stride by 8 in llr dfs

Snakefile

@@ -134,35 +134,35 @@ TESTSET_FAST = [
 
 TESTSET_LOW_LEVEL_REPRESENTATION = [
     *expand(
-        "{kernel}/20x{N}xf32/{variant}",
+        "{kernel}/40x{N}xf32/{variant}",
         kernel=[
             "relu",
         ],
-        N=[4, 8, 12, 16, 20],
+        N=[8, 16, 24, 32, 40],
         variant=["snitch_stream"],
     ),
     *expand(
-        "{kernel}/{M}x20xf32/{variant}",
+        "{kernel}/{M}x40xf32/{variant}",
         kernel=[
             "relu",
         ],
-        M=[4, 8, 12, 16, 20],
+        M=[8, 16, 24, 32, 40],
         variant=["snitch_stream"],
     ),
     *expand(
-        "{kernel}/20x{N}xf32/{variant}",
+        "{kernel}/40x{N}xf32/{variant}",
         kernel=[
             "sum",
         ],
-        N=[4, 8, 12, 16, 20],
+        N=[8, 16, 24, 32, 40],
         variant=["linalg_xdsl"],
     ),
     *expand(
-        "{kernel}/{M}x20xf32/{variant}",
+        "{kernel}/{M}x40xf32/{variant}",
         kernel=[
             "sum",
         ],
-        M=[4, 8, 12, 16, 20],
+        M=[8, 16, 24, 32, 40],
         variant=["linalg_xdsl"],
     ),
     *expand(

results/kernels.low_level_representation.csv

@@ -8,21 +8,21 @@ matmul_transb,1x20x40xf32,snitch_stream,660,1459,1456,2.701195219123506,0,0,502,
 matmul_transb,1x20x8xf32,snitch_stream,219,1001,998,2.6666666666666665,0,0,102,272,0.4657534246575342,0.9444444444444444,108,0,0,0.4931506849315068,0,4,2.4,2.4,1,0,45,0.46875,51,0,0,0.2328767123287671,0,783,0.0,0.726027397260274,0.0
 matmul_transb,1x4x40xf32,snitch_stream,359,1138,1135,2.159340659340659,0,0,182,393,0.5069637883008357,0.8921568627450981,204,0,0,0.5682451253481894,0,20,0.9760765550239234,0.9760765550239234,1,0,209,0.7683823529411765,63,0,0,0.17548746518105848,0,780,0.0,0.7437325905292479,0.0
 matmul_transb,1x8x40xf32,snitch_stream,421,1212,1209,2.427480916030534,0,0,262,636,0.6223277909738717,0.9225352112676056,284,0,0,0.6745843230403801,0,20,1.3588516746411485,1.3588516746411483,1,0,209,0.7545126353790613,68,0,0,0.16152019002375298,0,792,0.0,0.836104513064133,0.0
-relu,12x20xf32,snitch_stream,171,915,912,0.991869918699187,0,0,123,122,0.7192982456140351,0.984,125,0,0,0.7309941520467836,0,0,17.857142857142858,17.857142857142858,1,0,7,0.2916666666666667,17,0,0,0.09941520467836257,0,745,0.0,0.8304093567251463,0.0
-relu,16x20xf32,snitch_stream,211,960,957,0.9938650306748467,0,0,163,162,0.7725118483412322,0.9878787878787879,165,0,0,0.7819905213270142,0,0,23.57142857142857,23.571428571428573,1,0,7,0.2916666666666667,17,0,0,0.08056872037914692,0,750,0.0,0.8625592417061612,0.0
-relu,20x12xf32,snitch_stream,171,915,912,0.991869918699187,0,0,123,122,0.7192982456140351,0.984,125,0,0,0.7309941520467836,0,0,17.857142857142858,17.857142857142858,1,0,7,0.2916666666666667,17,0,0,0.09941520467836257,0,745,0.0,0.8304093567251463,0.0
