## 串行：数值检查通过；每遍 103127 个周期
  DMA 指令 2 条，vwait.ge 3 条，vld 64、vst 64
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [504, 509)
{ misc: vtrace 0x80000000 }
{ s0: simm.s32 s1, 0 }  # materialized constant
# entry bundle: %transform.1 = custom-call(%x_hbm.1)
{ s1: sld s6, [smem:0x1] }
{ s0: sne.s32 p0, s6, 0 }
{ s0: @p0 sbr.rel L_024e }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]
# ... 省略 runtime 代码 1 个 bundle
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [510, 589)
{ s0: simm.s32 s7, 0 }  # copy for cssa
L_01ff:
{ s0: sshll.u32 s8, s7, 0x9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:47:60-47:76 <- 58:24-58:37 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 311] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:47:60-47:76 <- 59:24-59:37 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 311] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:50:77-50:93 <- 61:24-61:38 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 311] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:50:77-50:93 <- 62:24-62:38 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 311]
  s1: simm.s32 s12, 0 ;  # materialized constant
  .encoding { s1.y = lo(imm1) } }
{ s0: sadd.s32 s11, s8, s0 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:58:24-58:45 <- 56:21-56:38 <- 53:13-53:32 [dma_start; LLO 20] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:59:24-59:44 <- 56:21-56:38 <- 53:13-53:32 [dma_wait; LLO 20]
{ s0: dma.simple [vmem:s12], [hbm:s11], length=512, dst_flag=[sflag:52] ;
  .encoding { s0.length = lo(imm0) } }
{ misc: vwait.ge [sflag:52], 512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:59:24-59:44 <- 56:21-56:38 <- 53:13-53:32 [dma_wait; LLO 326]
  .encoding { misc.value = imm2 } }
{ s0: sadd.s32 s15, s8, s1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:61:24-61:46 <- 56:21-56:38 <- 53:13-53:32 [dma_start; LLO 289] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:62:24-62:45 <- 56:21-56:38 <- 53:13-53:32 [dma_wait; LLO 289]
  s1: simm.s32 s16, 1024 ;  # materialized constant
  vld: vld.8x128 v0, [vmem:0x0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 33]
  misc: vsyncadd.s32 [sflag:52], -512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:59:24-59:44 <- 56:21-56:38 <- 53:13-53:32 [dma_wait; LLO 327]
  .encoding { misc.value = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 97]
  vld: vld.8x128 v2, [vmem:0x8] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 34]
{ va0: vmul.8x128.f32 v3, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 98]
  vld: vld.8x128 v4, [vmem:0x10] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 35]
{ va0: vmul.8x128.f32 v6, 2.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 99]
  va1: vadd.8x128.f32 v5, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 161]
  vld: vld.8x128 v7, [vmem:0x18] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 36]
{ va0: vmul.8x128.f32 v9, 2.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 100]
  va1: vadd.8x128.f32 v8, 1.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 162]
  vld: vld.8x128 v10, [vmem:0x20] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 37]
{ va0: vmul.8x128.f32 v12, 2.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 101]
  va1: vadd.8x128.f32 v11, 1.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 163]
  vst: vst.8x128 [vmem:0x400], v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 225]
  vld: vld.8x128 v13, [vmem:0x28] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 38]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v15, 2.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 102]
  va1: vadd.8x128.f32 v14, 1.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 164]
  vst: vst.8x128 [vmem:0x408], v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 226]
  vld: vld.8x128 v16, [vmem:0x30] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 39]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v18, 2.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 103]
  va1: vadd.8x128.f32 v17, 1.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 165]
  vst: vst.8x128 [vmem:0x410], v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 227]
  vld: vld.8x128 v19, [vmem:0x38] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 40]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v21, 2.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 104]
  va1: vadd.8x128.f32 v20, 1.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 166]
  vst: vst.8x128 [vmem:0x418], v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 228]
  vld: vld.8x128 v22, [vmem:0x40] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 41]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v24, 2.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 105]
  va1: vadd.8x128.f32 v23, 1.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 167]
  vst: vst.8x128 [vmem:0x420], v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 229]
  vld: vld.8x128 v25, [vmem:0x48] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 42]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v27, 2.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 106]
  va1: vadd.8x128.f32 v26, 1.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 168]
  vst: vst.8x128 [vmem:0x428], v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 230]
  vld: vld.8x128 v28, [vmem:0x50] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 43]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v30, 2.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 107]
  va1: vadd.8x128.f32 v29, 1.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 169]
  vst: vst.8x128 [vmem:0x430], v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 231]
  vld: vld.8x128 v31, [vmem:0x58] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 44]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 108]
  va1: vadd.8x128.f32 v0, 1.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 170]
  vst: vst.8x128 [vmem:0x438], v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 232]
  vld: vld.8x128 v2, [vmem:0x60] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 45]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v4, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 109]
  va1: vadd.8x128.f32 v3, 1.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 171]
  vst: vst.8x128 [vmem:0x440], v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 233]
  vld: vld.8x128 v5, [vmem:0x68] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 46]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v7, 2.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 110]
  va1: vadd.8x128.f32 v6, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 172]
  vst: vst.8x128 [vmem:0x448], v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 234]
  vld: vld.8x128 v8, [vmem:0x70] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 47]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v10, 2.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 111]
  va1: vadd.8x128.f32 v9, 1.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 173]
  vst: vst.8x128 [vmem:0x450], v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 235]
  vld: vld.8x128 v11, [vmem:0x78] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 48]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v13, 2.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 112]
  va1: vadd.8x128.f32 v12, 1.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 174]
  vst: vst.8x128 [vmem:0x458], v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 236]
  vld: vld.8x128 v14, [vmem:0x80] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 49]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v16, 2.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 113]
  va1: vadd.8x128.f32 v15, 1.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 175]
  vst: vst.8x128 [vmem:0x460], v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 237]
  vld: vld.8x128 v17, [vmem:0x88] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 50]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v19, 2.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 114]
  va1: vadd.8x128.f32 v18, 1.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 176]
  vst: vst.8x128 [vmem:0x468], v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 238]
  vld: vld.8x128 v20, [vmem:0x90] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 51]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v22, 2.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 115]
  va1: vadd.8x128.f32 v21, 1.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 177]
  vst: vst.8x128 [vmem:0x470], v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 239]
  vld: vld.8x128 v23, [vmem:0x98] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 52]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v25, 2.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 116]
  va1: vadd.8x128.f32 v24, 1.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 178]
  vst: vst.8x128 [vmem:0x478], v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 240]
  vld: vld.8x128 v26, [vmem:0xa0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 53]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v28, 2.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 117]
  va1: vadd.8x128.f32 v27, 1.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 179]
  vst: vst.8x128 [vmem:0x480], v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 241]
  vld: vld.8x128 v29, [vmem:0xa8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 54]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v31, 2.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 118]
  va1: vadd.8x128.f32 v30, 1.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 180]
  vst: vst.8x128 [vmem:0x488], v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 242]
  vld: vld.8x128 v0, [vmem:0xb0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 55]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v2, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 119]
  va1: vadd.8x128.f32 v1, 1.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 181]
  vst: vst.8x128 [vmem:0x490], v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 243]
  vld: vld.8x128 v3, [vmem:0xb8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 56]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v5, 2.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 120]
  va1: vadd.8x128.f32 v4, 1.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 182]
  vst: vst.8x128 [vmem:0x498], v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 244]
  vld: vld.8x128 v6, [vmem:0xc0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 57]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v8, 2.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 121]
  va1: vadd.8x128.f32 v7, 1.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 183]
  vst: vst.8x128 [vmem:0x4a0], v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 245]
  vld: vld.8x128 v9, [vmem:0xc8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 58]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v11, 2.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 122]
  va1: vadd.8x128.f32 v10, 1.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 184]
  vst: vst.8x128 [vmem:0x4a8], v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 246]
