    1. store_src_rg                   coords = src.rg
    2. init_lane_masks                CondMask = LoopMask = RetMask = true
    3. store_condition_mask           $8 = CondMask
    4. store_condition_mask           $43 = CondMask
    5. store_condition_mask           $65 = CondMask
    6. store_condition_mask           $27 = CondMask
    7. store_condition_mask           $88 = CondMask
    8. store_condition_mask           $79 = CondMask
    9. branch_if_no_active_lanes      branch_if_no_active_lanes +57 (#66)
   10. store_return_mask              $80 = RetMask
   11. zero_4_slots_unmasked          v(0..3) = 0
   12. copy_2_constants               $81..82 = testInputs(0..1)
   13. copy_2_slots_masked            v = Mask($81..82)
   14. copy_2_constants               $81..82 = testInputs(0..1)
   15. copy_2_slots_masked            v = Mask($81..82)
   16. store_condition_mask           $81 = CondMask
   17. copy_2_slots_unmasked          $82..83 = v
   18. copy_2_constants               $84..85 = testInputs(0..1)
   19. cmpne_2_floats                 $82..83 = notEqual($82..83, $84..85)
   20. bitwise_or_int                 $82 |= $83
   21. merge_condition_mask           CondMask = $81 & $82
   22. zero_slot_unmasked             $83 = 0
   23. copy_slot_masked               [test_no_op_mat2_X_vec2].result = Mask($83)
   24. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   25. load_condition_mask            CondMask = $81
   26. store_condition_mask           $81 = CondMask
   27. copy_2_slots_unmasked          $82..83 = v
   28. copy_2_constants               $84..85 = testInputs(0..1)
   29. cmpne_2_floats                 $82..83 = notEqual($82..83, $84..85)
   30. bitwise_or_int                 $82 |= $83
   31. merge_condition_mask           CondMask = $81 & $82
   32. zero_slot_unmasked             $83 = 0
   33. copy_slot_masked               [test_no_op_mat2_X_vec2].result = Mask($83)
   34. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   35. load_condition_mask            CondMask = $81
   36. zero_2_slots_unmasked          $81..82 = 0
   37. copy_2_constants               $83..84 = testInputs(0..1)
   38. sub_2_floats                   $81..82 -= $83..84
   39. copy_2_slots_masked            v = Mask($81..82)
   40. zero_2_slots_unmasked          $81..82 = 0
   41. copy_2_constants               $83..84 = testInputs(0..1)
   42. sub_2_floats                   $81..82 -= $83..84
   43. copy_2_slots_masked            v = Mask($81..82)
   44. store_condition_mask           $81 = CondMask
   45. copy_2_slots_unmasked          $82..83 = v
   46. zero_2_slots_unmasked          $84..85 = 0
   47. copy_2_constants               $86..87 = testInputs(0..1)
   48. sub_2_floats                   $84..85 -= $86..87
   49. cmpne_2_floats                 $82..83 = notEqual($82..83, $84..85)
   50. bitwise_or_int                 $82 |= $83
   51. merge_condition_mask           CondMask = $81 & $82
   52. zero_slot_unmasked             $83 = 0
   53. copy_slot_masked               [test_no_op_mat2_X_vec2].result = Mask($83)
   54. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   55. load_condition_mask            CondMask = $81
   56. zero_2_slots_unmasked          $81..82 = 0
   57. copy_2_slots_masked            vv = Mask($81..82)
   58. zero_2_slots_unmasked          $81..82 = 0
   59. copy_2_slots_masked            vv = Mask($81..82)
   60. zero_2_slots_unmasked          $83..84 = 0
   61. cmpeq_2_floats                 $81..82 = equal($81..82, $83..84)
   62. bitwise_and_int                $81 &= $82
   63. copy_slot_masked               [test_no_op_mat2_X_vec2].result = Mask($81)
   64. load_return_mask               RetMask = $80
   65. copy_slot_unmasked             $80 = [test_no_op_mat2_X_vec2].result
   66. zero_slot_unmasked             $89 = 0
   67. merge_condition_mask           CondMask = $79 & $80
   68. branch_if_no_active_lanes      branch_if_no_active_lanes +63 (#131)
   69. store_return_mask              $90 = RetMask
   70. zero_4_slots_unmasked          v(0..3) = 0
   71. zero_2_slots_unmasked          vv(1..2) = 0
   72. copy_3_constants               $91..93 = testInputs(0..2)
   73. copy_3_slots_masked            v = Mask($91..93)
   74. copy_3_constants               $91..93 = testInputs(0..2)
   75. copy_3_slots_masked            v = Mask($91..93)
   76. store_condition_mask           $91 = CondMask
   77. copy_3_slots_unmasked          $92..94 = v
   78. copy_3_constants               $95..97 = testInputs(0..2)
   79. cmpne_3_floats                 $92..94 = notEqual($92..94, $95..97)
   80. bitwise_or_int                 $93 |= $94
   81. bitwise_or_int                 $92 |= $93
   82. merge_condition_mask           CondMask = $91 & $92
   83. zero_slot_unmasked             $93 = 0
