en Copyright 2015 Java http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Assembler/bfloat.ll#a2588948 Fix convertBFloatAPFloatToAPInt for NaN/Inf valuesBfloat type has an 8-bit exponent so the exponent of NaN/Inf numbersmust be 0xff instead of 0x1f. This is probably a copy-paste mistakefrom the half float type.Reviewed By: lattnerDifferential Revision: https://reviews.llvm.org/D81302 List of files: /llvm-project-15.0.7/llvm/test/Assembler/bfloat.ll Sat, 06 Jun 2020 00:06:42 +0000 Diego Caballero <diego.caballero@intel.com> http://172.16.0.5:8080/history/llvm-project-15.0.7/llvm/test/Assembler/bfloat.ll#8c24f331 [IR][BFloat] Add BFloat IR typeSummary:The BFloat IR type is introduced to provide support for, initially, the BFloat16datatype introduced with the Armv8.6 architecture (optional from Armv8.2onwards). It has an 8-bit exponent and a 7-bit mantissa and behaves like an IEEE754 floating point IR type.This is part of a patch series upstreaming Armv8.6 features. Subsequent patcheswill upstream intrinsics support and C-lang support for BFloat.Reviewers: SjoerdMeijer, rjmccall, rsmith, liutianle, RKSimon, craig.topper, jfb, LukeGeeson, sdesmalen, deadalnix, ctetreauSubscribers: hiraditya, llvm-commits, danielkiss, arphaman, kristof.beyls, dexonsmithTags: #llvmDifferential Revision: https://reviews.llvm.org/D78190 List of files: /llvm-project-15.0.7/llvm/test/Assembler/bfloat.ll Tue, 31 Mar 2020 22:49:38 +0000 Ties Stuij <ties.stuij@arm.com>