  vld: vld.8x128 v12, [vmem:0xd0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 59]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v14, 2.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 123]
  va1: vadd.8x128.f32 v13, 1.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 185]
  vst: vst.8x128 [vmem:0x4b0], v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 247]
  vld: vld.8x128 v15, [vmem:0xd8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 60]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v17, 2.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 124]
  va1: vadd.8x128.f32 v16, 1.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 186]
  vst: vst.8x128 [vmem:0x4b8], v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 248]
  vld: vld.8x128 v18, [vmem:0xe0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 61]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v20, 2.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 125]
  va1: vadd.8x128.f32 v19, 1.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 187]
  vst: vst.8x128 [vmem:0x4c0], v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 249]
  vld: vld.8x128 v21, [vmem:0xe8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 62]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v23, 2.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 126]
  va1: vadd.8x128.f32 v22, 1.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 188]
  vst: vst.8x128 [vmem:0x4c8], v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 250]
  vld: vld.8x128 v24, [vmem:0xf0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 63]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v26, 2.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 127]
  va1: vadd.8x128.f32 v25, 1.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 189]
  vst: vst.8x128 [vmem:0x4d0], v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 251]
  vld: vld.8x128 v27, [vmem:0xf8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 64]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v29, 2.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 128]
  va1: vadd.8x128.f32 v28, 1.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 190]
  vst: vst.8x128 [vmem:0x4d8], v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 252]
  vld: vld.8x128 v30, [vmem:0x100] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 65]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v0, 2.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 129]
  va1: vadd.8x128.f32 v31, 1.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 191]
  vst: vst.8x128 [vmem:0x4e0], v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 253]
  vld: vld.8x128 v1, [vmem:0x108] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 66]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v3, 2.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 130]
  va1: vadd.8x128.f32 v2, 1.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 192]
  vst: vst.8x128 [vmem:0x4e8], v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 254]
  vld: vld.8x128 v4, [vmem:0x110] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 67]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v6, 2.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 131]
  va1: vadd.8x128.f32 v5, 1.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 193]
  vst: vst.8x128 [vmem:0x4f0], v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 255]
  vld: vld.8x128 v7, [vmem:0x118] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 68]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v9, 2.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 132]
  va1: vadd.8x128.f32 v8, 1.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 194]
  vst: vst.8x128 [vmem:0x4f8], v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 256]
  vld: vld.8x128 v10, [vmem:0x120] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 69]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v12, 2.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 133]
  va1: vadd.8x128.f32 v11, 1.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 195]
  vst: vst.8x128 [vmem:0x500], v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 257]
  vld: vld.8x128 v13, [vmem:0x128] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 70]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v15, 2.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 134]
  va1: vadd.8x128.f32 v14, 1.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 196]
  vst: vst.8x128 [vmem:0x508], v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 258]
  vld: vld.8x128 v16, [vmem:0x130] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 71]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v18, 2.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 135]
  va1: vadd.8x128.f32 v17, 1.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 197]
  vst: vst.8x128 [vmem:0x510], v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 259]
  vld: vld.8x128 v19, [vmem:0x138] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 72]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v21, 2.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 136]
  va1: vadd.8x128.f32 v20, 1.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 198]
  vst: vst.8x128 [vmem:0x518], v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 260]
  vld: vld.8x128 v22, [vmem:0x140] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 73]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v24, 2.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 137]
  va1: vadd.8x128.f32 v23, 1.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 199]
  vst: vst.8x128 [vmem:0x520], v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 261]
  vld: vld.8x128 v25, [vmem:0x148] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 74]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v27, 2.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 138]
  va1: vadd.8x128.f32 v26, 1.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 200]
  vst: vst.8x128 [vmem:0x528], v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 262]
  vld: vld.8x128 v28, [vmem:0x150] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 75]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v30, 2.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 139]
  va1: vadd.8x128.f32 v29, 1.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 201]
  vst: vst.8x128 [vmem:0x530], v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 263]
  vld: vld.8x128 v31, [vmem:0x158] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 76]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 140]
  va1: vadd.8x128.f32 v0, 1.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 202]
  vst: vst.8x128 [vmem:0x538], v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 264]
  vld: vld.8x128 v2, [vmem:0x160] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 77]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v4, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 141]
  va1: vadd.8x128.f32 v3, 1.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 203]
  vst: vst.8x128 [vmem:0x540], v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 265]
  vld: vld.8x128 v5, [vmem:0x168] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 78]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v7, 2.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 142]
  va1: vadd.8x128.f32 v6, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 204]
  vst: vst.8x128 [vmem:0x548], v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 266]
  vld: vld.8x128 v8, [vmem:0x170] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 79]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v10, 2.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 143]
  va1: vadd.8x128.f32 v9, 1.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 205]
  vst: vst.8x128 [vmem:0x550], v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 267]
  vld: vld.8x128 v11, [vmem:0x178] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 80]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v13, 2.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 144]
  va1: vadd.8x128.f32 v12, 1.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 206]
  vst: vst.8x128 [vmem:0x558], v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 268]
  vld: vld.8x128 v14, [vmem:0x180] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 81]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v16, 2.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 145]
  va1: vadd.8x128.f32 v15, 1.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 207]
  vst: vst.8x128 [vmem:0x560], v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 269]
  vld: vld.8x128 v17, [vmem:0x188] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 82]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v19, 2.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 146]
  va1: vadd.8x128.f32 v18, 1.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 208]
  vst: vst.8x128 [vmem:0x568], v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 270]
  vld: vld.8x128 v20, [vmem:0x190] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 83]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v22, 2.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 147]
  va1: vadd.8x128.f32 v21, 1.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 209]
  vst: vst.8x128 [vmem:0x570], v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 271]
  vld: vld.8x128 v23, [vmem:0x198] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 84]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v25, 2.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 148]
  va1: vadd.8x128.f32 v24, 1.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 210]
  vst: vst.8x128 [vmem:0x578], v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 272]
  vld: vld.8x128 v26, [vmem:0x1a0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 85]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v28, 2.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 149]
  va1: vadd.8x128.f32 v27, 1.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 211]
  vst: vst.8x128 [vmem:0x580], v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 273]
  vld: vld.8x128 v29, [vmem:0x1a8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 86]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v31, 2.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 150]
  va1: vadd.8x128.f32 v30, 1.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 212]
  vst: vst.8x128 [vmem:0x588], v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 274]
  vld: vld.8x128 v0, [vmem:0x1b0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 87]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v2, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 151]
  va1: vadd.8x128.f32 v1, 1.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 213]
  vst: vst.8x128 [vmem:0x590], v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 275]
  vld: vld.8x128 v3, [vmem:0x1b8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 88]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v5, 2.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 152]
  va1: vadd.8x128.f32 v4, 1.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 214]