   84. copy_slot_masked               [test_no_op_mat3_X_vec3].result = Mask($93)
   85. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   86. load_condition_mask            CondMask = $91
   87. store_condition_mask           $91 = CondMask
   88. copy_3_slots_unmasked          $92..94 = v
   89. copy_3_constants               $95..97 = testInputs(0..2)
   90. cmpne_3_floats                 $92..94 = notEqual($92..94, $95..97)
   91. bitwise_or_int                 $93 |= $94
   92. bitwise_or_int                 $92 |= $93
   93. merge_condition_mask           CondMask = $91 & $92
   94. zero_slot_unmasked             $93 = 0
   95. copy_slot_masked               [test_no_op_mat3_X_vec3].result = Mask($93)
   96. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   97. load_condition_mask            CondMask = $91
   98. zero_3_slots_unmasked          $91..93 = 0
   99. copy_3_constants               $94..96 = testInputs(0..2)
  100. sub_3_floats                   $91..93 -= $94..96
  101. copy_3_slots_masked            v = Mask($91..93)
  102. zero_3_slots_unmasked          $91..93 = 0
  103. copy_3_constants               $94..96 = testInputs(0..2)
  104. sub_3_floats                   $91..93 -= $94..96
  105. copy_3_slots_masked            v = Mask($91..93)
  106. store_condition_mask           $91 = CondMask
  107. copy_3_slots_unmasked          $92..94 = v
  108. zero_3_slots_unmasked          $95..97 = 0
  109. copy_3_constants               $98..100 = testInputs(0..2)
  110. sub_3_floats                   $95..97 -= $98..100
  111. cmpne_3_floats                 $92..94 = notEqual($92..94, $95..97)
  112. bitwise_or_int                 $93 |= $94
  113. bitwise_or_int                 $92 |= $93
  114. merge_condition_mask           CondMask = $91 & $92
  115. zero_slot_unmasked             $93 = 0
  116. copy_slot_masked               [test_no_op_mat3_X_vec3].result = Mask($93)
  117. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  118. load_condition_mask            CondMask = $91
  119. zero_3_slots_unmasked          $91..93 = 0
  120. copy_3_slots_masked            vv = Mask($91..93)
  121. zero_3_slots_unmasked          $91..93 = 0
  122. copy_3_slots_masked            vv = Mask($91..93)
  123. zero_3_slots_unmasked          $94..96 = 0
  124. cmpeq_3_floats                 $91..93 = equal($91..93, $94..96)
  125. bitwise_and_int                $92 &= $93
  126. bitwise_and_int                $91 &= $92
  127. copy_slot_masked               [test_no_op_mat3_X_vec3].result = Mask($91)
  128. load_return_mask               RetMask = $90
  129. copy_slot_unmasked             $90 = [test_no_op_mat3_X_vec3].result
  130. copy_slot_masked               $89 = Mask($90)
  131. load_condition_mask            CondMask = $79
  132. zero_slot_unmasked             $28 = 0
  133. merge_condition_mask           CondMask = $88 & $89
  134. branch_if_no_active_lanes      branch_if_no_active_lanes +63 (#197)
  135. store_return_mask              $29 = RetMask
  136. zero_4_slots_unmasked          v = 0
  137. zero_4_slots_unmasked          vv = 0
  138. copy_4_constants               $30..33 = testInputs
  139. copy_4_slots_masked            v = Mask($30..33)
  140. copy_4_constants               $30..33 = testInputs
  141. copy_4_slots_masked            v = Mask($30..33)
  142. store_condition_mask           $30 = CondMask
  143. copy_4_slots_unmasked          $31..34 = v
  144. copy_4_constants               $35..38 = testInputs
  145. cmpne_4_floats                 $31..34 = notEqual($31..34, $35..38)
  146. bitwise_or_2_ints              $31..32 |= $33..34
  147. bitwise_or_int                 $31 |= $32
  148. merge_condition_mask           CondMask = $30 & $31
  149. zero_slot_unmasked             $32 = 0
  150. copy_slot_masked               [test_no_op_mat4_X_vec4].result = Mask($32)
  151. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  152. load_condition_mask            CondMask = $30
  153. store_condition_mask           $30 = CondMask
  154. copy_4_slots_unmasked          $31..34 = v
  155. copy_4_constants               $35..38 = testInputs
  156. cmpne_4_floats                 $31..34 = notEqual($31..34, $35..38)
  157. bitwise_or_2_ints              $31..32 |= $33..34
  158. bitwise_or_int                 $31 |= $32
  159. merge_condition_mask           CondMask = $30 & $31
  160. zero_slot_unmasked             $32 = 0
  161. copy_slot_masked               [test_no_op_mat4_X_vec4].result = Mask($32)
  162. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  163. load_condition_mask            CondMask = $30
  164. zero_4_slots_unmasked          $30..33 = 0
  165. copy_4_constants               $34..37 = testInputs
  166. sub_4_floats                   $30..33 -= $34..37
  167. copy_4_slots_masked            v = Mask($30..33)
  168. zero_4_slots_unmasked          $30..33 = 0
