    1. store_src_rg                   coords = src.rg
    2. init_lane_masks                CondMask = LoopMask = RetMask = true
    3. store_condition_mask           $165 = CondMask
    4. store_condition_mask           $149 = CondMask
    5. store_condition_mask           $97 = CondMask
    6. store_condition_mask           $66 = CondMask
    7. store_condition_mask           $51 = CondMask
    8. branch_if_no_lanes_active      branch_if_no_lanes_active +74 (label 6 at #82)
    9. store_return_mask              $52 = RetMask
   10. zero_4_slots_unmasked          m = 0
   11. zero_4_slots_unmasked          mm = 0
   12. zero_2_slots_unmasked          $53..54 = 0
   13. swizzle_4                      $53..56 = ($53..56).yxxy
   14. copy_4_slots_unmasked          z = $53..56
   15. copy_4_constants               $53..56 = testMatrix2x2
   16. copy_4_slots_masked            m = Mask($53..56)
   17. copy_4_constants               $53..56 = testMatrix2x2
   18. copy_4_slots_masked            m = Mask($53..56)
   19. store_condition_mask           $53 = CondMask
   20. copy_4_slots_unmasked          $54..57 = m
   21. copy_4_constants               $58..61 = testMatrix2x2
   22. cmpne_4_floats                 $54..57 = notEqual($54..57, $58..61)
   23. bitwise_or_2_ints              $54..55 |= $56..57
   24. bitwise_or_int                 $54 |= $55
   25. merge_condition_mask           CondMask = $53 & $54
   26. zero_slot_unmasked             $55 = 0
   27. copy_slot_masked               [test_no_op_scalar_X_mat2].result = Mask($55)
   28. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   29. load_condition_mask            CondMask = $53
   30. store_condition_mask           $53 = CondMask
   31. copy_4_slots_unmasked          $54..57 = m
   32. copy_4_constants               $58..61 = testMatrix2x2
   33. cmpne_4_floats                 $54..57 = notEqual($54..57, $58..61)
   34. bitwise_or_2_ints              $54..55 |= $56..57
   35. bitwise_or_int                 $54 |= $55
   36. merge_condition_mask           CondMask = $53 & $54
   37. zero_slot_unmasked             $55 = 0
   38. copy_slot_masked               [test_no_op_scalar_X_mat2].result = Mask($55)
   39. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   40. load_condition_mask            CondMask = $53
   41. store_condition_mask           $53 = CondMask
   42. copy_4_slots_unmasked          $54..57 = m
   43. copy_4_constants               $58..61 = testMatrix2x2
   44. cmpne_4_floats                 $54..57 = notEqual($54..57, $58..61)
   45. bitwise_or_2_ints              $54..55 |= $56..57
   46. bitwise_or_int                 $54 |= $55
   47. merge_condition_mask           CondMask = $53 & $54
   48. zero_slot_unmasked             $55 = 0
   49. copy_slot_masked               [test_no_op_scalar_X_mat2].result = Mask($55)
   50. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   51. load_condition_mask            CondMask = $53
   52. zero_4_slots_unmasked          $53..56 = 0
   53. copy_4_slots_unmasked          $57..60 = m
   54. sub_4_floats                   $53..56 -= $57..60
   55. copy_4_slots_masked            m = Mask($53..56)
   56. store_condition_mask           $53 = CondMask
   57. copy_4_slots_unmasked          $54..57 = m
   58. zero_4_slots_unmasked          $58..61 = 0
   59. copy_4_constants               $62..65 = testMatrix2x2
   60. sub_4_floats                   $58..61 -= $62..65
   61. cmpne_4_floats                 $54..57 = notEqual($54..57, $58..61)
   62. bitwise_or_2_ints              $54..55 |= $56..57
   63. bitwise_or_int                 $54 |= $55
   64. merge_condition_mask           CondMask = $53 & $54
   65. zero_slot_unmasked             $55 = 0
   66. copy_slot_masked               [test_no_op_scalar_X_mat2].result = Mask($55)
   67. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
   68. load_condition_mask            CondMask = $53
   69. zero_2_slots_unmasked          $53..54 = 0
   70. swizzle_4                      $53..56 = ($53..56).yxxy
   71. copy_4_slots_masked            mm = Mask($53..56)
   72. zero_2_slots_unmasked          $53..54 = 0
   73. swizzle_4                      $53..56 = ($53..56).yxxy
   74. copy_4_slots_masked            mm = Mask($53..56)
   75. copy_4_slots_unmasked          $57..60 = z
   76. cmpeq_4_floats                 $53..56 = equal($53..56, $57..60)
   77. bitwise_and_2_ints             $53..54 &= $55..56
   78. bitwise_and_int                $53 &= $54
   79. copy_slot_masked               [test_no_op_scalar_X_mat2].result = Mask($53)
   80. load_return_mask               RetMask = $52
   81. copy_slot_unmasked             $52 = [test_no_op_scalar_X_mat2].result
   82. label                          label 0x00000006
   83. zero_slot_unmasked             $67 = 0
   84. merge_condition_mask           CondMask = $51 & $52
   85. branch_if_no_lanes_active      branch_if_no_lanes_active +129 (label 5 at #214)
   86. store_return_mask              $68 = RetMask
   87. zero_4_slots_unmasked          m₁(0..3) = 0
   88. zero_4_slots_unmasked          m₁(4..7) = 0
   89. zero_4_slots_unmasked          m₁(8), mm₁(0..2) = 0
   90. zero_4_slots_unmasked          mm₁(3..6) = 0
   91. zero_2_slots_unmasked          mm₁(7..8) = 0
   92. zero_2_slots_unmasked          $69..70 = 0
   93. shuffle                        $69..77 = ($69..77)[1 0 0 0 1 0 0 0 1]
   94. copy_4_slots_unmasked          z₁(0..3) = $69..72
   95. copy_4_slots_unmasked          z₁(4..7) = $73..76
   96. copy_slot_unmasked             z₁(8) = $77
   97. copy_4_constants               $69..72 = testMatrix3x3(0..3)
   98. copy_4_constants               $73..76 = testMatrix3x3(4..7)
   99. copy_constant                  $77 = testMatrix3x3(8)
  100. copy_4_slots_masked            m₁(0..3) = Mask($69..72)
  101. copy_4_slots_masked            m₁(4..7) = Mask($73..76)
  102. copy_slot_masked               m₁(8) = Mask($77)
  103. copy_4_constants               $69..72 = testMatrix3x3(0..3)
  104. copy_4_constants               $73..76 = testMatrix3x3(4..7)
  105. copy_constant                  $77 = testMatrix3x3(8)
  106. copy_4_slots_masked            m₁(0..3) = Mask($69..72)
  107. copy_4_slots_masked            m₁(4..7) = Mask($73..76)
  108. copy_slot_masked               m₁(8) = Mask($77)
  109. store_condition_mask           $69 = CondMask
  110. copy_4_slots_unmasked          $70..73 = m₁(0..3)
  111. copy_4_slots_unmasked          $74..77 = m₁(4..7)
  112. copy_slot_unmasked             $78 = m₁(8)
  113. copy_4_constants               $79..82 = testMatrix3x3(0..3)
  114. copy_4_constants               $83..86 = testMatrix3x3(4..7)
  115. copy_constant                  $87 = testMatrix3x3(8)
  116. cmpne_n_floats                 $70..78 = notEqual($70..78, $79..87)
  117. bitwise_or_4_ints              $71..74 |= $75..78
  118. bitwise_or_2_ints              $71..72 |= $73..74