  vst: vst.8x128 [vmem:0x598], v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 276]
  vld: vld.8x128 v6, [vmem:0x1c0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 89]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v8, 2.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 153]
  va1: vadd.8x128.f32 v7, 1.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 215]
  vst: vst.8x128 [vmem:0x5a0], v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 277]
  vld: vld.8x128 v9, [vmem:0x1c8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 90]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v11, 2.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 154]
  va1: vadd.8x128.f32 v10, 1.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 216]
  vst: vst.8x128 [vmem:0x5a8], v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 278]
  vld: vld.8x128 v12, [vmem:0x1d0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 91]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v14, 2.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 155]
  va1: vadd.8x128.f32 v13, 1.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 217]
  vst: vst.8x128 [vmem:0x5b0], v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 279]
  vld: vld.8x128 v15, [vmem:0x1d8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 92]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v17, 2.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 156]
  va1: vadd.8x128.f32 v16, 1.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 218]
  vst: vst.8x128 [vmem:0x5b8], v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 280]
  vld: vld.8x128 v18, [vmem:0x1e0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 93]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v20, 2.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 157]
  va1: vadd.8x128.f32 v19, 1.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 219]
  vst: vst.8x128 [vmem:0x5c0], v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 281]
  vld: vld.8x128 v21, [vmem:0x1e8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 94]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v23, 2.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 158]
  va1: vadd.8x128.f32 v22, 1.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 220]
  vst: vst.8x128 [vmem:0x5c8], v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 282]
  vld: vld.8x128 v24, [vmem:0x1f0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 95]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v26, 2.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 159]
  va1: vadd.8x128.f32 v25, 1.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 221]
  vst: vst.8x128 [vmem:0x5d0], v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 283]
  vld: vld.8x128 v27, [vmem:0x1f8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:45 <- 56:21-56:38 <- 53:13-53:32 [get; LLO 96]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v29, 2.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:51 <- 56:21-56:38 <- 53:13-53:32 [mul; LLO 160]
  va1: vadd.8x128.f32 v28, 1.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 222]
  vst: vst.8x128 [vmem:0x5d8], v22 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 284]
{ va1: vadd.8x128.f32 v30, 1.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 223]
  vst: vst.8x128 [vmem:0x5e0], v25 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 285]
{ va1: vadd.8x128.f32 v31, 1.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:36-60:57 <- 56:21-56:38 <- 53:13-53:32 [add; LLO 224]
  vst: vst.8x128 [vmem:0x5e8], v28 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 286]
{ vst: vst.8x128 [vmem:0x5f0], v30 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 287]
{ vst: vst.8x128 [vmem:0x5f8], v31 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:60:24-60:33 <- 56:21-56:38 <- 53:13-53:32 [swap; LLO 288]
{ s0: dma.simple [hbm:s15], [vmem:s16], length=512, dst_flag=[sflag:54] ;
  .encoding { s0.length = lo(imm0) } }
{ misc: vwait.ge [sflag:54], 512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:62:24-62:45 <- 56:21-56:38 <- 53:13-53:32 [dma_wait; LLO 328]
  .encoding { misc.value = imm2 } }
{ s0: sadd.s32 s7, 1, s7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:56:21-56:38 | copy for cssa
  misc: vsyncadd.s32 [sflag:54], -512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:62:24-62:45 <- 56:21-56:38 <- 53:13-53:32 [dma_wait; LLO 329]
  .encoding { misc.value = imm2 } }
{ s0: sge.s32 p1, s7, 64 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:56:21-56:38 | loop exit test
{ s0: @!p1 sbr.rel L_01ff }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:56:21-56:38
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]
# ... 省略 runtime 代码 2 个 bundle
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [591, 607)
{ s1: sld s20, [smem:0x0] }
{ s1: sld s21, [smem:0x1] }
{ s0: sand.u32 s24, 0xfff, s20 }
{ s0: sadd.s32 s22, 1, s21 ;
  s1: sshll.u32 s27, s24, 0x12 }
{ s0: sand.u32 s23, 0x1, s22 }  # smod.u32 w/div 2
{ s0: sand.u32 s25, 0x3, s23 }
{ s0: sshll.u32 s26, s25, 0x10 }
{ s0: sadd.s32 s28, 131072, s26 }
{ s0: sshrl.u32 s29, s28, 0x2 }
{ s0: sor.u32 s30, 0x2d, s29 }
{ s0: sor.u32 s17, s30, s27 }
{ misc: vsyncadd.remote.s32 [sflag:s17], 1 }
{ misc: vwait.ge [sflag:45], 1 ;
  .encoding { misc.value = imm2 } }
{ misc: vsyncadd.s32 [sflag:45], -1 ;
  .encoding { misc.value = imm2 } }
{ misc: cfence }
{ misc: vtrace 0x90000000 }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]

## 输入双缓冲：数值检查通过；每遍 67603 个周期
  DMA 指令 3 条，vwait.ge 3 条，vld 64、vst 64
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [504, 509)
{ misc: vtrace 0x80000000 }
{ s0: simm.s32 s1, 0 }  # materialized constant
# entry bundle: %transform.1 = custom-call(%x_hbm.1)
{ s1: sld s6, [smem:0x1] }
{ s0: sne.s32 p0, s6, 0 }
{ s0: @p0 sbr.rel L_0259 }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]
# ... 省略 runtime 代码 1 个 bundle
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [510, 600)
{ s0: simm.s32 s7, 0 ;  # materialized constant
  s1: simm.s32 s10, 0 ;  # copy for cssa
  .encoding { s0.y = lo(imm0) ; s1.y = const(0) } }
{ s0: dma.simple [vmem:s7], [hbm:s0], length=512, dst_flag=[sflag:52] ;
  .encoding { s0.length = lo(imm0) } }
L_0200:
{ s0: slt.s32 p1, s10, 0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 30]
  s1: ssub.s32 s11, 0, s10 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 31]
{ s0: @!p1 smov s11, s10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35
  s1: sshll.u32 s12, s10, 0x9 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:47:60-47:76 <- 73:20-73:36 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 355] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:50:77-50:93 <- 81:24-81:38 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 355] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:50:77-50:93 <- 82:24-82:38 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 355] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56
{ s0: sand.u32 s13, 0x1, s11 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 33] | smod.u32 w/div 2
{ s0: ssub.s32 s14, 0, s13 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 34]
{ s0: @!p1 smov s14, s13 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35
{ s0: slt.s32 p2, s14, 0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [ne, lt, and; LLO 354]
  s1: sadd.s32 s15, 2, s14 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 40]
{ s0: @!p2 smov s15, s14 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35
{ s0: sshll.u32 s16, s15, 0x9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 356] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 356]
  s1: sadd.s32 s17, 52, s15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 43]
  .encoding { s1.y = lo(imm1) } }
{ s0: sadd.s32 s18, 0, s16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43
  .encoding { s0.y = lo(imm0) } }
{ misc: vwait.ge [sflag:s17], 512 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 377]
{ s0: sadd.s32 s19, 1, s10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:69:17-69:34 | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:37 | copy for cssa
  s1: ssub.s32 s20, 1, s15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:39-77:47 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [sub; LLO 52]
  misc: vsyncadd.s32 [sflag:s17], -512 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 378]
{ s0: sge.s32 p3, s19, 64 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:45 <- 69:17-69:34 <- 53:13-53:32 [lt; LLO 357] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:21-75:46 | loop exit test
  s1: slt.s32 p4, s19, 64 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:45 <- 69:17-69:34 <- 53:13-53:32 [lt; LLO 362]
{ s0: sshll.u32 s21, s20, 0x9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 358]
  s1: sadd.s32 s24, s12, s0 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56
{ s0: sadd.s32 s25, 512, s24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56
  s1: sadd.s32 s26, 0, s21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 56]
  .encoding { s1.y = lo(imm1) } }
{ s0: sadd.s32 s27, 52, s20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 54]
  s1: sadd.s32 s30, s12, s1 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:81:24-81:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 324] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:82:24-82:45 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 324]
{ s0: @p4 dma.simple [vmem:s26], [hbm:s25], length=512, dst_flag=[sflag:s27] }
{ s0: simm.s32 s2, 1024 ;  # materialized constant
  vld: vld.8x128 v0, [vmem:s18] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 68]
{ va0: vmul.8x128.f32 v1, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 132]
  vld: vld.8x128 v2, [vmem:s18 + 0x8] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 69]
{ va0: vmul.8x128.f32 v3, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 133]
  vld: vld.8x128 v4, [vmem:s18 + 0x10] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 70]
{ va0: vmul.8x128.f32 v6, 2.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 134]
  va1: vadd.8x128.f32 v5, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 196]
  vld: vld.8x128 v7, [vmem:s18 + 0x18] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 71]
{ va0: vmul.8x128.f32 v9, 2.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 135]
  va1: vadd.8x128.f32 v8, 1.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 197]
  vld: vld.8x128 v10, [vmem:s18 + 0x20] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 72]
{ va0: vmul.8x128.f32 v12, 2.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 136]
  va1: vadd.8x128.f32 v11, 1.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 198]
  vst: vst.8x128 [vmem:0x400], v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 260]
  vld: vld.8x128 v13, [vmem:s18 + 0x28] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 73]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v15, 2.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 137]
  va1: vadd.8x128.f32 v14, 1.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 199]
  vst: vst.8x128 [vmem:0x408], v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 261]
  vld: vld.8x128 v16, [vmem:s18 + 0x30] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 74]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v18, 2.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 138]
