Commit Graph

  • 963d903ae4 Merge branch 'elliot/report+video' into 'main' main [Quy Anh] «Elliot» Nguyen 2026-06-19 12:53:27 +02:00
  • 931d2127b3 Final [Quy Anh] «Elliot» Nguyen 2026-06-19 12:53:27 +02:00
  • 77c294214d Merge branch 'elliot/report+video' into 'main' Linus Baumberger 2026-06-19 12:52:20 +02:00
  • acf38ac195 Final [Quy Anh] «Elliot» Nguyen 2026-06-19 12:52:20 +02:00
  • e9aff24767 Merge branch 'elliot/final' into 'main' Ahmed Kallala 2026-06-19 12:10:27 +02:00
  • cd94ec06eb Coup de grâce [Quy Anh] «Elliot» Nguyen 2026-06-19 12:06:29 +02:00
  • 22052263c9 Merge branch 'elliot/remove-duplicates' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-18 22:33:37 +02:00
  • be263a947a Clean up [Quy Anh] «Elliot» Nguyen 2026-06-18 22:33:37 +02:00
  • 0ff7e829de Merge branch 'elliot/scalar-benchmark' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-18 22:25:19 +02:00
  • 822e9d0ef3 Best Non-Vectorized Version Benchmark + Duplicate Files Removal [Quy Anh] «Elliot» Nguyen 2026-06-18 22:25:19 +02:00
  • bd7728015a Merge branch 'elliot/corrections' into 'main' Linus Baumberger 2026-06-18 11:59:50 +02:00
  • e7af3103e8 Perf plot [Quy Anh] «Elliot» Nguyen 2026-06-18 11:59:50 +02:00
  • ca969564a6 Merge branch 'ahmed-moreilp' into 'main' Linus Baumberger 2026-06-18 08:36:11 +02:00
  • 460c445671 Ahmed moreilp Ahmed Kallala 2026-06-18 08:36:11 +02:00
  • 6cbb8121a7 Merge branch 'elliot/flop-count' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-17 07:28:12 +02:00
  • a7d45665da Flop Count Calculations + x86 Measurements + c13 Fix [Quy Anh] «Elliot» Nguyen 2026-06-17 07:28:11 +02:00
  • 0849aa7307 Merge branch 'ahmed-moreilp' into 'main' Ahmed Kallala 2026-06-17 03:01:20 +02:00
  • 816626664c Ahmed moreilp Ahmed Kallala 2026-06-17 03:01:19 +02:00
  • 42b3c7097e Merge branch 'ahmed-moreilp' into 'main' Linus Baumberger 2026-06-15 13:59:29 +02:00
  • 00b6a58d71 Ahmed moreilp Ahmed Kallala 2026-06-15 13:59:29 +02:00
  • 6189784c30 Merge branch 'more_flopcount' into 'main' Linus Baumberger 2026-06-15 13:49:31 +02:00
  • e869831109 more flopcounts salijagic 2026-06-15 13:49:31 +02:00
  • e4b9087e8b Merge branch 'ahmed-moreilp' into 'main' Linus Baumberger 2026-06-15 11:16:03 +02:00
  • acb53e071e Ahmed moreilp Ahmed Kallala 2026-06-15 11:16:03 +02:00
  • d72b87daf1 Merge branch 'linus/flopcount_lut_exp' into 'main' salijagic 2026-06-14 21:33:56 +02:00
  • 386bdd1536 Merge branch 'linus/flopcount_inline_soa_symmetry' into 'main' salijagic 2026-06-14 21:19:44 +02:00
  • c75cdb8ccc Merge branch 'linus/flopcount_inline_soa_symmetry_ILP_16' into 'main' salijagic 2026-06-14 21:04:00 +02:00
  • f448e9b761 Merge branch 'linus/flopcount_inline_soa_symmetry_ILP_8' into 'main' salijagic 2026-06-14 20:58:47 +02:00
  • 71a630894d Merge branch 'linus/flopcount_inline_soa_symmetry_ILP_4' into 'main' salijagic 2026-06-14 20:58:30 +02:00
  • b21eeb08f2 Merge branch 'flop_smail' into 'main' Linus Baumberger 2026-06-14 16:17:27 +02:00
  • 26e4e52705 flopcounts, more is coming salijagic 2026-06-14 16:17:27 +02:00
  • ed69ff010f Merge branch 'linus/flopcount_inline_soa_symmetry_ILP_2' into 'main' salijagic 2026-06-14 16:15:56 +02:00
  • a883b243b1 Merge branch 'linus/flopcount_inline_soa_avx2' into 'main' salijagic 2026-06-14 16:03:19 +02:00
