From 8e1be8f5632cb6c247c8b42786c913a0422289ff Mon Sep 17 00:00:00 2001 From: Sasha Lopoukhine Date: Sat, 16 Mar 2024 16:25:07 +0000 Subject: [PATCH] sort columns in csv --- .../conv2d_d1_s1_3x3/1x1x8x8xf64/tests.csv | 10 +- kernels/conv2d_d1_s1_3x3/params.csv | 10 +- kernels/ddot/128xf64/tests.csv | 14 +-- kernels/ddot/params.csv | 14 +-- kernels/dense/8x8xf64/tests.csv | 12 +-- kernels/dense/params.csv | 12 +-- kernels/dsum/8x16xf32/tests.csv | 16 ++-- kernels/dsum/params.csv | 16 ++-- kernels/fill/16x16xf64/tests.csv | 12 +-- kernels/fill/params.csv | 12 +-- kernels/kernels.csv | 94 +++++++++---------- kernels/matmul/8x8xf64/baseline.csv | 4 +- kernels/matmul/8x8xf64/linalg.csv | 4 +- kernels/matmul/8x8xf64/snitch_stream.csv | 4 +- kernels/matmul/8x8xf64/snrt.csv | 4 +- kernels/matmul/8x8xf64/tests.csv | 10 +- .../1x1x16x16xf64/baseline.csv | 4 +- .../1x1x16x16xf64/linalg.csv | 4 +- .../1x1x16x16xf64/snitch_stream.csv | 4 +- .../1x1x16x16xf64/snrt.csv | 4 +- .../1x1x16x16xf64/tests.csv | 10 +- kernels/pooling_nchw_max_d1_s2_3x3/params.csv | 10 +- .../1x1x16x16xf64/baseline.csv | 4 +- .../1x1x16x16xf64/linalg.csv | 4 +- .../1x1x16x16xf64/snitch_stream.csv | 4 +- .../1x1x16x16xf64/snrt.csv | 4 +- .../1x1x16x16xf64/tests.csv | 10 +- kernels/pooling_nchw_sum_d1_s2_3x3/params.csv | 10 +- kernels/relu/16x16xf64/baseline.csv | 4 +- kernels/relu/16x16xf64/linalg.csv | 4 +- kernels/relu/16x16xf64/scf_xdsl.csv | 4 +- kernels/relu/16x16xf64/snitch_stream.csv | 4 +- kernels/relu/16x16xf64/snrt.csv | 4 +- kernels/relu/16x16xf64/ssr.csv | 4 +- kernels/relu/16x16xf64/ssr_frep.csv | 4 +- kernels/relu/16x16xf64/tests.csv | 16 ++-- kernels/relu/params.csv | 16 ++-- snitch/Makefile.rules | 2 +- 38 files changed, 191 insertions(+), 191 deletions(-) diff --git a/kernels/conv2d_d1_s1_3x3/1x1x8x8xf64/tests.csv b/kernels/conv2d_d1_s1_3x3/1x1x8x8xf64/tests.csv index f66adf0e..71903bb5 100644 --- a/kernels/conv2d_d1_s1_3x3/1x1x8x8xf64/tests.csv +++ b/kernels/conv2d_d1_s1_3x3/1x1x8x8xf64/tests.csv @@ -1,5 +1,5 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,761,33,765,2319,765,369,459,354,1080,36,0.0,2297,0,0,0,0.0211674150096215,0.9586466165413534,1.0,1.0,22,3.0508474576271185,1.2439024390243902,0.490699166132136,0.2270686337395766,0.4627450980392157,1559,0.5118665811417575 -linalg,0.0,754,34,729,2242,729,369,561,319,972,36,0.0,2221,0,0,0,0.0228341168569509,0.9554390563564876,1.0,1.0,21,3.047021943573668,1.5203252032520325,0.4895903290799194,0.2142377434519811,0.4375857338820302,1489,0.5124244459368704 -snitch_stream,0.0,743,150,146,2245,398,36,36,274,972,36,0.0,2029,0,0,0,0.0998003992015968,0.4932432432432432,2.7260273972602738,2.7260273972602738,216,3.5474452554744524,1.0,0.2648037258815702,0.1823020625415835,0.6884422110552764,1503,0.364604125083167 -snrt,0.0,756,66,146,2241,398,36,36,272,972,36,0.0,2018,0,0,0,0.044414535666218,0.6886792452830188,2.7260273972602738,2.7260273972602738,223,3.573529411764706,1.0,0.2678331090174966,0.1830417227456258,0.6834170854271356,1486,0.3122476446837147 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1559,2297,2319,3.0508474576271185,1.2439024390243902,354,1080,0.2270686337395766,0.4627450980392157,765,459,369,0.490699166132136,22,36,1.0,1.0,baseline,0,765,0.9586466165413534,33,0,0,0.0211674150096215,761,0.0,0.5118665811417575,0.0 +1489,2221,2242,3.047021943573668,1.5203252032520325,319,972,0.2142377434519811,0.4375857338820302,729,561,369,0.4895903290799194,21,36,1.0,1.0,linalg,0,729,0.9554390563564876,34,0,0,0.0228341168569509,754,0.0,0.5124244459368704,0.0 +1503,2029,2245,3.5474452554744524,1.0,274,972,0.1823020625415835,0.6884422110552764,398,36,36,0.2648037258815702,216,36,2.7260273972602738,2.7260273972602738,snitch_stream,0,146,0.4932432432432432,150,0,0,0.0998003992015968,743,0.0,0.364604125083167,0.0 +1486,2018,2241,3.573529411764706,1.0,272,972,0.1830417227456258,0.6834170854271356,398,36,36,0.2678331090174966,223,36,2.7260273972602738,2.7260273972602738,snrt,0,146,0.6886792452830188,66,0,0,0.044414535666218,756,0.0,0.3122476446837147,0.0 diff --git a/kernels/conv2d_d1_s1_3x3/params.csv b/kernels/conv2d_d1_s1_3x3/params.csv index b4dead45..543d7246 100644 --- a/kernels/conv2d_d1_s1_3x3/params.csv +++ b/kernels/conv2d_d1_s1_3x3/params.csv @@ -1,5 +1,5 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -1x1x8x8xf64,baseline,0.0,761,33,765,2319,765,369,459,354,1080,36,0.0,2297,0,0,0,0.0211674150096215,0.9586466165413534,1.0,1.0,22,3.0508474576271185,1.2439024390243902,0.490699166132136,0.2270686337395766,0.4627450980392157,1559,0.5118665811417575 -1x1x8x8xf64,linalg,0.0,754,34,729,2242,729,369,561,319,972,36,0.0,2221,0,0,0,0.0228341168569509,0.9554390563564876,1.0,1.0,21,3.047021943573668,1.5203252032520325,0.4895903290799194,0.2142377434519811,0.4375857338820302,1489,0.5124244459368704 -1x1x8x8xf64,snitch_stream,0.0,743,150,146,2245,398,36,36,274,972,36,0.0,2029,0,0,0,0.0998003992015968,0.4932432432432432,2.7260273972602738,2.7260273972602738,216,3.5474452554744524,1.0,0.2648037258815702,0.1823020625415835,0.6884422110552764,1503,0.364604125083167 -1x1x8x8xf64,snrt,0.0,756,66,146,2241,398,36,36,272,972,36,0.0,2018,0,0,0,0.044414535666218,0.6886792452830188,2.7260273972602738,2.7260273972602738,223,3.573529411764706,1.0,0.2678331090174966,0.1830417227456258,0.6834170854271356,1486,0.3122476446837147 +params,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1x1x8x8xf64,1559,2297,2319,3.0508474576271185,1.2439024390243902,354,1080,0.2270686337395766,0.4627450980392157,765,459,369,0.490699166132136,22,36,1.0,1.0,baseline,0,765,0.9586466165413534,33,0,0,0.0211674150096215,761,0.0,0.5118665811417575,0.0 +1x1x8x8xf64,1489,2221,2242,3.047021943573668,1.5203252032520325,319,972,0.2142377434519811,0.4375857338820302,729,561,369,0.4895903290799194,21,36,1.0,1.0,linalg,0,729,0.9554390563564876,34,0,0,0.0228341168569509,754,0.0,0.5124244459368704,0.0 +1x1x8x8xf64,1503,2029,2245,3.5474452554744524,1.0,274,972,0.1823020625415835,0.6884422110552764,398,36,36,0.2648037258815702,216,36,2.7260273972602738,2.7260273972602738,snitch_stream,0,146,0.4932432432432432,150,0,0,0.0998003992015968,743,0.0,0.364604125083167,0.0 +1x1x8x8xf64,1486,2018,2241,3.573529411764706,1.0,272,972,0.1830417227456258,0.6834170854271356,398,36,36,0.2678331090174966,223,36,2.7260273972602738,2.7260273972602738,snrt,0,146,0.6886792452830188,66,0,0,0.044414535666218,756,0.0,0.3122476446837147,0.0 diff --git a/kernels/ddot/128xf64/tests.csv b/kernels/ddot/128xf64/tests.csv index 3cf29e5e..b978af79 100644 --- a/kernels/ddot/128xf64/tests.csv +++ b/kernels/ddot/128xf64/tests.csv @@ -1,7 +1,7 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,752,520,386,1672,386,257,257,128,384,1,0.0,1680,0,0,0,0.5597416576964478,0.4260485651214128,1.0,1.0,0,3.0,1.0,0.4155005382131324,0.1377825618945102,0.3316062176165803,929,0.9752421959095802 -linalg,0.0,758,521,386,1685,386,257,257,128,384,1,0.0,1694,0,0,0,0.5560298826040555,0.4255788313120176,1.0,1.0,0,3.0,1.0,0.4119530416221985,0.1366061899679829,0.3316062176165803,937,0.967982924226254 -snitch_stream,0.0,762,16,6,1331,132,1,1,0,384,1,0.0,819,0,0,0,0.0280701754385964,0.2727272727272727,22.0,0.0,512,0.0,1.0,0.231578947368421,0.0,0.0,570,0.2596491228070175 -snrt,0.0,758,12,6,1307,132,1,1,2,384,1,0.0,804,0,0,0,0.0218181818181818,0.3333333333333333,22.0,22.0,503,192.0,1.0,0.24,0.0036363636363636,0.0151515151515151,550,0.2618181818181818 -scf_xdsl,0.0,757,524,386,1717,386,257,257,127,384,1,0.0,1715,0,0,0,0.5452653485952134,0.4241758241758241,1.0,1.0,2,3.0236220472440944,1.0,0.4016649323621228,0.1321540062434963,0.3290155440414508,961,0.946930280957336 -scf,0.0,758,521,386,1685,386,257,257,128,384,1,0.0,1694,0,0,0,0.5560298826040555,0.4255788313120176,1.0,1.0,0,3.0,1.0,0.4119530416221985,0.1366061899679829,0.3316062176165803,937,0.967982924226254 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +929,1680,1672,3.0,1.0,128,384,0.1377825618945102,0.3316062176165803,386,257,257,0.4155005382131324,0,1,1.0,1.0,baseline,0,386,0.4260485651214128,520,0,0,0.5597416576964478,752,0.0,0.9752421959095802,0.0 +937,1694,1685,3.0,1.0,128,384,0.1366061899679829,0.3316062176165803,386,257,257,0.4119530416221985,0,1,1.0,1.0,linalg,0,386,0.4255788313120176,521,0,0,0.5560298826040555,758,0.0,0.967982924226254,0.0 +570,819,1331,0.0,1.0,0,384,0.0,0.0,132,1,1,0.231578947368421,512,1,0.0,22.0,snitch_stream,0,6,0.2727272727272727,16,0,0,0.0280701754385964,762,0.0,0.2596491228070175,0.0 +550,804,1307,192.0,1.0,2,384,0.0036363636363636,0.0151515151515151,132,1,1,0.24,503,1,22.0,22.0,snrt,0,6,0.3333333333333333,12,0,0,0.0218181818181818,758,0.0,0.2618181818181818,0.0 +961,1715,1717,3.0236220472440944,1.0,127,384,0.1321540062434963,0.3290155440414508,386,257,257,0.4016649323621228,2,1,1.0,1.0,scf_xdsl,0,386,0.4241758241758241,524,0,0,0.5452653485952134,757,0.0,0.946930280957336,0.0 +937,1694,1685,3.0,1.0,128,384,0.1366061899679829,0.3316062176165803,386,257,257,0.4119530416221985,0,1,1.0,1.0,scf,0,386,0.4255788313120176,521,0,0,0.5560298826040555,758,0.0,0.967982924226254,0.0 diff --git a/kernels/ddot/params.csv b/kernels/ddot/params.csv index f7cb6294..2103de4e 100644 --- a/kernels/ddot/params.csv +++ b/kernels/ddot/params.csv @@ -1,7 +1,7 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -128xf64,baseline,0.0,752,520,386,1672,386,257,257,128,384,1,0.0,1680,0,0,0,0.5597416576964478,0.4260485651214128,1.0,1.0,0,3.0,1.0,0.4155005382131324,0.1377825618945102,0.3316062176165803,929,0.9752421959095802 -128xf64,linalg,0.0,758,521,386,1685,386,257,257,128,384,1,0.0,1694,0,0,0,0.5560298826040555,0.4255788313120176,1.0,1.0,0,3.0,1.0,0.4119530416221985,0.1366061899679829,0.3316062176165803,937,0.967982924226254 -128xf64,snitch_stream,0.0,762,16,6,1331,132,1,1,0,384,1,0.0,819,0,0,0,0.0280701754385964,0.2727272727272727,22.0,0.0,512,0.0,1.0,0.231578947368421,0.0,0.0,570,0.2596491228070175 -128xf64,snrt,0.0,758,12,6,1307,132,1,1,2,384,1,0.0,804,0,0,0,0.0218181818181818,0.3333333333333333,22.0,22.0,503,192.0,1.0,0.24,0.0036363636363636,0.0151515151515151,550,0.2618181818181818 -128xf64,scf_xdsl,0.0,757,524,386,1717,386,257,257,127,384,1,0.0,1715,0,0,0,0.5452653485952134,0.4241758241758241,1.0,1.0,2,3.0236220472440944,1.0,0.4016649323621228,0.1321540062434963,0.3290155440414508,961,0.946930280957336 -128xf64,scf,0.0,758,521,386,1685,386,257,257,128,384,1,0.0,1694,0,0,0,0.5560298826040555,0.4255788313120176,1.0,1.0,0,3.0,1.0,0.4119530416221985,0.1366061899679829,0.3316062176165803,937,0.967982924226254 +params,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +128xf64,929,1680,1672,3.0,1.0,128,384,0.1377825618945102,0.3316062176165803,386,257,257,0.4155005382131324,0,1,1.0,1.0,baseline,0,386,0.4260485651214128,520,0,0,0.5597416576964478,752,0.0,0.9752421959095802,0.0 +128xf64,937,1694,1685,3.0,1.0,128,384,0.1366061899679829,0.3316062176165803,386,257,257,0.4119530416221985,0,1,1.0,1.0,linalg,0,386,0.4255788313120176,521,0,0,0.5560298826040555,758,0.0,0.967982924226254,0.0 +128xf64,570,819,1331,0.0,1.0,0,384,0.0,0.0,132,1,1,0.231578947368421,512,1,0.0,22.0,snitch_stream,0,6,0.2727272727272727,16,0,0,0.0280701754385964,762,0.0,0.2596491228070175,0.0 +128xf64,550,804,1307,192.0,1.0,2,384,0.0036363636363636,0.0151515151515151,132,1,1,0.24,503,1,22.0,22.0,snrt,0,6,0.3333333333333333,12,0,0,0.0218181818181818,758,0.0,0.2618181818181818,0.0 +128xf64,961,1715,1717,3.0236220472440944,1.0,127,384,0.1321540062434963,0.3290155440414508,386,257,257,0.4016649323621228,2,1,1.0,1.0,scf_xdsl,0,386,0.4241758241758241,524,0,0,0.5452653485952134,757,0.0,0.946930280957336,0.0 +128xf64,937,1694,1685,3.0,1.0,128,384,0.1366061899679829,0.3316062176165803,386,257,257,0.4119530416221985,0,1,1.0,1.0,scf,0,386,0.4255788313120176,521,0,0,0.5560298826040555,758,0.0,0.967982924226254,0.0 diff --git a/kernels/dense/8x8xf64/tests.csv b/kernels/dense/8x8xf64/tests.csv index 0df02b4d..7d92d819 100644 --- a/kernels/dense/8x8xf64/tests.csv +++ b/kernels/dense/8x8xf64/tests.csv @@ -1,6 +1,6 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,752,79,1628,3981,1628,247,740,852,641,1793,0.0,3990,0,0,0,0.024390243902439,0.9537199765670769,1.0,1.0,0,2.797191887675507,1.1513513513513514,0.5026242667489966,0.1979005866008027,0.3937346437346437,3239,0.5270145106514357 -fused,0.0,770,54,1388,3733,1388,120,620,659,646,1808,0.0,3715,0,0,0,0.0182186234817813,0.9625520110957004,1.0,1.0,18,2.7987616099071206,1.0629032258064517,0.4682860998650472,0.2179487179487179,0.4654178674351585,2964,0.4865047233468286 -linalg,0.0,757,80,1628,3986,1628,247,740,852,641,1793,0.0,3994,0,0,0,0.0247066090179122,0.9531615925058547,1.0,1.0,0,2.797191887675507,1.1513513513513514,0.5027794935145151,0.1979617047560222,0.3937346437346437,3238,0.5274861025324274 -snrt,0.0,757,114,387,3465,771,64,64,64,612,1793,0.0,3349,0,0,0,0.0420819490586932,0.7724550898203593,1.992248062015504,1.992248062015504,116,2.929738562091503,1.0,0.2846068660022148,0.2259136212624584,0.7937743190661478,2709,0.3266888150609081 -snitch_stream,0.0,762,300,387,3491,771,64,64,64,612,1793,0.0,3375,0,0,0,0.1098901098901098,0.5633187772925764,1.992248062015504,1.992248062015504,116,2.929738562091503,1.0,0.2824175824175824,0.2241758241758241,0.7937743190661478,2730,0.3923076923076923 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +3239,3990,3981,2.797191887675507,1.1513513513513514,641,1793,0.1979005866008027,0.3937346437346437,1628,852,740,0.5026242667489966,0,247,1.0,1.0,baseline,0,1628,0.9537199765670769,79,0,0,0.024390243902439,752,0.0,0.5270145106514357,0.0 +2964,3715,3733,2.7987616099071206,1.0629032258064517,646,1808,0.2179487179487179,0.4654178674351585,1388,659,620,0.4682860998650472,18,120,1.0,1.0,fused,0,1388,0.9625520110957004,54,0,0,0.0182186234817813,770,0.0,0.4865047233468286,0.0 +3238,3994,3986,2.797191887675507,1.1513513513513514,641,1793,0.1979617047560222,0.3937346437346437,1628,852,740,0.5027794935145151,0,247,1.0,1.0,linalg,0,1628,0.9531615925058547,80,0,0,0.0247066090179122,757,0.0,0.5274861025324274,0.0 +2709,3349,3465,2.929738562091503,1.0,612,1793,0.2259136212624584,0.7937743190661478,771,64,64,0.2846068660022148,116,64,1.992248062015504,1.992248062015504,snrt,0,387,0.7724550898203593,114,0,0,0.0420819490586932,757,0.0,0.3266888150609081,0.0 +2730,3375,3491,2.929738562091503,1.0,612,1793,0.2241758241758241,0.7937743190661478,771,64,64,0.2824175824175824,116,64,1.992248062015504,1.992248062015504,snitch_stream,0,387,0.5633187772925764,300,0,0,0.1098901098901098,762,0.0,0.3923076923076923,0.0 diff --git a/kernels/dense/params.csv b/kernels/dense/params.csv index cd1e6aba..a4cc4a81 100644 --- a/kernels/dense/params.csv +++ b/kernels/dense/params.csv @@ -1,6 +1,6 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -8x8xf64,baseline,0.0,752,79,1628,3981,1628,247,740,852,641,1793,0.0,3990,0,0,0,0.024390243902439,0.9537199765670769,1.0,1.0,0,2.797191887675507,1.1513513513513514,0.5026242667489966,0.1979005866008027,0.3937346437346437,3239,0.5270145106514357 -8x8xf64,fused,0.0,770,54,1388,3733,1388,120,620,659,646,1808,0.0,3715,0,0,0,0.0182186234817813,0.9625520110957004,1.0,1.0,18,2.7987616099071206,1.0629032258064517,0.4682860998650472,0.2179487179487179,0.4654178674351585,2964,0.4865047233468286 -8x8xf64,linalg,0.0,757,80,1628,3986,1628,247,740,852,641,1793,0.0,3994,0,0,0,0.0247066090179122,0.9531615925058547,1.0,1.0,0,2.797191887675507,1.1513513513513514,0.5027794935145151,0.1979617047560222,0.3937346437346437,3238,0.5274861025324274 -8x8xf64,snrt,0.0,757,114,387,3465,771,64,64,64,612,1793,0.0,3349,0,0,0,0.0420819490586932,0.7724550898203593,1.992248062015504,1.992248062015504,116,2.929738562091503,1.0,0.2846068660022148,0.2259136212624584,0.7937743190661478,2709,0.3266888150609081 -8x8xf64,snitch_stream,0.0,762,300,387,3491,771,64,64,64,612,1793,0.0,3375,0,0,0,0.1098901098901098,0.5633187772925764,1.992248062015504,1.992248062015504,116,2.929738562091503,1.0,0.2824175824175824,0.2241758241758241,0.7937743190661478,2730,0.3923076923076923 +params,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +8x8xf64,3239,3990,3981,2.797191887675507,1.1513513513513514,641,1793,0.1979005866008027,0.3937346437346437,1628,852,740,0.5026242667489966,0,247,1.0,1.0,baseline,0,1628,0.9537199765670769,79,0,0,0.024390243902439,752,0.0,0.5270145106514357,0.0 +8x8xf64,2964,3715,3733,2.7987616099071206,1.0629032258064517,646,1808,0.2179487179487179,0.4654178674351585,1388,659,620,0.4682860998650472,18,120,1.0,1.0,fused,0,1388,0.9625520110957004,54,0,0,0.0182186234817813,770,0.0,0.4865047233468286,0.0 +8x8xf64,3238,3994,3986,2.797191887675507,1.1513513513513514,641,1793,0.1979617047560222,0.3937346437346437,1628,852,740,0.5027794935145151,0,247,1.0,1.0,linalg,0,1628,0.9531615925058547,80,0,0,0.0247066090179122,757,0.0,0.5274861025324274,0.0 +8x8xf64,2709,3349,3465,2.929738562091503,1.0,612,1793,0.2259136212624584,0.7937743190661478,771,64,64,0.2846068660022148,116,64,1.992248062015504,1.992248062015504,snrt,0,387,0.7724550898203593,114,0,0,0.0420819490586932,757,0.0,0.3266888150609081,0.0 +8x8xf64,2730,3375,3491,2.929738562091503,1.0,612,1793,0.2241758241758241,0.7937743190661478,771,64,64,0.2824175824175824,116,64,1.992248062015504,1.992248062015504,snitch_stream,0,387,0.5633187772925764,300,0,0,0.1098901098901098,762,0.0,0.3923076923076923,0.0 diff --git a/kernels/dsum/8x16xf32/tests.csv b/kernels/dsum/8x16xf32/tests.csv index ac6b2562..61158488 100644 --- a/kernels/dsum/8x16xf32/tests.csv +++ b/kernels/dsum/8x16xf32/tests.csv @@ -1,8 +1,8 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,713,649,512,1905,512,256,256,128,384,128.0,0.0,1914,0,0,0,0.5399334442595674,0.4409991386735573,1.0,1.0,0,3.0,1.0,0.4259567387687188,0.1064891846921797,0.25,1202,0.9658901830282862 -linalg,0.0,713,53,512,1504,512,256,424,128,384,128.0,0.0,1506,0,0,0,0.0667506297229219,0.9061946902654868,1.0,1.0,0,3.0,1.65625,0.6448362720403022,0.1612090680100755,0.25,794,0.7115869017632241 -snrt,0.0,732,15,5,918,131,0,0,129,385,,0.0,918,0,0,0,0.0802139037433155,0.25,26.2,26.200000000000003,0,2.984496124031008,0.0,0.7005347593582888,0.6898395721925134,0.9847328244274808,187,0.7807486631016043 -ssr1d,0.0,733,14,130,979,130,0,0,128,384,,0.0,985,0,0,0,0.0553359683794466,0.9027777777777778,1.0,1.0,0,3.0,0.0,0.5138339920948617,0.5059288537549407,0.9846153846153848,253,0.5691699604743083 -ssr2d,0.0,733,18,130,995,130,0,0,128,384,,0.0,1005,0,0,0,0.0659340659340659,0.8783783783783784,1.0,1.0,0,3.0,0.0,0.4761904761904761,0.4688644688644688,0.9846153846153848,273,0.5421245421245421 -snitch_stream,0.0,729,19,4,937,130,0,0,0,381,,0.0,810,0,0,0,0.0909090909090909,0.1739130434782608,32.5,0.0,127,0.0,0.0,0.6220095693779905,0.0,0.0,209,0.7129186602870814 -scf_xdsl,0.0,719,1062,512,2736,512,256,256,128,384,128.0,0.0,2737,0,0,0,0.5260029717682021,0.3252858958068615,1.0,1.0,0,3.0,1.0,0.2535908865775136,0.0633977216443784,0.25,2019,0.7795938583457157 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1202,1914,1905,3.0,1.0,128,384,0.1064891846921797,0.25,512,256,256,0.4259567387687188,0,128.0,1.0,1.0,baseline,0,512,0.4409991386735573,649,0,0,0.5399334442595674,713,0.0,0.9658901830282862,0.0 +794,1506,1504,3.0,1.65625,128,384,0.1612090680100755,0.25,512,424,256,0.6448362720403022,0,128.0,1.0,1.0,linalg,0,512,0.9061946902654868,53,0,0,0.0667506297229219,713,0.0,0.7115869017632241,0.0 +187,918,918,2.984496124031008,0.0,129,385,0.6898395721925134,0.9847328244274808,131,0,0,0.7005347593582888,0,,26.200000000000003,26.2,snrt,0,5,0.25,15,0,0,0.0802139037433155,732,0.0,0.7807486631016043,0.0 +253,985,979,3.0,0.0,128,384,0.5059288537549407,0.9846153846153848,130,0,0,0.5138339920948617,0,,1.0,1.0,ssr1d,0,130,0.9027777777777778,14,0,0,0.0553359683794466,733,0.0,0.5691699604743083,0.0 +273,1005,995,3.0,0.0,128,384,0.4688644688644688,0.9846153846153848,130,0,0,0.4761904761904761,0,,1.0,1.0,ssr2d,0,130,0.8783783783783784,18,0,0,0.0659340659340659,733,0.0,0.5421245421245421,0.0 +209,810,937,0.0,0.0,0,381,0.0,0.0,130,0,0,0.6220095693779905,127,,0.0,32.5,snitch_stream,0,4,0.1739130434782608,19,0,0,0.0909090909090909,729,0.0,0.7129186602870814,0.0 +2019,2737,2736,3.0,1.0,128,384,0.0633977216443784,0.25,512,256,256,0.2535908865775136,0,128.0,1.0,1.0,scf_xdsl,0,512,0.3252858958068615,1062,0,0,0.5260029717682021,719,0.0,0.7795938583457157,0.0 diff --git a/kernels/dsum/params.csv b/kernels/dsum/params.csv index 3605efaa..c699b772 100644 --- a/kernels/dsum/params.csv +++ b/kernels/dsum/params.csv @@ -1,8 +1,8 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -8x16xf32,baseline,0.0,713,649,512,1905,512,256,256,128,384,128.0,0.0,1914,0,0,0,0.5399334442595674,0.4409991386735573,1.0,1.0,0,3.0,1.0,0.4259567387687188,0.1064891846921797,0.25,1202,0.9658901830282862 -8x16xf32,linalg,0.0,713,53,512,1504,512,256,424,128,384,128.0,0.0,1506,0,0,0,0.0667506297229219,0.9061946902654868,1.0,1.0,0,3.0,1.65625,0.6448362720403022,0.1612090680100755,0.25,794,0.7115869017632241 -8x16xf32,snrt,0.0,732,15,5,918,131,0,0,129,385,,0.0,918,0,0,0,0.0802139037433155,0.25,26.2,26.200000000000003,0,2.984496124031008,0.0,0.7005347593582888,0.6898395721925134,0.9847328244274808,187,0.7807486631016043 -8x16xf32,ssr1d,0.0,733,14,130,979,130,0,0,128,384,,0.0,985,0,0,0,0.0553359683794466,0.9027777777777778,1.0,1.0,0,3.0,0.0,0.5138339920948617,0.5059288537549407,0.9846153846153848,253,0.5691699604743083 -8x16xf32,ssr2d,0.0,733,18,130,995,130,0,0,128,384,,0.0,1005,0,0,0,0.0659340659340659,0.8783783783783784,1.0,1.0,0,3.0,0.0,0.4761904761904761,0.4688644688644688,0.9846153846153848,273,0.5421245421245421 -8x16xf32,snitch_stream,0.0,729,19,4,937,130,0,0,0,381,,0.0,810,0,0,0,0.0909090909090909,0.1739130434782608,32.5,0.0,127,0.0,0.0,0.6220095693779905,0.0,0.0,209,0.7129186602870814 -8x16xf32,scf_xdsl,0.0,719,1062,512,2736,512,256,256,128,384,128.0,0.0,2737,0,0,0,0.5260029717682021,0.3252858958068615,1.0,1.0,0,3.0,1.0,0.2535908865775136,0.0633977216443784,0.25,2019,0.7795938583457157 +params,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +8x16xf32,1202,1914,1905,3.0,1.0,128,384,0.1064891846921797,0.25,512,256,256,0.4259567387687188,0,128.0,1.0,1.0,baseline,0,512,0.4409991386735573,649,0,0,0.5399334442595674,713,0.0,0.9658901830282862,0.0 +8x16xf32,794,1506,1504,3.0,1.65625,128,384,0.1612090680100755,0.25,512,424,256,0.6448362720403022,0,128.0,1.0,1.0,linalg,0,512,0.9061946902654868,53,0,0,0.0667506297229219,713,0.0,0.7115869017632241,0.0 +8x16xf32,187,918,918,2.984496124031008,0.0,129,385,0.6898395721925134,0.9847328244274808,131,0,0,0.7005347593582888,0,,26.200000000000003,26.2,snrt,0,5,0.25,15,0,0,0.0802139037433155,732,0.0,0.7807486631016043,0.0 +8x16xf32,253,985,979,3.0,0.0,128,384,0.5059288537549407,0.9846153846153848,130,0,0,0.5138339920948617,0,,1.0,1.0,ssr1d,0,130,0.9027777777777778,14,0,0,0.0553359683794466,733,0.0,0.5691699604743083,0.0 +8x16xf32,273,1005,995,3.0,0.0,128,384,0.4688644688644688,0.9846153846153848,130,0,0,0.4761904761904761,0,,1.0,1.0,ssr2d,0,130,0.8783783783783784,18,0,0,0.0659340659340659,733,0.0,0.5421245421245421,0.0 +8x16xf32,209,810,937,0.0,0.0,0,381,0.0,0.0,130,0,0,0.6220095693779905,127,,0.0,32.5,snitch_stream,0,4,0.1739130434782608,19,0,0,0.0909090909090909,729,0.0,0.7129186602870814,0.0 +8x16xf32,2019,2737,2736,3.0,1.0,128,384,0.0633977216443784,0.25,512,256,256,0.2535908865775136,0,128.0,1.0,1.0,scf_xdsl,0,512,0.3252858958068615,1062,0,0,0.5260029717682021,719,0.0,0.7795938583457157,0.0 diff --git a/kernels/fill/16x16xf64/tests.csv b/kernels/fill/16x16xf64/tests.csv index 7d7810fd..6b44bdb2 100644 --- a/kernels/fill/16x16xf64/tests.csv +++ b/kernels/fill/16x16xf64/tests.csv @@ -1,6 +1,6 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,tend,end,fpss_fpu_issues,snitch_load_latency,snitch_loads,fpss_fpu_latency,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,690,59,256,1057,256,256.0,0.0,1059,0,0,0,0,0,0,0,0.1594594594594594,0.8126984126984127,1.0,0.0,0,0.0,0,0.6918918918918919,0.0,0.0,370,0.8513513513513514 -linalg,0.0,686,58,256,1030,256,256.0,0.0,1032,0,0,0,0,0,0,0,0.1671469740634005,0.8152866242038217,1.0,0.0,0,0.0,0,0.7377521613832853,0.0,0.0,347,0.9048991354466858 -snitch_stream,0.0,694,14,5,987,259,,0.0,737,7,0,0,257,0,0,0,0.0476190476190476,0.2631578947368421,51.8,51.8,250,36.71428571428572,0,0.8809523809523809,0.0238095238095238,0.027027027027027,294,0.9285714285714286 -snrt,0.0,694,13,6,990,260,,0.0,992,258,0,0,257,0,0,0,0.0434782608695652,0.3157894736842105,43.333333333333336,43.33333333333333,0,0.996124031007752,0,0.8695652173913043,0.862876254180602,0.9923076923076924,299,0.9130434782608696 -scf_xdsl,0.0,694,1596,257,3352,257,256.0,0.0,3358,1,0,0,1,0,0,0,0.598874296435272,0.1386940097139773,1.0,1.0,0,1.0,0,0.0964352720450281,0.0003752345215759,0.0038910505836575,2665,0.6953095684803001 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +370,1059,1057,0.0,0,0,0,0.0,0.0,256,0,0,0.6918918918918919,0,256.0,0.0,1.0,baseline,0,256,0.8126984126984127,59,0,0,0.1594594594594594,690,0.0,0.8513513513513514,0.0 +347,1032,1030,0.0,0,0,0,0.0,0.0,256,0,0,0.7377521613832853,0,256.0,0.0,1.0,linalg,0,256,0.8152866242038217,58,0,0,0.1671469740634005,686,0.0,0.9048991354466858,0.0 +294,737,987,36.71428571428572,0,7,257,0.0238095238095238,0.027027027027027,259,0,0,0.8809523809523809,250,,51.8,51.8,snitch_stream,0,5,0.2631578947368421,14,0,0,0.0476190476190476,694,0.0,0.9285714285714286,0.0 +299,992,990,0.996124031007752,0,258,257,0.862876254180602,0.9923076923076924,260,0,0,0.8695652173913043,0,,43.33333333333333,43.333333333333336,snrt,0,6,0.3157894736842105,13,0,0,0.0434782608695652,694,0.0,0.9130434782608696,0.0 +2665,3358,3352,1.0,0,1,1,0.0003752345215759,0.0038910505836575,257,0,0,0.0964352720450281,0,256.0,1.0,1.0,scf_xdsl,0,257,0.1386940097139773,1596,0,0,0.598874296435272,694,0.0,0.6953095684803001,0.0 diff --git a/kernels/fill/params.csv b/kernels/fill/params.csv index b8481fe6..5e5f14aa 100644 --- a/kernels/fill/params.csv +++ b/kernels/fill/params.csv @@ -1,6 +1,6 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,tend,end,fpss_fpu_issues,snitch_load_latency,snitch_loads,fpss_fpu_latency,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -16x16xf64,baseline,0.0,690,59,256,1057,256,256.0,0.0,1059,0,0,0,0,0,0,0,0.1594594594594594,0.8126984126984127,1.0,0.0,0,0.0,0,0.6918918918918919,0.0,0.0,370,0.8513513513513514 -16x16xf64,linalg,0.0,686,58,256,1030,256,256.0,0.0,1032,0,0,0,0,0,0,0,0.1671469740634005,0.8152866242038217,1.0,0.0,0,0.0,0,0.7377521613832853,0.0,0.0,347,0.9048991354466858 -16x16xf64,snitch_stream,0.0,694,14,5,987,259,,0.0,737,7,0,0,257,0,0,0,0.0476190476190476,0.2631578947368421,51.8,51.8,250,36.71428571428572,0,0.8809523809523809,0.0238095238095238,0.027027027027027,294,0.9285714285714286 -16x16xf64,snrt,0.0,694,13,6,990,260,,0.0,992,258,0,0,257,0,0,0,0.0434782608695652,0.3157894736842105,43.333333333333336,43.33333333333333,0,0.996124031007752,0,0.8695652173913043,0.862876254180602,0.9923076923076924,299,0.9130434782608696 -16x16xf64,scf_xdsl,0.0,694,1596,257,3352,257,256.0,0.0,3358,1,0,0,1,0,0,0,0.598874296435272,0.1386940097139773,1.0,1.0,0,1.0,0,0.0964352720450281,0.0003752345215759,0.0038910505836575,2665,0.6953095684803001 +params,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +16x16xf64,370,1059,1057,0.0,0,0,0,0.0,0.0,256,0,0,0.6918918918918919,0,256.0,0.0,1.0,baseline,0,256,0.8126984126984127,59,0,0,0.1594594594594594,690,0.0,0.8513513513513514,0.0 +16x16xf64,347,1032,1030,0.0,0,0,0,0.0,0.0,256,0,0,0.7377521613832853,0,256.0,0.0,1.0,linalg,0,256,0.8152866242038217,58,0,0,0.1671469740634005,686,0.0,0.9048991354466858,0.0 +16x16xf64,294,737,987,36.71428571428572,0,7,257,0.0238095238095238,0.027027027027027,259,0,0,0.8809523809523809,250,,51.8,51.8,snitch_stream,0,5,0.2631578947368421,14,0,0,0.0476190476190476,694,0.0,0.9285714285714286,0.0 +16x16xf64,299,992,990,0.996124031007752,0,258,257,0.862876254180602,0.9923076923076924,260,0,0,0.8695652173913043,0,,43.33333333333333,43.333333333333336,snrt,0,6,0.3157894736842105,13,0,0,0.0434782608695652,694,0.0,0.9130434782608696,0.0 +16x16xf64,2665,3358,3352,1.0,0,1,1,0.0003752345215759,0.0038910505836575,257,0,0,0.0964352720450281,0,256.0,1.0,1.0,scf_xdsl,0,257,0.1386940097139773,1596,0,0,0.598874296435272,694,0.0,0.6953095684803001,0.0 diff --git a/kernels/kernels.csv b/kernels/kernels.csv index 825b3bdc..c3ea7fa0 100644 --- a/kernels/kernels.csv +++ b/kernels/kernels.csv @@ -1,47 +1,47 @@ -test,params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc,snitch_stores -conv2d_d1_s1_3x3,1x1x8x8xf64,baseline,0.0,761,33,765,2319,765,369,459,354,1080,36.0,0.0,2297,0,0,0.0,0.0211674150096215,0.9586466165413534,1.0,1.0,22,3.0508474576271185,1.2439024390243902,0.490699166132136,0.2270686337395766,0.4627450980392157,1559,0.5118665811417575, -conv2d_d1_s1_3x3,1x1x8x8xf64,linalg,0.0,754,34,729,2242,729,369,561,319,972,36.0,0.0,2221,0,0,0.0,0.0228341168569509,0.9554390563564876,1.0,1.0,21,3.047021943573668,1.5203252032520325,0.4895903290799194,0.2142377434519811,0.4375857338820302,1489,0.5124244459368704, -conv2d_d1_s1_3x3,1x1x8x8xf64,snitch_stream,0.0,743,150,146,2245,398,36,36,274,972,36.0,0.0,2029,0,0,0.0,0.0998003992015968,0.4932432432432432,2.7260273972602738,2.7260273972602738,216,3.5474452554744524,1.0,0.2648037258815702,0.1823020625415835,0.6884422110552764,1503,0.364604125083167, -conv2d_d1_s1_3x3,1x1x8x8xf64,snrt,0.0,756,66,146,2241,398,36,36,272,972,36.0,0.0,2018,0,0,0.0,0.044414535666218,0.6886792452830188,2.7260273972602738,2.7260273972602738,223,3.573529411764706,1.0,0.2678331090174966,0.1830417227456258,0.6834170854271356,1486,0.3122476446837147, -ddot,128xf64,baseline,0.0,752,520,386,1672,386,257,257,128,384,1.0,0.0,1680,0,0,0.0,0.5597416576964478,0.4260485651214128,1.0,1.0,0,3.0,1.0,0.4155005382131324,0.1377825618945102,0.3316062176165803,929,0.9752421959095802, -ddot,128xf64,linalg,0.0,758,521,386,1685,386,257,257,128,384,1.0,0.0,1694,0,0,0.0,0.5560298826040555,0.4255788313120176,1.0,1.0,0,3.0,1.0,0.4119530416221985,0.1366061899679829,0.3316062176165803,937,0.967982924226254, -ddot,128xf64,snitch_stream,0.0,762,16,6,1331,132,1,1,0,384,1.0,0.0,819,0,0,0.0,0.0280701754385964,0.2727272727272727,22.0,0.0,512,0.0,1.0,0.231578947368421,0.0,0.0,570,0.2596491228070175, -ddot,128xf64,snrt,0.0,758,12,6,1307,132,1,1,2,384,1.0,0.0,804,0,0,0.0,0.0218181818181818,0.3333333333333333,22.0,22.0,503,192.0,1.0,0.24,0.0036363636363636,0.0151515151515151,550,0.2618181818181818, -ddot,128xf64,scf_xdsl,0.0,757,524,386,1717,386,257,257,127,384,1.0,0.0,1715,0,0,0.0,0.5452653485952134,0.4241758241758241,1.0,1.0,2,3.0236220472440944,1.0,0.4016649323621228,0.1321540062434963,0.3290155440414508,961,0.946930280957336, -ddot,128xf64,scf,0.0,758,521,386,1685,386,257,257,128,384,1.0,0.0,1694,0,0,0.0,0.5560298826040555,0.4255788313120176,1.0,1.0,0,3.0,1.0,0.4119530416221985,0.1366061899679829,0.3316062176165803,937,0.967982924226254, -dense,8x8xf64,baseline,0.0,752,79,1628,3981,1628,740,852,641,1793,247.0,0.0,3990,0,0,0.0,0.024390243902439,0.9537199765670769,1.0,1.0,0,2.797191887675507,1.1513513513513514,0.5026242667489966,0.1979005866008027,0.3937346437346437,3239,0.5270145106514357, -dense,8x8xf64,fused,0.0,770,54,1388,3733,1388,620,659,646,1808,120.0,0.0,3715,0,0,0.0,0.0182186234817813,0.9625520110957004,1.0,1.0,18,2.7987616099071206,1.0629032258064517,0.4682860998650472,0.2179487179487179,0.4654178674351585,2964,0.4865047233468286, -dense,8x8xf64,linalg,0.0,757,80,1628,3986,1628,740,852,641,1793,247.0,0.0,3994,0,0,0.0,0.0247066090179122,0.9531615925058547,1.0,1.0,0,2.797191887675507,1.1513513513513514,0.5027794935145151,0.1979617047560222,0.3937346437346437,3238,0.5274861025324274, -dense,8x8xf64,snrt,0.0,757,114,387,3465,771,64,64,612,1793,64.0,0.0,3349,0,0,0.0,0.0420819490586932,0.7724550898203593,1.992248062015504,1.992248062015504,116,2.929738562091503,1.0,0.2846068660022148,0.2259136212624584,0.7937743190661478,2709,0.3266888150609081, -dense,8x8xf64,snitch_stream,0.0,762,300,387,3491,771,64,64,612,1793,64.0,0.0,3375,0,0,0.0,0.1098901098901098,0.5633187772925764,1.992248062015504,1.992248062015504,116,2.929738562091503,1.0,0.2824175824175824,0.2241758241758241,0.7937743190661478,2730,0.3923076923076923, -dsum,8x16xf32,baseline,0.0,713,649,512,1905,512,256,256,128,384,128.0,0.0,1914,0,0,0.0,0.5399334442595674,0.4409991386735573,1.0,1.0,0,3.0,1.0,0.4259567387687188,0.1064891846921797,0.25,1202,