For example, when loading 1.0, use lu52i.d $a0, $zero, 0x3ff movgr2fr.d $fa0, $a0 instead of addi.d $a0, $zero, 1 movgr2fr.d $fa0, $a0 ffint.d.l $fa0, $fa0 Thanks to @xry111 for the suggestion: https://reviews.llvm.org/D128898#3632140

For example, when loading 3.1415926535897931159979634685441851615905761718750, use lu12i.w $a0, 345154 ori $a0, $a0, 3352 lu32i.d $a0, -450053 lu52i.d $a0, $a0, 1024 movgr2fr.d $fa0, $a0 instead of pcalau12i $a0, .LCPI2_0 addi.d $a0, $a0, .LCPI2_0 fld.d $fa0, $a0, 0 Test this patch with 3A5000 on llvm13, the result shows that SPEC CPU2006 FP score increases 1% in average, 470.lbm score increases 8.9%. Thanks to @xry111 for the suggestion: https://reviews.llvm.org/D128898#3632140

For example, when loading 1.03.1415926535897931159979634685441851615905761718750, use lu12i.w $a0, 345154 ori $a0, $a0, 3352 lu32i.d $a0, -450053 lu52i.d $a0, $zero$a0, 0x3ff1024 movgr2fr.d $fa0, $a0 instead of addi.dpcalau12i $a0, $zero, 1.LCPI2_0 movgr2fr.d $fa0addi.d $a0, $a0, $a0.LCPI2_0 ffint.d.l $fa0ld.d $fa0, $a0, 0 Test this patch with 3A5000 on llvm13, the result shows that SPEC CPU2006 FP score increases 1% in average, $fa0470.lbm score increases 8.9%. Thanks to @xry111 for the suggestion: https://reviews.llvm.org/D128898#3632140