  va1: vadd.8x128.f32 v17, 1.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 200]
  vst: vst.8x128 [vmem:0x410], v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 262]
  vld: vld.8x128 v19, [vmem:s18 + 0x38] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 75]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v21, 2.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 139]
  va1: vadd.8x128.f32 v20, 1.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 201]
  vst: vst.8x128 [vmem:0x418], v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 263]
  vld: vld.8x128 v22, [vmem:s18 + 0x40] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 76]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v24, 2.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 140]
  va1: vadd.8x128.f32 v23, 1.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 202]
  vst: vst.8x128 [vmem:0x420], v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 264]
  vld: vld.8x128 v25, [vmem:s18 + 0x48] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 77]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v27, 2.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 141]
  va1: vadd.8x128.f32 v26, 1.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 203]
  vst: vst.8x128 [vmem:0x428], v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 265]
  vld: vld.8x128 v28, [vmem:s18 + 0x50] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 78]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v30, 2.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 142]
  va1: vadd.8x128.f32 v29, 1.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 204]
  vst: vst.8x128 [vmem:0x430], v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 266]
  vld: vld.8x128 v31, [vmem:s18 + 0x58] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 79]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 143]
  va1: vadd.8x128.f32 v0, 1.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 205]
  vst: vst.8x128 [vmem:0x438], v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 267]
  vld: vld.8x128 v2, [vmem:s18 + 0x60] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 80]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v4, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 144]
  va1: vadd.8x128.f32 v3, 1.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 206]
  vst: vst.8x128 [vmem:0x440], v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 268]
  vld: vld.8x128 v5, [vmem:s18 + 0x68] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 81]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v7, 2.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 145]
  va1: vadd.8x128.f32 v6, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 207]
  vst: vst.8x128 [vmem:0x448], v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 269]
  vld: vld.8x128 v8, [vmem:s18 + 0x70] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 82]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v10, 2.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 146]
  va1: vadd.8x128.f32 v9, 1.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 208]
  vst: vst.8x128 [vmem:0x450], v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 270]
  vld: vld.8x128 v11, [vmem:s18 + 0x78] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 83]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v13, 2.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 147]
  va1: vadd.8x128.f32 v12, 1.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 209]
  vst: vst.8x128 [vmem:0x458], v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 271]
  vld: vld.8x128 v14, [vmem:s18 + 0x80] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 84]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v16, 2.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 148]
  va1: vadd.8x128.f32 v15, 1.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 210]
  vst: vst.8x128 [vmem:0x460], v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 272]
  vld: vld.8x128 v17, [vmem:s18 + 0x88] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 85]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v19, 2.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 149]
  va1: vadd.8x128.f32 v18, 1.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 211]
  vst: vst.8x128 [vmem:0x468], v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 273]
  vld: vld.8x128 v20, [vmem:s18 + 0x90] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 86]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v22, 2.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 150]
  va1: vadd.8x128.f32 v21, 1.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 212]
  vst: vst.8x128 [vmem:0x470], v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 274]
  vld: vld.8x128 v23, [vmem:s18 + 0x98] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 87]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v25, 2.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 151]
  va1: vadd.8x128.f32 v24, 1.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 213]
  vst: vst.8x128 [vmem:0x478], v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 275]
  vld: vld.8x128 v26, [vmem:s18 + 0xa0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 88]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v28, 2.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 152]
  va1: vadd.8x128.f32 v27, 1.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 214]
  vst: vst.8x128 [vmem:0x480], v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 276]
  vld: vld.8x128 v29, [vmem:s18 + 0xa8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 89]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v31, 2.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 153]
  va1: vadd.8x128.f32 v30, 1.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 215]
  vst: vst.8x128 [vmem:0x488], v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 277]
  vld: vld.8x128 v0, [vmem:s18 + 0xb0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 90]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v2, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 154]
  va1: vadd.8x128.f32 v1, 1.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 216]
  vst: vst.8x128 [vmem:0x490], v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 278]
  vld: vld.8x128 v3, [vmem:s18 + 0xb8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 91]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v5, 2.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 155]
  va1: vadd.8x128.f32 v4, 1.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 217]
  vst: vst.8x128 [vmem:0x498], v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 279]
  vld: vld.8x128 v6, [vmem:s18 + 0xc0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 92]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v8, 2.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 156]
  va1: vadd.8x128.f32 v7, 1.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 218]
  vst: vst.8x128 [vmem:0x4a0], v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 280]
  vld: vld.8x128 v9, [vmem:s18 + 0xc8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 93]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v11, 2.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 157]
  va1: vadd.8x128.f32 v10, 1.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 219]
  vst: vst.8x128 [vmem:0x4a8], v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 281]
  vld: vld.8x128 v12, [vmem:s18 + 0xd0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 94]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v14, 2.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 158]
  va1: vadd.8x128.f32 v13, 1.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 220]
  vst: vst.8x128 [vmem:0x4b0], v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 282]
  vld: vld.8x128 v15, [vmem:s18 + 0xd8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 95]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v17, 2.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 159]
  va1: vadd.8x128.f32 v16, 1.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 221]
  vst: vst.8x128 [vmem:0x4b8], v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 283]
  vld: vld.8x128 v18, [vmem:s18 + 0xe0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 96]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v20, 2.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 160]
  va1: vadd.8x128.f32 v19, 1.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 222]
  vst: vst.8x128 [vmem:0x4c0], v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 284]
  vld: vld.8x128 v21, [vmem:s18 + 0xe8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 97]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v23, 2.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 161]
  va1: vadd.8x128.f32 v22, 1.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 223]
  vst: vst.8x128 [vmem:0x4c8], v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 285]
  vld: vld.8x128 v24, [vmem:s18 + 0xf0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 98]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v26, 2.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 162]
  va1: vadd.8x128.f32 v25, 1.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 224]
  vst: vst.8x128 [vmem:0x4d0], v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 286]
  vld: vld.8x128 v27, [vmem:s18 + 0xf8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 99]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v29, 2.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 163]
  va1: vadd.8x128.f32 v28, 1.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 225]
  vst: vst.8x128 [vmem:0x4d8], v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 287]
  vld: vld.8x128 v30, [vmem:s18 + 0x100] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 100]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v0, 2.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 164]
  va1: vadd.8x128.f32 v31, 1.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 226]
  vst: vst.8x128 [vmem:0x4e0], v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 288]
  vld: vld.8x128 v1, [vmem:s18 + 0x108] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 101]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v3, 2.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 165]
  va1: vadd.8x128.f32 v2, 1.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 227]
  vst: vst.8x128 [vmem:0x4e8], v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 289]
  vld: vld.8x128 v4, [vmem:s18 + 0x110] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 102]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v6, 2.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 166]
  va1: vadd.8x128.f32 v5, 1.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 228]
  vst: vst.8x128 [vmem:0x4f0], v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 290]
  vld: vld.8x128 v7, [vmem:s18 + 0x118] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 103]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v9, 2.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 167]
  va1: vadd.8x128.f32 v8, 1.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 229]
  vst: vst.8x128 [vmem:0x4f8], v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 291]