  • fd340a2580 Merge branch 'linus/flopcount_inline_soa_avx2_polyexp' into 'main' salijagic 2026-06-14 16:03:03 +02:00
  • 92021020ce flopcount for the double avx2 with polyexp Linus Baumberger 2026-06-14 16:03:03 +02:00
  • 79c0093861 Merge branch 'linus/flopcount_soa_avx2' into 'main' salijagic 2026-06-14 16:02:22 +02:00
  • 3992237051 Merge branch 'linus/flopcount_soa_avx2_polyexp' into 'main' salijagic 2026-06-14 16:01:56 +02:00
  • cede05925d Merge branch 'linus/flopcount_soa_avx2_polyexp_rsqrt' into 'main' salijagic 2026-06-14 16:01:38 +02:00
  • b56f035369 flopcount for the avx2_polyexp_rsqrt Linus Baumberger 2026-06-14 16:01:38 +02:00
  • 585360915e Merge branch 'linus/flopcount_soa_avx2_polyexp_rsqrt_loopopt' into 'main' salijagic 2026-06-14 16:01:17 +02:00
  • baba2d5bdb flopcount of the best avx2 version Linus Baumberger 2026-06-14 16:01:17 +02:00
  • f50ddcdfea flopcount lut exp lbaumberger 2026-06-14 15:25:56 +02:00
  • 1ebbf35296 flopcount soa_symmetry lbaumberger 2026-06-14 15:21:19 +02:00
  • bc190cf86e flopcount for ILP16 lbaumberger 2026-06-14 15:19:52 +02:00
  • 52ded11271 flopcount for ILP8 lbaumberger 2026-06-14 15:18:56 +02:00
  • 2cc051f0bb flopcount for ILP4 lbaumberger 2026-06-14 15:17:50 +02:00
  • 1de9bff92f flopcount for ILP 2 lbaumberger 2026-06-14 15:14:11 +02:00
  • 315295d599 flopcounts for basic avx2 with doubles lbaumberger 2026-06-12 17:27:54 +02:00
  • 6c339dcc3b flopcount for the initial float soa avx2 variant lbaumberger 2026-06-12 16:45:05 +02:00
  • c142a58f6e flopcount and expfix for soa_avx2_polyexp lbaumberger 2026-06-12 16:36:36 +02:00
  • a87863096f Merge branch 'elliot/flop-count-correction' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-09 00:01:08 +02:00
  • d8d075f1d7 Flop Count Correction [Quy Anh] «Elliot» Nguyen 2026-06-09 00:01:08 +02:00
  • 6d249951eb Merge branch 'elliot/visual-correction' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-05 11:05:04 +02:00
  • 421b34e7cf Visual Correction [Quy Anh] «Elliot» Nguyen 2026-06-05 11:03:43 +02:00
  • 886693cfd3 Merge branch 'elliot/emergency' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-05 06:44:00 +02:00
  • b1b3e09908 Emergency Corrections [Quy Anh] «Elliot» Nguyen 2026-06-05 06:43:31 +02:00
  • 980f6841ea Merge branch 'elliot/add-pptx' into 'main' Linus Baumberger 2026-06-04 16:16:10 +02:00
  • d5843ab44e Added Presentation [Quy Anh] «Elliot» Nguyen 2026-06-04 16:11:48 +02:00
  • 1873f3e3ff Merge branch 'elliot/plot-update' into 'main' Linus Baumberger 2026-06-04 12:32:41 +02:00
  • f50a1c44ea Plot Update [Quy Anh] «Elliot» Nguyen 2026-06-04 12:32:41 +02:00
  • 66bcdae6f9 Merge branch 'elliot/mem-bench' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-03 22:37:08 +02:00
  • 0d43a08edf Memory Read Bandwidth for x86 [Quy Anh] «Elliot» Nguyen 2026-06-03 22:37:08 +02:00
  • e496bcd08e Merge branch 'linus/updated_analysis_v2' into 'main' [Quy Anh] «Elliot» Nguyen 2026-06-03 22:36:44 +02:00
  • ac9ef23559 updated create csv and updated plots Linus Baumberger 2026-06-03 22:36:44 +02:00
  • daf75faaf3 Merge branch 'plots_3' into 'main' Linus Baumberger 2026-06-01 14:29:26 +02:00
  • 3b6e3d4bf3 fixed plot.py salijagic 2026-06-01 14:29:26 +02:00
  • cec3fccbc8 Merge branch 'new_comments' into 'main' Linus Baumberger 2026-06-01 14:28:14 +02:00
  • 05f96917cd added comments to fastest version, maybe skim them salijagic 2026-06-01 14:28:14 +02:00