  169. copy_4_constants               $34..37 = testInputs
  170. sub_4_floats                   $30..33 -= $34..37
  171. copy_4_slots_masked            v = Mask($30..33)
  172. store_condition_mask           $30 = CondMask
  173. copy_4_slots_unmasked          $31..34 = v
  174. zero_4_slots_unmasked          $35..38 = 0
  175. copy_4_constants               $39..42 = testInputs
  176. sub_4_floats                   $35..38 -= $39..42
  177. cmpne_4_floats                 $31..34 = notEqual($31..34, $35..38)
  178. bitwise_or_2_ints              $31..32 |= $33..34
  179. bitwise_or_int                 $31 |= $32
  180. merge_condition_mask           CondMask = $30 & $31
  181. zero_slot_unmasked             $32 = 0
  182. copy_slot_masked               [test_no_op_mat4_X_vec4].result = Mask($32)
  183. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  184. load_condition_mask            CondMask = $30
  185. zero_4_slots_unmasked          $30..33 = 0
  186. copy_4_slots_masked            vv = Mask($30..33)
  187. zero_4_slots_unmasked          $30..33 = 0
  188. copy_4_slots_masked            vv = Mask($30..33)
  189. zero_4_slots_unmasked          $34..37 = 0
  190. cmpeq_4_floats                 $30..33 = equal($30..33, $34..37)
  191. bitwise_and_2_ints             $30..31 &= $32..33
  192. bitwise_and_int                $30 &= $31
  193. copy_slot_masked               [test_no_op_mat4_X_vec4].result = Mask($30)
  194. load_return_mask               RetMask = $29
  195. copy_slot_unmasked             $29 = [test_no_op_mat4_X_vec4].result
  196. copy_slot_masked               $28 = Mask($29)
  197. load_condition_mask            CondMask = $88
  198. zero_slot_unmasked             $66 = 0
  199. merge_condition_mask           CondMask = $27 & $28
  200. branch_if_no_active_lanes      branch_if_no_active_lanes +111 (#311)
  201. store_return_mask              $67 = RetMask
  202. copy_constant                  $68 = 0xBF800000 (-1.0)
  203. copy_slot_unmasked             $69 = $68
  204. copy_2_slots_unmasked          n = $68..69
  205. copy_constant                  $68 = 0x3F800000 (1.0)
  206. copy_slot_unmasked             $69 = $68
  207. copy_2_slots_unmasked          i = $68..69
  208. zero_4_slots_unmasked          z(0..3) = 0
  209. zero_2_slots_unmasked          vv = 0
  210. zero_2_slots_unmasked          $68..69 = 0
  211. copy_2_slots_masked            vv = Mask($68..69)
  212. zero_2_slots_unmasked          $68..69 = 0
  213. copy_2_slots_masked            vv = Mask($68..69)
  214. store_condition_mask           $68 = CondMask
  215. copy_2_slots_unmasked          $69..70 = vv
  216. copy_2_slots_unmasked          $71..72 = z
  217. cmpne_2_floats                 $69..70 = notEqual($69..70, $71..72)
  218. bitwise_or_int                 $69 |= $70
  219. merge_condition_mask           CondMask = $68 & $69
  220. zero_slot_unmasked             $70 = 0
  221. copy_slot_masked               [test_no_op_vec2_X_mat2].result = Mask($70)
  222. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  223. load_condition_mask            CondMask = $68
  224. copy_2_slots_unmasked          $29..30 = i
  225. copy_4_constants               $31..34 = testMatrix2x2
  226. copy_2_slots_unmasked          $68..69 = $29..30
  227. copy_2_slots_unmasked          $70..71 = $31..32
  228. mul_2_floats                   $68..69 *= $70..71
  229. add_float                      $68 += $69
  230. copy_2_slots_unmasked          $69..70 = $29..30
  231. copy_2_slots_unmasked          $71..72 = $33..34
  232. mul_2_floats                   $69..70 *= $71..72
  233. add_float                      $69 += $70
  234. copy_2_slots_masked            v = Mask($68..69)
  235. store_condition_mask           $68 = CondMask
  236. copy_2_slots_unmasked          $69..70 = v
  237. copy_constant                  $71 = 0x40400000 (3.0)
  238. copy_constant                  $72 = 0x40E00000 (7.0)
  239. cmpne_2_floats                 $69..70 = notEqual($69..70, $71..72)
  240. bitwise_or_int                 $69 |= $70
  241. merge_condition_mask           CondMask = $68 & $69
  242. zero_slot_unmasked             $70 = 0
  243. copy_slot_masked               [test_no_op_vec2_X_mat2].result = Mask($70)
  244. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  245. load_condition_mask            CondMask = $68
  246. copy_4_constants               $29..32 = testMatrix2x2
  247. swizzle_3                      $30..32 = ($30..32).yxz
  248. copy_2_slots_unmasked          $33..34 = i
  249. copy_2_slots_unmasked          $68..69 = $29..30
  250. copy_2_slots_unmasked          $70..71 = $33..34
  251. mul_2_floats                   $68..69 *= $70..71
  252. add_float                      $68 += $69
  253. copy_2_slots_unmasked          $69..70 = $31..32
  254. copy_2_slots_unmasked          $71..72 = $33..34
