..31]`<br>`s[0..31], f32, s[0..31]` |
| `ssub.ov.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssub.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `stileid.u32` | `s[0..31]` |
| `sweird.f32` | `p[0..15], s[0..31]` |
| `sxor.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `wrcbreg` | `[cbreg:0x6 metadata:u2], u32`<br>`[cbreg:0x6 metadata:u2], s[0..31]` |
| `wrdreg` | `[dreg:u5], u32`<br>`[dreg:u5], s[0..31]` |

## dma

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `dma.local` | `[timem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:u32], [simem:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:u32], [smem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:u32], [simem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:u32], [smem:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:u32], [simem:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:u32], [smem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:u32], [simem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:u32], [smem:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:u32], [timem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:u32], [simem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:u32], [smem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:u32], [simem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:u32], [smem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:u32], [timem:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:u32], [timem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:u32], [simem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:u32], [smem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:u32], [simem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:u32], [smem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:u32], [timem:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32`<br>`[timem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], u32`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31]`<br>`[hbm:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31]`<br>`[timem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:u32], [timem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:u32], [timem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:u32], [simem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:u32], [smem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:u32], [timem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:u32], [timem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:u32], [simem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:u32], [smem:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], u32`<br>`[smem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], u32`<br>`[simem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31]`<br>`[simem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], s[0..31]`<br>`[smem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [simem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [smem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:u32], [timem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:u32], [timem:s[0..31]], s[0..31]`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], u32`<br>`[spmem:s[0..31]], [sflag:s[0..31]], [timem:s[0..31]], s[0..31]` |
| `general_dma` | `dma_vreg_dest_offset=(u6), dma_vreg_source_offset=(u6), dma_sreg_source_offset=(u5), dma_length=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32), thread_select=(u1), trace_en=(u1), dst_opcode=(write / write_4b / compare_and_swap / read_and_add), dst_mem_core_id=(reserved_self / non_core / tensor_core_0 / reserved_tc_1 / sparse_core_0 / sparse_core_1 / reserved_sc_2 / reserved_sc_3), dst_mem_mem_id=(hbm / host / imem / timem), sync_mode=(32b_word / descriptor), relaxed_ordering_or_thread_select_1=(u1), src_mem_core_id=(reserved_self / non_core / tensor_core_0 / reserved_tc_1 / sparse_core_0 / sparse_core_1 / reserved_sc_2 / reserved_sc_3), src_mem_mem_id=(hbm / host / imem / timem), dma_sreg_dest_offset=(u5), source_in_vreg=(u1), dest_in_vreg=(u1), nondefault_stride_dimensions=(u3), source_sync_flag_number_valid=(u1), source_sync_flag_number=(u5), dest_sync_flags_valid=(u1), dest_sync_flags_vs1=(u5), destination_id_valid=(u1), destination_id=(u5), upper_host_addr_valid=(u1), upper_host_addr=(u5)` |
| `single_strided_dma` | `dma_sreg_source_offset=(u5), dma_length=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32), thread_select=(u1), trace_en=(u1), dst_opcode=(write / write_4b / compare_and_swap / read_and_add), dst_mem_core_id=(reserved_self / non_core / tensor_core_0 / reserved_tc_1 / sparse_core_0 / sparse_core_1 / reserved_sc_2 / reserved_sc_3), dst_mem_mem_id=(hbm / host / imem / timem), sync_mode=(32b_word / descriptor), relaxed_ordering_or_thread_select_1=(u1), src_mem_core_id=(reserved_self / non_core / tensor_core_0 / reserved_tc_1 / sparse_core_0 / sparse_core_1 / reserved_sc_2 / reserved_sc_3), src_mem_mem_id=(hbm / host / imem / timem), dma_sreg_dest_offset=(u5), dest_sync_flags=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32), inner_vector_length=(u5), source_stride_valid=(u1), source_stride=(u5), destination_stride_valid=(u1), destination_stride=(u5), elements_per_stride_valid=(u1), elements_per_stride=(u5), upper_host_addr_valid=(u1), upper_host_addr=(u5)` |