  • 2c0aa094e3 Merge branch 'linus/updated_analysis' into 'main' salijagic 2026-06-01 10:57:34 +02:00
  • df4fb4a154 additional plot, and short script to do a nonvectorized benchmark (like they... Linus Baumberger 2026-06-01 10:57:34 +02:00
  • fcff5b0389 Merge branch 'linus/analysis' into 'main' salijagic 2026-05-31 22:06:14 +02:00
  • f7504b8238 Linus/analysis Linus Baumberger 2026-05-31 22:06:14 +02:00
  • 0ef7a28965 Merge branch 'measurements_arm' into 'main' Linus Baumberger 2026-05-31 19:42:21 +02:00
  • f668e3c5bf Measurements arm Ahmed Kallala 2026-05-31 19:42:21 +02:00
  • e76a850b31 Merge branch 'elliot/measurements' into 'main' Linus Baumberger 2026-05-30 21:26:19 +02:00
  • 3aa84d6b8f x86 measurements [Quy Anh] «Elliot» Nguyen 2026-05-30 21:26:19 +02:00
  • c16a67a1f8 Merge branch 'elliot/emergency' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-30 00:14:55 +02:00
  • f1b7f41077 Padding Fix + Float RSQRT Patch [Quy Anh] «Elliot» Nguyen 2026-05-30 00:14:32 +02:00
  • 520e7ccc1f Merge branch 'linus/main_file_fixes' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-28 22:54:36 +02:00
  • be0753f9d8 fix for the two main files lbaumberger 2026-05-28 22:49:08 +02:00
  • 29e4c72c4a Merge branch 'elliot/ilp' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-28 22:07:18 +02:00
  • a783661dd6 Added ILP after AVX-512 [Quy Anh] «Elliot» Nguyen 2026-05-28 22:07:18 +02:00
  • 46e8765997 Merge branch 'elliot/avx-512' into 'main' Linus Baumberger 2026-05-28 21:58:04 +02:00
  • dec2fabfba Added AVX-512 for float version + Vectorised second pass [Quy Anh] «Elliot» Nguyen 2026-05-28 21:58:04 +02:00
  • 2949366a57 Merge branch 'linus/variant_avx2_float_loopsplit' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-28 18:55:25 +02:00
  • 7f235ebb20 last minor loop optimization Linus Baumberger 2026-05-28 18:55:25 +02:00
  • 37ce96041d Merge branch 'rsqrt-clean' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-27 17:17:27 +02:00
  • 9d8d63f1bf variant with rsqrt Linus Baumberger 2026-05-27 17:17:26 +02:00
  • c684c9e780 Merge branch 'symmetry' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-27 09:33:36 +02:00
  • 46ded05a72 Symmetry salijagic 2026-05-27 09:33:36 +02:00
  • bd371cd8e5 Merge branch 'elliot/measure.sh' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-26 21:18:42 +02:00
  • 122c6f0428 Added measurement script [Quy Anh] «Elliot» Nguyen 2026-05-26 21:18:42 +02:00
  • 18d9177ef6 Merge branch 'elliot/avx-512' into 'main' Linus Baumberger 2026-05-25 14:01:10 +02:00
  • 7448281515 variant_inlining_8_SoA_AVX512_polyexp_fastinvsqrt_sr.cpp [Quy Anh] «Elliot» Nguyen 2026-05-25 14:01:10 +02:00
  • e7c63ea40a Merge branch 'linus/variant_avx2_float' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-25 12:07:46 +02:00
  • f7bea275b3 SoA implementation for float and AVX2 includind the polyexp fix Linus Baumberger 2026-05-25 12:07:46 +02:00
  • 1d585ffba7 Merge branch 'linus/variant_avx2_impl' into 'main' [Quy Anh] «Elliot» Nguyen 2026-05-25 11:43:58 +02:00
  • 764f07cbe1 AVX2 implementation Linus Baumberger 2026-05-25 11:43:58 +02:00
  • ecaa670ca5 Merge branch 'elliot/avx-512' into 'main' Linus Baumberger 2026-05-25 00:59:19 +02:00
  • 7b3d554117 AVX-512 First Attempt [Quy Anh] «Elliot» Nguyen 2026-05-25 00:59:19 +02:00