0.9658901830282862, -dsum,8x16xf32,linalg,0.0,713,53,512,1504,512,256,424,128,384,128.0,0.0,1506,0,0,0.0,0.0667506297229219,0.9061946902654868,1.0,1.0,0,3.0,1.65625,0.6448362720403022,0.1612090680100755,0.25,794,0.7115869017632241, -dsum,8x16xf32,snrt,0.0,732,15,5,918,131,0,0,129,385,,0.0,918,0,0,0.0,0.0802139037433155,0.25,26.2,26.200000000000003,0,2.984496124031008,0.0,0.7005347593582888,0.6898395721925134,0.9847328244274808,187,0.7807486631016043, -dsum,8x16xf32,ssr1d,0.0,733,14,130,979,130,0,0,128,384,,0.0,985,0,0,0.0,0.0553359683794466,0.9027777777777778,1.0,1.0,0,3.0,0.0,0.5138339920948617,0.5059288537549407,0.9846153846153848,253,0.5691699604743083, -dsum,8x16xf32,ssr2d,0.0,733,18,130,995,130,0,0,128,384,,0.0,1005,0,0,0.0,0.0659340659340659,0.8783783783783784,1.0,1.0,0,3.0,0.0,0.4761904761904761,0.4688644688644688,0.9846153846153848,273,0.5421245421245421, -dsum,8x16xf32,snitch_stream,0.0,729,19,4,937,130,0,0,0,381,,0.0,810,0,0,0.0,0.0909090909090909,0.1739130434782608,32.5,0.0,127,0.0,0.0,0.6220095693779905,0.0,0.0,209,0.7129186602870814, -dsum,8x16xf32,scf_xdsl,0.0,719,1062,512,2736,512,256,256,128,384,128.0,0.0,2737,0,0,0.0,0.5260029717682021,0.3252858958068615,1.0,1.0,0,3.0,1.0,0.2535908865775136,0.0633977216443784,0.25,2019,0.7795938583457157, -fill,16x16xf64,baseline,0.0,690,59,256,1057,256,0,0,0,0,256.0,0.0,1059,0,0,0.0,0.1594594594594594,0.8126984126984127,1.0,0.0,0,0.0,0.0,0.6918918918918919,0.0,0.0,370,0.8513513513513514, -fill,16x16xf64,linalg,0.0,686,58,256,1030,256,0,0,0,0,256.0,0.0,1032,0,0,0.0,0.1671469740634005,0.8152866242038217,1.0,0.0,0,0.0,0.0,0.7377521613832853,0.0,0.0,347,0.9048991354466858, -fill,16x16xf64,snitch_stream,0.0,694,14,5,987,259,0,0,7,257,,0.0,737,0,0,0.0,0.0476190476190476,0.2631578947368421,51.8,51.8,250,36.71428571428572,0.0,0.8809523809523809,0.0238095238095238,0.027027027027027,294,0.9285714285714286, -fill,16x16xf64,snrt,0.0,694,13,6,990,260,0,0,258,257,,0.0,992,0,0,0.0,0.0434782608695652,0.3157894736842105,43.333333333333336,43.33333333333333,0,0.996124031007752,0.0,0.8695652173913043,0.862876254180602,0.9923076923076924,299,0.9130434782608696, -fill,16x16xf64,scf_xdsl,0.0,694,1596,257,3352,257,0,0,1,1,256.0,0.0,3358,0,0,0.0,0.598874296435272,0.1386940097139773,1.0,1.0,0,1.0,0.0,0.0964352720450281,0.0003752345215759,0.0038910505836575,2665,0.6953095684803001, -matmul,8x8xf64,baseline,0.0,755,44,1106,3269,1106,476,475,508,1536,118.0,0.0,3241,0,0,0.0,0.0174950298210735,0.9617391304347828,1.0,1.0,28,3.0236220472440944,0.9978991596638656,0.4397614314115308,0.2019880715705765,0.4593128390596745,2515,0.4572564612326043, -matmul,8x8xf64,linalg,0.0,754,265,1179,3576,1179,548,587,512,1536,119.0,0.0,3565,15,5,3.0,0.093871767623096,0.8164819944598338,1.0,1.0,11,3.0,1.0711678832116789,0.4176408076514346,0.1813673397095288,0.4342663273960984,2823,0.5115125752745306,133.0 -matmul,8x8xf64,snitch_stream,0.0,736,175,259,3071,643,0,0,586,1665,,0.0,2879,0,0,0.0,0.0749143835616438,0.5967741935483871,2.4826254826254828,2.4826254826254823,192,2.841296928327645,0.0,0.2752568493150685,0.2508561643835616,0.911353032659409,2336,0.3501712328767123, -matmul,8x8xf64,snrt,0.0,741,31,259,3063,643,0,0,585,1665,,0.0,2869,0,0,0.0,0.0133448127421437,0.8931034482758621,2.4826254826254828,2.4826254826254828,194,2.8461538461538463,0.0,0.2767972449418855,0.2518295307791648,0.9097978227060654,2323,0.2901420576840293, -pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,baseline,0.0,716,260,980,2142,980,490,490,441,441,49.0,0.0,2150,0,0,0.0,0.181184668989547,0.7903225806451613,1.0,1.0,0,1.0,1.0,0.6829268292682927,0.3073170731707317,0.45,1435,0.8641114982578397, -pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,linalg,0.0,738,37,854,1938,854,364,364,439,441,49.0,0.0,1933,0,0,0.0,0.0308076602830974,0.9584736251402918,1.0,1.0,5,1.0045558086560364,1.0,0.7110741049125728,0.3655287260616153,0.5140515222482436,1201,0.7418817651956703, -pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,snitch_stream,0.0,742,174,198,1856,541,49,49,389,441,49.0,0.0,1736,0,0,0.0,0.1560538116591928,0.532258064516129,2.7323232323232323,2.7323232323232323,120,1.1336760925449871,1.0,0.4852017937219731,0.3488789237668161,0.7190388170055453,1115,0.6412556053811659, -pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,snrt,0.0,753,68,198,1846,541,49,49,394,441,49.0,0.0,1735,0,0,0.0,0.0621572212065813,0.7443609022556391,2.7323232323232323,2.7323232323232323,111,1.119289340101523,1.0,0.4945155393053016,0.360146252285192,0.7282809611829945,1094,0.556672760511883, -pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,baseline,0.0,738,37,854,2685,854,364,490,435,1323,49.0,0.0,2663,0,0,0.0,0.0189938398357289,0.9584736251402918,1.0,1.0,22,3.041379310344828,1.3461538461538465,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.4573921971252566, -pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,linalg,0.0,738,37,854,2685,854,364,490,435,1323,49.0,0.0,2662,0,0,0.0,0.0189938398357289,0.9584736251402918,1.0,1.0,23,3.041379310344828,1.3461538461538465,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.4573921971252566, -pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,snitch_stream,0.0,742,174,198,2738,541,49,49,388,1323,49.0,0.0,2510,0,0,0.0,0.0871306960440661,0.532258064516129,2.7323232323232323,2.7323232323232323,228,3.4097938144329896,1.0,0.2709063595393089,0.1942914371557336,0.7171903881700554,1997,0.358037055583375, -pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,snrt,0.0,753,68,198,2728,541,49,49,390,1323,49.0,0.0,2509,0,0,0.0,0.034412955465587,0.7443609022556391,2.7323232323232323,2.7323232323232323,219,3.392307692307692,1.0,0.2737854251012145,0.1973684210526315,0.7208872458410351,1976,0.3081983805668016, -relu,16x16xf64,baseline,0.0,697,78,769,1777,769,256,256,256,257,256.0,0.0,1773,0,0,0.0,0.0721554116558741,0.9079102715466352,1.0,1.0,4,1.00390625,1.0,0.7113783533765032,0.2368177613320999,0.3328998699609883,1081,0.7835337650323775, -relu,16x16xf64,ssr,0.0,703,528,259,1546,259,0,0,257,257,,0.0,1548,0,0,0.0,0.624113475177305,0.3290978398983482,1.0,1.0,0,1.0,0.0,0.3061465721040189,0.3037825059101655,0.9922779922779924,846,0.9302600472813238, -relu,16x16xf64,ssr_frep,0.0,698,15,5,1024,259,0,0,12,257,,0.0,779,0,0,0.0,0.0458715596330275,0.25,51.8,51.8,245,21.416666666666668,0.0,0.7920489296636085,0.036697247706422,0.0463320463320463,327,0.8379204892966361, -relu,16x16xf64,snrt,0.0,703,16,6,1036,259,0,0,4,257,,0.0,783,0,0,0.0,0.0479041916167664,0.2727272727272727,43.16666666666666,43.16666666666666,253,64.25,0.0,0.7754491017964071,0.0119760479041916,0.0154440154440154,334,0.8233532934131736, -relu,16x16xf64,linalg,0.0,720,78,769,1793,769,256,256,256,257,256.0,0.0,1790,0,0,0.0,0.0726256983240223,0.9079102715466352,1.0,1.0,3,1.00390625,1.0,0.7160148975791434,0.2383612662942272,0.3328998699609883,1074,0.7886405959031657, -relu,16x16xf64,snitch_stream,0.0,720,22,5,1041,259,0,0,3,257,,0.0,787,0,0,0.0,0.0683229813664596,0.1851851851851851,51.8,51.8,254,85.66666666666667,0.0,0.8043478260869565,0.0093167701863354,0.0115830115830115,322,0.8726708074534162, -relu,16x16xf64,scf_xdsl,0.0,707,1857,769,4161,769,257,257,256,256,256.0,0.0,4166,0,0,0.0,0.5367052023121387,0.2928408225437928,1.0,1.0,0,1.0,1.0,0.2222543352601156,0.0739884393063583,0.3328998699609883,3460,0.7589595375722543,2.0 +test,params,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,impl,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart,snitch_stores +conv2d_d1_s1_3x3,1x1x8x8xf64,1559,2297,2319,3.0508474576271185,1.2439024390243902,354,1080,0.2270686337395766,0.4627450980392157,765,459,369,0.490699166132136,22,36.0,1.0,1.0,baseline,0.0,765,0.9586466165413534,33,0,0,0.0211674150096215,761,0.0,0.5118665811417575,0.0, +conv2d_d1_s1_3x3,1x1x8x8xf64,1489,2221,2242,3.047021943573668,1.5203252032520325,319,972,0.2142377434519811,0.4375857338820302,729,561,369,0.4895903290799194,21,36.0,1.0,1.0,linalg,0.0,729,0.9554390563564876,34,0,0,0.0228341168569509,754,0.0,0.5124244459368704,0.0, +conv2d_d1_s1_3x3,1x1x8x8xf64,1503,2029,2245,3.5474452554744524,1.0,274,972,0.1823020625415835,0.6884422110552764,398,36,36,0.2648037258815702,216,36.0,2.7260273972602738,2.7260273972602738,snitch_stream,0.0,146,0.4932432432432432,150,0,0,0.0998003992015968,743,0.0,0.364604125083167,0.0, +conv2d_d1_s1_3x3,1x1x8x8xf64,1486,2018,2241,3.573529411764706,1.0,272,972,0.1830417227456258,0.6834170854271356,398,36,36,0.2678331090174966,223,36.0,2.7260273972602738,2.7260273972602738,snrt,0.0,146,0.6886792452830188,66,0,0,0.044414535666218,756,0.0,0.3122476446837147,0.0, +ddot,128xf64,929,1680,1672,3.0,1.0,128,384,0.1377825618945102,0.3316062176165803,386,257,257,0.4155005382131324,0,1.0,1.0,1.0,baseline,0.0,386,0.4260485651214128,520,0,0,0.5597416576964478,752,0.0,0.9752421959095802,0.0, +ddot,128xf64,937,1694,1685,3.0,1.0,128,384,0.1366061899679829,0.3316062176165803,386,257,257,0.4119530416221985,0,1.0,1.0,1.0,linalg,0.0,386,0.4255788313120176,521,0,0,0.5560298826040555,758,0.0,0.967982924226254,0.0, +ddot,128xf64,570,819,1331,0.0,1.0,0,384,0.0,0.0,132,1,1,0.231578947368421,512,1.0,0.0,22.0,snitch_stream,0.0,6,0.2727272727272727,16,0,0,0.0280701754385964,762,0.0,0.2596491228070175,0.0, +ddot,128xf64,550,804,1307,192.0,1.0,2,384,0.0036363636363636,0.0151515151515151,132,1,1,0.24,503,1.0,22.0,22.0,snrt,0.0,6,0.3333333333333333,12,0,0,0.0218181818181818,758,0.0,0.2618181818181818,0.0, +ddot,128xf64,961,1715,1717,3.0236220472440944,1.0,127,384,0.1321540062434963,0.3290155440414508,386,257,257,0.4016649323621228,2,1.0,1.0,1.0,scf_xdsl,0.0,386,0.4241758241758241,524,0,0,0.5452653485952134,757,0.0,0.946930280957336,0.0, +ddot,128xf64,937,1694,1685,3.0,1.0,128,384,0.1366061899679829,0.3316062176165803,386,257,257,0.4119530416221985,0,1.0,1.0,1.0,scf,0.0,386,0.4255788313120176,521,0,0,0.5560298826040555,758,0.0,0.967982924226254,0.0, +dense,8x8xf64,3239,3990,3981,2.797191887675507,1.1513513513513514,641,1793,0.1979005866008027,0.3937346437346437,1628,852,740,0.5026242667489966,0,247.0,1.0,1.0,baseline,0.0,1628,0.9537199765670769,79,0,0,0.024390243902439,752,0.0,0.5270145106514357,0.0, +dense,8x8xf64,2964,3715,3733,2.7987616099071206,1.0629032258064517,646,1808,0.2179487179487179,0.4654178674351585,1388,659,620,0.4682860998650472,18,120.0,1.0,1.0,fused,0.0,1388,0.9625520110957004,54,0,0,0.0182186234817813,770,0.0,0.4865047233468286,0.0, +dense,8x8xf64,3238,3994,3986,2.797191887675507,1.1513513513513514,641,1793,0.1979617047560222,0.3937346437346437,1628,852,740,0.5027794935145151,0,247.0,1.0,1.0,linalg,0.0,1628,0.9531615925058547,80,0,0,0.0247066090179122,757,0.0,0.5274861025324274,0.0, +dense,8x8xf64,2709,3349,3465,2.929738562091503,1.0,612,1793,0.2259136212624584,0.7937743190661478,771,64,64,0.2846068660022148,116,64.0,1.992248062015504,