-relu,20x16xf32,snitch_stream,211,960,957,0.9938650306748467,0,0,163,162,0.7725118483412322,0.9878787878787879,165,0,0,0.7819905213270142,0,0,23.57142857142857,23.571428571428573,1,0,7,0.2916666666666667,17,0,0,0.08056872037914692,0,750,0.0,0.8625592417061612,0.0
-relu,20x20xf32,snitch_stream,251,1005,1002,0.9950738916256158,0,0,203,202,0.8087649402390438,0.9902439024390244,205,0,0,0.8167330677290837,0,0,29.285714285714285,29.285714285714285,1,0,7,0.2916666666666667,17,0,0,0.06772908366533864,0,755,0.0,0.8844621513944224,0.0
-relu,20x4xf32,snitch_stream,91,825,822,0.9767441860465116,0,0,43,42,0.4725274725274725,0.9555555555555556,45,0,0,0.4945054945054945,0,0,6.428571428571429,6.428571428571429,1,0,7,0.2916666666666667,17,0,0,0.18681318681318682,0,735,0.0,0.6813186813186813,0.0
-relu,20x8xf32,snitch_stream,131,870,867,0.9879518072289156,0,0,83,82,0.6335877862595419,0.9764705882352941,85,0,0,0.648854961832061,0,0,12.142857142857144,12.142857142857142,1,0,7,0.2916666666666667,17,0,0,0.1297709923664122,0,740,0.0,0.7786259541984732,0.0
-relu,4x20xf32,snitch_stream,91,825,822,0.9767441860465116,0,0,43,42,0.4725274725274725,0.9555555555555556,45,0,0,0.4945054945054945,0,0,6.428571428571429,6.428571428571429,1,0,7,0.2916666666666667,17,0,0,0.18681318681318682,0,735,0.0,0.6813186813186813,0.0
-relu,8x20xf32,snitch_stream,131,870,867,0.9879518072289156,0,0,83,82,0.6335877862595419,0.9764705882352941,85,0,0,0.648854961832061,0,0,12.142857142857144,12.142857142857142,1,0,7,0.2916666666666667,17,0,0,0.1297709923664122,0,740,0.0,0.7786259541984732,0.0
-sum,12x20xf32,linalg_xdsl,191,980,976,2.975206611570248,0,0,121,360,0.6335078534031413,0.983739837398374,123,0,0,0.643979057591623,0,0,24.599999999999998,24.6,1,0,5,0.19230769230769232,21,0,0,0.1099476439790576,0,790,0.0,0.7539267015706805,0.0
-sum,16x20xf32,linalg_xdsl,231,1030,1026,2.981366459627329,0,0,161,480,0.696969696969697,0.9877300613496932,163,0,0,0.7056277056277056,0,0,32.6,32.6,1,0,5,0.19230769230769232,21,0,0,0.09090909090909091,0,800,0.0,0.7965367965367965,0.0
-sum,20x12xf32,linalg_xdsl,191,980,976,2.975206611570248,0,0,121,360,0.6335078534031413,0.983739837398374,123,0,0,0.643979057591623,0,0,24.599999999999998,24.6,1,0,5,0.19230769230769232,21,0,0,0.1099476439790576,0,790,0.0,0.7539267015706805,0.0
-sum,20x16xf32,linalg_xdsl,231,1030,1026,2.981366459627329,0,0,161,480,0.696969696969697,0.9877300613496932,163,0,0,0.7056277056277056,0,0,32.6,32.6,1,0,5,0.19230769230769232,21,0,0,0.09090909090909091,0,800,0.0,0.7965367965367965,0.0