  119. bitwise_or_int                 $71 |= $72
  120. bitwise_or_int                 $70 |= $71
  121. merge_condition_mask           CondMask = $69 & $70
  122. zero_slot_unmasked             $71 = 0
  123. copy_slot_masked               [test_no_op_scalar_X_mat3].result = Mask($71)
  124. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  125. load_condition_mask            CondMask = $69
  126. store_condition_mask           $69 = CondMask
  127. copy_4_slots_unmasked          $70..73 = m₁(0..3)
  128. copy_4_slots_unmasked          $74..77 = m₁(4..7)
  129. copy_slot_unmasked             $78 = m₁(8)
  130. copy_4_constants               $79..82 = testMatrix3x3(0..3)
  131. copy_4_constants               $83..86 = testMatrix3x3(4..7)
  132. copy_constant                  $87 = testMatrix3x3(8)
  133. cmpne_n_floats                 $70..78 = notEqual($70..78, $79..87)
  134. bitwise_or_4_ints              $71..74 |= $75..78
  135. bitwise_or_2_ints              $71..72 |= $73..74
  136. bitwise_or_int                 $71 |= $72
  137. bitwise_or_int                 $70 |= $71
  138. merge_condition_mask           CondMask = $69 & $70
  139. zero_slot_unmasked             $71 = 0
  140. copy_slot_masked               [test_no_op_scalar_X_mat3].result = Mask($71)
  141. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  142. load_condition_mask            CondMask = $69
  143. store_condition_mask           $69 = CondMask
  144. copy_4_slots_unmasked          $70..73 = m₁(0..3)
  145. copy_4_slots_unmasked          $74..77 = m₁(4..7)
  146. copy_slot_unmasked             $78 = m₁(8)
  147. copy_4_constants               $79..82 = testMatrix3x3(0..3)
  148. copy_4_constants               $83..86 = testMatrix3x3(4..7)
  149. copy_constant                  $87 = testMatrix3x3(8)
  150. cmpne_n_floats                 $70..78 = notEqual($70..78, $79..87)
  151. bitwise_or_4_ints              $71..74 |= $75..78
  152. bitwise_or_2_ints              $71..72 |= $73..74
  153. bitwise_or_int                 $71 |= $72
  154. bitwise_or_int                 $70 |= $71
  155. merge_condition_mask           CondMask = $69 & $70
  156. zero_slot_unmasked             $71 = 0
  157. copy_slot_masked               [test_no_op_scalar_X_mat3].result = Mask($71)
  158. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  159. load_condition_mask            CondMask = $69
  160. zero_4_slots_unmasked          $69..72 = 0
  161. zero_4_slots_unmasked          $73..76 = 0
  162. zero_slot_unmasked             $77 = 0
  163. copy_4_slots_unmasked          $78..81 = m₁(0..3)
  164. copy_4_slots_unmasked          $82..85 = m₁(4..7)
  165. copy_slot_unmasked             $86 = m₁(8)
  166. sub_n_floats                   $69..77 -= $78..86
  167. copy_4_slots_masked            m₁(0..3) = Mask($69..72)
  168. copy_4_slots_masked            m₁(4..7) = Mask($73..76)
  169. copy_slot_masked               m₁(8) = Mask($77)
  170. store_condition_mask           $69 = CondMask
  171. copy_4_slots_unmasked          $70..73 = m₁(0..3)
  172. copy_4_slots_unmasked          $74..77 = m₁(4..7)
  173. copy_slot_unmasked             $78 = m₁(8)
  174. zero_4_slots_unmasked          $79..82 = 0
  175. zero_4_slots_unmasked          $83..86 = 0
  176. zero_slot_unmasked             $87 = 0
  177. copy_4_constants               $88..91 = testMatrix3x3(0..3)
  178. copy_4_constants               $92..95 = testMatrix3x3(4..7)
  179. copy_constant                  $96 = testMatrix3x3(8)
  180. sub_n_floats                   $79..87 -= $88..96
  181. cmpne_n_floats                 $70..78 = notEqual($70..78, $79..87)
  182. bitwise_or_4_ints              $71..74 |= $75..78
  183. bitwise_or_2_ints              $71..72 |= $73..74
  184. bitwise_or_int                 $71 |= $72
  185. bitwise_or_int                 $70 |= $71
  186. merge_condition_mask           CondMask = $69 & $70
  187. zero_slot_unmasked             $71 = 0
  188. copy_slot_masked               [test_no_op_scalar_X_mat3].result = Mask($71)
  189. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  190. load_condition_mask            CondMask = $69
  191. zero_2_slots_unmasked          $69..70 = 0
  192. shuffle                        $69..77 = ($69..77)[1 0 0 0 1 0 0 0 1]
  193. copy_4_slots_masked            mm₁(0..3) = Mask($69..72)
  194. copy_4_slots_masked            mm₁(4..7) = Mask($73..76)
  195. copy_slot_masked               mm₁(8) = Mask($77)
  196. zero_2_slots_unmasked          $69..70 = 0
  197. shuffle                        $69..77 = ($69..77)[1 0 0 0 1 0 0 0 1]
  198. copy_4_slots_masked            mm₁(0..3) = Mask($69..72)
  199. copy_4_slots_masked            mm₁(4..7) = Mask($73..76)
  200. copy_slot_masked               mm₁(8) = Mask($77)
  201. copy_4_slots_unmasked          $78..81 = z₁(0..3)
  202. copy_4_slots_unmasked          $82..85 = z₁(4..7)
  203. copy_slot_unmasked             $86 = z₁(8)
  204. cmpeq_n_floats                 $69..77 = equal($69..77, $78..86)
  205. bitwise_and_4_ints             $70..73 &= $74..77
  206. bitwise_and_2_ints             $70..71 &= $72..73
  207. bitwise_and_int                $70 &= $71
  208. bitwise_and_int                $69 &= $70
  209. copy_slot_masked               [test_no_op_scalar_X_mat3].result = Mask($69)
  210. load_return_mask               RetMask = $68
  211. copy_slot_unmasked             $68 = [test_no_op_scalar_X_mat3].result
  212. label                          label 0x00000007
  213. copy_slot_masked               $67 = Mask($68)
  214. label                          label 0x00000005
  215. load_condition_mask            CondMask = $51
  216. zero_slot_unmasked             $98 = 0
  217. merge_condition_mask           CondMask = $66 & $67
  218. branch_if_no_lanes_active      branch_if_no_lanes_active +165 (label 4 at #383)
  219. store_return_mask              $99 = RetMask
  220. copy_4_constants               $100..103 = testInputs
  221. copy_4_constants               $104..107 = testInputs
  222. copy_4_constants               $108..111 = testInputs
  223. copy_4_constants               $112..115 = testInputs
  224. copy_4_slots_unmasked          testMatrix4x4(0..3) = $100..103
  225. copy_4_slots_unmasked          testMatrix4x4(4..7) = $104..107
  226. copy_4_slots_unmasked          testMatrix4x4(8..11) = $108..111
  227. copy_4_slots_unmasked          testMatrix4x4(12..15) = $112..115
  228. zero_4_slots_unmasked          m₂(0..3) = 0
  229. zero_4_slots_unmasked          m₂(4..7) = 0
  230. zero_4_slots_unmasked          m₂(8..11) = 0
  231. zero_4_slots_unmasked          m₂(12..15) = 0
  232. zero_4_slots_unmasked          mm₂(0..3) = 0
  233. zero_4_slots_unmasked          mm₂(4..7) = 0
  234. zero_4_slots_unmasked          mm₂(8..11) = 0
  235. zero_4_slots_unmasked          mm₂(12..15) = 0
  236. zero_2_slots_unmasked          $100..101 = 0
