Lines Matching refs:L1
104 | 3.1 R := X + N*L1, where L1 := single-precision(-log2/64).
105 | 3.2 R := R + N*L2, L2 := extended-precision(-log2/64 - L1).
106 | Notes: a) The way L1 and L2 are chosen ensures L1+L2 approximate
108 | b) N*L1 is exact because N is no longer than 22 bits and
109 | L1 is no longer than 24 bits.
110 | c) The calculation X+N*L1 is also exact due to cancellation.
111 | Thus, R is practically X+N(L1+L2) to full 64 bits.
506 fmuls #0xBC317218,%fp0 | ...N * L1, L1 = lead(-log2/64)
507 fmulx L2,%fp2 | ...N * L2, L1+L2 = -log2/64
508 faddx %fp1,%fp0 | ...X + N*L1
672 fmuls #0xBC317218,%fp0 | ...N * L1, L1 = lead(-log2/64)
673 fmulx L2,%fp2 | ...N * L2, L1+L2 = -log2/64
674 faddx %fp1,%fp0 | ...X + N*L1