1.992248062015504,snrt,0.0,387,0.7724550898203593,114,0,0,0.0420819490586932,757,0.0,0.3266888150609081,0.0, +dense,8x8xf64,2730,3375,3491,2.929738562091503,1.0,612,1793,0.2241758241758241,0.7937743190661478,771,64,64,0.2824175824175824,116,64.0,1.992248062015504,1.992248062015504,snitch_stream,0.0,387,0.5633187772925764,300,0,0,0.1098901098901098,762,0.0,0.3923076923076923,0.0, +dsum,8x16xf32,1202,1914,1905,3.0,1.0,128,384,0.1064891846921797,0.25,512,256,256,0.4259567387687188,0,128.0,1.0,1.0,baseline,0.0,512,0.4409991386735573,649,0,0,0.5399334442595674,713,0.0,0.9658901830282862,0.0, +dsum,8x16xf32,794,1506,1504,3.0,1.65625,128,384,0.1612090680100755,0.25,512,424,256,0.6448362720403022,0,128.0,1.0,1.0,linalg,0.0,512,0.9061946902654868,53,0,0,0.0667506297229219,713,0.0,0.7115869017632241,0.0, +dsum,8x16xf32,187,918,918,2.984496124031008,0.0,129,385,0.6898395721925134,0.9847328244274808,131,0,0,0.7005347593582888,0,,26.200000000000003,26.2,snrt,0.0,5,0.25,15,0,0,0.0802139037433155,732,0.0,0.7807486631016043,0.0, +dsum,8x16xf32,253,985,979,3.0,0.0,128,384,0.5059288537549407,0.9846153846153848,130,0,0,0.5138339920948617,0,,1.0,1.0,ssr1d,0.0,130,0.9027777777777778,14,0,0,0.0553359683794466,733,0.0,0.5691699604743083,0.0, +dsum,8x16xf32,273,1005,995,3.0,0.0,128,384,0.4688644688644688,0.9846153846153848,130,0,0,0.4761904761904761,0,,1.0,1.0,ssr2d,0.0,130,0.8783783783783784,18,0,0,0.0659340659340659,733,0.0,0.5421245421245421,0.0, +dsum,8x16xf32,209,810,937,0.0,0.0,0,381,0.0,0.0,130,0,0,0.6220095693779905,127,,0.0,32.5,snitch_stream,0.0,4,0.1739130434782608,19,0,0,0.0909090909090909,729,0.0,0.7129186602870814,0.0, +dsum,8x16xf32,2019,2737,2736,3.0,1.0,128,384,0.0633977216443784,0.25,512,256,256,0.2535908865775136,0,128.0,1.0,1.0,scf_xdsl,0.0,512,0.3252858958068615,1062,0,0,0.5260029717682021,719,0.0,0.7795938583457157,0.0, +fill,16x16xf64,370,1059,1057,0.0,0.0,0,0,0.0,0.0,256,0,0,0.6918918918918919,0,256.0,0.0,1.0,baseline,0.0,256,0.8126984126984127,59,0,0,0.1594594594594594,690,0.0,0.8513513513513514,0.0, +fill,16x16xf64,347,1032,1030,0.0,0.0,0,0,0.0,0.0,256,0,0,0.7377521613832853,0,256.0,0.0,1.0,linalg,0.0,256,0.8152866242038217,58,0,0,0.1671469740634005,686,0.0,0.9048991354466858,0.0, +fill,16x16xf64,294,737,987,36.71428571428572,0.0,7,257,0.0238095238095238,0.027027027027027,259,0,0,0.8809523809523809,250,,51.8,51.8,snitch_stream,0.0,5,0.2631578947368421,14,0,0,0.0476190476190476,694,0.0,0.9285714285714286,0.0, +fill,16x16xf64,299,992,990,0.996124031007752,0.0,258,257,0.862876254180602,0.9923076923076924,260,0,0,0.8695652173913043,0,,43.33333333333333,43.333333333333336,snrt,0.0,6,0.3157894736842105,13,0,0,0.0434782608695652,694,0.0,0.9130434782608696,0.0, +fill,16x16xf64,2665,3358,3352,1.0,0.0,1,1,0.0003752345215759,0.0038910505836575,257,0,0,0.0964352720450281,0,256.0,1.0,1.0,scf_xdsl,0.0,257,0.1386940097139773,1596,0,0,0.598874296435272,694,0.0,0.6953095684803001,0.0, +matmul,8x8xf64,2515,3241,3269,3.0236220472440944,0.9978991596638656,508,1536,0.2019880715705765,0.4593128390596745,1106,475,476,0.4397614314115308,28,118.0,1.0,1.0,baseline,0.0,1106,0.9617391304347828,44,0,0,0.0174950298210735,755,0.0,0.4572564612326043,0.0, +matmul,8x8xf64,2823,3565,3576,3.0,1.0711678832116789,512,1536,0.1813673397095288,0.4342663273960984,1179,587,548,0.4176408076514346,11,119.0,1.0,1.0,linalg,3.0,1179,0.8164819944598338,265,15,5,0.093871767623096,754,0.0,0.5115125752745306,0.0,133.0 +matmul,8x8xf64,2336,2879,3071,2.841296928327645,0.0,586,1665,0.2508561643835616,0.911353032659409,643,0,0,0.2752568493150685,192,,2.4826254826254823,2.4826254826254828,snitch_stream,0.0,259,0.5967741935483871,175,0,0,0.0749143835616438,736,0.0,0.3501712328767123,0.0, +matmul,8x8xf64,2323,2869,3063,2.8461538461538463,0.0,585,1665,0.2518295307791648,0.9097978227060654,643,0,0,0.2767972449418855,194,,2.4826254826254828,2.4826254826254828,snrt,0.0,259,0.8931034482758621,31,0,0,0.0133448127421437,741,0.0,0.2901420576840293,0.0, +pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,1435,2150,2142,1.0,1.0,441,441,0.3073170731707317,0.45,980,490,490,0.6829268292682927,0,49.0,1.0,1.0,baseline,0.0,980,0.7903225806451613,260,0,0,0.181184668989547,716,0.0,0.8641114982578397,0.0, +pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,1201,1933,1938,1.0045558086560364,1.0,439,441,0.3655287260616153,0.5140515222482436,854,364,364,0.7110741049125728,5,49.0,1.0,1.0,linalg,0.0,854,0.9584736251402918,37,0,0,0.0308076602830974,738,0.0,0.7418817651956703,0.0, +pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,1115,1736,1856,1.1336760925449871,1.0,389,441,0.3488789237668161,0.7190388170055453,541,49,49,0.4852017937219731,120,49.0,2.7323232323232323,2.7323232323232323,snitch_stream,0.0,198,0.532258064516129,174,0,0,0.1560538116591928,742,0.0,0.6412556053811659,0.0, +pooling_nchw_max_d1_s2_3x3,1x1x16x16xf64,1094,1735,1846,1.119289340101523,1.0,394,441,0.360146252285192,0.7282809611829945,541,49,49,0.4945155393053016,111,49.0,2.7323232323232323,2.7323232323232323,snrt,0.0,198,0.7443609022556391,68,0,0,0.0621572212065813,753,0.0,0.556672760511883,0.0, +pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,1948,2663,2685,3.041379310344828,1.3461538461538465,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,22,49.0,1.0,1.0,baseline,0.0,854,0.9584736251402918,37,0,0,0.0189938398357289,738,0.0,0.4573921971252566,0.0, +pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,1948,2662,2685,3.041379310344828,1.3461538461538465,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,23,49.0,1.0,1.0,linalg,0.0,854,0.9584736251402918,37,0,0,0.0189938398357289,738,0.0,0.4573921971252566,0.0, +pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,1997,2510,2738,3.4097938144329896,1.0,388,1323,0.1942914371557336,0.7171903881700554,541,49,49,0.2709063595393089,228,49.0,2.7323232323232323,2.7323232323232323,snitch_stream,0.0,198,0.532258064516129,174,0,0,0.0871306960440661,742,0.0,0.358037055583375,0.0, +pooling_nchw_sum_d1_s2_3x3,1x1x16x16xf64,1976,2509,2728,3.392307692307692,1.0,390,1323,0.1973684210526315,0.7208872458410351,541,49,49,0.2737854251012145,219,49.0,2.7323232323232323,2.7323232323232323,snrt,0.0,198,0.7443609022556391,68,0,0,0.034412955465587,753,0.0,0.3081983805668016,0.0, +relu,16x16xf64,1081,1773,1777,1.00390625,1.0,256,257,0.2368177613320999,0.3328998699609883,769,256,256,0.7113783533765032,4,256.0,1.0,1.0,baseline,0.0,769,0.9079102715466352,78,0,0,0.0721554116558741,697,0.0,0.7835337650323775,0.0, +relu,16x16xf64,846,1548,1546,1.0,0.0,257,257,0.3037825059101655,0.9922779922779924,259,0,0,0.3061465721040189,0,,1.0,1.0,ssr,0.0,259,0.3290978398983482,528,0,0,0.624113475177305,703,0.0,0.9302600472813238,0.0, +relu,16x16xf64,327,779,1024,21.416666666666668,0.0,12,257,0.036697247706422,0.0463320463320463,259,0,0,0.7920489296636085,245,,51.8,51.8,ssr_frep,0.0,5,0.25,15,0,0,0.0458715596330275,698,0.0,0.8379204892966361,0.0, +relu,16x16xf64,334,783,1036,64.25,0.0,4,257,0.0119760479041916,0.0154440154440154,259,0,0,0.7754491017964071,253,,43.16666666666666,43.16666666666666,snrt,0.0,6,0.2727272727272727,16,0,0,0.0479041916167664,703,0.0,0.8233532934131736,0.0, +relu,16x16xf64,1074,1790,1793,1.00390625,1.0,256,257,0.2383612662942272,0.3328998699609883,769,256,256,0.7160148975791434,3,256.0,1.0,1.0,linalg,0.0,769,0.9079102715466352,78,0,0,0.0726256983240223,720,0.0,0.7886405959031657,0.0, +relu,16x16xf64,322,787,1041,85.66666666666667,0.0,3,257,0.0093167701863354,0.0115830115830115,259,0,0,0.8043478260869565,254,,51.8,51.8,snitch_stream,0.0,5,0.1851851851851851,22,0,0,0.0683229813664596,720,0.0,0.8726708074534162,0.0, +relu,16x16xf64,3460,4166,4161,1.0,1.0,256,256,0.0739884393063583,0.3328998699609883,769,257,257,0.2222543352601156,0,256.0,1.0,1.0,scf_xdsl,0.0,769,0.2928408225437928,1857,0,0,0.5367052023121387,707,0.0,0.7589595375722543,0.0,2.0 diff --git a/kernels/matmul/8x8xf64/baseline.csv b/kernels/matmul/8x8xf64/baseline.csv index aea15042..a4996460 100644 --- a/kernels/matmul/8x8xf64/baseline.csv +++ b/kernels/matmul/8x8xf64/baseline.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,755,44,1106,3269,1106,118,476,475,508,1536,0.0,3241,0,0,0,0.01749502982107356,0.9617391304347827,1.0,1.0,28,3.0236220472440944,0.9978991596638656,0.4397614314115308,0.20198807157057655,0.4593128390596745,2515,0.4572564612326043 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +2515,3241,3269,3.0236220472440944,0.9978991596638656,508,1536,0.20198807157057655,0.4593128390596745,1106,475,476,0.4397614314115308,28,118,1.0,1.0,0,1106,0.9617391304347827,44,0,0,0.01749502982107356,755,0.0,0.4572564612326043,0.0 diff --git a/kernels/matmul/8x8xf64/linalg.csv b/kernels/matmul/8x8xf64/linalg.csv index 75c03d84..5634785a 100644 --- a/kernels/matmul/8x8xf64/linalg.csv +++ b/kernels/matmul/8x8xf64/linalg.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_stores,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,snitch_loads,snitch_load_latency,tend,end,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,754,265,133,1179,3576,1179,119,548,587,512,1536,5,15,0.0,3565,3.0,0.093871767623096,0.8164819944598338,1.0,1.0,11,3.0,1.0711678832116789,0.41764080765143463,0.18136733970952887,0.4342663273960984,2823,0.5115125752745306 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,snitch_stores,start,tend,total_ipc,tstart +2823,3565,3576,3.0,1.0711678832116789,512,1536,0.18136733970952887,0.4342663273960984,1179,587,548,0.41764080765143463,11,119,1.0,1.0,3.0,1179,0.8164819944598338,265,15,5,0.093871767623096,133,754,0.0,0.5115125752745306,0.0 diff --git a/kernels/matmul/8x8xf64/snitch_stream.csv b/kernels/matmul/8x8xf64/snitch_stream.csv index d8ff31ef..cbc59afa 100644 --- a/kernels/matmul/8x8xf64/snitch_stream.csv +++ b/kernels/matmul/8x8xf64/snitch_stream.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,736,175,259,3071,643,586,1665,0.0,2879,0,0,0,0,0,0.07491438356164383,0.5967741935483871,2.4826254826254828,2.4826254826254823,192,2.8412969283276452,0,0.2752568493150685,0.2508561643835616,0.911353032659409,2336,0.3501712328767123 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +2336,2879,3071,2.8412969283276452,0,586,1665,0.2508561643835616,0.911353032659409,643,0,0,0.2752568493150685,192,2.4826254826254823,2.4826254826254828,0,259,0.5967741935483871,175,0,0,0.07491438356164383,736,0.0,0.3501712328767123,0.0 diff --git a/kernels/matmul/8x8xf64/snrt.csv b/kernels/matmul/8x8xf64/snrt.csv index a3a1c96d..12816b05 100644 --- a/kernels/matmul/8x8xf64/snrt.csv +++ b/kernels/matmul/8x8xf64/snrt.