  237. shuffle                        $100..115 = ($100..115)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  238. copy_4_slots_unmasked          z₂(0..3) = $100..103
  239. copy_4_slots_unmasked          z₂(4..7) = $104..107
  240. copy_4_slots_unmasked          z₂(8..11) = $108..111
  241. copy_4_slots_unmasked          z₂(12..15) = $112..115
  242. copy_4_slots_unmasked          $100..103 = testMatrix4x4(0..3)
  243. copy_4_slots_unmasked          $104..107 = testMatrix4x4(4..7)
  244. copy_4_slots_unmasked          $108..111 = testMatrix4x4(8..11)
  245. copy_4_slots_unmasked          $112..115 = testMatrix4x4(12..15)
  246. copy_4_slots_masked            m₂(0..3) = Mask($100..103)
  247. copy_4_slots_masked            m₂(4..7) = Mask($104..107)
  248. copy_4_slots_masked            m₂(8..11) = Mask($108..111)
  249. copy_4_slots_masked            m₂(12..15) = Mask($112..115)
  250. copy_4_slots_unmasked          $100..103 = testMatrix4x4(0..3)
  251. copy_4_slots_unmasked          $104..107 = testMatrix4x4(4..7)
  252. copy_4_slots_unmasked          $108..111 = testMatrix4x4(8..11)
  253. copy_4_slots_unmasked          $112..115 = testMatrix4x4(12..15)
  254. copy_4_slots_masked            m₂(0..3) = Mask($100..103)
  255. copy_4_slots_masked            m₂(4..7) = Mask($104..107)
  256. copy_4_slots_masked            m₂(8..11) = Mask($108..111)
  257. copy_4_slots_masked            m₂(12..15) = Mask($112..115)
  258. store_condition_mask           $100 = CondMask
  259. copy_4_slots_unmasked          $101..104 = m₂(0..3)
  260. copy_4_slots_unmasked          $105..108 = m₂(4..7)
  261. copy_4_slots_unmasked          $109..112 = m₂(8..11)
  262. copy_4_slots_unmasked          $113..116 = m₂(12..15)
  263. copy_4_slots_unmasked          $117..120 = testMatrix4x4(0..3)
  264. copy_4_slots_unmasked          $121..124 = testMatrix4x4(4..7)
  265. copy_4_slots_unmasked          $125..128 = testMatrix4x4(8..11)
  266. copy_4_slots_unmasked          $129..132 = testMatrix4x4(12..15)
  267. cmpne_n_floats                 $101..116 = notEqual($101..116, $117..132)
  268. bitwise_or_4_ints              $109..112 |= $113..116
  269. bitwise_or_4_ints              $105..108 |= $109..112
  270. bitwise_or_4_ints              $101..104 |= $105..108
  271. bitwise_or_2_ints              $101..102 |= $103..104
  272. bitwise_or_int                 $101 |= $102
  273. merge_condition_mask           CondMask = $100 & $101
  274. zero_slot_unmasked             $102 = 0
  275. copy_slot_masked               [test_no_op_scalar_X_mat4].result = Mask($102)
  276. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  277. load_condition_mask            CondMask = $100
  278. store_condition_mask           $100 = CondMask
  279. copy_4_slots_unmasked          $101..104 = m₂(0..3)
  280. copy_4_slots_unmasked          $105..108 = m₂(4..7)
  281. copy_4_slots_unmasked          $109..112 = m₂(8..11)
  282. copy_4_slots_unmasked          $113..116 = m₂(12..15)
  283. copy_4_slots_unmasked          $117..120 = testMatrix4x4(0..3)
  284. copy_4_slots_unmasked          $121..124 = testMatrix4x4(4..7)
  285. copy_4_slots_unmasked          $125..128 = testMatrix4x4(8..11)
  286. copy_4_slots_unmasked          $129..132 = testMatrix4x4(12..15)
  287. cmpne_n_floats                 $101..116 = notEqual($101..116, $117..132)
  288. bitwise_or_4_ints              $109..112 |= $113..116
  289. bitwise_or_4_ints              $105..108 |= $109..112
  290. bitwise_or_4_ints              $101..104 |= $105..108
  291. bitwise_or_2_ints              $101..102 |= $103..104
  292. bitwise_or_int                 $101 |= $102
  293. merge_condition_mask           CondMask = $100 & $101
  294. zero_slot_unmasked             $102 = 0
  295. copy_slot_masked               [test_no_op_scalar_X_mat4].result = Mask($102)
  296. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  297. load_condition_mask            CondMask = $100
  298. store_condition_mask           $100 = CondMask
  299. copy_4_slots_unmasked          $101..104 = m₂(0..3)
  300. copy_4_slots_unmasked          $105..108 = m₂(4..7)
  301. copy_4_slots_unmasked          $109..112 = m₂(8..11)
  302. copy_4_slots_unmasked          $113..116 = m₂(12..15)
  303. copy_4_slots_unmasked          $117..120 = testMatrix4x4(0..3)
  304. copy_4_slots_unmasked          $121..124 = testMatrix4x4(4..7)
  305. copy_4_slots_unmasked          $125..128 = testMatrix4x4(8..11)
  306. copy_4_slots_unmasked          $129..132 = testMatrix4x4(12..15)
  307. cmpne_n_floats                 $101..116 = notEqual($101..116, $117..132)
  308. bitwise_or_4_ints              $109..112 |= $113..116
  309. bitwise_or_4_ints              $105..108 |= $109..112
  310. bitwise_or_4_ints              $101..104 |= $105..108
  311. bitwise_or_2_ints              $101..102 |= $103..104
  312. bitwise_or_int                 $101 |= $102
  313. merge_condition_mask           CondMask = $100 & $101
  314. zero_slot_unmasked             $102 = 0
  315. copy_slot_masked               [test_no_op_scalar_X_mat4].result = Mask($102)
  316. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  317. load_condition_mask            CondMask = $100
  318. zero_4_slots_unmasked          $100..103 = 0
  319. zero_4_slots_unmasked          $104..107 = 0
  320. zero_4_slots_unmasked          $108..111 = 0
  321. zero_4_slots_unmasked          $112..115 = 0
  322. copy_4_slots_unmasked          $116..119 = m₂(0..3)
  323. copy_4_slots_unmasked          $120..123 = m₂(4..7)
  324. copy_4_slots_unmasked          $124..127 = m₂(8..11)
  325. copy_4_slots_unmasked          $128..131 = m₂(12..15)
  326. sub_n_floats                   $100..115 -= $116..131
  327. copy_4_slots_masked            m₂(0..3) = Mask($100..103)
  328. copy_4_slots_masked            m₂(4..7) = Mask($104..107)
  329. copy_4_slots_masked            m₂(8..11) = Mask($108..111)
  330. copy_4_slots_masked            m₂(12..15) = Mask($112..115)
  331. store_condition_mask           $100 = CondMask
  332. copy_4_slots_unmasked          $101..104 = m₂(0..3)
  333. copy_4_slots_unmasked          $105..108 = m₂(4..7)
  334. copy_4_slots_unmasked          $109..112 = m₂(8..11)
  335. copy_4_slots_unmasked          $113..116 = m₂(12..15)
  336. zero_4_slots_unmasked          $117..120 = 0
  337. zero_4_slots_unmasked          $121..124 = 0
  338. zero_4_slots_unmasked          $125..128 = 0
  339. zero_4_slots_unmasked          $129..132 = 0
  340. copy_4_slots_unmasked          $133..136 = testMatrix4x4(0..3)
  341. copy_4_slots_unmasked          $137..140 = testMatrix4x4(4..7)
  342. copy_4_slots_unmasked          $141..144 = testMatrix4x4(8..11)
  343. copy_4_slots_unmasked          $145..148 = testMatrix4x4(12..15)
  344. sub_n_floats                   $117..132 -= $133..148
  345. cmpne_n_floats                 $101..116 = notEqual($101..116, $117..132)
  346. bitwise_or_4_ints              $109..112 |= $113..116