-sum,20x20xf32,linalg_xdsl,271,1080,1076,2.985074626865672,0,0,201,600,0.7416974169741697,0.9901477832512315,203,0,0,0.7490774907749077,0,0,40.6,40.6,1,0,5,0.19230769230769232,21,0,0,0.07749077490774908,0,810,0.0,0.8265682656826567,0.0
-sum,20x4xf32,linalg_xdsl,111,880,876,2.926829268292683,0,0,41,120,0.36936936936936937,0.9534883720930233,43,0,0,0.38738738738738737,0,0,8.6,8.6,1,0,5,0.19230769230769232,21,0,0,0.1891891891891892,0,770,0.0,0.5765765765765766,0.0
-sum,20x8xf32,linalg_xdsl,151,930,926,2.962962962962963,0,0,81,240,0.5364238410596026,0.9759036144578314,83,0,0,0.5496688741721855,0,0,16.599999999999998,16.6,1,0,5,0.19230769230769232,21,0,0,0.1390728476821192,0,780,0.0,0.6887417218543046,0.0
-sum,4x20xf32,linalg_xdsl,111,880,876,2.926829268292683,0,0,41,120,0.36936936936936937,0.9534883720930233,43,0,0,0.38738738738738737,0,0,8.6,8.6,1,0,5,0.19230769230769232,21,0,0,0.1891891891891892,0,770,0.0,0.5765765765765766,0.0
-sum,8x20xf32,linalg_xdsl,151,930,926,2.962962962962963,0,0,81,240,0.5364238410596026,0.9759036144578314,83,0,0,0.5496688741721855,0,0,16.599999999999998,16.6,1,0,5,0.19230769230769232,21,0,0,0.1390728476821192,0,780,0.0,0.6887417218543046,0.0
+relu,16x40xf32,snitch_stream,371,1151,1148,0.9969040247678018,0,0,323,322,0.8706199460916442,0.9938461538461538,325,0,0,0.876010781671159,0,0,46.42857142857143,46.42857142857143,1,0,7,0.28,18,0,0,0.04851752021563342,0,781,0.0,0.9245283018867924,0.0
+relu,24x40xf32,snitch_stream,531,1331,1328,0.9979296066252588,0,0,483,482,0.9096045197740112,0.9958762886597938,485,0,0,0.9133709981167608,0,0,69.28571428571428,69.28571428571429,1,0,7,0.28,18,0,0,0.03389830508474576,0,801,0.0,0.9472693032015066,0.0
+relu,32x40xf32,snitch_stream,688,1502,1499,0.9984447900466563,0,0,643,642,0.934593023255814,0.9968992248062015,645,0,0,0.9375,0,0,92.14285714285715,92.14285714285714,1,0,7,0.2916666666666667,17,0,0,0.024709302325581394,0,815,0.0,0.9622093023255814,0.0
+relu,40x16xf32,snitch_stream,371,1151,1148,0.9969040247678018,0,0,323,322,0.8706199460916442,0.9938461538461538,325,0,0,0.876010781671159,0,0,46.42857142857143,46.42857142857143,1,0,7,0.28,18,0,0,0.04851752021563342,0,781,0.0,0.9245283018867924,0.0
+relu,40x24xf32,snitch_stream,531,1331,1328,0.9979296066252588,0,0,483,482,0.9096045197740112,0.9958762886597938,485,0,0,0.9133709981167608,0,0,69.28571428571428,69.28571428571429,1,0,7,0.28,18,0,0,0.03389830508474576,0,801,0.0,0.9472693032015066,0.0
+relu,40x32xf32,snitch_stream,688,1502,1499,0.9984447900466563,0,0,643,642,0.934593023255814,0.9968992248062015,645,0,0,0.9375,0,0,92.14285714285715,92.14285714285714,1,0,7,0.2916666666666667,17,0,0,0.024709302325581394,0,815,0.0,0.9622093023255814,0.0
+relu,40x40xf32,snitch_stream,848,1682,1679,0.9987546699875467,0,0,803,802,0.9469339622641509,0.9975155279503105,805,0,0,0.9492924528301887,0,0,115.0,115.0,1,0,7,0.2916666666666667,17,0,0,0.020047169811320754,0,835,0.0,0.9693396226415094,0.0