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,741,31,259,3063,643,585,1665,0.0,2869,0,0,0,0,0,0.01334481274214378,0.8931034482758621,2.4826254826254828,2.4826254826254828,194,2.8461538461538463,0,0.2767972449418855,0.25182953077916487,0.9097978227060654,2323,0.2901420576840293 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +2323,2869,3063,2.8461538461538463,0,585,1665,0.25182953077916487,0.9097978227060654,643,0,0,0.2767972449418855,194,2.4826254826254828,2.4826254826254828,0,259,0.8931034482758621,31,0,0,0.01334481274214378,741,0.0,0.2901420576840293,0.0 diff --git a/kernels/matmul/8x8xf64/tests.csv b/kernels/matmul/8x8xf64/tests.csv index 8b4ea7f5..8eeca201 100644 --- a/kernels/matmul/8x8xf64/tests.csv +++ b/kernels/matmul/8x8xf64/tests.csv @@ -1,5 +1,5 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_stores,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc,snitch_stores -baseline,0.0,755,44,1106,3269,1106,118.0,476,475,508,1536,0.0,3241,0,0,0.0,0.0174950298210735,0.9617391304347828,1.0,1.0,28,3.0236220472440944,0.9978991596638656,0.4397614314115308,0.2019880715705765,0.4593128390596745,2515,0.4572564612326043, -linalg,0.0,754,265,1179,3576,1179,119.0,548,587,512,1536,0.0,3565,15,5,3.0,0.093871767623096,0.8164819944598338,1.0,1.0,11,3.0,1.0711678832116789,0.4176408076514346,0.1813673397095288,0.4342663273960984,2823,0.5115125752745306,133.0 -snitch_stream,0.0,736,175,259,3071,643,,0,0,586,1665,0.0,2879,0,0,0.0,0.0749143835616438,0.5967741935483871,2.4826254826254828,2.4826254826254823,192,2.841296928327645,0.0,0.2752568493150685,0.2508561643835616,0.911353032659409,2336,0.3501712328767123, -snrt,0.0,741,31,259,3063,643,,0,0,585,1665,0.0,2869,0,0,0.0,0.0133448127421437,0.8931034482758621,2.4826254826254828,2.4826254826254828,194,2.8461538461538463,0.0,0.2767972449418855,0.2518295307791648,0.9097978227060654,2323,0.2901420576840293, +impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart,snitch_stores +baseline,2515,3241,3269,3.0236220472440944,0.9978991596638656,508,1536,0.2019880715705765,0.4593128390596745,1106,475,476,0.4397614314115308,28,118.0,1.0,1.0,0.0,1106,0.9617391304347828,44,0,0,0.0174950298210735,755,0.0,0.4572564612326043,0.0, +linalg,2823,3565,3576,3.0,1.0711678832116789,512,1536,0.1813673397095288,0.4342663273960984,1179,587,548,0.4176408076514346,11,119.0,1.0,1.0,3.0,1179,0.8164819944598338,265,15,5,0.093871767623096,754,0.0,0.5115125752745306,0.0,133.0 +snitch_stream,2336,2879,3071,2.841296928327645,0.0,586,1665,0.2508561643835616,0.911353032659409,643,0,0,0.2752568493150685,192,,2.4826254826254823,2.4826254826254828,0.0,259,0.5967741935483871,175,0,0,0.0749143835616438,736,0.0,0.3501712328767123,0.0, +snrt,2323,2869,3063,2.8461538461538463,0.0,585,1665,0.2518295307791648,0.9097978227060654,643,0,0,0.2767972449418855,194,,2.4826254826254828,2.4826254826254828,0.0,259,0.8931034482758621,31,0,0,0.0133448127421437,741,0.0,0.2901420576840293,0.0, diff --git a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/baseline.csv b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/baseline.csv index 23a1678a..ce5d244d 100644 --- a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/baseline.csv +++ b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/baseline.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,716,260,980,2142,980,490,490,441,441,49,0.0,2150,0,0,0,0.18118466898954705,0.7903225806451613,1.0,1.0,0,1.0,1.0,0.6829268292682927,0.3073170731707317,0.45,1435,0.8641114982578397 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1435,2150,2142,1.0,1.0,441,441,0.3073170731707317,0.45,980,490,490,0.6829268292682927,0,49,1.0,1.0,0,980,0.7903225806451613,260,0,0,0.18118466898954705,716,0.0,0.8641114982578397,0.0 diff --git a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/linalg.csv b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/linalg.csv index c960d9da..396c6d0c 100644 --- a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/linalg.csv +++ b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/linalg.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,738,37,854,1938,854,364,364,439,441,49,0.0,1933,0,0,0,0.030807660283097418,0.9584736251402918,1.0,1.0,5,1.0045558086560364,1.0,0.7110741049125728,0.3655287260616153,0.5140515222482436,1201,0.7418817651956703 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1201,1933,1938,1.0045558086560364,1.0,439,441,0.3655287260616153,0.5140515222482436,854,364,364,0.7110741049125728,5,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.030807660283097418,738,0.0,0.7418817651956703,0.0 diff --git a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv index 7f01c737..7b8f152a 100644 --- a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv +++ b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,742,174,198,1856,541,49,49,389,441,49,0.0,1736,0,0,0,0.15605381165919283,0.532258064516129,2.7323232323232323,2.7323232323232323,120,1.1336760925449871,1.0,0.4852017937219731,0.3488789237668161,0.7190388170055453,1115,0.6412556053811659 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1115,1736,1856,1.1336760925449871,1.0,389,441,0.3488789237668161,0.7190388170055453,541,49,49,0.4852017937219731,120,49,2.7323232323232323,2.7323232323232323,0,198,0.532258064516129,174,0,0,0.15605381165919283,742,0.0,0.6412556053811659,0.0 diff --git a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snrt.csv b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snrt.csv index 6a3e3b58..0c191abc 100644 --- a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snrt.csv +++ b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/snrt.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,753,68,198,1846,541,49,49,394,441,49,0.0,1735,0,0,0,0.062157221206581355,0.7443609022556391,2.7323232323232323,2.7323232323232323,111,1.119289340101523,1.0,0.4945155393053016,0.360146252285192,0.7282809611829945,1094,0.556672760511883 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1094,1735,1846,1.119289340101523,1.0,394,441,0.360146252285192,0.7282809611829945,541,49,49,0.4945155393053016,111,49,2.7323232323232323,2.7323232323232323,0,198,0.7443609022556391,68,0,0,0.062157221206581355,753,0.0,0.556672760511883,0.0 diff --git a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/tests.csv b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/tests.csv index 37641b1f..e2db6957 100644 --- a/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/tests.csv +++ b/kernels/pooling_nchw_max_d1_s2_3x3/1x1x16x16xf64/tests.csv @@ -1,5 +1,5 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,716,260,980,2142,980,490,490,441,441,49,0.0,2150,0,0,0,0.181184668989547,0.7903225806451613,1.0,1.0,0,1.0,1.0,0.6829268292682927,0.3073170731707317,0.45,1435,0.8641114982578397 -linalg,0.0,738,37,854,1938,854,364,364,439,441,49,0.0,1933,0,0,0,0.0308076602830974,0.9584736251402918,1.0,1.0,5,1.0045558086560364,1.0,0.7110741049125728,0.3655287260616153,0.5140515222482436,1201,0.7418817651956703 -snitch_stream,0.0,742,174,198,1856,541,49,49,389,441,49,0.0,1736,0,0,0,0.1560538116591928,0.532258064516129,2.7323232323232323,2.7323232323232323,120,1.1336760925449871,1.0,0.4852017937219731,0.3488789237668161,0.7190388170055453,1115,0.6412556053811659 -snrt,0.0,753,68,198,1846,541,49,49,394,441,49,0.0,1735,0,0,0,0.0621572212065813,0.7443609022556391,2.7323232323232323,2.7323232323232323,111,1.119289340101523,1.0,0.4945155393053016,0.360146252285192,0.7282809611829945,1094,0.556672760511883 +impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +baseline,1435,2150,2142,1.0,1.0,441,441,0.3073170731707317,0.45,980,490,490,0.6829268292682927,0,49,1.0,1.0,0,980,0.7903225806451613,260,0,0,0.181184668989547,716,0.0,0.8641114982578397,0.0 +linalg,1201,1933,1938,1.0045558086560364,1.0,439,441,0.3655287260616153,0.5140515222482436,854,364,364,0.7110741049125728,5,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.0308076602830974,738,0.0,0.7418817651956703,0.0 +snitch_stream,1115,1736,1856,1.1336760925449871,1.0,389,441,0.3488789237668161,0.7190388170055453,541,49,49,0.4852017937219731,120,49,2.7323232323232323,2.7323232323232323,0,198,0.532258064516129,174,0,0,0.1560538116591928,742,0.0,0.6412556053811659,0.0 +snrt,1094,1735,1846,1.119289340101523,1.0,394,441,0.360146252285192,0.7282809611829945,541,49,49,0.4945155393053016,111,49,2.7323232323232323,2.7323232323232323,0,198,0.7443609022556391,68,0,0,0.0621572212065813,753,0.0,0.556672760511883,0.0 diff --git a/kernels/pooling_nchw_max_d1_s2_3x3/params.csv b/kernels/pooling_nchw_max_d1_s2_3x3/params.csv index e8ed98ce..413b861b 100644 --- a/kernels/pooling_nchw_max_d1_s2_3x3/params.csv +++ b/kernels/pooling_nchw_max_d1_s2_3x3/params.csv @@ -1,5 +1,5 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -1x1x16x16xf64,baseline,0.0,716,260,980,2142,980,490,490,441,441,49,0.0,2150,0,0,0,0.181184668989547,0.7903225806451613,1.0,1.0,0,1.0,1.0,0.6829268292682927,0.3073170731707317,0.45,1435,0.8641114982578397 -1x1x16x16xf64,linalg,0.0,738,37,854,1938,854,364,364,439,441,49,0.0,1933,0,0,0,0.0308076602830974,0.9584736251402918,1.0,1.0,5,1.0045558086560364,1.0,0.7110741049125728,0.3655287260616153,0.5140515222482436,1201,0.7418817651956703 -1x1x16x16xf64,snitch_stream,0.0,742,174,198,1856,541,49,49,389,441,49,0.0,1736,0,0,0,0.1560538116591928,0.532258064516129,2.7323232323232323,2.7323232323232323,120,1.1336760925449871,1.0,0.4852017937219731,0.3488789237668161,0.7190388170055453,1115,0.6412556053811659 -1x1x16x16xf64,snrt,0.0,753,68,198,1846,541,49,49,394,441,49,0.0,1735,0,0,0,0.0621572212065813,0.7443609022556391,2.7323232323232323,2.7323232323232323,111,1.119289340101523,1.0,0.4945155393053016,0.360146252285192,0.7282809611829945,1094,0.556672760511883 +params,impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1x1x16x16xf64,baseline,1435,2150,2142,1.0,1.0,441,441,0.3073170731707317,0.45,980,490,490,0.6829268292682927,0,49,1.0,1.0,0,980,0.7903225806451613,260,0,0,0.181184668989547,716,0.0,0.8641114982578397,0.0 +1x1x16x16xf64,linalg,1201,1933,1938,1.0045558086560364,1.0,439,441,0.3655287260616153,0.5140515222482436,854,364,364,0.7110741049125728,5,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.0308076602830974,738,0.0,0.7418817651956703,0.0 +1x1x16x16xf64,snitch_stream,1115,1736,1856,1.1336760925449871,1.0,389,441,0.3488789237668161,0.7190388170055453,541,49,49,0.4852017937219731,120,49,2.7323232323232323,2.7323232323232323,0,198,0.532258064516129,174,0,0,0.1560538116591928,742,0.0,0.6412556053811659,0.0 +1x1x16x16xf64,snrt,1094,1735,1846,1.119289340101523,1.0,394,441,0.360146252285192,0.7282809611829945,541,49,49,0.4945155393053016,111,49,2.7323232323232323,2.7323232323232323,0,198,0.7443609022556391,68,0,0,0.0621572212065813,753,0.0,0.556672760511883,0.0 diff --git a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/baseline.csv b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/baseline.csv index b84076a6..cfd6e6a0 100644 --- a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/baseline.csv +++ b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/baseline.