  347. bitwise_or_4_ints              $105..108 |= $109..112
  348. bitwise_or_4_ints              $101..104 |= $105..108
  349. bitwise_or_2_ints              $101..102 |= $103..104
  350. bitwise_or_int                 $101 |= $102
  351. merge_condition_mask           CondMask = $100 & $101
  352. zero_slot_unmasked             $102 = 0
  353. copy_slot_masked               [test_no_op_scalar_X_mat4].result = Mask($102)
  354. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  355. load_condition_mask            CondMask = $100
  356. zero_2_slots_unmasked          $100..101 = 0
  357. shuffle                        $100..115 = ($100..115)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  358. copy_4_slots_masked            mm₂(0..3) = Mask($100..103)
  359. copy_4_slots_masked            mm₂(4..7) = Mask($104..107)
  360. copy_4_slots_masked            mm₂(8..11) = Mask($108..111)
  361. copy_4_slots_masked            mm₂(12..15) = Mask($112..115)
  362. zero_2_slots_unmasked          $100..101 = 0
  363. shuffle                        $100..115 = ($100..115)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  364. copy_4_slots_masked            mm₂(0..3) = Mask($100..103)
  365. copy_4_slots_masked            mm₂(4..7) = Mask($104..107)
  366. copy_4_slots_masked            mm₂(8..11) = Mask($108..111)
  367. copy_4_slots_masked            mm₂(12..15) = Mask($112..115)
  368. copy_4_slots_unmasked          $116..119 = z₂(0..3)
  369. copy_4_slots_unmasked          $120..123 = z₂(4..7)
  370. copy_4_slots_unmasked          $124..127 = z₂(8..11)
  371. copy_4_slots_unmasked          $128..131 = z₂(12..15)
  372. cmpeq_n_floats                 $100..115 = equal($100..115, $116..131)
  373. bitwise_and_4_ints             $108..111 &= $112..115
  374. bitwise_and_4_ints             $104..107 &= $108..111
  375. bitwise_and_4_ints             $100..103 &= $104..107
  376. bitwise_and_2_ints             $100..101 &= $102..103
  377. bitwise_and_int                $100 &= $101
  378. copy_slot_masked               [test_no_op_scalar_X_mat4].result = Mask($100)
  379. load_return_mask               RetMask = $99
  380. copy_slot_unmasked             $99 = [test_no_op_scalar_X_mat4].result
  381. label                          label 0x00000008
  382. copy_slot_masked               $98 = Mask($99)
  383. label                          label 0x00000004
  384. load_condition_mask            CondMask = $66
  385. zero_slot_unmasked             $150 = 0
  386. merge_condition_mask           CondMask = $97 & $98
  387. branch_if_no_lanes_active      branch_if_no_lanes_active +118 (label 3 at #505)
  388. store_return_mask              $151 = RetMask
  389. zero_4_slots_unmasked          m₃ = 0
  390. zero_4_slots_unmasked          mm₃ = 0
  391. zero_2_slots_unmasked          $152..153 = 0
  392. swizzle_4                      $152..155 = ($152..155).yxxy
  393. copy_4_slots_unmasked          z₃ = $152..155
  394. copy_constant                  $152 = 0x3F800000 (1.0)
  395. swizzle_4                      $152..155 = ($152..155).xxxx
  396. copy_4_slots_unmasked          s = $152..155
  397. copy_constant                  $152 = testInputs(0)
  398. copy_slot_unmasked             scalar = $152
  399. zero_slot_unmasked             $152 = 0
  400. copy_slot_unmasked             $153 = scalar
  401. swizzle_4                      $152..155 = ($152..155).yxxy
  402. copy_4_slots_masked            m₃ = Mask($152..155)
  403. zero_slot_unmasked             $152 = 0
  404. copy_slot_unmasked             $153 = scalar
  405. swizzle_4                      $152..155 = ($152..155).yxxy
  406. copy_4_slots_masked            m₃ = Mask($152..155)
  407. store_condition_mask           $152 = CondMask
  408. copy_4_slots_unmasked          $153..156 = m₃
  409. zero_slot_unmasked             $157 = 0
  410. copy_slot_unmasked             $158 = scalar
  411. swizzle_4                      $157..160 = ($157..160).yxxy
  412. cmpne_4_floats                 $153..156 = notEqual($153..156, $157..160)
  413. bitwise_or_2_ints              $153..154 |= $155..156
  414. bitwise_or_int                 $153 |= $154
  415. merge_condition_mask           CondMask = $152 & $153
  416. zero_slot_unmasked             $154 = 0
  417. copy_slot_masked               [test_no_op_mat2_X_scalar].result = Mask($154)
  418. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  419. load_condition_mask            CondMask = $152
  420. copy_slot_unmasked             $152 = scalar
  421. swizzle_4                      $152..155 = ($152..155).xxxx
  422. copy_4_slots_unmasked          $156..159 = s
  423. div_4_floats                   $152..155 /= $156..159
  424. copy_4_slots_masked            m₃ = Mask($152..155)
  425. store_condition_mask           $152 = CondMask
  426. copy_4_slots_unmasked          $153..156 = m₃
  427. copy_slot_unmasked             $157 = scalar
  428. copy_slot_unmasked             $158 = scalar
  429. copy_slot_unmasked             $159 = scalar
  430. copy_slot_unmasked             $160 = scalar
  431. cmpne_4_floats                 $153..156 = notEqual($153..156, $157..160)
  432. bitwise_or_2_ints              $153..154 |= $155..156
  433. bitwise_or_int                 $153 |= $154
  434. merge_condition_mask           CondMask = $152 & $153
  435. zero_slot_unmasked             $154 = 0
  436. copy_slot_masked               [test_no_op_mat2_X_scalar].result = Mask($154)
  437. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  438. load_condition_mask            CondMask = $152
  439. copy_slot_unmasked             $152 = scalar
  440. swizzle_4                      $152..155 = ($152..155).xxxx
  441. copy_4_slots_unmasked          $156..159 = z₃
  442. add_4_floats                   $152..155 += $156..159
  443. copy_4_slots_masked            m₃ = Mask($152..155)
  444. copy_4_slots_unmasked          $152..155 = z₃
  445. copy_slot_unmasked             $156 = scalar
  446. swizzle_4                      $156..159 = ($156..159).xxxx
  447. add_4_floats                   $152..155 += $156..159
  448. copy_4_slots_masked            m₃ = Mask($152..155)
  449. store_condition_mask           $152 = CondMask
  450. copy_4_slots_unmasked          $153..156 = m₃
  451. copy_slot_unmasked             $157 = scalar
  452. copy_slot_unmasked             $158 = scalar
  453. copy_slot_unmasked             $159 = scalar
  454. copy_slot_unmasked             $160 = scalar
  455. cmpne_4_floats                 $153..156 = notEqual($153..156, $157..160)
  456. bitwise_or_2_ints              $153..154 |= $155..156
  457. bitwise_or_int                 $153 |= $154
  458. merge_condition_mask           CondMask = $152 & $153
  459. zero_slot_unmasked             $154 = 0
  460. copy_slot_masked               [test_no_op_mat2_X_scalar].result = Mask($154)
  461. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  462. load_condition_mask            CondMask = $152