  vld: vld.8x128 v10, [vmem:s18 + 0x120] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 104]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v12, 2.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 168]
  va1: vadd.8x128.f32 v11, 1.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 230]
  vst: vst.8x128 [vmem:0x500], v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 292]
  vld: vld.8x128 v13, [vmem:s18 + 0x128] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 105]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v15, 2.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 169]
  va1: vadd.8x128.f32 v14, 1.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 231]
  vst: vst.8x128 [vmem:0x508], v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 293]
  vld: vld.8x128 v16, [vmem:s18 + 0x130] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 106]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v18, 2.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 170]
  va1: vadd.8x128.f32 v17, 1.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 232]
  vst: vst.8x128 [vmem:0x510], v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 294]
  vld: vld.8x128 v19, [vmem:s18 + 0x138] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 107]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v21, 2.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 171]
  va1: vadd.8x128.f32 v20, 1.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 233]
  vst: vst.8x128 [vmem:0x518], v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 295]
  vld: vld.8x128 v22, [vmem:s18 + 0x140] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 108]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v24, 2.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 172]
  va1: vadd.8x128.f32 v23, 1.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 234]
  vst: vst.8x128 [vmem:0x520], v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 296]
  vld: vld.8x128 v25, [vmem:s18 + 0x148] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 109]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v27, 2.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 173]
  va1: vadd.8x128.f32 v26, 1.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 235]
  vst: vst.8x128 [vmem:0x528], v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 297]
  vld: vld.8x128 v28, [vmem:s18 + 0x150] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 110]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v30, 2.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 174]
  va1: vadd.8x128.f32 v29, 1.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 236]
  vst: vst.8x128 [vmem:0x530], v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 298]
  vld: vld.8x128 v31, [vmem:s18 + 0x158] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 111]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 175]
  va1: vadd.8x128.f32 v0, 1.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 237]
  vst: vst.8x128 [vmem:0x538], v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 299]
  vld: vld.8x128 v2, [vmem:s18 + 0x160] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 112]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v4, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 176]
  va1: vadd.8x128.f32 v3, 1.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 238]
  vst: vst.8x128 [vmem:0x540], v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 300]
  vld: vld.8x128 v5, [vmem:s18 + 0x168] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 113]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v7, 2.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 177]
  va1: vadd.8x128.f32 v6, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 239]
  vst: vst.8x128 [vmem:0x548], v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 301]
  vld: vld.8x128 v8, [vmem:s18 + 0x170] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 114]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v10, 2.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 178]
  va1: vadd.8x128.f32 v9, 1.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 240]
  vst: vst.8x128 [vmem:0x550], v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 302]
  vld: vld.8x128 v11, [vmem:s18 + 0x178] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 115]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v13, 2.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 179]
  va1: vadd.8x128.f32 v12, 1.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 241]
  vst: vst.8x128 [vmem:0x558], v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 303]
  vld: vld.8x128 v14, [vmem:s18 + 0x180] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 116]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v16, 2.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 180]
  va1: vadd.8x128.f32 v15, 1.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 242]
  vst: vst.8x128 [vmem:0x560], v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 304]
  vld: vld.8x128 v17, [vmem:s18 + 0x188] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 117]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v19, 2.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 181]
  va1: vadd.8x128.f32 v18, 1.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 243]
  vst: vst.8x128 [vmem:0x568], v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 305]
  vld: vld.8x128 v20, [vmem:s18 + 0x190] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 118]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v22, 2.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 182]
  va1: vadd.8x128.f32 v21, 1.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 244]
  vst: vst.8x128 [vmem:0x570], v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 306]
  vld: vld.8x128 v23, [vmem:s18 + 0x198] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 119]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v25, 2.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 183]
  va1: vadd.8x128.f32 v24, 1.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 245]
  vst: vst.8x128 [vmem:0x578], v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 307]
  vld: vld.8x128 v26, [vmem:s18 + 0x1a0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 120]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v28, 2.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 184]
  va1: vadd.8x128.f32 v27, 1.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 246]
  vst: vst.8x128 [vmem:0x580], v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 308]
  vld: vld.8x128 v29, [vmem:s18 + 0x1a8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 121]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v31, 2.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 185]
  va1: vadd.8x128.f32 v30, 1.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 247]
  vst: vst.8x128 [vmem:0x588], v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 309]
  vld: vld.8x128 v0, [vmem:s18 + 0x1b0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 122]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v2, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 186]
  va1: vadd.8x128.f32 v1, 1.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 248]
  vst: vst.8x128 [vmem:0x590], v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 310]
  vld: vld.8x128 v3, [vmem:s18 + 0x1b8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 123]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v5, 2.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 187]
  va1: vadd.8x128.f32 v4, 1.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 249]
  vst: vst.8x128 [vmem:0x598], v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 311]
  vld: vld.8x128 v6, [vmem:s18 + 0x1c0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 124]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v8, 2.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 188]
  va1: vadd.8x128.f32 v7, 1.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 250]
  vst: vst.8x128 [vmem:0x5a0], v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 312]
  vld: vld.8x128 v9, [vmem:s18 + 0x1c8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 125]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v11, 2.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 189]
  va1: vadd.8x128.f32 v10, 1.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 251]
  vst: vst.8x128 [vmem:0x5a8], v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 313]
  vld: vld.8x128 v12, [vmem:s18 + 0x1d0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 126]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v14, 2.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 190]
  va1: vadd.8x128.f32 v13, 1.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 252]
  vst: vst.8x128 [vmem:0x5b0], v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 314]
  vld: vld.8x128 v15, [vmem:s18 + 0x1d8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 127]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v17, 2.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 191]
  va1: vadd.8x128.f32 v16, 1.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 253]
  vst: vst.8x128 [vmem:0x5b8], v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 315]
  vld: vld.8x128 v18, [vmem:s18 + 0x1e0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 128]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v20, 2.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 192]
  va1: vadd.8x128.f32 v19, 1.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 254]
  vst: vst.8x128 [vmem:0x5c0], v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 316]
  vld: vld.8x128 v21, [vmem:s18 + 0x1e8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 129]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v23, 2.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 193]
  va1: vadd.8x128.f32 v22, 1.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 255]
  vst: vst.8x128 [vmem:0x5c8], v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 317]
  vld: vld.8x128 v24, [vmem:s18 + 0x1f0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 130]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v26, 2.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 194]
  va1: vadd.8x128.f32 v25, 1.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 256]
  vst: vst.8x128 [vmem:0x5d0], v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 318]
  vld: vld.8x128 v27, [vmem:s18 + 0x1f8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:48 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 131]
  .encoding { vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v29, 2.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:54 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 195]
  va1: vadd.8x128.f32 v28, 1.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 257]
  vst: vst.8x128 [vmem:0x5d8], v22 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 319]
{ va1: vadd.8x128.f32 v30, 1.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 258]
  vst: vst.8x128 [vmem:0x5e0], v25 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 320]
{ va1: vadd.8x128.f32 v31, 1.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:36-80:60 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 259]