  255. mul_2_floats                   $69..70 *= $71..72
  256. add_float                      $69 += $70
  257. copy_2_slots_masked            v = Mask($68..69)
  258. store_condition_mask           $68 = CondMask
  259. copy_2_slots_unmasked          $69..70 = v
  260. copy_constant                  $71 = 0x40800000 (4.0)
  261. copy_constant                  $72 = 0x40C00000 (6.0)
  262. cmpne_2_floats                 $69..70 = notEqual($69..70, $71..72)
  263. bitwise_or_int                 $69 |= $70
  264. merge_condition_mask           CondMask = $68 & $69
  265. zero_slot_unmasked             $70 = 0
  266. copy_slot_masked               [test_no_op_vec2_X_mat2].result = Mask($70)
  267. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  268. load_condition_mask            CondMask = $68
  269. copy_2_slots_unmasked          $29..30 = n
  270. copy_4_constants               $31..34 = testMatrix2x2
  271. copy_2_slots_unmasked          $68..69 = $29..30
  272. copy_2_slots_unmasked          $70..71 = $31..32
  273. mul_2_floats                   $68..69 *= $70..71
  274. add_float                      $68 += $69
  275. copy_2_slots_unmasked          $69..70 = $29..30
  276. copy_2_slots_unmasked          $71..72 = $33..34
  277. mul_2_floats                   $69..70 *= $71..72
  278. add_float                      $69 += $70
  279. copy_2_slots_masked            v = Mask($68..69)
  280. store_condition_mask           $68 = CondMask
  281. copy_2_slots_unmasked          $69..70 = v
  282. copy_constant                  $71 = 0xC0400000 (-3.0)
  283. copy_constant                  $72 = 0xC0E00000 (-7.0)
  284. cmpne_2_floats                 $69..70 = notEqual($69..70, $71..72)
  285. bitwise_or_int                 $69 |= $70
  286. merge_condition_mask           CondMask = $68 & $69
  287. zero_slot_unmasked             $70 = 0
  288. copy_slot_masked               [test_no_op_vec2_X_mat2].result = Mask($70)
  289. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  290. load_condition_mask            CondMask = $68
  291. copy_4_constants               $29..32 = testMatrix2x2
  292. swizzle_3                      $30..32 = ($30..32).yxz
  293. copy_2_slots_unmasked          $33..34 = n
  294. copy_2_slots_unmasked          $68..69 = $29..30
  295. copy_2_slots_unmasked          $70..71 = $33..34
  296. mul_2_floats                   $68..69 *= $70..71
  297. add_float                      $68 += $69
  298. copy_2_slots_unmasked          $69..70 = $31..32
  299. copy_2_slots_unmasked          $71..72 = $33..34
  300. mul_2_floats                   $69..70 *= $71..72
  301. add_float                      $69 += $70
  302. copy_2_slots_masked            v = Mask($68..69)
  303. copy_constant                  $70 = 0xC0800000 (-4.0)
  304. copy_constant                  $71 = 0xC0C00000 (-6.0)
  305. cmpeq_2_floats                 $68..69 = equal($68..69, $70..71)
  306. bitwise_and_int                $68 &= $69
  307. copy_slot_masked               [test_no_op_vec2_X_mat2].result = Mask($68)
  308. load_return_mask               RetMask = $67
  309. copy_slot_unmasked             $67 = [test_no_op_vec2_X_mat2].result
  310. copy_slot_masked               $66 = Mask($67)
  311. load_condition_mask            CondMask = $27
  312. zero_slot_unmasked             $44 = 0
  313. merge_condition_mask           CondMask = $65 & $66
  314. branch_if_no_active_lanes      branch_if_no_active_lanes +157 (#471)
  315. store_return_mask              $45 = RetMask
  316. copy_constant                  $46 = 0xBF800000 (-1.0)
  317. swizzle_3                      $46..48 = ($46..48).xxx
  318. copy_3_slots_unmasked          n = $46..48
  319. copy_constant                  $46 = 0x3F800000 (1.0)
  320. swizzle_3                      $46..48 = ($46..48).xxx
  321. copy_3_slots_unmasked          i = $46..48
  322. zero_4_slots_unmasked          z(0..3) = 0
  323. zero_4_slots_unmasked          v(1..4) = 0
  324. zero_slot_unmasked             vv(2) = 0
  325. zero_3_slots_unmasked          $46..48 = 0
  326. copy_3_slots_masked            vv = Mask($46..48)
  327. zero_3_slots_unmasked          $46..48 = 0
  328. copy_3_slots_masked            vv = Mask($46..48)
  329. store_condition_mask           $46 = CondMask
  330. copy_3_slots_unmasked          $47..49 = vv
  331. copy_3_slots_unmasked          $50..52 = z
  332. cmpne_3_floats                 $47..49 = notEqual($47..49, $50..52)
  333. bitwise_or_int                 $48 |= $49
  334. bitwise_or_int                 $47 |= $48
  335. merge_condition_mask           CondMask = $46 & $47
  336. zero_slot_unmasked             $48 = 0
  337. copy_slot_masked               [test_no_op_vec3_X_mat3].result = Mask($48)
  338. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  339. load_condition_mask            CondMask = $46