  463. copy_slot_unmasked             $152 = scalar
  464. swizzle_4                      $152..155 = ($152..155).xxxx
  465. copy_4_slots_unmasked          $156..159 = z₃
  466. sub_4_floats                   $152..155 -= $156..159
  467. copy_4_slots_masked            m₃ = Mask($152..155)
  468. copy_4_slots_unmasked          $152..155 = z₃
  469. copy_slot_unmasked             $156 = scalar
  470. swizzle_4                      $156..159 = ($156..159).xxxx
  471. sub_4_floats                   $152..155 -= $156..159
  472. copy_4_slots_masked            m₃ = Mask($152..155)
  473. store_condition_mask           $152 = CondMask
  474. copy_4_slots_unmasked          $153..156 = m₃
  475. zero_4_slots_unmasked          $157..160 = 0
  476. copy_slot_unmasked             $161 = scalar
  477. copy_slot_unmasked             $162 = scalar
  478. copy_slot_unmasked             $163 = scalar
  479. copy_slot_unmasked             $164 = scalar
  480. sub_4_floats                   $157..160 -= $161..164
  481. cmpne_4_floats                 $153..156 = notEqual($153..156, $157..160)
  482. bitwise_or_2_ints              $153..154 |= $155..156
  483. bitwise_or_int                 $153 |= $154
  484. merge_condition_mask           CondMask = $152 & $153
  485. zero_slot_unmasked             $154 = 0
  486. copy_slot_masked               [test_no_op_mat2_X_scalar].result = Mask($154)
  487. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  488. load_condition_mask            CondMask = $152
  489. zero_2_slots_unmasked          $152..153 = 0
  490. swizzle_4                      $152..155 = ($152..155).yxxy
  491. copy_4_slots_masked            mm₃ = Mask($152..155)
  492. zero_2_slots_unmasked          $152..153 = 0
  493. swizzle_4                      $152..155 = ($152..155).yxxy
  494. copy_4_slots_masked            mm₃ = Mask($152..155)
  495. copy_4_slots_unmasked          $156..159 = z₃
  496. cmpeq_4_floats                 $152..155 = equal($152..155, $156..159)
  497. bitwise_and_2_ints             $152..153 &= $154..155
  498. bitwise_and_int                $152 &= $153
  499. copy_slot_masked               [test_no_op_mat2_X_scalar].result = Mask($152)
  500. load_return_mask               RetMask = $151
  501. copy_slot_unmasked             $151 = [test_no_op_mat2_X_scalar].result
  502. stack_rewind
  503. label                          label 0x00000009
  504. copy_slot_masked               $150 = Mask($151)
  505. label                          label 0x00000003
  506. load_condition_mask            CondMask = $97
  507. zero_slot_unmasked             $166 = 0
  508. merge_condition_mask           CondMask = $149 & $150
  509. branch_if_no_lanes_active      branch_if_no_lanes_active +187 (label 2 at #696)
  510. store_return_mask              $167 = RetMask
  511. zero_4_slots_unmasked          m₄(0..3) = 0
  512. zero_4_slots_unmasked          m₄(4..7) = 0
  513. zero_4_slots_unmasked          m₄(8), mm₄(0..2) = 0
  514. zero_4_slots_unmasked          mm₄(3..6) = 0
  515. zero_2_slots_unmasked          mm₄(7..8) = 0
  516. zero_2_slots_unmasked          $168..169 = 0
  517. shuffle                        $168..176 = ($168..176)[1 0 0 0 1 0 0 0 1]
  518. copy_4_slots_unmasked          z₄(0..3) = $168..171
  519. copy_4_slots_unmasked          z₄(4..7) = $172..175
  520. copy_slot_unmasked             z₄(8) = $176
  521. copy_constant                  $168 = 0x3F800000 (1.0)
  522. swizzle_3                      $168..170 = ($168..170).xxx
  523. copy_constant                  $171 = 0x3F800000 (1.0)
  524. swizzle_3                      $171..173 = ($171..173).xxx
  525. copy_constant                  $174 = 0x3F800000 (1.0)
  526. swizzle_3                      $174..176 = ($174..176).xxx
  527. copy_4_slots_unmasked          s₁(0..3) = $168..171
  528. copy_4_slots_unmasked          s₁(4..7) = $172..175
  529. copy_slot_unmasked             s₁(8) = $176
  530. copy_constant                  $168 = testInputs(0)
  531. copy_slot_unmasked             scalar₁ = $168
  532. swizzle_3                      $168..170 = ($168..170).xxx
  533. copy_3_slots_unmasked          scalar3 = $168..170
  534. zero_slot_unmasked             $168 = 0
  535. copy_slot_unmasked             $169 = scalar₁
  536. shuffle                        $168..176 = ($168..176)[1 0 0 0 1 0 0 0 1]
  537. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  538. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  539. copy_slot_masked               m₄(8) = Mask($176)
  540. zero_slot_unmasked             $168 = 0
  541. copy_slot_unmasked             $169 = scalar₁
  542. shuffle                        $168..176 = ($168..176)[1 0 0 0 1 0 0 0 1]
  543. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  544. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  545. copy_slot_masked               m₄(8) = Mask($176)
  546. store_condition_mask           $168 = CondMask
  547. copy_4_slots_unmasked          $169..172 = m₄(0..3)
  548. copy_4_slots_unmasked          $173..176 = m₄(4..7)
  549. copy_slot_unmasked             $177 = m₄(8)
  550. zero_slot_unmasked             $178 = 0
  551. copy_slot_unmasked             $179 = scalar₁
  552. shuffle                        $178..186 = ($178..186)[1 0 0 0 1 0 0 0 1]
  553. cmpne_n_floats                 $169..177 = notEqual($169..177, $178..186)
  554. bitwise_or_4_ints              $170..173 |= $174..177
  555. bitwise_or_2_ints              $170..171 |= $172..173
  556. bitwise_or_int                 $170 |= $171
  557. bitwise_or_int                 $169 |= $170
  558. merge_condition_mask           CondMask = $168 & $169
  559. zero_slot_unmasked             $170 = 0
  560. copy_slot_masked               [test_no_op_mat3_X_scalar].result = Mask($170)
  561. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  562. load_condition_mask            CondMask = $168
  563. copy_slot_unmasked             $168 = scalar₁
  564. swizzle_4                      $168..171 = ($168..171).xxxx
  565. copy_4_slots_unmasked          $172..175 = $168..171
  566. copy_slot_unmasked             $176 = $175
  567. copy_4_slots_unmasked          $177..180 = s₁(0..3)
  568. copy_4_slots_unmasked          $181..184 = s₁(4..7)
  569. copy_slot_unmasked             $185 = s₁(8)
  570. div_n_floats                   $168..176 /= $177..185
  571. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  572. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  573. copy_slot_masked               m₄(8) = Mask($176)
  574. store_condition_mask           $168 = CondMask
  575. copy_4_slots_unmasked          $169..172 = m₄(0..3)
  576. copy_4_slots_unmasked          $173..176 = m₄(4..7)
  577. copy_slot_unmasked             $177 = m₄(8)
  578. copy_3_slots_unmasked          $178..180 = scalar3
  579. copy_3_slots_unmasked          $181..183 = scalar3