  vst: vst.8x128 [vmem:0x5e8], v28 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 321]
{ vst: vst.8x128 [vmem:0x5f0], v30 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 322]
{ vst: vst.8x128 [vmem:0x5f8], v31 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:80:24-80:33 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 323]
{ s0: dma.simple [hbm:s30], [vmem:s2], length=512, dst_flag=[sflag:54] ;
  .encoding { s0.length = lo(imm0) } }
{ misc: vwait.ge [sflag:54], 512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:82:24-82:45 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 379]
  .encoding { misc.value = imm2 } }
{ s0: @!p3 sbr.rel L_0200 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:69:17-69:34
  s1: smov s10, s19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:69:17-69:34 | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:37 | copy for cssa
  misc: vsyncadd.s32 [sflag:54], -512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:82:24-82:45 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 380]
  .encoding { misc.value = imm2 } }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]
# ... 省略 runtime 代码 2 个 bundle
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [602, 618)
{ s1: sld s20, [smem:0x0] }
{ s1: sld s21, [smem:0x1] }
{ s0: sand.u32 s24, 0xfff, s20 }
{ s0: sadd.s32 s22, 1, s21 ;
  s1: sshll.u32 s27, s24, 0x12 }
{ s0: sand.u32 s23, 0x1, s22 }  # smod.u32 w/div 2
{ s0: sand.u32 s25, 0x3, s23 }
{ s0: sshll.u32 s26, s25, 0x10 }
{ s0: sadd.s32 s28, 131072, s26 }
{ s0: sshrl.u32 s29, s28, 0x2 }
{ s0: sor.u32 s30, 0x2d, s29 }
{ s0: sor.u32 s17, s30, s27 }
{ misc: vsyncadd.remote.s32 [sflag:s17], 1 }
{ misc: vwait.ge [sflag:45], 1 ;
  .encoding { misc.value = imm2 } }
{ misc: vsyncadd.s32 [sflag:45], -1 ;
  .encoding { misc.value = imm2 } }
{ misc: cfence }
{ misc: vtrace 0x90000000 }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]

## 输入输出双缓冲：数值检查通过；每遍 67653 个周期
  DMA 指令 3 条，vwait.ge 5 条，vld 64、vst 64
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [504, 509)
{ misc: vtrace 0x80000000 }
{ s0: simm.s32 s1, 0 }  # materialized constant
# entry bundle: %transform.1 = custom-call(%x_hbm.1)
{ s1: sld s6, [smem:0x1] }
{ s0: sne.s32 p0, s6, 0 }
{ s0: @p0 sbr.rel L_025c }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]
# ... 省略 runtime 代码 1 个 bundle
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [510, 604)
{ s0: simm.s32 s7, 0 ;  # materialized constant
  s1: simm.s32 s10, 0 ;  # copy for cssa
  .encoding { s0.y = lo(imm0) ; s1.y = const(0) } }
{ s0: dma.simple [vmem:s7], [hbm:s0], length=512, dst_flag=[sflag:52] ;
  .encoding { s0.length = lo(imm0) } }
L_0200:
{ s0: slt.s32 p1, s10, 0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 30]
  s1: ssub.s32 s11, 0, s10 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 31]
{ s0: @!p1 smov s11, s10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35
  s1: sshll.u32 s12, s10, 0x9 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:47:60-47:76 <- 73:20-73:36 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 369] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:50:77-50:93 <- 90:24-90:41 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 369] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56
{ s0: sand.u32 s13, 0x1, s11 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 33] | smod.u32 w/div 2
{ s0: ssub.s32 s14, 0, s13 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [rem; LLO 34]
{ s0: @!p1 smov s14, s13 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35
{ s0: slt.s32 p2, s14, 0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [ne, lt, and; LLO 368]
  s1: sadd.s32 s15, 2, s14 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 40]
{ s0: @!p2 smov s15, s14 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:71:27-71:35
{ s0: sshll.u32 s16, s15, 0x9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:90:24-90:49 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 370] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 370] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 370] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 370]
  s1: sadd.s32 s17, 52, s15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 43]
  .encoding { s1.y = lo(imm1) } }
{ s0: sadd.s32 s18, 0, s16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43
  .encoding { s0.y = lo(imm0) } }
{ misc: vwait.ge [sflag:s17], 512 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 394]
{ s0: sadd.s32 s19, 1, s10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:69:17-69:34 | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:37 | copy for cssa
  s1: ssub.s32 s20, 1, s15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:39-77:47 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [sub; LLO 52]
  misc: vsyncadd.s32 [sflag:s17], -512 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:73:20-73:43 <- 69:17-69:34 <- 53:13-53:32 [dma_wait; LLO 395]
{ s0: sge.s32 p3, s19, 64 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:45 <- 69:17-69:34 <- 53:13-53:32 [lt; LLO 371] | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:21-75:46 | loop exit test
  s1: slt.s32 p4, s19, 64 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:45 <- 69:17-69:34 <- 53:13-53:32 [lt; LLO 379]
{ s0: sshll.u32 s21, s20, 0x9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 372]
  s1: sadd.s32 s24, s12, s0 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56
{ s0: sadd.s32 s25, 512, s24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56
  s1: sge.s32 p5, s10, 2 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:85:33-85:42 <- 69:17-69:34 <- 53:13-53:32 [ge; LLO 380]
{ s0: sadd.s32 s26, 0, s21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 56]
  s1: sadd.s32 s27, 52, s20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:77:24-77:56 <- 75:21-75:46 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 54]
  .encoding { s0.y = lo(imm0) } }
{ s0: @p4 dma.simple [vmem:s26], [hbm:s25], length=512, dst_flag=[sflag:s27] }
{ s0: sadd.s32 s28, 54, s15 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:90:24-90:49 | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:87:28-87:56
{ misc: @p5 vwait.ge [sflag:s28], 512 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:87:28-87:56
{ s0: sadd.s32 s29, 1024, s16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36
  s1: sadd.s32 s3, s12, s1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:90:24-90:49 <- 69:17-69:34 <- 53:13-53:32 [dma_start; LLO 332]
  vld: vld.8x128 v0, [vmem:s18] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 74]
  misc: @p5 vsyncadd.s32 [sflag:s28], -512 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:87:28-87:56
{ s0: smov s10, s19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:69:17-69:34 | chapter02/02_dma_pipeline/01_pallas_double_buffer.py:75:29-75:37 | copy for cssa
  va0: vmul.8x128.f32 v1, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 138]
  vld: vld.8x128 v2, [vmem:s18 + 0x8] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 75]
{ va0: vmul.8x128.f32 v3, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 139]
  vld: vld.8x128 v4, [vmem:s18 + 0x10] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 76]
{ va0: vmul.8x128.f32 v6, 2.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 140]
  va1: vadd.8x128.f32 v5, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 202]
  vld: vld.8x128 v7, [vmem:s18 + 0x18] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 77]
{ va0: vmul.8x128.f32 v9, 2.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 141]
  va1: vadd.8x128.f32 v8, 1.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 203]
  vld: vld.8x128 v10, [vmem:s18 + 0x20] }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 78]
{ va0: vmul.8x128.f32 v12, 2.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 142]
  va1: vadd.8x128.f32 v11, 1.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 204]
  vst: vst.8x128 [vmem:s29], v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 267]
  vld: vld.8x128 v13, [vmem:s18 + 0x28] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 79]
  .encoding { vst.base = vs0 } }
{ va0: vmul.8x128.f32 v15, 2.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 143]
  va1: vadd.8x128.f32 v14, 1.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 205]
  vst: vst.8x128 [vmem:s29 + 0x8], v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 268]
  vld: vld.8x128 v16, [vmem:s18 + 0x30] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 80]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v18, 2.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 144]
  va1: vadd.8x128.f32 v17, 1.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 206]
  vst: vst.8x128 [vmem:s29 + 0x10], v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 269]
  vld: vld.8x128 v19, [vmem:s18 + 0x38] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 81]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v21, 2.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 145]
  va1: vadd.8x128.f32 v20, 1.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 207]
  vst: vst.8x128 [vmem:s29 + 0x18], v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 270]
  vld: vld.8x128 v22, [vmem:s18 + 0x40] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 82]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v24, 2.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 146]
  va1: vadd.8x128.f32 v23, 1.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 208]
  vst: vst.8x128 [vmem:s29 + 0x20], v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 271]
  vld: vld.8x128 v25, [vmem:s18 + 0x48] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 83]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v27, 2.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 147]
  va1: vadd.8x128.f32 v26, 1.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 209]
  vst: vst.8x128 [vmem:s29 + 0x28], v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 272]
  vld: vld.8x128 v28, [vmem:s18 + 0x50] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 84]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v30, 2.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 148]
  va1: vadd.8x128.f32 v29, 1.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 210]
  vst: vst.8x128 [vmem:s29 + 0x30], v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 273]
  vld: vld.8x128 v31, [vmem:s18 + 0x58] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 85]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 149]
  va1: vadd.8x128.f32 v0, 1.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 211]
  vst: vst.8x128 [vmem:s29 + 0x38], v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 274]
  vld: vld.8x128 v2, [vmem:s18 + 0x60] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 86]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v4, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 150]
  va1: vadd.8x128.f32 v3, 1.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 212]
  vst: vst.8x128 [vmem:s29 + 0x40], v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 275]
  vld: vld.8x128 v5, [vmem:s18 + 0x68] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 87]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v7, 2.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 151]
  va1: vadd.8x128.f32 v6, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 213]