  340. copy_3_slots_unmasked          $67..69 = i
  341. copy_4_constants               $70..73 = testMatrix3x3(0..3)
  342. copy_4_constants               $74..77 = testMatrix3x3(4..7)
  343. copy_constant                  $78 = testMatrix3x3(8)
  344. copy_3_slots_unmasked          $46..48 = $67..69
  345. copy_3_slots_unmasked          $49..51 = $70..72
  346. mul_3_floats                   $46..48 *= $49..51
  347. add_float                      $47 += $48
  348. add_float                      $46 += $47
  349. copy_3_slots_unmasked          $47..49 = $67..69
  350. copy_3_slots_unmasked          $50..52 = $73..75
  351. mul_3_floats                   $47..49 *= $50..52
  352. add_float                      $48 += $49
  353. add_float                      $47 += $48
  354. copy_3_slots_unmasked          $48..50 = $67..69
  355. copy_3_slots_unmasked          $51..53 = $76..78
  356. mul_3_floats                   $48..50 *= $51..53
  357. add_float                      $49 += $50
  358. add_float                      $48 += $49
  359. copy_3_slots_masked            v = Mask($46..48)
  360. store_condition_mask           $46 = CondMask
  361. copy_3_slots_unmasked          $47..49 = v
  362. copy_constant                  $50 = 0x40C00000 (6.0)
  363. copy_constant                  $51 = 0x41700000 (15.0)
  364. copy_constant                  $52 = 0x41C00000 (24.0)
  365. cmpne_3_floats                 $47..49 = notEqual($47..49, $50..52)
  366. bitwise_or_int                 $48 |= $49
  367. bitwise_or_int                 $47 |= $48
  368. merge_condition_mask           CondMask = $46 & $47
  369. zero_slot_unmasked             $48 = 0
  370. copy_slot_masked               [test_no_op_vec3_X_mat3].result = Mask($48)
  371. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  372. load_condition_mask            CondMask = $46
  373. copy_4_constants               $67..70 = testMatrix3x3(0..3)
  374. copy_4_constants               $71..74 = testMatrix3x3(4..7)
  375. copy_constant                  $75 = testMatrix3x3(8)
  376. shuffle                        $68..75 = ($68..75)[2 5 0 3 6 1 4 7]
  377. copy_3_slots_unmasked          $76..78 = i
  378. copy_3_slots_unmasked          $46..48 = $67..69
  379. copy_3_slots_unmasked          $49..51 = $76..78
  380. mul_3_floats                   $46..48 *= $49..51
  381. add_float                      $47 += $48
  382. add_float                      $46 += $47
  383. copy_3_slots_unmasked          $47..49 = $70..72
  384. copy_3_slots_unmasked          $50..52 = $76..78
  385. mul_3_floats                   $47..49 *= $50..52
  386. add_float                      $48 += $49
  387. add_float                      $47 += $48
  388. copy_3_slots_unmasked          $48..50 = $73..75
  389. copy_3_slots_unmasked          $51..53 = $76..78
  390. mul_3_floats                   $48..50 *= $51..53
  391. add_float                      $49 += $50
  392. add_float                      $48 += $49
  393. copy_3_slots_masked            v = Mask($46..48)
  394. store_condition_mask           $46 = CondMask
  395. copy_3_slots_unmasked          $47..49 = v
  396. copy_constant                  $50 = 0x41400000 (12.0)
  397. copy_constant                  $51 = 0x41700000 (15.0)
  398. copy_constant                  $52 = 0x41900000 (18.0)
  399. cmpne_3_floats                 $47..49 = notEqual($47..49, $50..52)
  400. bitwise_or_int                 $48 |= $49
  401. bitwise_or_int                 $47 |= $48
  402. merge_condition_mask           CondMask = $46 & $47
  403. zero_slot_unmasked             $48 = 0
  404. copy_slot_masked               [test_no_op_vec3_X_mat3].result = Mask($48)
  405. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  406. load_condition_mask            CondMask = $46
  407. copy_3_slots_unmasked          $67..69 = n
  408. copy_4_constants               $70..73 = testMatrix3x3(0..3)
  409. copy_4_constants               $74..77 = testMatrix3x3(4..7)
  410. copy_constant                  $78 = testMatrix3x3(8)
  411. copy_3_slots_unmasked          $46..48 = $67..69
  412. copy_3_slots_unmasked          $49..51 = $70..72
  413. mul_3_floats                   $46..48 *= $49..51
  414. add_float                      $47 += $48
  415. add_float                      $46 += $47
  416. copy_3_slots_unmasked          $47..49 = $67..69
  417. copy_3_slots_unmasked          $50..52 = $73..75
  418. mul_3_floats                   $47..49 *= $50..52
  419. add_float                      $48 += $49
  420. add_float                      $47 += $48
  421. copy_3_slots_unmasked          $48..50 = $67..69
  422. copy_3_slots_unmasked          $51..53 = $76..78
  423. mul_3_floats                   $48..50 *= $51..53
  424. add_float                      $49 += $50