  580. copy_3_slots_unmasked          $184..186 = scalar3
  581. cmpne_n_floats                 $169..177 = notEqual($169..177, $178..186)
  582. bitwise_or_4_ints              $170..173 |= $174..177
  583. bitwise_or_2_ints              $170..171 |= $172..173
  584. bitwise_or_int                 $170 |= $171
  585. bitwise_or_int                 $169 |= $170
  586. merge_condition_mask           CondMask = $168 & $169
  587. zero_slot_unmasked             $170 = 0
  588. copy_slot_masked               [test_no_op_mat3_X_scalar].result = Mask($170)
  589. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  590. load_condition_mask            CondMask = $168
  591. copy_slot_unmasked             $168 = scalar₁
  592. swizzle_4                      $168..171 = ($168..171).xxxx
  593. copy_4_slots_unmasked          $172..175 = $168..171
  594. copy_slot_unmasked             $176 = $175
  595. copy_4_slots_unmasked          $177..180 = z₄(0..3)
  596. copy_4_slots_unmasked          $181..184 = z₄(4..7)
  597. copy_slot_unmasked             $185 = z₄(8)
  598. add_n_floats                   $168..176 += $177..185
  599. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  600. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  601. copy_slot_masked               m₄(8) = Mask($176)
  602. copy_4_slots_unmasked          $168..171 = z₄(0..3)
  603. copy_4_slots_unmasked          $172..175 = z₄(4..7)
  604. copy_slot_unmasked             $176 = z₄(8)
  605. copy_slot_unmasked             $177 = scalar₁
  606. swizzle_4                      $177..180 = ($177..180).xxxx
  607. copy_4_slots_unmasked          $181..184 = $177..180
  608. copy_slot_unmasked             $185 = $184
  609. add_n_floats                   $168..176 += $177..185
  610. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  611. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  612. copy_slot_masked               m₄(8) = Mask($176)
  613. store_condition_mask           $168 = CondMask
  614. copy_4_slots_unmasked          $169..172 = m₄(0..3)
  615. copy_4_slots_unmasked          $173..176 = m₄(4..7)
  616. copy_slot_unmasked             $177 = m₄(8)
  617. copy_3_slots_unmasked          $178..180 = scalar3
  618. copy_3_slots_unmasked          $181..183 = scalar3
  619. copy_3_slots_unmasked          $184..186 = scalar3
  620. cmpne_n_floats                 $169..177 = notEqual($169..177, $178..186)
  621. bitwise_or_4_ints              $170..173 |= $174..177
  622. bitwise_or_2_ints              $170..171 |= $172..173
  623. bitwise_or_int                 $170 |= $171
  624. bitwise_or_int                 $169 |= $170
  625. merge_condition_mask           CondMask = $168 & $169
  626. zero_slot_unmasked             $170 = 0
  627. copy_slot_masked               [test_no_op_mat3_X_scalar].result = Mask($170)
  628. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  629. load_condition_mask            CondMask = $168
  630. copy_slot_unmasked             $168 = scalar₁
  631. swizzle_4                      $168..171 = ($168..171).xxxx
  632. copy_4_slots_unmasked          $172..175 = $168..171
  633. copy_slot_unmasked             $176 = $175
  634. copy_4_slots_unmasked          $177..180 = z₄(0..3)
  635. copy_4_slots_unmasked          $181..184 = z₄(4..7)
  636. copy_slot_unmasked             $185 = z₄(8)
  637. sub_n_floats                   $168..176 -= $177..185
  638. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  639. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  640. copy_slot_masked               m₄(8) = Mask($176)
  641. copy_4_slots_unmasked          $168..171 = z₄(0..3)
  642. copy_4_slots_unmasked          $172..175 = z₄(4..7)
  643. copy_slot_unmasked             $176 = z₄(8)
  644. copy_slot_unmasked             $177 = scalar₁
  645. swizzle_4                      $177..180 = ($177..180).xxxx
  646. copy_4_slots_unmasked          $181..184 = $177..180
  647. copy_slot_unmasked             $185 = $184
  648. sub_n_floats                   $168..176 -= $177..185
  649. copy_4_slots_masked            m₄(0..3) = Mask($168..171)
  650. copy_4_slots_masked            m₄(4..7) = Mask($172..175)
  651. copy_slot_masked               m₄(8) = Mask($176)
  652. store_condition_mask           $168 = CondMask
  653. copy_4_slots_unmasked          $169..172 = m₄(0..3)
  654. copy_4_slots_unmasked          $173..176 = m₄(4..7)
  655. copy_slot_unmasked             $177 = m₄(8)
  656. zero_4_slots_unmasked          $178..181 = 0
  657. zero_4_slots_unmasked          $182..185 = 0
  658. zero_slot_unmasked             $186 = 0
  659. copy_3_slots_unmasked          $187..189 = scalar3
  660. copy_3_slots_unmasked          $190..192 = scalar3
  661. copy_3_slots_unmasked          $193..195 = scalar3
  662. sub_n_floats                   $178..186 -= $187..195
  663. cmpne_n_floats                 $169..177 = notEqual($169..177, $178..186)
  664. bitwise_or_4_ints              $170..173 |= $174..177
  665. bitwise_or_2_ints              $170..171 |= $172..173
  666. bitwise_or_int                 $170 |= $171
  667. bitwise_or_int                 $169 |= $170
  668. merge_condition_mask           CondMask = $168 & $169
  669. zero_slot_unmasked             $170 = 0
  670. copy_slot_masked               [test_no_op_mat3_X_scalar].result = Mask($170)
  671. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  672. load_condition_mask            CondMask = $168
  673. zero_2_slots_unmasked          $168..169 = 0
  674. shuffle                        $168..176 = ($168..176)[1 0 0 0 1 0 0 0 1]
  675. copy_4_slots_masked            mm₄(0..3) = Mask($168..171)
  676. copy_4_slots_masked            mm₄(4..7) = Mask($172..175)
  677. copy_slot_masked               mm₄(8) = Mask($176)
  678. zero_2_slots_unmasked          $168..169 = 0
  679. shuffle                        $168..176 = ($168..176)[1 0 0 0 1 0 0 0 1]
  680. copy_4_slots_masked            mm₄(0..3) = Mask($168..171)
  681. copy_4_slots_masked            mm₄(4..7) = Mask($172..175)
  682. copy_slot_masked               mm₄(8) = Mask($176)
  683. copy_4_slots_unmasked          $177..180 = z₄(0..3)
  684. copy_4_slots_unmasked          $181..184 = z₄(4..7)
  685. copy_slot_unmasked             $185 = z₄(8)
  686. cmpeq_n_floats                 $168..176 = equal($168..176, $177..185)
  687. bitwise_and_4_ints             $169..172 &= $173..176
  688. bitwise_and_2_ints             $169..170 &= $171..172
  689. bitwise_and_int                $169 &= $170
  690. bitwise_and_int                $168 &= $169
  691. copy_slot_masked               [test_no_op_mat3_X_scalar].result = Mask($168)
  692. load_return_mask               RetMask = $167
  693. copy_slot_unmasked             $167 = [test_no_op_mat3_X_scalar].result
  694. label                          label 0x0000000A
  695. copy_slot_masked               $166 = Mask($167)
  696. label                          label 0x00000002