  vst: vst.8x128 [vmem:s29 + 0x48], v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 276]
  vld: vld.8x128 v8, [vmem:s18 + 0x70] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 88]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v10, 2.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 152]
  va1: vadd.8x128.f32 v9, 1.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 214]
  vst: vst.8x128 [vmem:s29 + 0x50], v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 277]
  vld: vld.8x128 v11, [vmem:s18 + 0x78] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 89]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v13, 2.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 153]
  va1: vadd.8x128.f32 v12, 1.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 215]
  vst: vst.8x128 [vmem:s29 + 0x58], v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 278]
  vld: vld.8x128 v14, [vmem:s18 + 0x80] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 90]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v16, 2.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 154]
  va1: vadd.8x128.f32 v15, 1.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 216]
  vst: vst.8x128 [vmem:s29 + 0x60], v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 279]
  vld: vld.8x128 v17, [vmem:s18 + 0x88] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 91]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v19, 2.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 155]
  va1: vadd.8x128.f32 v18, 1.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 217]
  vst: vst.8x128 [vmem:s29 + 0x68], v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 280]
  vld: vld.8x128 v20, [vmem:s18 + 0x90] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 92]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v22, 2.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 156]
  va1: vadd.8x128.f32 v21, 1.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 218]
  vst: vst.8x128 [vmem:s29 + 0x70], v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 281]
  vld: vld.8x128 v23, [vmem:s18 + 0x98] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 93]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v25, 2.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 157]
  va1: vadd.8x128.f32 v24, 1.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 219]
  vst: vst.8x128 [vmem:s29 + 0x78], v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 282]
  vld: vld.8x128 v26, [vmem:s18 + 0xa0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 94]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v28, 2.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 158]
  va1: vadd.8x128.f32 v27, 1.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 220]
  vst: vst.8x128 [vmem:s29 + 0x80], v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 283]
  vld: vld.8x128 v29, [vmem:s18 + 0xa8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 95]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v31, 2.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 159]
  va1: vadd.8x128.f32 v30, 1.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 221]
  vst: vst.8x128 [vmem:s29 + 0x88], v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 284]
  vld: vld.8x128 v0, [vmem:s18 + 0xb0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 96]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v2, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 160]
  va1: vadd.8x128.f32 v1, 1.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 222]
  vst: vst.8x128 [vmem:s29 + 0x90], v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 285]
  vld: vld.8x128 v3, [vmem:s18 + 0xb8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 97]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v5, 2.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 161]
  va1: vadd.8x128.f32 v4, 1.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 223]
  vst: vst.8x128 [vmem:s29 + 0x98], v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 286]
  vld: vld.8x128 v6, [vmem:s18 + 0xc0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 98]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v8, 2.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 162]
  va1: vadd.8x128.f32 v7, 1.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 224]
  vst: vst.8x128 [vmem:s29 + 0xa0], v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 287]
  vld: vld.8x128 v9, [vmem:s18 + 0xc8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 99]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v11, 2.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 163]
  va1: vadd.8x128.f32 v10, 1.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 225]
  vst: vst.8x128 [vmem:s29 + 0xa8], v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 288]
  vld: vld.8x128 v12, [vmem:s18 + 0xd0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 100]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v14, 2.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 164]
  va1: vadd.8x128.f32 v13, 1.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 226]
  vst: vst.8x128 [vmem:s29 + 0xb0], v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 289]
  vld: vld.8x128 v15, [vmem:s18 + 0xd8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 101]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v17, 2.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 165]
  va1: vadd.8x128.f32 v16, 1.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 227]
  vst: vst.8x128 [vmem:s29 + 0xb8], v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 290]
  vld: vld.8x128 v18, [vmem:s18 + 0xe0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 102]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v20, 2.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 166]
  va1: vadd.8x128.f32 v19, 1.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 228]
  vst: vst.8x128 [vmem:s29 + 0xc0], v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 291]
  vld: vld.8x128 v21, [vmem:s18 + 0xe8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 103]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v23, 2.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 167]
  va1: vadd.8x128.f32 v22, 1.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 229]
  vst: vst.8x128 [vmem:s29 + 0xc8], v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 292]
  vld: vld.8x128 v24, [vmem:s18 + 0xf0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 104]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v26, 2.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 168]
  va1: vadd.8x128.f32 v25, 1.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 230]
  vst: vst.8x128 [vmem:s29 + 0xd0], v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 293]
  vld: vld.8x128 v27, [vmem:s18 + 0xf8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 105]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v29, 2.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 169]
  va1: vadd.8x128.f32 v28, 1.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 231]
  vst: vst.8x128 [vmem:s29 + 0xd8], v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 294]
  vld: vld.8x128 v30, [vmem:s18 + 0x100] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 106]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v0, 2.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 170]
  va1: vadd.8x128.f32 v31, 1.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 232]
  vst: vst.8x128 [vmem:s29 + 0xe0], v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 295]
  vld: vld.8x128 v1, [vmem:s18 + 0x108] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 107]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v3, 2.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 171]
  va1: vadd.8x128.f32 v2, 1.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 233]
  vst: vst.8x128 [vmem:s29 + 0xe8], v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 296]
  vld: vld.8x128 v4, [vmem:s18 + 0x110] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 108]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v6, 2.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 172]
  va1: vadd.8x128.f32 v5, 1.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 234]
  vst: vst.8x128 [vmem:s29 + 0xf0], v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 297]
  vld: vld.8x128 v7, [vmem:s18 + 0x118] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 109]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v9, 2.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 173]
  va1: vadd.8x128.f32 v8, 1.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 235]
  vst: vst.8x128 [vmem:s29 + 0xf8], v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 298]
  vld: vld.8x128 v10, [vmem:s18 + 0x120] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 110]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v12, 2.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 174]
  va1: vadd.8x128.f32 v11, 1.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 236]
  vst: vst.8x128 [vmem:s29 + 0x100], v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 299]
  vld: vld.8x128 v13, [vmem:s18 + 0x128] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 111]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v15, 2.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 175]
  va1: vadd.8x128.f32 v14, 1.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 237]
  vst: vst.8x128 [vmem:s29 + 0x108], v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 300]
  vld: vld.8x128 v16, [vmem:s18 + 0x130] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 112]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v18, 2.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 176]
  va1: vadd.8x128.f32 v17, 1.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 238]
  vst: vst.8x128 [vmem:s29 + 0x110], v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 301]
  vld: vld.8x128 v19, [vmem:s18 + 0x138] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 113]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v21, 2.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 177]
  va1: vadd.8x128.f32 v20, 1.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 239]
  vst: vst.8x128 [vmem:s29 + 0x118], v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 302]
  vld: vld.8x128 v22, [vmem:s18 + 0x140] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 114]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v24, 2.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 178]
  va1: vadd.8x128.f32 v23, 1.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 240]
  vst: vst.8x128 [vmem:s29 + 0x120], v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 303]
  vld: vld.8x128 v25, [vmem:s18 + 0x148] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 115]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v27, 2.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 179]
  va1: vadd.8x128.f32 v26, 1.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 241]
  vst: vst.8x128 [vmem:s29 + 0x128], v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 304]
  vld: vld.8x128 v28, [vmem:s18 + 0x150] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 116]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v30, 2.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 180]