  425. add_float                      $48 += $49
  426. copy_3_slots_masked            v = Mask($46..48)
  427. store_condition_mask           $46 = CondMask
  428. copy_3_slots_unmasked          $47..49 = v
  429. copy_constant                  $50 = 0xC0C00000 (-6.0)
  430. copy_constant                  $51 = 0xC1700000 (-15.0)
  431. copy_constant                  $52 = 0xC1C00000 (-24.0)
  432. cmpne_3_floats                 $47..49 = notEqual($47..49, $50..52)
  433. bitwise_or_int                 $48 |= $49
  434. bitwise_or_int                 $47 |= $48
  435. merge_condition_mask           CondMask = $46 & $47
  436. zero_slot_unmasked             $48 = 0
  437. copy_slot_masked               [test_no_op_vec3_X_mat3].result = Mask($48)
  438. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  439. load_condition_mask            CondMask = $46
  440. copy_4_constants               $67..70 = testMatrix3x3(0..3)
  441. copy_4_constants               $71..74 = testMatrix3x3(4..7)
  442. copy_constant                  $75 = testMatrix3x3(8)
  443. shuffle                        $68..75 = ($68..75)[2 5 0 3 6 1 4 7]
  444. copy_3_slots_unmasked          $76..78 = n
  445. copy_3_slots_unmasked          $46..48 = $67..69
  446. copy_3_slots_unmasked          $49..51 = $76..78
  447. mul_3_floats                   $46..48 *= $49..51
  448. add_float                      $47 += $48
  449. add_float                      $46 += $47
  450. copy_3_slots_unmasked          $47..49 = $70..72
  451. copy_3_slots_unmasked          $50..52 = $76..78
  452. mul_3_floats                   $47..49 *= $50..52
  453. add_float                      $48 += $49
  454. add_float                      $47 += $48
  455. copy_3_slots_unmasked          $48..50 = $73..75
  456. copy_3_slots_unmasked          $51..53 = $76..78
  457. mul_3_floats                   $48..50 *= $51..53
  458. add_float                      $49 += $50
  459. add_float                      $48 += $49
  460. copy_3_slots_masked            v = Mask($46..48)
  461. copy_constant                  $49 = 0xC1400000 (-12.0)
  462. copy_constant                  $50 = 0xC1700000 (-15.0)
  463. copy_constant                  $51 = 0xC1900000 (-18.0)
  464. cmpeq_3_floats                 $46..48 = equal($46..48, $49..51)
  465. bitwise_and_int                $47 &= $48
  466. bitwise_and_int                $46 &= $47
  467. copy_slot_masked               [test_no_op_vec3_X_mat3].result = Mask($46)
  468. load_return_mask               RetMask = $45
  469. copy_slot_unmasked             $45 = [test_no_op_vec3_X_mat3].result
  470. copy_slot_masked               $44 = Mask($45)
  471. load_condition_mask            CondMask = $65
  472. zero_slot_unmasked             $9 = 0
  473. merge_condition_mask           CondMask = $43 & $44
  474. branch_if_no_active_lanes      branch_if_no_active_lanes +194 (#668)
  475. store_return_mask              $10 = RetMask
  476. copy_constant                  $11 = 0xBF800000 (-1.0)
  477. swizzle_4                      $11..14 = ($11..14).xxxx
  478. copy_4_slots_unmasked          n = $11..14
  479. copy_constant                  $11 = 0x3F800000 (1.0)
  480. swizzle_4                      $11..14 = ($11..14).xxxx
  481. copy_4_slots_unmasked          i = $11..14
  482. zero_4_slots_unmasked          z = 0
  483. copy_4_constants               $11..14 = testMatrix2x2
  484. copy_4_constants               $15..18 = testMatrix2x2
  485. copy_4_constants               $19..22 = testMatrix2x2
  486. copy_4_constants               $23..26 = testMatrix2x2
  487. copy_4_slots_unmasked          testMatrix4x4(0..3) = $11..14
  488. copy_4_slots_unmasked          testMatrix4x4(4..7) = $15..18
  489. copy_4_slots_unmasked          testMatrix4x4(8..11) = $19..22
  490. copy_4_slots_unmasked          testMatrix4x4(12..15) = $23..26
  491. zero_4_slots_unmasked          v = 0
  492. zero_4_slots_unmasked          vv = 0
  493. zero_4_slots_unmasked          $11..14 = 0
  494. copy_4_slots_masked            vv = Mask($11..14)
  495. zero_4_slots_unmasked          $11..14 = 0
  496. copy_4_slots_masked            vv = Mask($11..14)
  497. store_condition_mask           $11 = CondMask
  498. copy_4_slots_unmasked          $12..15 = vv
  499. copy_4_slots_unmasked          $16..19 = z
  500. cmpne_4_floats                 $12..15 = notEqual($12..15, $16..19)
  501. bitwise_or_2_ints              $12..13 |= $14..15
  502. stack_rewind
  503. bitwise_or_int                 $12 |= $13
  504. merge_condition_mask           CondMask = $11 & $12
  505. zero_slot_unmasked             $13 = 0
  506. copy_slot_masked               [test_no_op_vec4_X_mat4].result = Mask($13)
  507. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  508. load_condition_mask            CondMask = $11
  509. copy_4_slots_unmasked          $45..48 = i