  697. load_condition_mask            CondMask = $149
  698. zero_slot_unmasked             $0 = 0
  699. merge_condition_mask           CondMask = $165 & $166
  700. branch_if_no_lanes_active      branch_if_no_lanes_active +227 (label 1 at #927)
  701. store_return_mask              $1 = RetMask
  702. zero_4_slots_unmasked          m₅(0..3) = 0
  703. zero_4_slots_unmasked          m₅(4..7) = 0
  704. zero_4_slots_unmasked          m₅(8..11) = 0
  705. zero_4_slots_unmasked          m₅(12..15) = 0
  706. zero_4_slots_unmasked          mm₅(0..3) = 0
  707. zero_4_slots_unmasked          mm₅(4..7) = 0
  708. zero_4_slots_unmasked          mm₅(8..11) = 0
  709. zero_4_slots_unmasked          mm₅(12..15) = 0
  710. zero_2_slots_unmasked          $2..3 = 0
  711. shuffle                        $2..17 = ($2..17)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  712. copy_4_slots_unmasked          z₅(0..3) = $2..5
  713. copy_4_slots_unmasked          z₅(4..7) = $6..9
  714. copy_4_slots_unmasked          z₅(8..11) = $10..13
  715. copy_4_slots_unmasked          z₅(12..15) = $14..17
  716. copy_constant                  $2 = 0x3F800000 (1.0)
  717. swizzle_4                      $2..5 = ($2..5).xxxx
  718. copy_constant                  $6 = 0x3F800000 (1.0)
  719. swizzle_4                      $6..9 = ($6..9).xxxx
  720. copy_constant                  $10 = 0x3F800000 (1.0)
  721. swizzle_4                      $10..13 = ($10..13).xxxx
  722. copy_constant                  $14 = 0x3F800000 (1.0)
  723. swizzle_4                      $14..17 = ($14..17).xxxx
  724. copy_4_slots_unmasked          s₂(0..3) = $2..5
  725. copy_4_slots_unmasked          s₂(4..7) = $6..9
  726. copy_4_slots_unmasked          s₂(8..11) = $10..13
  727. copy_4_slots_unmasked          s₂(12..15) = $14..17
  728. copy_constant                  $2 = testInputs(0)
  729. copy_slot_unmasked             scalar₂ = $2
  730. swizzle_4                      $2..5 = ($2..5).xxxx
  731. copy_4_slots_unmasked          scalar4 = $2..5
  732. zero_slot_unmasked             $2 = 0
  733. copy_slot_unmasked             $3 = scalar₂
  734. shuffle                        $2..17 = ($2..17)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  735. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  736. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  737. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  738. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  739. zero_slot_unmasked             $2 = 0
  740. copy_slot_unmasked             $3 = scalar₂
  741. shuffle                        $2..17 = ($2..17)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  742. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  743. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  744. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  745. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  746. store_condition_mask           $2 = CondMask
  747. copy_4_slots_unmasked          $3..6 = m₅(0..3)
  748. copy_4_slots_unmasked          $7..10 = m₅(4..7)
  749. copy_4_slots_unmasked          $11..14 = m₅(8..11)
  750. copy_4_slots_unmasked          $15..18 = m₅(12..15)
  751. zero_slot_unmasked             $19 = 0
  752. copy_slot_unmasked             $20 = scalar₂
  753. shuffle                        $19..34 = ($19..34)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  754. cmpne_n_floats                 $3..18 = notEqual($3..18, $19..34)
  755. bitwise_or_4_ints              $11..14 |= $15..18
  756. bitwise_or_4_ints              $7..10 |= $11..14
  757. bitwise_or_4_ints              $3..6 |= $7..10
  758. bitwise_or_2_ints              $3..4 |= $5..6
  759. bitwise_or_int                 $3 |= $4
  760. merge_condition_mask           CondMask = $2 & $3
  761. zero_slot_unmasked             $4 = 0
  762. copy_slot_masked               [test_no_op_mat4_X_scalar].result = Mask($4)
  763. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  764. load_condition_mask            CondMask = $2
  765. copy_slot_unmasked             $2 = scalar₂
  766. swizzle_4                      $2..5 = ($2..5).xxxx
  767. copy_4_slots_unmasked          $6..9 = $2..5
  768. copy_4_slots_unmasked          $10..13 = $6..9
  769. copy_4_slots_unmasked          $14..17 = $10..13
  770. copy_4_slots_unmasked          $18..21 = s₂(0..3)
  771. copy_4_slots_unmasked          $22..25 = s₂(4..7)
  772. copy_4_slots_unmasked          $26..29 = s₂(8..11)
  773. copy_4_slots_unmasked          $30..33 = s₂(12..15)
  774. div_n_floats                   $2..17 /= $18..33
  775. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  776. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  777. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  778. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  779. store_condition_mask           $2 = CondMask
  780. copy_4_slots_unmasked          $3..6 = m₅(0..3)
  781. copy_4_slots_unmasked          $7..10 = m₅(4..7)
  782. copy_4_slots_unmasked          $11..14 = m₅(8..11)
  783. copy_4_slots_unmasked          $15..18 = m₅(12..15)
  784. copy_4_slots_unmasked          $19..22 = scalar4
  785. copy_4_slots_unmasked          $23..26 = scalar4
  786. copy_4_slots_unmasked          $27..30 = scalar4
  787. copy_4_slots_unmasked          $31..34 = scalar4
  788. cmpne_n_floats                 $3..18 = notEqual($3..18, $19..34)
  789. bitwise_or_4_ints              $11..14 |= $15..18
  790. bitwise_or_4_ints              $7..10 |= $11..14
  791. bitwise_or_4_ints              $3..6 |= $7..10
  792. bitwise_or_2_ints              $3..4 |= $5..6
  793. bitwise_or_int                 $3 |= $4
  794. merge_condition_mask           CondMask = $2 & $3
  795. zero_slot_unmasked             $4 = 0
  796. copy_slot_masked               [test_no_op_mat4_X_scalar].result = Mask($4)
  797. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  798. load_condition_mask            CondMask = $2
  799. copy_slot_unmasked             $2 = scalar₂
  800. swizzle_4                      $2..5 = ($2..5).xxxx
  801. copy_4_slots_unmasked          $6..9 = $2..5
  802. copy_4_slots_unmasked          $10..13 = $6..9
  803. copy_4_slots_unmasked          $14..17 = $10..13
  804. copy_4_slots_unmasked          $18..21 = z₅(0..3)
  805. copy_4_slots_unmasked          $22..25 = z₅(4..7)
  806. copy_4_slots_unmasked          $26..29 = z₅(8..11)
  807. copy_4_slots_unmasked          $30..33 = z₅(12..15)
  808. add_n_floats                   $2..17 += $18..33
  809. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  810. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  811. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  812. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  813. copy_4_slots_unmasked          $2..5 = z₅(0..3)
  814. copy_4_slots_unmasked          $6..9 = z₅(4..7)
  815. copy_4_slots_unmasked          $10..13 = z₅(8..11)
