en Copyright 2015 Java http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Analysis/CostModel/X86/interleaved-load-float.ll#198aa849 [X86][Costmodel] Load/store i32/f32 Stride=3 VF=8 interleaving costsThe only sched models that for cpu's that support avx2but not avx512 are: haswell, broadwell, skylake, zen1-3For load we have:https://godbolt.org/z/zdz5Ga6fs - for intels `Block RThroughput: =7.0`; for ryzens, `Block RThroughput: <=6.0`So pick cost of `7`.For store we have:https://godbolt.org/z/qn71513ac - for intels `Block RThroughput: =11.0`; for ryzens, `Block RThroughput: <=8.0`So pick cost of `11`.I'm directly using the shuffling asm the llc produced,without any manual fixups that may be neededto ensure sequential execution.Reviewed By: RKSimonDifferential Revision: https://reviews.llvm.org/D111021 List of files: /llvm-project-15.0.7/llvm/test/Analysis/CostModel/X86/interleaved-load-float.ll Mon, 04 Oct 2021 11:23:13 +0000 Roman Lebedev <lebedev.ri@gmail.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Analysis/CostModel/X86/interleaved-load-float.ll#66917583 [LV][X86] update the cost of interleaving mem. access of floatsRecommit:This patch contains update of the costs of interleaved loads of v8f32 of stride 3 and 8.fixed the location of the lit test it works with make check-all.Differential Revision: https://reviews.llvm.org/D39403llvm-svn: 317471 List of files: /llvm-project-15.0.7/llvm/test/Analysis/CostModel/X86/interleaved-load-float.ll Mon, 06 Nov 2017 10:56:20 +0000 Mohammed Agabaria <mohammed.agabaria@intel.com>