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,738,37,854,2685,854,364,490,435,1323,49,0.0,2663,0,0,0,0.018993839835728953,0.9584736251402918,1.0,1.0,22,3.0413793103448277,1.3461538461538463,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.45739219712525664 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1948,2663,2685,3.0413793103448277,1.3461538461538463,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,22,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.018993839835728953,738,0.0,0.45739219712525664,0.0 diff --git a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/linalg.csv b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/linalg.csv index 2fd4aa47..e7e78ed7 100644 --- a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/linalg.csv +++ b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/linalg.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,738,37,854,2685,854,364,490,435,1323,49,0.0,2662,0,0,0,0.018993839835728953,0.9584736251402918,1.0,1.0,23,3.0413793103448277,1.3461538461538463,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.45739219712525664 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1948,2662,2685,3.0413793103448277,1.3461538461538463,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,23,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.018993839835728953,738,0.0,0.45739219712525664,0.0 diff --git a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv index 7a15cc8b..07100198 100644 --- a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv +++ b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snitch_stream.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,742,174,198,2738,541,49,49,388,1323,49,0.0,2510,0,0,0,0.0871306960440661,0.532258064516129,2.7323232323232323,2.7323232323232323,228,3.4097938144329896,1.0,0.27090635953930897,0.1942914371557336,0.7171903881700554,1997,0.35803705558337506 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1997,2510,2738,3.4097938144329896,1.0,388,1323,0.1942914371557336,0.7171903881700554,541,49,49,0.27090635953930897,228,49,2.7323232323232323,2.7323232323232323,0,198,0.532258064516129,174,0,0,0.0871306960440661,742,0.0,0.35803705558337506,0.0 diff --git a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snrt.csv b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snrt.csv index d3cbce18..7701e50a 100644 --- a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snrt.csv +++ b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/snrt.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,753,68,198,2728,541,49,49,390,1323,49,0.0,2509,0,0,0,0.03441295546558704,0.7443609022556391,2.7323232323232323,2.7323232323232323,219,3.3923076923076922,1.0,0.27378542510121456,0.19736842105263158,0.7208872458410351,1976,0.3081983805668016 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1976,2509,2728,3.3923076923076922,1.0,390,1323,0.19736842105263158,0.7208872458410351,541,49,49,0.27378542510121456,219,49,2.7323232323232323,2.7323232323232323,0,198,0.7443609022556391,68,0,0,0.03441295546558704,753,0.0,0.3081983805668016,0.0 diff --git a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/tests.csv b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/tests.csv index 80be04c8..ebc624fd 100644 --- a/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/tests.csv +++ b/kernels/pooling_nchw_sum_d1_s2_3x3/1x1x16x16xf64/tests.csv @@ -1,5 +1,5 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -baseline,0.0,738,37,854,2685,854,364,490,435,1323,49,0.0,2663,0,0,0,0.0189938398357289,0.9584736251402918,1.0,1.0,22,3.041379310344828,1.3461538461538465,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.4573921971252566 -linalg,0.0,738,37,854,2685,854,364,490,435,1323,49,0.0,2662,0,0,0,0.0189938398357289,0.9584736251402918,1.0,1.0,23,3.041379310344828,1.3461538461538465,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.4573921971252566 -snitch_stream,0.0,742,174,198,2738,541,49,49,388,1323,49,0.0,2510,0,0,0,0.0871306960440661,0.532258064516129,2.7323232323232323,2.7323232323232323,228,3.4097938144329896,1.0,0.2709063595393089,0.1942914371557336,0.7171903881700554,1997,0.358037055583375 -snrt,0.0,753,68,198,2728,541,49,49,390,1323,49,0.0,2509,0,0,0,0.034412955465587,0.7443609022556391,2.7323232323232323,2.7323232323232323,219,3.392307692307692,1.0,0.2737854251012145,0.1973684210526315,0.7208872458410351,1976,0.3081983805668016 +impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +baseline,1948,2663,2685,3.041379310344828,1.3461538461538465,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,22,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.0189938398357289,738,0.0,0.4573921971252566,0.0 +linalg,1948,2662,2685,3.041379310344828,1.3461538461538465,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,23,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.0189938398357289,738,0.0,0.4573921971252566,0.0 +snitch_stream,1997,2510,2738,3.4097938144329896,1.0,388,1323,0.1942914371557336,0.7171903881700554,541,49,49,0.2709063595393089,228,49,2.7323232323232323,2.7323232323232323,0,198,0.532258064516129,174,0,0,0.0871306960440661,742,0.0,0.358037055583375,0.0 +snrt,1976,2509,2728,3.392307692307692,1.0,390,1323,0.1973684210526315,0.7208872458410351,541,49,49,0.2737854251012145,219,49,2.7323232323232323,2.7323232323232323,0,198,0.7443609022556391,68,0,0,0.034412955465587,753,0.0,0.3081983805668016,0.0 diff --git a/kernels/pooling_nchw_sum_d1_s2_3x3/params.csv b/kernels/pooling_nchw_sum_d1_s2_3x3/params.csv index 806e2a16..02fff055 100644 --- a/kernels/pooling_nchw_sum_d1_s2_3x3/params.csv +++ b/kernels/pooling_nchw_sum_d1_s2_3x3/params.csv @@ -1,5 +1,5 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -1x1x16x16xf64,baseline,0.0,738,37,854,2685,854,364,490,435,1323,49,0.0,2663,0,0,0,0.0189938398357289,0.9584736251402918,1.0,1.0,22,3.041379310344828,1.3461538461538465,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.4573921971252566 -1x1x16x16xf64,linalg,0.0,738,37,854,2685,854,364,490,435,1323,49,0.0,2662,0,0,0,0.0189938398357289,0.9584736251402918,1.0,1.0,23,3.041379310344828,1.3461538461538465,0.4383983572895277,0.223305954825462,0.509367681498829,1948,0.4573921971252566 -1x1x16x16xf64,snitch_stream,0.0,742,174,198,2738,541,49,49,388,1323,49,0.0,2510,0,0,0,0.0871306960440661,0.532258064516129,2.7323232323232323,2.7323232323232323,228,3.4097938144329896,1.0,0.2709063595393089,0.1942914371557336,0.7171903881700554,1997,0.358037055583375 -1x1x16x16xf64,snrt,0.0,753,68,198,2728,541,49,49,390,1323,49,0.0,2509,0,0,0,0.034412955465587,0.7443609022556391,2.7323232323232323,2.7323232323232323,219,3.392307692307692,1.0,0.2737854251012145,0.1973684210526315,0.7208872458410351,1976,0.3081983805668016 +params,impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1x1x16x16xf64,baseline,1948,2663,2685,3.041379310344828,1.3461538461538465,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,22,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.0189938398357289,738,0.0,0.4573921971252566,0.0 +1x1x16x16xf64,linalg,1948,2662,2685,3.041379310344828,1.3461538461538465,435,1323,0.223305954825462,0.509367681498829,854,490,364,0.4383983572895277,23,49,1.0,1.0,0,854,0.9584736251402918,37,0,0,0.0189938398357289,738,0.0,0.4573921971252566,0.0 +1x1x16x16xf64,snitch_stream,1997,2510,2738,3.4097938144329896,1.0,388,1323,0.1942914371557336,0.7171903881700554,541,49,49,0.2709063595393089,228,49,2.7323232323232323,2.7323232323232323,0,198,0.532258064516129,174,0,0,0.0871306960440661,742,0.0,0.358037055583375,0.0 +1x1x16x16xf64,snrt,1976,2509,2728,3.392307692307692,1.0,390,1323,0.1973684210526315,0.7208872458410351,541,49,49,0.2737854251012145,219,49,2.7323232323232323,2.7323232323232323,0,198,0.7443609022556391,68,0,0,0.034412955465587,753,0.0,0.3081983805668016,0.0 diff --git a/kernels/relu/16x16xf64/baseline.csv b/kernels/relu/16x16xf64/baseline.csv index 042060d0..db69240d 100644 --- a/kernels/relu/16x16xf64/baseline.csv +++ b/kernels/relu/16x16xf64/baseline.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,fpss_loads,fpss_load_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,697,78,769,1777,769,256,257,256,256,256,0.0,1773,0,0,0,0.07215541165587419,0.9079102715466352,1.0,1.0,4,1.00390625,1.0,0.7113783533765032,0.23681776133209992,0.3328998699609883,1081,0.7835337650323775 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1081,1773,1777,1.00390625,1.0,256,257,0.23681776133209992,0.3328998699609883,769,256,256,0.7113783533765032,4,256,1.0,1.0,0,769,0.9079102715466352,78,0,0,0.07215541165587419,697,0.0,0.7835337650323775,0.0 diff --git a/kernels/relu/16x16xf64/linalg.csv b/kernels/relu/16x16xf64/linalg.csv index 667f78c6..f88fd565 100644 --- a/kernels/relu/16x16xf64/linalg.csv +++ b/kernels/relu/16x16xf64/linalg.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,fpss_loads,fpss_load_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,720,78,769,1793,769,256,257,256,256,256,0.0,1790,0,0,0,0.07262569832402235,0.9079102715466352,1.0,1.0,3,1.00390625,1.0,0.7160148975791434,0.2383612662942272,0.3328998699609883,1074,0.7886405959031657 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +1074,1790,1793,1.00390625,1.0,256,257,0.2383612662942272,0.3328998699609883,769,256,256,0.7160148975791434,3,256,1.0,1.0,0,769,0.9079102715466352,78,0,0,0.07262569832402235,720,0.0,0.7886405959031657,0.0 diff --git a/kernels/relu/16x16xf64/scf_xdsl.csv b/kernels/relu/16x16xf64/scf_xdsl.csv index 0129bbb1..2179edef 100644 --- a/kernels/relu/16x16xf64/scf_xdsl.csv +++ b/kernels/relu/16x16xf64/scf_xdsl.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_stores,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_loads,fpss_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,707,1857,2,769,4161,769,257,257,256,256,256,0.0,4166,0,0,0,0.5367052023121387,0.29284082254379284,1.0,1.0,0,1.0,1.0,0.22225433526011562,0.07398843930635839,0.3328998699609883,3460,0.7589595375722543 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,snitch_stores,start,tend,total_ipc,tstart +3460,4166,4161,1.0,1.0,256,256,0.07398843930635839,0.3328998699609883,769,257,257,0.22225433526011562,0,256,1.0,1.0,0,769,0.29284082254379284,1857,0,0,0.5367052023121387,2,707,0.0,0.7589595375722543,0.0 diff --git a/kernels/relu/16x16xf64/snitch_stream.csv b/kernels/relu/16x16xf64/snitch_stream.csv index b2c00f8c..9029637e 100644 --- a/kernels/relu/16x16xf64/snitch_stream.csv +++ b/kernels/relu/16x16xf64/snitch_stream.