  816. copy_4_slots_unmasked          $14..17 = z₅(12..15)
  817. copy_slot_unmasked             $18 = scalar₂
  818. swizzle_4                      $18..21 = ($18..21).xxxx
  819. copy_4_slots_unmasked          $22..25 = $18..21
  820. copy_4_slots_unmasked          $26..29 = $22..25
  821. copy_4_slots_unmasked          $30..33 = $26..29
  822. add_n_floats                   $2..17 += $18..33
  823. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  824. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  825. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  826. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  827. store_condition_mask           $2 = CondMask
  828. copy_4_slots_unmasked          $3..6 = m₅(0..3)
  829. copy_4_slots_unmasked          $7..10 = m₅(4..7)
  830. copy_4_slots_unmasked          $11..14 = m₅(8..11)
  831. copy_4_slots_unmasked          $15..18 = m₅(12..15)
  832. copy_4_slots_unmasked          $19..22 = scalar4
  833. copy_4_slots_unmasked          $23..26 = scalar4
  834. copy_4_slots_unmasked          $27..30 = scalar4
  835. copy_4_slots_unmasked          $31..34 = scalar4
  836. cmpne_n_floats                 $3..18 = notEqual($3..18, $19..34)
  837. bitwise_or_4_ints              $11..14 |= $15..18
  838. bitwise_or_4_ints              $7..10 |= $11..14
  839. bitwise_or_4_ints              $3..6 |= $7..10
  840. bitwise_or_2_ints              $3..4 |= $5..6
  841. bitwise_or_int                 $3 |= $4
  842. merge_condition_mask           CondMask = $2 & $3
  843. zero_slot_unmasked             $4 = 0
  844. copy_slot_masked               [test_no_op_mat4_X_scalar].result = Mask($4)
  845. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  846. load_condition_mask            CondMask = $2
  847. copy_slot_unmasked             $2 = scalar₂
  848. swizzle_4                      $2..5 = ($2..5).xxxx
  849. copy_4_slots_unmasked          $6..9 = $2..5
  850. copy_4_slots_unmasked          $10..13 = $6..9
  851. copy_4_slots_unmasked          $14..17 = $10..13
  852. copy_4_slots_unmasked          $18..21 = z₅(0..3)
  853. copy_4_slots_unmasked          $22..25 = z₅(4..7)
  854. copy_4_slots_unmasked          $26..29 = z₅(8..11)
  855. copy_4_slots_unmasked          $30..33 = z₅(12..15)
  856. sub_n_floats                   $2..17 -= $18..33
  857. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  858. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  859. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  860. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  861. copy_4_slots_unmasked          $2..5 = z₅(0..3)
  862. copy_4_slots_unmasked          $6..9 = z₅(4..7)
  863. copy_4_slots_unmasked          $10..13 = z₅(8..11)
  864. copy_4_slots_unmasked          $14..17 = z₅(12..15)
  865. copy_slot_unmasked             $18 = scalar₂
  866. swizzle_4                      $18..21 = ($18..21).xxxx
  867. copy_4_slots_unmasked          $22..25 = $18..21
  868. copy_4_slots_unmasked          $26..29 = $22..25
  869. copy_4_slots_unmasked          $30..33 = $26..29
  870. sub_n_floats                   $2..17 -= $18..33
  871. copy_4_slots_masked            m₅(0..3) = Mask($2..5)
  872. copy_4_slots_masked            m₅(4..7) = Mask($6..9)
  873. copy_4_slots_masked            m₅(8..11) = Mask($10..13)
  874. copy_4_slots_masked            m₅(12..15) = Mask($14..17)
  875. store_condition_mask           $2 = CondMask
  876. copy_4_slots_unmasked          $3..6 = m₅(0..3)
  877. copy_4_slots_unmasked          $7..10 = m₅(4..7)
  878. copy_4_slots_unmasked          $11..14 = m₅(8..11)
  879. copy_4_slots_unmasked          $15..18 = m₅(12..15)
  880. zero_4_slots_unmasked          $19..22 = 0
  881. zero_4_slots_unmasked          $23..26 = 0
  882. zero_4_slots_unmasked          $27..30 = 0
  883. zero_4_slots_unmasked          $31..34 = 0
  884. copy_4_slots_unmasked          $35..38 = scalar4
  885. copy_4_slots_unmasked          $39..42 = scalar4
  886. copy_4_slots_unmasked          $43..46 = scalar4
  887. copy_4_slots_unmasked          $47..50 = scalar4
  888. sub_n_floats                   $19..34 -= $35..50
  889. cmpne_n_floats                 $3..18 = notEqual($3..18, $19..34)
  890. bitwise_or_4_ints              $11..14 |= $15..18
  891. bitwise_or_4_ints              $7..10 |= $11..14
  892. bitwise_or_4_ints              $3..6 |= $7..10
  893. bitwise_or_2_ints              $3..4 |= $5..6
  894. bitwise_or_int                 $3 |= $4
  895. merge_condition_mask           CondMask = $2 & $3
  896. zero_slot_unmasked             $4 = 0
  897. copy_slot_masked               [test_no_op_mat4_X_scalar].result = Mask($4)
  898. mask_off_return_mask           RetMask &= ~(CondMask & LoopMask & RetMask)
  899. load_condition_mask            CondMask = $2
  900. zero_2_slots_unmasked          $2..3 = 0
  901. shuffle                        $2..17 = ($2..17)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  902. copy_4_slots_masked            mm₅(0..3) = Mask($2..5)
  903. copy_4_slots_masked            mm₅(4..7) = Mask($6..9)
  904. copy_4_slots_masked            mm₅(8..11) = Mask($10..13)
  905. copy_4_slots_masked            mm₅(12..15) = Mask($14..17)
  906. zero_2_slots_unmasked          $2..3 = 0
  907. shuffle                        $2..17 = ($2..17)[1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1]
  908. copy_4_slots_masked            mm₅(0..3) = Mask($2..5)
  909. copy_4_slots_masked            mm₅(4..7) = Mask($6..9)
  910. copy_4_slots_masked            mm₅(8..11) = Mask($10..13)
  911. copy_4_slots_masked            mm₅(12..15) = Mask($14..17)
  912. copy_4_slots_unmasked          $18..21 = z₅(0..3)
  913. copy_4_slots_unmasked          $22..25 = z₅(4..7)
  914. copy_4_slots_unmasked          $26..29 = z₅(8..11)
  915. copy_4_slots_unmasked          $30..33 = z₅(12..15)
  916. cmpeq_n_floats                 $2..17 = equal($2..17, $18..33)
  917. bitwise_and_4_ints             $10..13 &= $14..17
  918. bitwise_and_4_ints             $6..9 &= $10..13
  919. bitwise_and_4_ints             $2..5 &= $6..9
  920. bitwise_and_2_ints             $2..3 &= $4..5
  921. bitwise_and_int                $2 &= $3
  922. copy_slot_masked               [test_no_op_mat4_X_scalar].result = Mask($2)
  923. load_return_mask               RetMask = $1
  924. copy_slot_unmasked             $1 = [test_no_op_mat4_X_scalar].result
  925. label                          label 0x0000000B
  926. copy_slot_masked               $0 = Mask($1)
  927. label                          label 0x00000001
  928. load_condition_mask            CondMask = $165
  929. swizzle_4                      $0..3 = ($0..3).xxxx
  930. copy_4_constants               $4..7 = colorRed
  931. copy_4_constants               $8..11 = colorGreen
  932. mix_4_ints                     $0..3 = mix($4..7, $8..11, $0..3)
  933. load_src                       src.rgba = $0..3