  510. copy_4_slots_unmasked          $49..52 = testMatrix4x4(0..3)
  511. copy_4_slots_unmasked          $53..56 = testMatrix4x4(4..7)
  512. copy_4_slots_unmasked          $57..60 = testMatrix4x4(8..11)
  513. copy_4_slots_unmasked          $61..64 = testMatrix4x4(12..15)
  514. copy_4_slots_unmasked          $11..14 = $45..48
  515. copy_4_slots_unmasked          $15..18 = $49..52
  516. mul_4_floats                   $11..14 *= $15..18
  517. add_2_floats                   $11..12 += $13..14
  518. add_float                      $11 += $12
  519. copy_4_slots_unmasked          $12..15 = $45..48
  520. copy_4_slots_unmasked          $16..19 = $53..56
  521. mul_4_floats                   $12..15 *= $16..19
  522. add_2_floats                   $12..13 += $14..15
  523. add_float                      $12 += $13
  524. copy_4_slots_unmasked          $13..16 = $45..48
  525. copy_4_slots_unmasked          $17..20 = $57..60
  526. mul_4_floats                   $13..16 *= $17..20
  527. add_2_floats                   $13..14 += $15..16
  528. add_float                      $13 += $14
  529. copy_4_slots_unmasked          $14..17 = $45..48
  530. copy_4_slots_unmasked          $18..21 = $61..64
  531. mul_4_floats                   $14..17 *= $18..21
  532. add_2_floats                   $14..15 += $16..17
  533. add_float                      $14 += $15
  534. copy_4_slots_masked            v = Mask($11..14)
  535. store_condition_mask           $11 = CondMask
  536. copy_4_slots_unmasked          $12..15 = v
  537. copy_constant                  $16 = 0x41200000 (10.0)
  538. copy_constant                  $17 = 0x41200000 (10.0)
  539. copy_constant                  $18 = 0x41200000 (10.0)
  540. copy_constant                  $19 = 0x41200000 (10.0)
  541. cmpne_4_floats                 $12..15 = notEqual($12..15, $16..19)
  542. bitwise_or_2_ints              $12..13 |= $14..15
  543. bitwise_or_int                 $12 |= $13
  544. merge_condition_mask           CondMask = $11 & $12
  545. zero_slot_unmasked             $13 = 0
  546. copy_slot_masked               [test_no_op_vec4_X_mat4].result = Mask($13)
  547. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  548. load_condition_mask            CondMask = $11
  549. copy_4_slots_unmasked          $45..48 = testMatrix4x4(0..3)
  550. copy_4_slots_unmasked          $49..52 = testMatrix4x4(4..7)
  551. copy_4_slots_unmasked          $53..56 = testMatrix4x4(8..11)
  552. copy_4_slots_unmasked          $57..60 = testMatrix4x4(12..15)
  553. shuffle                        $46..60 = ($46..60)[3 7 11 0 4 8 12 1 5 9 13 2 6 10 14]
  554. copy_4_slots_unmasked          $61..64 = i
  555. copy_4_slots_unmasked          $11..14 = $45..48
  556. copy_4_slots_unmasked          $15..18 = $61..64
  557. mul_4_floats                   $11..14 *= $15..18
  558. add_2_floats                   $11..12 += $13..14
  559. add_float                      $11 += $12
  560. copy_4_slots_unmasked          $12..15 = $49..52
  561. copy_4_slots_unmasked          $16..19 = $61..64
  562. mul_4_floats                   $12..15 *= $16..19
  563. add_2_floats                   $12..13 += $14..15
  564. add_float                      $12 += $13
  565. copy_4_slots_unmasked          $13..16 = $53..56
  566. copy_4_slots_unmasked          $17..20 = $61..64
  567. mul_4_floats                   $13..16 *= $17..20
  568. add_2_floats                   $13..14 += $15..16
  569. add_float                      $13 += $14
  570. copy_4_slots_unmasked          $14..17 = $57..60
  571. copy_4_slots_unmasked          $18..21 = $61..64
  572. mul_4_floats                   $14..17 *= $18..21
  573. add_2_floats                   $14..15 += $16..17
  574. add_float                      $14 += $15
  575. copy_4_slots_masked            v = Mask($11..14)
  576. store_condition_mask           $11 = CondMask
  577. copy_4_slots_unmasked          $12..15 = v
  578. copy_constant                  $16 = 0x40800000 (4.0)
  579. copy_constant                  $17 = 0x41000000 (8.0)
  580. copy_constant                  $18 = 0x41400000 (12.0)
  581. copy_constant                  $19 = 0x41800000 (16.0)
  582. cmpne_4_floats                 $12..15 = notEqual($12..15, $16..19)
  583. bitwise_or_2_ints              $12..13 |= $14..15
  584. bitwise_or_int                 $12 |= $13
  585. merge_condition_mask           CondMask = $11 & $12
  586. zero_slot_unmasked             $13 = 0
  587. copy_slot_masked               [test_no_op_vec4_X_mat4].result = Mask($13)
  588. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  589. load_condition_mask            CondMask = $11
  590. copy_4_slots_unmasked          $45..48 = n
  591. copy_4_slots_unmasked          $49..52 = testMatrix4x4(0..3)
  592. copy_4_slots_unmasked          $53..56 = testMatrix4x4(4..7)