## misc

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `atomic_tile_add_set_done` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `atomic_tile_add_set_done_inverted` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `por` | `p[0..15], p[0..31], p[0..31]`<br>`p[0..15], @first_id_occurrence, p[0..31]`<br>`p[0..15], p[0..31], @first_id_occurrence` |
| `push_scmf` | `y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `read_sync_state_done` | `pdst=(preg_0 / preg_1 / preg_2 / preg_3 / preg_4 / preg_5 / preg_6 / preg_7 / preg_8 / preg_9 / preg_10 / preg_11 / preg_12 / preg_13 / reserved_0 / reserved_1)` |
| `read_sync_state_value` | `dest=(u5)` |
| `sadd.ov.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sadd.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sand.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `sbarrier.arrive` | `[bflag:u2], u32`<br>`[bflag:u2], s[0..31]` |
| `sc.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sclz.u32` | `s[0..31], s[0..31]` |
| `seq.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sge.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sge.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sgt.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sgt.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `simm.f32` | `s[0..31], f32` |
| `simm.s32` | `s[0..31], u32` |
| `sint` | `u32`<br>`s[0..31]` |
| `sle.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sle.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `slt.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `slt.u32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `smax.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `smem.add.s32` | `[smem:s[0..31]], [sflag:u5], u32` |
| `smin.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `smov.u32` | `s[0..31], s[0..31]` |
| `sne.s32` | `p[0..15], s[0..31], u32`<br>`p[0..15], s[0..31], s[0..31]` |
| `sor.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssettm` | `u32`<br>`s[0..31]` |
| `sshla.ov.s32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshll.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshra.s32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `sshrl.u32` | `s[0..31], s[0..31], u32`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssub.ov.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssub.s32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `ssyncadd.both.s32` | `[sflag:s[0..31]], [sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], [sflag:s[0..31]], s[0..31]` |
| `ssyncadd.other.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncadd.remote.done.s32` | `[sflag:s[0..31]@s[0..31]], u32`<br>`[sflag:s[0..31]@s[0..31]], s[0..31]`<br>`[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncadd.remote.doneinv.s32` | `[sflag:s[0..31]@s[0..31]], u32`<br>`[sflag:s[0..31]@s[0..31]], s[0..31]`<br>`[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncadd.remote.s32` | `[sflag:s[0..31]@s[0..31]], u32`<br>`[sflag:s[0..31]@s[0..31]], s[0..31]`<br>`[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncadd.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncadd.tile.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncpa.u1` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncread` | `s[0..31], [sflag:u32]`<br>`s[0..31], [sflag:s[0..31]]` |
| `ssyncread.done` | `s[0..31], [sflag:u32]`<br>`s[0..31], [sflag:s[0..31]]` |
| `ssyncread.pa` | `s[0..31], [sflag:u32]`<br>`s[0..31], [sflag:s[0..31]]` |
| `ssyncset.both.done` | `[sflag:s[0..31]], [sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], [sflag:s[0..31]], s[0..31]` |
| `ssyncset.both.s32` | `[sflag:s[0..31]], [sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], [sflag:s[0..31]], s[0..31]` |
| `ssyncset.done` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.other.done` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.other.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.remote.done` | `[sflag:s[0..31]@s[0..31]], u32`<br>`[sflag:s[0..31]@s[0..31]], s[0..31]`<br>`[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.remote.doneinv` | `[sflag:s[0..31]@s[0..31]], u32`<br>`[sflag:s[0..31]@s[0..31]], s[0..31]`<br>`[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.remote.s32` | `[sflag:s[0..31]@s[0..31]], u32`<br>`[sflag:s[0..31]@s[0..31]], s[0..31]`<br>`[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.tile.done.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.tile.doneinv.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `ssyncset.tile.s32` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `strace` | `u32`<br>`s[0..31]` |
| `swait.done` | `[sflag:s[0..31]]` |
| `swait.eq` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.eqordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.ge` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.geordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.gt` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.gtordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.lt` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.ltordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.ne` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.neordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swait.notdone` | `[sflag:s[0..31]]` |
| `swatch.done` | `[sflag:s[0..31]]` |
| `swatch.end` | `无操作数` |
| `swatch.end.sel` | `u32`<br>`s[0..31]` |
| `swatch.eq` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.eqordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.ge` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.geordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.gt` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.gtordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.lt` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.ltordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.ne` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.neordone` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `swatch.notdone` | `[sflag:s[0..31]]` |
| `swatch.wait` | `[sflag:s[0..31]]` |
| `swatch.wait.sel` | `[sflag:s[0..31]], u32`<br>`[sflag:s[0..31]], s[0..31]` |
| `sxor.u32` | `s[0..31], u32, s[0..31]`<br>`s[0..31], s[0..31], s[0..31]` |
| `yieldable_sync_done` | `x=(u5)` |
| `yieldable_sync_equal` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_equal_or_done` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_greater` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_greater_or_done` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_greater_or_equal` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_greater_or_equal_or_done` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_less` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_less_or_done` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_not_done` | `x=(u5)` |
| `yieldable_sync_not_equal` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `yieldable_sync_not_equal_or_done` | `x=(u5), y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |

## va0

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `eup_push` | `x=(u6)` |
| `unpack_interleaved_compressed_lower_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_s8_to_s16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_u8_to_u16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_s8_to_s16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_u8_to_u16` | `dest=(u6), x=(u6)` |
| `vadd.s16` | `v[0..63], v[0..63], v[0..63]` |
| `vadd.s32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vand.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vbcast.xlane.b16` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vbroadcast` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vc.u16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vc.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vceil.bf16` | `v[0..63], v[0..63]` |
| `vceil.f32` | `v[0..63], v[0..63]` |
| `vclamp.gez.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vclamp.gez.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vclamps.bf16` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vclamps.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vclass.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vclass.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vclz` | `v[0..63], v[0..63]` |
| `vcmask` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, s[0..31]` |
| `vcombine.high` | `v[0..63], v[0..63], v[0..63]` |
| `vcombine.low` | `v[0..63], v[0..63], v[0..63]` |
| `vcosq.bf16` | `erf, v[0..63]` |
| `vcosq.f32` | `erf, v[0..63]` |
| `vcvt.f32.f8e5m2` | `v[0..63], v[0..63]` |
| `vcvt.f32.hf16` | `v[0..63], v[0..63]` |
| `vcvt.f32.if8` | `v[0..63], v[0..63]` |
| `vcvt.f32.s32` | `v[0..63], v[0..63]` |
| `vcvt.s32.f32` | `v[0..63], v[0..63]` |
| `vcvt.sr.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vcvt.sr.f32.bf8` | `v[0..63], v[0..63], v[0..63]` |
| `vcvt.sr.f32.hf16` | `v[0..63], v[0..63], v[0..63]` |
| `vcvt.sr.f32.if8` | `v[0..63], v[0..63], v[0..63]` |
| `vector_clamp_asymmetric_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_clamp_asymmetric_f32` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_convert_bf16_to_s4` | `dest=(u6), x=(u6)` |
| `vector_convert_bf16_to_s8` | `dest=(u6), x=(u6)` |
| `vector_convert_bf16_to_u4` | `dest=(u6), x=(u6)` |
| `vector_convert_bf16_to_u8` | `dest=(u6), x=(u6)` |
| `vector_convert_f32_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_s4_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_u4_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_gt_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_gte_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_lane_left_shift_insert_b16` | `dest=(u6), x=(u6), y_vreg=(u6), shift_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30)` |
| `vector_lt_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_lte_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_remap_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_remap_f32` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `veq.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `veq.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `veq.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `veq.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `verf.bf16` | `erf, v[0..63]` |
| `verf.f32` | `erf, v[0..63]` |
| `vfloor.bf16` | `v[0..63], v[0..63]` |
| `vfloor.f32` | `v[0..63], v[0..63]` |
| `vge.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vge.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vgt.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vimm.bf16` | `v[0..63], u32` |
| `vimm.f32` | `v[0..63], f32` |
| `vimm.hf16` | `v[0..63], u32` |
| `vimm.s32` | `v[0..63], u32` |
| `vlaneseq.c.b16` | `v[0..63]` |
| `vlaneseq.i.b16` | `v[0..63]` |
| `vlaneseq.u32` | `v[0..63]` |
| `vle.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vle.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.to.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.to.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vle.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlog2.bf16` | `erf, v[0..63]` |
| `vlog2.f32` | `erf, v[0..63]` |
| `vlt.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vlt.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.to.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.to.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vlt.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vmand` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmax.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmax.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vmax.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmax.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vmctz.xlane` | `v[0..63], vm[0..15]` |
| `vmctz.xlane.b16` | `v[0..63], vm[0..15]` |
| `vmin.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmin.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vmin.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmin.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vmmov` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmneg` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmor` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmov` | `v[0..63], v[0..63]`<br>`v[0..63], s[0..31]` |
| `vmpackc.even` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmpackc.low` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmpcnt.ones.xlane` | `v[0..63], vm[0..15]` |
| `vmpcnt.ones.xlane.b16` | `v[0..63], vm[0..15]` |
| `vmperm.xlane.b16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], s[0..31]` |
| `vmperm.xlane.b32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], s[0..31]` |
| `vmperm.xlane.b8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], s[0..31]` |
| `vmperm.xlane.c2i.b16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmperm.xlane.c2i.b8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmperm.xlane.i2c.b16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmperm.xlane.i2c.b8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmprefix.xlane` | `v[0..63], vm[0..15]` |
| `vmprefix.xlane.b16` | `v[0..63], vm[0..15]` |
| `vmul.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmul.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]`<br>`v[0..63], f32, v[0..63]` |
| `vmul.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmul.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vmulhi.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]` |
| `vmxor` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vne.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vne.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vne.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vne.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vnez.u8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, s[0..31]` |
| `vnsel` | `v[0..63], vm[0..15], u32, v[0..63]`<br>`v[0..63], vm[0..15], s[0..31], v[0..63]` |
| `vor.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vpack.c.b16.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b2.b1` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b32.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b4.b2` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b8.b4` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.bf16.s8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.bf16.u8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b16.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b32.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b8.b4` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.bf16.s8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.bf16.u8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vpcnt` | `v[0..63], v[0..63]` |
| `vperm.xlane` | `v[0..63], v[0..63], v[0..63]` |
| `vperm.xlane.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vperm.xlane.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vperm.xlane.c2i.b16` | `v[0..63], v[0..63]` |
| `vperm.xlane.c2i.b8` | `v[0..63], v[0..63]` |
| `vperm.xlane.i2c.b16` | `v[0..63], v[0..63]` |
| `vperm.xlane.i2c.b8` | `v[0..63], v[0..63]` |
| `vpow2.bf16` | `erf, v[0..63]` |
| `vpow2.f32` | `erf, v[0..63]` |
| `vpsel` | `v[0..63], p5, u32, v[0..63]`<br>`v[0..63], p5, v[0..63], v[0..63]`<br>`v[0..63], !p5, s[0..31], v[0..63]` |
| `vrcp.bf16` | `erf, v[0..63]` |
| `vrcp.f32` | `erf, v[0..63]` |
| `vrot.xlane.b16` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vrot.xlane.b32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vround.rtna.bf16` | `v[0..63], v[0..63]` |
| `vround.rtna.f32` | `v[0..63], v[0..63]` |
| `vrsqrt.bf16` | `erf, v[0..63]` |
| `vrsqrt.f32` | `erf, v[0..63]` |
| `vsel` | `v[0..63], vm[0..15], u32, v[0..63]`<br>`v[0..63], vm[0..15], v[0..63], v[0..63]`<br>`v[0..63], vm[0..15], s[0..31], v[0..63]` |
| `vshift.insert` | `v[0..63], v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63], s[0..31]` |
| `vshll.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vshra.s32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vshrl.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vsigshft.bf16` | `erf, v[0..63]` |
| `vsigshft.f32` | `erf, v[0..63]` |
| `vsinq.bf16` | `erf, v[0..63]` |
| `vsinq.f32` | `erf, v[0..63]` |
| `vsub.s16` | `v[0..63], v[0..63], v[0..63]` |
| `vsub.s32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vtanh.bf16` | `erf, v[0..63]` |
| `vtanh.f32` | `erf, v[0..63]` |
| `vtrunc.bf16` | `v[0..63], v[0..63]` |
| `vtrunc.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.hf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s1.s2` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s2.s4` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s4.s8` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.hf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s1.s2` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s2.s4` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s4.s8` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.u8.u16` | `v[0..63], v[0..63]` |
| `vweird.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vweird.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vxor.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |

## va1

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `eup_push` | `x=(u6)` |
| `unpack_interleaved_compressed_lower_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_s8_to_s16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_u8_to_u16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_s8_to_s16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_u8_to_u16` | `dest=(u6), x=(u6)` |
| `vadd.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vadd.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]`<br>`v[0..63], f32, v[0..63]` |
| `vadd.s16` | `v[0..63], v[0..63], v[0..63]` |
| `vadd.s32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vand.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vbcast.xlane.b16` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vbroadcast` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vc.u16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vc.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vceil.bf16` | `v[0..63], v[0..63]` |
| `vceil.f32` | `v[0..63], v[0..63]` |
| `vclamp.gez.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vclamp.gez.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vclamps.bf16` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vclamps.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vclass.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vclass.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vclz` | `v[0..63], v[0..63]` |
| `vcmask` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, s[0..31]` |
| `vcombine.high` | `v[0..63], v[0..63], v[0..63]` |
| `vcombine.low` | `v[0..63], v[0..63], v[0..63]` |
| `vcosq.bf16` | `erf, v[0..63]` |
| `vcosq.f32` | `erf, v[0..63]` |
| `vcvt.f32.f8e5m2` | `v[0..63], v[0..63]` |
| `vcvt.f32.hf16` | `v[0..63], v[0..63]` |
| `vcvt.f32.if8` | `v[0..63], v[0..63]` |
| `vcvt.f32.s32` | `v[0..63], v[0..63]` |
| `vcvt.s32.f32` | `v[0..63], v[0..63]` |
| `vcvt.sr.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vcvt.sr.f32.bf8` | `v[0..63], v[0..63], v[0..63]` |
| `vcvt.sr.f32.hf16` | `v[0..63], v[0..63], v[0..63]` |
| `vcvt.sr.f32.if8` | `v[0..63], v[0..63], v[0..63]` |
| `vector_add_high_f32_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_add_low_f32_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_clamp_asymmetric_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_clamp_asymmetric_f32` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_convert_bf16_to_s4` | `dest=(u6), x=(u6)` |
| `vector_convert_bf16_to_s8` | `dest=(u6), x=(u6)` |
| `vector_convert_bf16_to_u4` | `dest=(u6), x=(u6)` |
| `vector_convert_bf16_to_u8` | `dest=(u6), x=(u6)` |
| `vector_convert_f32_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_s4_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_u4_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_convert_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `vector_gt_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_gte_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_lane_left_shift_insert_b16` | `dest=(u6), x=(u6), y_vreg=(u6), shift_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30)` |
| `vector_lt_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_lte_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_remap_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_remap_f32` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `veq.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `veq.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `veq.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `veq.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `verf.bf16` | `erf, v[0..63]` |
| `verf.f32` | `erf, v[0..63]` |
| `vfloor.bf16` | `v[0..63], v[0..63]` |
| `vfloor.f32` | `v[0..63], v[0..63]` |
| `vge.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vge.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vgt.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vimm.bf16` | `v[0..63], u32` |
| `vimm.f32` | `v[0..63], f32` |
| `vimm.hf16` | `v[0..63], u32` |
| `vimm.s32` | `v[0..63], u32` |
| `vlaneseq.c.b16` | `v[0..63]` |
| `vlaneseq.i.b16` | `v[0..63]` |
| `vlaneseq.u32` | `v[0..63]` |
| `vle.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vle.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.to.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.to.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vle.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlog2.bf16` | `erf, v[0..63]` |
| `vlog2.f32` | `erf, v[0..63]` |
| `vlt.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vlt.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.to.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.to.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vlt.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vmand` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmax.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmax.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vmax.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmax.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vmctz.xlane` | `v[0..63], vm[0..15]` |
| `vmctz.xlane.b16` | `v[0..63], vm[0..15]` |
| `vmin.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmin.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vmin.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmin.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vmmov` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmneg` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmor` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmov` | `v[0..63], v[0..63]`<br>`v[0..63], s[0..31]` |
| `vmpackc.even` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmpackc.low` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmpcnt.ones.xlane` | `v[0..63], vm[0..15]` |
| `vmpcnt.ones.xlane.b16` | `v[0..63], vm[0..15]` |
| `vmperm.xlane.b16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], s[0..31]` |
| `vmperm.xlane.b32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], s[0..31]` |
| `vmperm.xlane.b8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15], s[0..31]` |
| `vmperm.xlane.c2i.b16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmperm.xlane.c2i.b8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmperm.xlane.i2c.b16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmperm.xlane.i2c.b8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm[0..15]` |
| `vmprefix.xlane` | `v[0..63], vm[0..15]` |
| `vmprefix.xlane.b16` | `v[0..63], vm[0..15]` |
| `vmul.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmul.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]`<br>`v[0..63], f32, v[0..63]` |
| `vmul.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmul.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vmxor` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vne.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vne.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vne.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vne.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vnez.u8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, s[0..31]` |
| `vnsel` | `v[0..63], vm[0..15], u32, v[0..63]`<br>`v[0..63], vm[0..15], s[0..31], v[0..63]` |
| `vor.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vpack.c.b16.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b2.b1` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b32.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b4.b2` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b8.b4` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.bf16.s8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.bf16.u8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b16.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b32.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b8.b4` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.bf16.s8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.bf16.u8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vpcnt` | `v[0..63], v[0..63]` |
| `vperm.xlane` | `v[0..63], v[0..63], v[0..63]` |
| `vperm.xlane.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vperm.xlane.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vperm.xlane.c2i.b16` | `v[0..63], v[0..63]` |
| `vperm.xlane.c2i.b8` | `v[0..63], v[0..63]` |
| `vperm.xlane.i2c.b16` | `v[0..63], v[0..63]` |
| `vperm.xlane.i2c.b8` | `v[0..63], v[0..63]` |
| `vpow2.bf16` | `erf, v[0..63]` |
| `vpow2.f32` | `erf, v[0..63]` |
| `vpsel` | `v[0..63], p5, u32, v[0..63]`<br>`v[0..63], p5, v[0..63], v[0..63]`<br>`v[0..63], !p5, s[0..31], v[0..63]` |
| `vrcp.bf16` | `erf, v[0..63]` |
| `vrcp.f32` | `erf, v[0..63]` |
| `vrot.xlane.b16` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vrot.xlane.b32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], s[0..31]` |
| `vround.rtna.bf16` | `v[0..63], v[0..63]` |
| `vround.rtna.f32` | `v[0..63], v[0..63]` |
| `vrsqrt.bf16` | `erf, v[0..63]` |
| `vrsqrt.f32` | `erf, v[0..63]` |
| `vsel` | `v[0..63], vm[0..15], u32, v[0..63]`<br>`v[0..63], vm[0..15], v[0..63], v[0..63]`<br>`v[0..63], vm[0..15], s[0..31], v[0..63]` |
| `vshift.insert` | `v[0..63], v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63], s[0..31]` |
| `vshll.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vshra.s32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vshrl.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vsigshft.bf16` | `erf, v[0..63]` |
| `vsigshft.f32` | `erf, v[0..63]` |
| `vsinq.bf16` | `erf, v[0..63]` |
| `vsinq.f32` | `erf, v[0..63]` |
| `vsub.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vsub.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]`<br>`v[0..63], f32, v[0..63]` |
| `vsub.s16` | `v[0..63], v[0..63], v[0..63]` |
| `vsub.s32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vtanh.bf16` | `erf, v[0..63]` |
| `vtanh.f32` | `erf, v[0..63]` |
| `vtrunc.bf16` | `v[0..63], v[0..63]` |
| `vtrunc.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.hf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s1.s2` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s2.s4` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s4.s8` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.hf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s1.s2` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s2.s4` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s4.s8` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.u8.u16` | `v[0..63], v[0..63]` |
| `vweird.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vweird.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vxor.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |

## va2

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `unpack_interleaved_compressed_lower_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_s8_to_s16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_lower_u8_to_u16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_s8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_s8_to_s16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_u8_to_bf16` | `dest=(u6), x=(u6)` |
| `unpack_interleaved_compressed_upper_u8_to_u16` | `dest=(u6), x=(u6)` |
| `vadd.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vadd.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]`<br>`v[0..63], f32, v[0..63]` |
| `vadd.s16` | `v[0..63], v[0..63], v[0..63]` |
| `vadd.s32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vand.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vc.u16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vc.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vclass.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vclass.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vclz` | `v[0..63], v[0..63]` |
| `vcmask` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, s[0..31]` |
| `vector_add_high_f32_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_add_low_f32_bf16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_gt_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_gte_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_lt_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `vector_lte_u16` | `dest=(u6), x=(u6), y_src=(vreg / lane_id / u32 / vs0 / vs1 / vs2 / vs3 / compressed_to_interleaved / interleaved_to_compressed / reserved_9 / reserved_10 / reserved_11 / reserved_12 / reserved_13 / reserved_14 / reserved_15 / reserved_16 / reserved_17 / reserved_18 / reserved_19 / reserved_20 / reserved_21 / reserved_22 / reserved_23 / reserved_24 / reserved_25 / reserved_26 / reserved_27 / reserved_28 / reserved_29 / reserved_30), y_vreg=(u6)` |
| `veq.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `veq.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `veq.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `veq.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vge.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vge.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vgt.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vgt.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vimm.bf16` | `v[0..63], u32` |
| `vimm.f32` | `v[0..63], f32` |
| `vimm.hf16` | `v[0..63], u32` |
| `vimm.s32` | `v[0..63], u32` |
| `vlaneseq.c.b16` | `v[0..63]` |
| `vlaneseq.i.b16` | `v[0..63]` |
| `vlaneseq.u32` | `v[0..63]` |
| `vle.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vle.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.to.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vle.to.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vle.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vlt.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.to.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vlt.to.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vlt.u32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vmand` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmax.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmax.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vmax.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmax.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vmin.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vmin.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]`<br>`v[0..63], v[0..63], f32` |
| `vmin.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmin.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vmmov` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmneg` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmor` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmov` | `v[0..63], v[0..63]`<br>`v[0..63], s[0..31]` |
| `vmpackc.even` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmpackc.low` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vmul.u16` | `v[0..63], v[0..63], v[0..63]` |
| `vmul.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vmxor` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vne.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vne.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], f32` |
| `vne.s16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vne.s32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], u32`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], s[0..31]` |
| `vnez.u8` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, s[0..31]` |
| `vnsel` | `v[0..63], vm[0..15], u32, v[0..63]`<br>`v[0..63], vm[0..15], s[0..31], v[0..63]` |
| `vor.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vpack.c.b16.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b2.b1` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b32.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b4.b2` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.b8.b4` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.bf16.s8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.bf16.u8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.c.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b16.b8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b32.b16` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.b8.b4` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.bf16.s8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.bf16.u8` | `v[0..63], v[0..63], v[0..63]` |
| `vpack.i.f32.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vpcnt` | `v[0..63], v[0..63]` |
| `vpsel` | `v[0..63], p5, u32, v[0..63]`<br>`v[0..63], p5, v[0..63], v[0..63]`<br>`v[0..63], !p5, s[0..31], v[0..63]` |
| `vpush` | `v2sf, v[0..63], u32`<br>`v2sf, v[0..63], s[0..31]` |
| `vsel` | `v[0..63], vm[0..15], u32, v[0..63]`<br>`v[0..63], vm[0..15], v[0..63], v[0..63]`<br>`v[0..63], vm[0..15], s[0..31], v[0..63]` |
| `vshll.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vshra.s32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vshrl.u32` | `v[0..63], v[0..63], u32`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], v[0..63], s[0..31]` |
| `vsub.bf16` | `v[0..63], v[0..63], v[0..63]` |
| `vsub.f32` | `v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]`<br>`v[0..63], f32, v[0..63]` |
| `vsub.s16` | `v[0..63], v[0..63], v[0..63]` |
| `vsub.s32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |
| `vunpack.c.0.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.0.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.1.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.2.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.bf8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.if8.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.3.s8.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.hf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s1.s2` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s2.s4` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s4.s8` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.l.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.hf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s1.s2` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s2.s4` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s4.s8` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.c.u.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.l.u8.u16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.bf16.f32` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s16.s32` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.s8.s16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.u8.bf16` | `v[0..63], v[0..63]` |
| `vunpack.i.u.u8.u16` | `v[0..63], v[0..63]` |
| `vweird.bf16` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vweird.f32` | `vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vxor.u32` | `v[0..63], u32, v[0..63]`<br>`v[0..63], v[0..63], v[0..63]`<br>`v[0..63], s[0..31], v[0..63]` |

## vld

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `vld` | `v[0..63], [tilespmem:u32]`<br>`v[0..63], [tilespmem:s[0..31]+u32]` |
| `vld.cb.msk` | `v[0..63], [tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vld.cb.upd.msk` | `v[0..63], [tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vld.idx.cb.msk` | `v[0..63], [tilespmem:v[0..63]+s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vld.idx.msk` | `v[0..63], [tilespmem:v[0..63]+s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:v[0..63]+s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:v[0..63]+s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |
| `vld.msk` | `v[0..63], [tilespmem:u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15`<br>`v[0..63], [tilespmem:s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15` |

## vst

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `tile_spmem_store_indexed_circular_buffer_return_value_add_bf16` | `source=(u6), base_address=(0x0 / vs0 / vs1 / vs2 / vs3 / reserved_0 / reserved_1 / reserved_2), cbreg=(u4), offset=(u32), stride=(vs0 / vs1 / vs2 / vs3 / u32), index=(u6), mask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7), dest=(u6)` |
| `tile_spmem_store_indexed_circular_buffer_return_value_add_f32` | `source=(u6), base_address=(0x0 / vs0 / vs1 / vs2 / vs3 / reserved_0 / reserved_1 / reserved_2), cbreg=(u4), offset=(u32), stride=(vs0 / vs1 / vs2 / vs3 / u32), index=(u6), mask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7), dest=(u6)` |
| `tile_spmem_store_indexed_circular_buffer_return_value_add_s16` | `source=(u6), base_address=(0x0 / vs0 / vs1 / vs2 / vs3 / reserved_0 / reserved_1 / reserved_2), cbreg=(u4), offset=(u32), stride=(vs0 / vs1 / vs2 / vs3 / u32), index=(u6), mask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7), dest=(u6)` |
| `tile_spmem_store_indexed_circular_buffer_return_value_add_s32` | `source=(u6), base_address=(0x0 / vs0 / vs1 / vs2 / vs3 / reserved_0 / reserved_1 / reserved_2), cbreg=(u4), offset=(u32), stride=(vs0 / vs1 / vs2 / vs3 / u32), index=(u6), mask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7), dest=(u6)` |
| `vst` | `[tilespmem:u32], v[0..63]`<br>`[tilespmem:s[0..31]+u32], v[0..63]` |
| `vst.add.bf16.msk` | `[tilespmem:s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.add.f32.msk` | `[tilespmem:s[0..31]+f32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+f32 ss:f32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+f32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.add.s16.msk` | `[tilespmem:s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.add.s32.msk` | `[tilespmem:s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.add.bf16.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.add.f32.msk` | `[tilespmem:s[0..31]+f32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+f32 ss:f32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+f32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.add.s16.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.add.s32.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.upd.add.bf16.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.upd.add.f32.msk` | `[tilespmem:s[0..31]+f32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+f32 ss:f32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+f32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.upd.add.s16.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.upd.add.s32.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.cb.upd.msk` | `[tilespmem:s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31] cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.add.bf16.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.add.f32.msk` | `[tilespmem:v[0..63]+s[0..31]+f32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.add.s16.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.add.s32.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.cb.add.bf16.msk` | `[tilespmem:v[0..63]+s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.cb.add.f32.msk` | `[tilespmem:v[0..63]+s[0..31]+f32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.cb.add.s16.msk` | `[tilespmem:v[0..63]+s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.cb.add.s32.msk` | `[tilespmem:v[0..63]+s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.cb.msk` | `[tilespmem:v[0..63]+s[0..31]+u32 cbreg:u4], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:v[0..63]+s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:v[0..63]+s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.ret.add.bf16.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], v[0..63], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.ret.add.f32.msk` | `[tilespmem:v[0..63]+s[0..31]+f32], v[0..63], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.ret.add.s16.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], v[0..63], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.idx.ret.add.s32.msk` | `[tilespmem:v[0..63]+s[0..31]+u32], v[0..63], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vst.msk` | `[tilespmem:s[0..31]+u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:u32], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]`<br>`[tilespmem:s[0..31]+u32 ss:s[0..31]], vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |

## stream

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `indirect_vreg_stream` | `indirect_offsets=(u6), valid_offset_mask=(u6), off_tile_start_offset_valid=(u1), off_tile_start_offset=(u5), sync_flag_core_type=(tec_scs / tac), sync_flag_count_type=(word_4b / descriptor), set_done_bit=(u1), tile_local_stride=(stride_32b / stride_64b / stride_128b / stride_256b / stride_512b / stride_1024b / stride_2048b / no_stride), post_update_circular_buffer=(u1), indirect_list_type=(word_offset / row_offset), indirect_list_stride=(u6), indirect_filter_en=(u1), trace_en=(u1), stream_opcode=(gather / gather_integer_add / gather_float_add / reserved_0 / scatter / scatter_integer_add / scatter_float_add / reserved_1), tile_local_memory_type=(smem / tile_spmem), tile_local_stream_type=(linear / circular_buffer), off_tile_memory_type=(spmem / tile_spmem_n / hbm / hbm_4b), gather_scatter_add_is_b16=(u1), s0_x=(u5), s0_y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32), s1_x=(u5), s1_y=(sreg0 / sreg1 / sreg2 / sreg3 / sreg4 / sreg5 / sreg6 / sreg7 / sreg8 / sreg9 / sreg10 / sreg11 / sreg12 / sreg13 / sreg14 / sreg15 / sreg16 / sreg17 / sreg18 / sreg19 / sreg20 / sreg21 / sreg22 / sreg23 / sreg24 / sreg25 / sreg26 / sreg27 / sreg28 / sreg29 / sreg30 / sreg31 / u32)` |
| `stream.indirect.gather` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.add.bf16` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.add.f32` | `[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.indirect.gather.add.s16` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.add.s32` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.cb` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.cb.add.bf16` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.cb.add.f32` | `[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.indirect.gather.cb.add.s16` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.gather.cb.add.s32` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.add.bf16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.add.f32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.add.s16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.add.s32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.cb` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.cb.add.bf16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.cb.add.f32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.cb.add.s16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.indirect.scatter.cb.add.s32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.linear.gather` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.add.bf16` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.add.f32` | `[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], f32, u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [spmem:s[0..31]], f32, u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], f32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], f32, u32` |
| `stream.linear.gather.add.s16` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.add.s32` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.cb` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.cb.add.bf16` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.cb.add.f32` | `[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], f32, u32`<br>`[tilespmem:u5], [sflag:f32], [spmem:s[0..31]], f32, u32`<br>`[tilespmem:u5], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], f32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], f32, u32` |
| `stream.linear.gather.cb.add.s16` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32` |
| `stream.linear.gather.cb.add.s32` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.add.bf16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.add.f32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:f32], f32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:f32], f32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:f32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:f32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], f32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], f32, u32` |
| `stream.linear.scatter.add.s16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.add.s32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.cb` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.cb.add.bf16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.cb.add.f32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:f32], f32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:f32], f32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:f32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:f32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], f32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], f32, u32` |
| `stream.linear.scatter.cb.add.s16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32` |
| `stream.linear.scatter.cb.add.s32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], u32` |
| `stream.strided.gather` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.add.bf16` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.add.f32` | `[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:f32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.strided.gather.add.s16` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.add.s32` | `[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:s[0..31]], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.cb` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.cb.add.bf16` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.cb.add.f32` | `[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:f32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.strided.gather.cb.add.s16` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.gather.cb.add.s32` | `[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:u32], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [hbm4b:s[0..31]+s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[tilespmem:u5], [sflag:s[0..31]], [spmem:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.add.bf16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.add.f32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.strided.scatter.add.s16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.add.s32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:s[0..31]], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.cb` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.cb.add.bf16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.cb.add.f32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:f32], f32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:f32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], f32, s[0..31], s[0..31], u32` |
| `stream.strided.scatter.cb.add.s16` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |
| `stream.strided.scatter.cb.add.s32` | `[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:u32], s[0..31], s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], u32, s[0..31], s[0..31], u32`<br>`[hbm4b:s[0..31]+s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32`<br>`[spmem:s[0..31]], [tilespmem:u5], [sflag:s[0..31]], s[0..31], s[0..31], s[0..31], u32` |