+relu,40x8xf32,snitch_stream,211,960,957,0.9938650306748467,0,0,163,162,0.7725118483412322,0.9878787878787879,165,0,0,0.7819905213270142,0,0,23.57142857142857,23.571428571428573,1,0,7,0.2916666666666667,17,0,0,0.08056872037914692,0,750,0.0,0.8625592417061612,0.0
+relu,8x40xf32,snitch_stream,211,960,957,0.9938650306748467,0,0,163,162,0.7725118483412322,0.9878787878787879,165,0,0,0.7819905213270142,0,0,23.57142857142857,23.571428571428573,1,0,7,0.2916666666666667,17,0,0,0.08056872037914692,0,750,0.0,0.8625592417061612,0.0
+sum,16x40xf32,linalg_xdsl,385,1219,1215,2.9906542056074765,0,0,321,960,0.8337662337662337,0.9938080495356038,323,0,0,0.8389610389610389,0,0,64.6,64.6,1,0,5,0.18518518518518517,22,0,0,0.05714285714285714,0,835,0.0,0.8961038961038961,0.0
+sum,24x40xf32,linalg_xdsl,556,1436,1432,2.993762993762994,0,0,481,1440,0.8651079136690647,0.9958592132505176,483,0,0,0.8687050359712231,0,0,96.6,96.6,1,0,5,0.17857142857142858,23,0,0,0.04136690647482014,0,881,0.0,0.9100719424460432,0.0
+sum,32x40xf32,linalg_xdsl,705,1628,1624,2.995319812792512,0,0,641,1920,0.9092198581560283,0.9968895800933126,643,0,0,0.9120567375886525,0,0,128.6,128.6,1,0,5,0.17857142857142858,23,0,0,0.032624113475177303,0,924,0.0,0.9446808510638298,0.0
+sum,40x16xf32,linalg_xdsl,385,1219,1215,2.9906542056074765,0,0,321,960,0.8337662337662337,0.9938080495356038,323,0,0,0.8389610389610389,0,0,64.6,64.6,1,0,5,0.18518518518518517,22,0,0,0.05714285714285714,0,835,0.0,0.8961038961038961,0.0
+sum,40x24xf32,linalg_xdsl,556,1436,1432,2.993762993762994,0,0,481,1440,0.8651079136690647,0.9958592132505176,483,0,0,0.8687050359712231,0,0,96.6,96.6,1,0,5,0.17857142857142858,23,0,0,0.04136690647482014,0,881,0.0,0.9100719424460432,0.0
+sum,40x32xf32,linalg_xdsl,705,1628,1624,2.995319812792512,0,0,641,1920,0.9092198581560283,0.9968895800933126,643,0,0,0.9120567375886525,0,0,128.6,128.6,1,0,5,0.17857142857142858,23,0,0,0.032624113475177303,0,924,0.0,0.9446808510638298,0.0
+sum,40x40xf32,linalg_xdsl,865,1833,1829,2.9962546816479403,0,0,801,2400,0.9260115606936417,0.9975093399750934,803,0,0,0.9283236994219654,0,0,160.6,160.6,1,0,5,0.17857142857142858,23,0,0,0.026589595375722544,0,969,0.0,0.9549132947976879,0.0
+sum,40x8xf32,linalg_xdsl,231,1030,1026,2.981366459627329,0,0,161,480,0.696969696969697,0.9877300613496932,163,0,0,0.7056277056277056,0,0,32.6,32.6,1,0,5,0.19230769230769232,21,0,0,0.09090909090909091,0,800,0.0,0.7965367965367965,0.0
+sum,8x40xf32,linalg_xdsl,231,1030,1026,2.981366459627329,0,0,161,480,0.696969696969697,0.9877300613496932,163,0,0,0.7056277056277056,0,0,32.6,32.6,1,0,5,0.19230769230769232,21,0,0,0.09090909090909091,0,800,0.0,0.7965367965367965,0.0