  593. copy_4_slots_unmasked          $57..60 = testMatrix4x4(8..11)
  594. copy_4_slots_unmasked          $61..64 = testMatrix4x4(12..15)
  595. copy_4_slots_unmasked          $11..14 = $45..48
  596. copy_4_slots_unmasked          $15..18 = $49..52
  597. mul_4_floats                   $11..14 *= $15..18
  598. add_2_floats                   $11..12 += $13..14
  599. add_float                      $11 += $12
  600. copy_4_slots_unmasked          $12..15 = $45..48
  601. copy_4_slots_unmasked          $16..19 = $53..56
  602. mul_4_floats                   $12..15 *= $16..19
  603. add_2_floats                   $12..13 += $14..15
  604. add_float                      $12 += $13
  605. copy_4_slots_unmasked          $13..16 = $45..48
  606. copy_4_slots_unmasked          $17..20 = $57..60
  607. mul_4_floats                   $13..16 *= $17..20
  608. add_2_floats                   $13..14 += $15..16
  609. add_float                      $13 += $14
  610. copy_4_slots_unmasked          $14..17 = $45..48
  611. copy_4_slots_unmasked          $18..21 = $61..64
  612. mul_4_floats                   $14..17 *= $18..21
  613. add_2_floats                   $14..15 += $16..17
  614. add_float                      $14 += $15
  615. copy_4_slots_masked            v = Mask($11..14)
  616. store_condition_mask           $11 = CondMask
  617. copy_4_slots_unmasked          $12..15 = v
  618. copy_constant                  $16 = 0xC1200000 (-10.0)
  619. copy_constant                  $17 = 0xC1200000 (-10.0)
  620. copy_constant                  $18 = 0xC1200000 (-10.0)
  621. copy_constant                  $19 = 0xC1200000 (-10.0)
  622. cmpne_4_floats                 $12..15 = notEqual($12..15, $16..19)
  623. bitwise_or_2_ints              $12..13 |= $14..15
  624. bitwise_or_int                 $12 |= $13
  625. merge_condition_mask           CondMask = $11 & $12
  626. zero_slot_unmasked             $13 = 0
  627. copy_slot_masked               [test_no_op_vec4_X_mat4].result = Mask($13)
  628. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  629. load_condition_mask            CondMask = $11
  630. copy_4_slots_unmasked          $45..48 = testMatrix4x4(0..3)
  631. copy_4_slots_unmasked          $49..52 = testMatrix4x4(4..7)
  632. copy_4_slots_unmasked          $53..56 = testMatrix4x4(8..11)
  633. copy_4_slots_unmasked          $57..60 = testMatrix4x4(12..15)
  634. shuffle                        $46..60 = ($46..60)[3 7 11 0 4 8 12 1 5 9 13 2 6 10 14]
  635. copy_4_slots_unmasked          $61..64 = n
  636. copy_4_slots_unmasked          $11..14 = $45..48
  637. copy_4_slots_unmasked          $15..18 = $61..64
  638. mul_4_floats                   $11..14 *= $15..18
  639. add_2_floats                   $11..12 += $13..14
  640. add_float                      $11 += $12
  641. copy_4_slots_unmasked          $12..15 = $49..52
  642. copy_4_slots_unmasked          $16..19 = $61..64
  643. mul_4_floats                   $12..15 *= $16..19
  644. add_2_floats                   $12..13 += $14..15
  645. add_float                      $12 += $13
  646. copy_4_slots_unmasked          $13..16 = $53..56
  647. copy_4_slots_unmasked          $17..20 = $61..64
  648. mul_4_floats                   $13..16 *= $17..20
  649. add_2_floats                   $13..14 += $15..16
  650. add_float                      $13 += $14
  651. copy_4_slots_unmasked          $14..17 = $57..60
  652. copy_4_slots_unmasked          $18..21 = $61..64
  653. mul_4_floats                   $14..17 *= $18..21
  654. add_2_floats                   $14..15 += $16..17
  655. add_float                      $14 += $15
  656. copy_4_slots_masked            v = Mask($11..14)
  657. copy_constant                  $15 = 0xC0800000 (-4.0)
  658. copy_constant                  $16 = 0xC1000000 (-8.0)
  659. copy_constant                  $17 = 0xC1400000 (-12.0)
  660. copy_constant                  $18 = 0xC1800000 (-16.0)
  661. cmpeq_4_floats                 $11..14 = equal($11..14, $15..18)
  662. bitwise_and_2_ints             $11..12 &= $13..14
  663. bitwise_and_int                $11 &= $12
  664. copy_slot_masked               [test_no_op_vec4_X_mat4].result = Mask($11)
  665. load_return_mask               RetMask = $10
  666. copy_slot_unmasked             $10 = [test_no_op_vec4_X_mat4].result
  667. copy_slot_masked               $9 = Mask($10)
  668. load_condition_mask            CondMask = $43
  669. copy_4_constants               $0..3 = colorRed
  670. merge_condition_mask           CondMask = $8 & $9
  671. copy_4_constants               $4..7 = colorGreen
  672. copy_4_slots_masked            $0..3 = Mask($4..7)
  673. load_condition_mask            CondMask = $8
  674. copy_4_slots_unmasked          [main].result = $0..3
  675. load_src                       src.rgba = [main].result