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,720,22,5,1041,259,3,257,0.0,787,0,0,0,0,0,0.06832298136645963,0.18518518518518517,51.8,51.800000000000004,254,85.66666666666667,0,0.8043478260869565,0.009316770186335404,0.011583011583011582,322,0.8726708074534162 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +322,787,1041,85.66666666666667,0,3,257,0.009316770186335404,0.011583011583011582,259,0,0,0.8043478260869565,254,51.800000000000004,51.8,0,5,0.18518518518518517,22,0,0,0.06832298136645963,720,0.0,0.8726708074534162,0.0 diff --git a/kernels/relu/16x16xf64/snrt.csv b/kernels/relu/16x16xf64/snrt.csv index 831fa963..103b3da3 100644 --- a/kernels/relu/16x16xf64/snrt.csv +++ b/kernels/relu/16x16xf64/snrt.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,703,16,6,1036,259,4,257,0.0,783,0,0,0,0,0,0.04790419161676647,0.2727272727272727,43.166666666666664,43.166666666666664,253,64.25,0,0.7754491017964071,0.011976047904191617,0.015444015444015444,334,0.8233532934131736 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +334,783,1036,64.25,0,4,257,0.011976047904191617,0.015444015444015444,259,0,0,0.7754491017964071,253,43.166666666666664,43.166666666666664,0,6,0.2727272727272727,16,0,0,0.04790419161676647,703,0.0,0.8233532934131736,0.0 diff --git a/kernels/relu/16x16xf64/ssr.csv b/kernels/relu/16x16xf64/ssr.csv index 34791b62..33826889 100644 --- a/kernels/relu/16x16xf64/ssr.csv +++ b/kernels/relu/16x16xf64/ssr.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,703,528,259,1546,259,257,257,0.0,1548,0,0,0,0,0,0.624113475177305,0.3290978398983482,1.0,1.0,0,1.0,0,0.3061465721040189,0.3037825059101655,0.9922779922779923,846,0.9302600472813238 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +846,1548,1546,1.0,0,257,257,0.3037825059101655,0.9922779922779923,259,0,0,0.3061465721040189,0,1.0,1.0,0,259,0.3290978398983482,528,0,0,0.624113475177305,703,0.0,0.9302600472813238,0.0 diff --git a/kernels/relu/16x16xf64/ssr_frep.csv b/kernels/relu/16x16xf64/ssr_frep.csv index e6e956f4..baa8c4ac 100644 --- a/kernels/relu/16x16xf64/ssr_frep.csv +++ b/kernels/relu/16x16xf64/ssr_frep.csv @@ -1,2 +1,2 @@ -tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,tend,end,snitch_load_latency,snitch_loads,fpss_load_latency,fpss_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc -0.0,698,15,5,1024,259,12,257,0.0,779,0,0,0,0,0,0.045871559633027525,0.25,51.8,51.800000000000004,245,21.416666666666668,0,0.7920489296636085,0.03669724770642202,0.04633204633204633,327,0.8379204892966361 +cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart +327,779,1024,21.416666666666668,0,12,257,0.03669724770642202,0.04633204633204633,259,0,0,0.7920489296636085,245,51.800000000000004,51.8,0,5,0.25,15,0,0,0.045871559633027525,698,0.0,0.8379204892966361,0.0 diff --git a/kernels/relu/16x16xf64/tests.csv b/kernels/relu/16x16xf64/tests.csv index 5e276afc..7a0af666 100644 --- a/kernels/relu/16x16xf64/tests.csv +++ b/kernels/relu/16x16xf64/tests.csv @@ -1,8 +1,8 @@ -impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,fpss_loads,fpss_load_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc,snitch_stores -baseline,0.0,697,78,769,1777,769,256,257,256,256,256.0,0.0,1773,0,0,0,0.0721554116558741,0.9079102715466352,1.0,1.0,4,1.00390625,1.0,0.7113783533765032,0.2368177613320999,0.3328998699609883,1081,0.7835337650323775, -ssr,0.0,703,528,259,1546,259,257,257,0,0,,0.0,1548,0,0,0,0.624113475177305,0.3290978398983482,1.0,1.0,0,1.0,0.0,0.3061465721040189,0.3037825059101655,0.9922779922779924,846,0.9302600472813238, -ssr_frep,0.0,698,15,5,1024,259,12,257,0,0,,0.0,779,0,0,0,0.0458715596330275,0.25,51.8,51.8,245,21.416666666666668,0.0,0.7920489296636085,0.036697247706422,0.0463320463320463,327,0.8379204892966361, -snrt,0.0,703,16,6,1036,259,4,257,0,0,,0.0,783,0,0,0,0.0479041916167664,0.2727272727272727,43.16666666666666,43.16666666666666,253,64.25,0.0,0.7754491017964071,0.0119760479041916,0.0154440154440154,334,0.8233532934131736, -linalg,0.0,720,78,769,1793,769,256,257,256,256,256.0,0.0,1790,0,0,0,0.0726256983240223,0.9079102715466352,1.0,1.0,3,1.00390625,1.0,0.7160148975791434,0.2383612662942272,0.3328998699609883,1074,0.7886405959031657, -snitch_stream,0.0,720,22,5,1041,259,3,257,0,0,,0.0,787,0,0,0,0.0683229813664596,0.1851851851851851,51.8,51.8,254,85.66666666666667,0.0,0.8043478260869565,0.0093167701863354,0.0115830115830115,322,0.8726708074534162, -scf_xdsl,0.0,707,1857,769,4161,769,256,256,257,257,256.0,0.0,4166,0,0,0,0.5367052023121387,0.2928408225437928,1.0,1.0,0,1.0,1.0,0.2222543352601156,0.0739884393063583,0.3328998699609883,3460,0.7589595375722543,2.0 +impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart,snitch_stores +baseline,1081,1773,1777,1.00390625,1.0,256,257,0.2368177613320999,0.3328998699609883,769,256,256,0.7113783533765032,4,256.0,1.0,1.0,0,769,0.9079102715466352,78,0,0,0.0721554116558741,697,0.0,0.7835337650323775,0.0, +ssr,846,1548,1546,1.0,0.0,257,257,0.3037825059101655,0.9922779922779924,259,0,0,0.3061465721040189,0,,1.0,1.0,0,259,0.3290978398983482,528,0,0,0.624113475177305,703,0.0,0.9302600472813238,0.0, +ssr_frep,327,779,1024,21.416666666666668,0.0,12,257,0.036697247706422,0.0463320463320463,259,0,0,0.7920489296636085,245,,51.8,51.8,0,5,0.25,15,0,0,0.0458715596330275,698,0.0,0.8379204892966361,0.0, +snrt,334,783,1036,64.25,0.0,4,257,0.0119760479041916,0.0154440154440154,259,0,0,0.7754491017964071,253,,43.16666666666666,43.16666666666666,0,6,0.2727272727272727,16,0,0,0.0479041916167664,703,0.0,0.8233532934131736,0.0, +linalg,1074,1790,1793,1.00390625,1.0,256,257,0.2383612662942272,0.3328998699609883,769,256,256,0.7160148975791434,3,256.0,1.0,1.0,0,769,0.9079102715466352,78,0,0,0.0726256983240223,720,0.0,0.7886405959031657,0.0, +snitch_stream,322,787,1041,85.66666666666667,0.0,3,257,0.0093167701863354,0.0115830115830115,259,0,0,0.8043478260869565,254,,51.8,51.8,0,5,0.1851851851851851,22,0,0,0.0683229813664596,720,0.0,0.8726708074534162,0.0, +scf_xdsl,3460,4166,4161,1.0,1.0,256,256,0.0739884393063583,0.3328998699609883,769,257,257,0.2222543352601156,0,256.0,1.0,1.0,0,769,0.2928408225437928,1857,0,0,0.5367052023121387,707,0.0,0.7589595375722543,0.0,2.0 diff --git a/kernels/relu/params.csv b/kernels/relu/params.csv index 4a12dc4d..f71b7a04 100644 --- a/kernels/relu/params.csv +++ b/kernels/relu/params.csv @@ -1,8 +1,8 @@ -params,impl,tstart,start,snitch_issues,snitch_fseq_offloads,end_fpss,fpss_issues,fpss_fpu_issues,fpss_fpu_latency,fpss_loads,fpss_load_latency,fpss_stores,tend,end,snitch_load_latency,snitch_loads,snitch_avg_load_latency,snitch_occupancy,snitch_fseq_rel_offloads,fseq_yield,fseq_fpu_yield,fpss_section_latency,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_occupancy,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,cycles,total_ipc,snitch_stores -16x16xf64,baseline,0.0,697,78,769,1777,769,256,257,256,256,256.0,0.0,1773,0,0,0,0.0721554116558741,0.9079102715466352,1.0,1.0,4,1.00390625,1.0,0.7113783533765032,0.2368177613320999,0.3328998699609883,1081,0.7835337650323775, -16x16xf64,ssr,0.0,703,528,259,1546,259,257,257,0,0,,0.0,1548,0,0,0,0.624113475177305,0.3290978398983482,1.0,1.0,0,1.0,0.0,0.3061465721040189,0.3037825059101655,0.9922779922779924,846,0.9302600472813238, -16x16xf64,ssr_frep,0.0,698,15,5,1024,259,12,257,0,0,,0.0,779,0,0,0,0.0458715596330275,0.25,51.8,51.8,245,21.416666666666668,0.0,0.7920489296636085,0.036697247706422,0.0463320463320463,327,0.8379204892966361, -16x16xf64,snrt,0.0,703,16,6,1036,259,4,257,0,0,,0.0,783,0,0,0,0.0479041916167664,0.2727272727272727,43.16666666666666,43.16666666666666,253,64.25,0.0,0.7754491017964071,0.0119760479041916,0.0154440154440154,334,0.8233532934131736, -16x16xf64,linalg,0.0,720,78,769,1793,769,256,257,256,256,256.0,0.0,1790,0,0,0,0.0726256983240223,0.9079102715466352,1.0,1.0,3,1.00390625,1.0,0.7160148975791434,0.2383612662942272,0.3328998699609883,1074,0.7886405959031657, -16x16xf64,snitch_stream,0.0,720,22,5,1041,259,3,257,0,0,,0.0,787,0,0,0,0.0683229813664596,0.1851851851851851,51.8,51.8,254,85.66666666666667,0.0,0.8043478260869565,0.0093167701863354,0.0115830115830115,322,0.8726708074534162, -16x16xf64,scf_xdsl,0.0,707,1857,769,4161,769,256,256,257,257,256.0,0.0,4166,0,0,0,0.5367052023121387,0.2928408225437928,1.0,1.0,0,1.0,1.0,0.2222543352601156,0.0739884393063583,0.3328998699609883,3460,0.7589595375722543,2.0 +params,impl,cycles,end,end_fpss,fpss_avg_fpu_latency,fpss_avg_load_latency,fpss_fpu_issues,fpss_fpu_latency,fpss_fpu_occupancy,fpss_fpu_rel_occupancy,fpss_issues,fpss_load_latency,fpss_loads,fpss_occupancy,fpss_section_latency,fpss_stores,fseq_fpu_yield,fseq_yield,snitch_avg_load_latency,snitch_fseq_offloads,snitch_fseq_rel_offloads,snitch_issues,snitch_load_latency,snitch_loads,snitch_occupancy,start,tend,total_ipc,tstart,snitch_stores +16x16xf64,baseline,1081,1773,1777,1.00390625,1.0,256,257,0.2368177613320999,0.3328998699609883,769,256,256,0.7113783533765032,4,256.0,1.0,1.0,0,769,0.9079102715466352,78,0,0,0.0721554116558741,697,0.0,0.7835337650323775,0.0, +16x16xf64,ssr,846,1548,1546,1.0,0.0,257,257,0.3037825059101655,0.9922779922779924,259,0,0,0.3061465721040189,0,,1.0,1.0,0,259,0.3290978398983482,528,0,0,0.624113475177305,703,0.0,0.9302600472813238,0.0, +16x16xf64,ssr_frep,327,779,1024,21.416666666666668,0.0,12,257,0.036697247706422,0.0463320463320463,259,0,0,0.7920489296636085,245,,51.8,51.8,0,5,0.25,15,0,0,0.0458715596330275,698,0.0,0.8379204892966361,0.0, +16x16xf64,snrt,334,783,1036,64.25,0.0,4,257,0.0119760479041916,0.0154440154440154,259,0,0,0.7754491017964071,253,,43.16666666666666,43.16666666666666,0,6,0.2727272727272727,16,0,0,0.0479041916167664,703,0.0,0.8233532934131736,0.0, +16x16xf64,linalg,1074,1790,1793,1.00390625,1.0,256,257,0.2383612662942272,0.3328998699609883,769,256,256,0.7160148975791434,3,256.0,1.0,1.0,0,769,0.9079102715466352,78,0,0,0.0726256983240223,720,0.0,0.7886405959031657,0.0, +16x16xf64,snitch_stream,322,787,1041,85.66666666666667,0.0,3,257,0.0093167701863354,0.0115830115830115,259,0,0,0.8043478260869565,254,,51.8,51.8,0,5,0.1851851851851851,22,0,0,0.0683229813664596,720,0.0,0.8726708074534162,0.0, +16x16xf64,scf_xdsl,3460,4166,4161,1.0,1.0,256,256,0.0739884393063583,0.3328998699609883,769,257,257,0.2222543352601156,0,256.0,1.0,1.0,0,769,0.2928408225437928,1857,0,0,0.5367052023121387,707,0.0,0.7589595375722543,0.0,2.0 diff --git a/snitch/Makefile.rules b/snitch/Makefile.rules index 9d38ff2b..e3161165 100644 --- a/snitch/Makefile.rules +++ b/snitch/Makefile.rules @@ -240,7 +240,7 @@ LOG_DIR = $<.logs $(DASM) < $< | $(GENTRACE) --permissive -d $*.trace.json > $*.trace.txt %.csv: %.x.logs/logs/trace_hart_00000000.trace.json - python3 -c "import json,sys; data = json.load(open(sys.argv[1])); keys = data[1].keys(); print(','.join(keys)); print(','.join(str(data[1][k]) for k in keys))" $< > $@ + python3 -c "import json,sys; data = json.load(open(sys.argv[1])); keys = sorted(data[1].keys()); print(','.join(keys)); print(','.join(str(data[1][k]) for k in keys))" $< > $@ define get_trace_targets