## vr

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `pop_xrf_write_partial_0` | `unit=(xrf0 / xrf1), dest_one=(u6)` |
| `pop_xrf_write_partial_1` | `unit=(xrf0 / xrf1), dest_one=(u6), dest_vm=(u4)` |
| `pop_xrf_write_partial_2` | `unit=(xrf0 / xrf1), dest_two=(u6)` |
| `pop_xrf_write_partial_3` | `unit=(xrf0 / xrf1), dest_two=(u6), dest_vm=(u4)` |
| `pop_xrf_write_partial_4` | `unit=(xrf0 / xrf1), dest_one=(u6), dest_two=(u6)` |
| `vmovc` | `v[0..63], v[0..63]` |
| `vpop` | `v[0..63], erf`<br>`v[0..63], v[0..63], vm[0..15], xrf` |

## vx

| 助记符 | 按顺序排列的操作数类型 |
|---|---|
| `max_index_scan_bf16` | `source_one=(vst_source / v0_y_vreg / v0_x / v1_y_vreg / v1_x / v2_y_vreg / v2_x / lane_id), vst_source=(u6), v0_y_vreg=(u6), v0_x=(u6), v1_y_vreg=(u6), v1_x=(u6), v2_y_vreg=(u6), v2_x=(u6), vmask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7)` |
| `max_index_scan_u16` | `source_one=(vst_source / v0_y_vreg / v0_x / v1_y_vreg / v1_x / v2_y_vreg / v2_x / lane_id), vst_source=(u6), v0_y_vreg=(u6), v0_x=(u6), v1_y_vreg=(u6), v1_x=(u6), v2_y_vreg=(u6), v2_x=(u6), vmask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7)` |
| `min_index_scan_bf16` | `source_one=(vst_source / v0_y_vreg / v0_x / v1_y_vreg / v1_x / v2_y_vreg / v2_x / lane_id), vst_source=(u6), v0_y_vreg=(u6), v0_x=(u6), v1_y_vreg=(u6), v1_x=(u6), v2_y_vreg=(u6), v2_x=(u6), vmask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7)` |
| `segmented_min_index_scan_bf16` | `source_one=(vst_source / v0_y_vreg / v0_x / v1_y_vreg / v1_x / v2_y_vreg / v2_x / lane_id), vst_source=(u6), v0_y_vreg=(u6), v0_x=(u6), v1_y_vreg=(u6), v1_x=(u6), v2_y_vreg=(u6), v2_x=(u6), vmask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7)` |
| `segmented_min_scan_bf16` | `source_one=(vst_source / v0_y_vreg / v0_x / v1_y_vreg / v1_x / v2_y_vreg / v2_x / lane_id), vst_source=(u6), v0_y_vreg=(u6), v0_x=(u6), v1_y_vreg=(u6), v1_x=(u6), v2_y_vreg=(u6), v2_x=(u6), vmask=(vmsk0 / vmsk1 / vmsk2 / vmsk3 / vmsk4 / vmsk5 / vmsk6 / vmsk7 / vmsk8 / vmsk9 / vmsk10 / vmsk11 / vmsk12 / vmsk13 / vmsk14 / vmsk15 / all_ones / 0x7f / 0x3f / 0x1f / 0xf / 0x7 / 0x3 / 0x1 / reserved_0 / reserved_1 / reserved_2 / reserved_3 / reserved_4 / reserved_5 / reserved_6 / reserved_7)` |
| `vadd.scan.msk.bf16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.scan.msk.bf16.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.scan.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.scan.msk.s16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.scan.msk.s16.s32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.scan.msk.s32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.seg.scan.bf16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.seg.scan.bf16.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.seg.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.seg.scan.s16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.seg.scan.s16.s32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vadd.seg.scan.s32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vdupcnt.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vdupcnt.msk.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.index.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.index.scan.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.scan.msk.bf16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.scan.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.scan.msk.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.scan.msk.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.index.scan.bf16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.index.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.index.scan.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.index.scan.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.scan.bf16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.scan.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmax.seg.scan.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.index.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.index.scan.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.index.scan.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.scan.msk.bf16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.scan.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.scan.msk.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.scan.msk.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.seg.index.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.seg.index.scan.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.seg.index.scan.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.seg.scan.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.seg.scan.u16` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vmin.seg.scan.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vsort.ascd.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]` |
| `vsort.ascd.msk.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]` |
| `vsort.dscd.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]` |
| `vsort.dscd.msk.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63], v[0..63]` |
| `vunique.msk.f32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
| `vunique.msk.u32` | `xrf, vm0 / vm1 / vm2 / vm3 / vm4 / vm5 / vm6 / vm7 / vm8 / vm9 / vm10 / vm11 / vm12 / vm13 / vm14 / vm15, v[0..63]` |