  va1: vadd.8x128.f32 v29, 1.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 242]
  vst: vst.8x128 [vmem:s29 + 0x130], v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 305]
  vld: vld.8x128 v31, [vmem:s18 + 0x158] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 117]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v1, 2.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 181]
  va1: vadd.8x128.f32 v0, 1.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 243]
  vst: vst.8x128 [vmem:s29 + 0x138], v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 306]
  vld: vld.8x128 v2, [vmem:s18 + 0x160] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 118]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v4, 2.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 182]
  va1: vadd.8x128.f32 v3, 1.0, v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 244]
  vst: vst.8x128 [vmem:s29 + 0x140], v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 307]
  vld: vld.8x128 v5, [vmem:s18 + 0x168] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 119]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v7, 2.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 183]
  va1: vadd.8x128.f32 v6, 1.0, v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 245]
  vst: vst.8x128 [vmem:s29 + 0x148], v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 308]
  vld: vld.8x128 v8, [vmem:s18 + 0x170] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 120]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v10, 2.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 184]
  va1: vadd.8x128.f32 v9, 1.0, v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 246]
  vst: vst.8x128 [vmem:s29 + 0x150], v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 309]
  vld: vld.8x128 v11, [vmem:s18 + 0x178] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 121]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v13, 2.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 185]
  va1: vadd.8x128.f32 v12, 1.0, v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 247]
  vst: vst.8x128 [vmem:s29 + 0x158], v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 310]
  vld: vld.8x128 v14, [vmem:s18 + 0x180] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 122]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v16, 2.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 186]
  va1: vadd.8x128.f32 v15, 1.0, v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 248]
  vst: vst.8x128 [vmem:s29 + 0x160], v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 311]
  vld: vld.8x128 v17, [vmem:s18 + 0x188] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 123]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v19, 2.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 187]
  va1: vadd.8x128.f32 v18, 1.0, v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 249]
  vst: vst.8x128 [vmem:s29 + 0x168], v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 312]
  vld: vld.8x128 v20, [vmem:s18 + 0x190] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 124]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v22, 2.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 188]
  va1: vadd.8x128.f32 v21, 1.0, v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 250]
  vst: vst.8x128 [vmem:s29 + 0x170], v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 313]
  vld: vld.8x128 v23, [vmem:s18 + 0x198] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 125]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v25, 2.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 189]
  va1: vadd.8x128.f32 v24, 1.0, v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 251]
  vst: vst.8x128 [vmem:s29 + 0x178], v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 314]
  vld: vld.8x128 v26, [vmem:s18 + 0x1a0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 126]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v28, 2.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 190]
  va1: vadd.8x128.f32 v27, 1.0, v22 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 252]
  vst: vst.8x128 [vmem:s29 + 0x180], v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 315]
  vld: vld.8x128 v29, [vmem:s18 + 0x1a8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 127]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v31, 2.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 191]
  va1: vadd.8x128.f32 v30, 1.0, v25 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 253]
  vst: vst.8x128 [vmem:s29 + 0x188], v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 316]
  vld: vld.8x128 v0, [vmem:s18 + 0x1b0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 128]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v2, 2.0, v0 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 192]
  va1: vadd.8x128.f32 v1, 1.0, v28 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 254]
  vst: vst.8x128 [vmem:s29 + 0x190], v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 317]
  vld: vld.8x128 v3, [vmem:s18 + 0x1b8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 129]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v5, 2.0, v3 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 193]
  va1: vadd.8x128.f32 v4, 1.0, v31 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 255]
  vst: vst.8x128 [vmem:s29 + 0x198], v30 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 318]
  vld: vld.8x128 v6, [vmem:s18 + 0x1c0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 130]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v8, 2.0, v6 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 194]
  va1: vadd.8x128.f32 v7, 1.0, v2 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 256]
  vst: vst.8x128 [vmem:s29 + 0x1a0], v1 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 319]
  vld: vld.8x128 v9, [vmem:s18 + 0x1c8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 131]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v11, 2.0, v9 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 195]
  va1: vadd.8x128.f32 v10, 1.0, v5 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 257]
  vst: vst.8x128 [vmem:s29 + 0x1a8], v4 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 320]
  vld: vld.8x128 v12, [vmem:s18 + 0x1d0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 132]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v14, 2.0, v12 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 196]
  va1: vadd.8x128.f32 v13, 1.0, v8 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 258]
  vst: vst.8x128 [vmem:s29 + 0x1b0], v7 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 321]
  vld: vld.8x128 v15, [vmem:s18 + 0x1d8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 133]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v17, 2.0, v15 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 197]
  va1: vadd.8x128.f32 v16, 1.0, v11 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 259]
  vst: vst.8x128 [vmem:s29 + 0x1b8], v10 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 322]
  vld: vld.8x128 v18, [vmem:s18 + 0x1e0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 134]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v20, 2.0, v18 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 198]
  va1: vadd.8x128.f32 v19, 1.0, v14 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 260]
  vst: vst.8x128 [vmem:s29 + 0x1c0], v13 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 323]
  vld: vld.8x128 v21, [vmem:s18 + 0x1e8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 135]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v23, 2.0, v21 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 199]
  va1: vadd.8x128.f32 v22, 1.0, v17 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 261]
  vst: vst.8x128 [vmem:s29 + 0x1c8], v16 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 324]
  vld: vld.8x128 v24, [vmem:s18 + 0x1f0] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 136]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v26, 2.0, v24 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 200]
  va1: vadd.8x128.f32 v25, 1.0, v20 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 262]
  vst: vst.8x128 [vmem:s29 + 0x1d0], v19 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 325]
  vld: vld.8x128 v27, [vmem:s18 + 0x1f8] ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:51 <- 69:17-69:34 <- 53:13-53:32 [get; LLO 137]
  .encoding { vst.base = vs0 ; vst.offset = imm2 } }
{ va0: vmul.8x128.f32 v29, 2.0, v27 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:57 <- 69:17-69:34 <- 53:13-53:32 [mul; LLO 201]
  va1: vadd.8x128.f32 v28, 1.0, v23 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 263]
  vst: vst.8x128 [vmem:s29 + 0x1d8], v22 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 326]
{ va1: vadd.8x128.f32 v30, 1.0, v26 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 264]
  vst: vst.8x128 [vmem:s29 + 0x1e0], v25 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 327]
{ va1: vadd.8x128.f32 v31, 1.0, v29 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:39-89:63 <- 69:17-69:34 <- 53:13-53:32 [add; LLO 265]
  vst: vst.8x128 [vmem:s29 + 0x1e8], v28 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 328]
{ vst: vst.8x128 [vmem:s29 + 0x1f0], v30 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 329]
{ s0: @!p3 sbr.rel L_0200 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:69:17-69:34
  vst: vst.8x128 [vmem:s29 + 0x1f8], v31 }  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:89:24-89:36 <- 69:17-69:34 <- 53:13-53:32 [swap; LLO 330]
{ s0: dma.simple [hbm:s3], [vmem:s29], length=512, dst_flag=[sflag:s28] }
{ misc: vwait.ge [sflag:54], 512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:94:20-94:54 <- 53:13-53:32 [dma_wait; LLO 400]
  .encoding { misc.value = imm2 } }
{ misc: vsyncadd.s32 [sflag:54], -512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:94:20-94:54 <- 53:13-53:32 [dma_wait; LLO 401]
  .encoding { misc.value = imm2 } }
{ misc: vwait.ge [sflag:55], 512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:95:20-95:60 <- 53:13-53:32 [dma_wait; LLO 402]
  .encoding { misc.value = imm2 } }
{ misc: vsyncadd.s32 [sflag:55], -512 ;  # chapter02/02_dma_pipeline/01_pallas_double_buffer.py:95:20-95:60 <- 53:13-53:32 [dma_wait; LLO 403]
  .encoding { misc.value = imm2 } }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]
# ... 省略 runtime 代码 1 个 bundle
# function jit(transform)/transform/mpmd_map [symbol 4, transform.1]; image bundles [605, 621)
{ s1: sld s20, [smem:0x0] }
{ s1: sld s21, [smem:0x1] }
{ s0: sand.u32 s24, 0xfff, s20 }
{ s0: sadd.s32 s22, 1, s21 ;
  s1: sshll.u32 s27, s24, 0x12 }
{ s0: sand.u32 s23, 0x1, s22 }  # smod.u32 w/div 2
{ s0: sand.u32 s25, 0x3, s23 }
{ s0: sshll.u32 s26, s25, 0x10 }
{ s0: sadd.s32 s28, 131072, s26 }
{ s0: sshrl.u32 s29, s28, 0x2 }
{ s0: sor.u32 s30, 0x2d, s29 }
{ s0: sor.u32 s17, s30, s27 }
{ misc: vsyncadd.remote.s32 [sflag:s17], 1 }
{ misc: vwait.ge [sflag:45], 1 ;
  .encoding { misc.value = imm2 } }
{ misc: vsyncadd.s32 [sflag:45], -1 ;
  .encoding { misc.value = imm2 } }
{ misc: cfence }
{ misc: vtrace 0x90000000 }
# end function jit(transform)/transform/mpmd_map [symbol 4, transform.1]

## 输入输出双缓冲，tile 为 f32[512,128]（256 KiB）：每遍 67782 个周期
## 输入输出双缓冲，tile 为 f32[1024,128]（512 KiB）：每遍 53332 个周期
## 输入输出双缓冲，tile 为 f32[2048,128]（1024 KiB）：每遍 43836 个周期