Lines Matching refs:Step
5481 #--case, reduce argument by one remainder step to make subsequent reduction
5912 #--case, reduce argument by one remainder step to make subsequent reduction
6063 # Step 1. If |X| >= 16 or |X| < 1/16, go to Step 5. #
6065 # Step 2. Let X = sgn * 2**k * 1.xxxxxxxx...x. #
6071 # Step 3. Approximate arctan(u) by a polynomial poly. #
6073 # Step 4. Return arctan(F) + poly, arctan(F) is fetched from a #
6076 # Step 5. If |X| >= 16, go to Step 7. #
6078 # Step 6. Approximate arctan(X) by an odd polynomial in X. Exit. #
6080 # Step 7. Define X' = -1/X. Approximate arctan(X') by an odd #
6713 # Step 1. Set ans := 1.0 #
6715 # Step 2. Return ans := ans + sign(X)*2^(-126). Exit. #
6722 # Step 1. Filter out extreme cases of input argument. #
6723 # 1.1 If |X| >= 2^(-65), go to Step 1.3. #
6724 # 1.2 Go to Step 7. #
6725 # 1.3 If |X| < 16380 log(2), go to Step 2. #
6726 # 1.4 Go to Step 8. #
6736 # 16380 log(2) used in Step 1.3 is also in the compact #
6737 # form. Thus taking the branch to Step 2 guarantees #
6740 # 16380 log(2) and the branch to Step 9 is taken. #
6742 # Step 2. Calculate N = round-to-nearest-int( X * 64/log2 ). #
6766 # Step 3. Calculate X - N*log2/64. #
6779 # after Step 3.2. #
6791 # This bound will be used in Step 4. #
6793 # Step 4. Approximate exp(R)-1 by a polynomial #
6808 # Step 5. Compute 2^(J/64)*exp(R) = 2^(J/64)*(1+p) by #
6820 # Step 6. Reconstruction of exp(X) #
6840 # Step 7. Return 1 + X. #
6847 # in Step 7.1 to avoid unnecessary trapping. (Although #
6853 # Step 8. Handle exp(X) where |X| >= 16380log2. #
6854 # 8.1 If |X| > 16480 log2, go to Step 9. #
6863 # 8.7 Go to Step 3. #
6866 # Step 9. Handle exp(X), |X| > 16480 log2. #
6881 # Step 1. Set ans := 0 #
6883 # Step 2. Return ans := X + ans. Exit. #
6890 # Step 1. Check |X| #
6891 # 1.1 If |X| >= 1/4, go to Step 1.3. #
6892 # 1.2 Go to Step 7. #
6893 # 1.3 If |X| < 70 log(2), go to Step 2. #
6894 # 1.4 Go to Step 10. #
6899 # the comparisons, see the notes on Step 1 of setox. #
6901 # Step 2. Calculate N = round-to-nearest-int( X * 64/log2 ). #
6910 # Notes: See the notes on Step 2 of setox. #
6912 # Step 3. Calculate X - N*log2/64. #
6917 # Notes: Applying the analysis of Step 3 of setox in this case #
6921 # Step 4. Approximate exp(R)-1 by a polynomial #
6936 # Step 5. Compute 2^(J/64)*p by #
6945 # be exploited in Step 6 below. The total relative error #
6949 # Step 6. Reconstruction of exp(X)-1 #
6951 # 6.1 If M <= 63, go to Step 6.3. #
6961 # Step 7. exp(X)-1 for |X| < 1/4. #
6962 # 7.1 If |X| >= 2^(-65), go to Step 9. #
6963 # 7.2 Go to Step 8. #
6965 # Step 8. Calculate exp(X)-1, |X| < 2^(-65). #
6979 # Step 9. Calculate exp(X)-1, |X| < 1/4, by a polynomial #
6998 # Step 10. Calculate exp(X)-1 for |X| >= 70 log 2. #
7000 # practical purposes. Therefore, go to Step 1 of setox. #
7113 #--Step 1.
7128 #--Step 2.
7150 #--Step 3.
7159 #--Step 4.
7193 #--Step 5
7204 #--Step 6
7216 #--Step 7
7223 #--Step 8
7251 bra.w EXPCONT1 # go back to Step 3
7254 #--Step 9
7276 #--Step 1.
7277 #--Step 1.1
7285 #--Step 1.3
7293 #--Step 2.
7312 #--Step 3.
7322 #--Step 4.
7365 #--Step 5
7370 #--Step 6
7371 #--Step 6.1
7375 #--Step 6.2 M >= 64
7382 #--Step 6.3 M <= 63
7386 #--Step 6.4 M <= -4
7392 #--Step 6.5 -3 <= M <= 63
7399 #--Step 6.6
7405 #--Step 7 |X| < 1/4.
7410 #--Step 8 |X| < 2^(-65)
7413 #--Step 8.2
7424 #--Step 8.3
7437 #--Step 9 exp(X)-1 by a simple polynomial
7488 #--Step 10 |X| > 70 log2
7492 #--Step 10.2
7501 #--Step 0.
7990 # Step 1. If |X-1| < 1/16, approximate log(X) by an odd #
7992 # move on to Step 2. #
7994 # Step 2. X = 2**k * Y where 1 <= Y < 2. Define F to be the first #
7999 # Step 3. Define u = (Y-F)/F. Approximate log(1+u) by a #
8002 # Step 4. Reconstruct #
8008 # Step 1: If |X| < 1/16, approximate log(1+X) by an odd #
8010 # to Step 2. #
8012 # Step 2: Let 1+X = 2**k * Y, where 1 <= Y < 2. Define F as done #
8013 # in Step 2 of the algorithm for LOGN and compute #
8023 # Note 2. In Step 2 of lognp1, in order to preserved accuracy, #
8677 # Step 0. If X < 0, create a NaN and raise the invalid operation #
8682 # Step 1. Call slognd to obtain Y = log(X), the natural log of X. #
8685 # Step 2. Compute log_10(X) = log(X) * (1/log(10)). #
8691 # Step 0. If X < 0, create a NaN and raise the invalid operation #
8696 # Step 1. Call sLogN to obtain Y = log(X), the natural log of X. #
8698 # Step 2. Compute log_10(X) = log(X) * (1/log(10)). #
8704 # Step 0. If X < 0, create a NaN and raise the invalid operation #
8709 # Step 1. Call slognd to obtain Y = log(X), the natural log of X. #
8712 # Step 2. Compute log_10(X) = log(X) * (1/log(2)). #
8718 # Step 0. If X < 0, create a NaN and raise the invalid operation #
8723 # Step 1. If X is not an integer power of two, i.e., X != 2^k, #
8724 # go to Step 3. #
8726 # Step 2. Return k. #
8731 # Step 3. Call sLogN to obtain Y = log(X), the natural log of X. #
8733 # Step 4. Compute log_2(X) = log(X) * (1/log(2)). #
9366 # Step 1. Save and strip signs of X and Y: signX := sign(X), #
9371 # Step 2. Set L := expo(X)-expo(Y), k := 0, Q := 0. #
9373 # R := X, go to Step 4. #
9378 # Step 3. Perform MOD(X,Y) #
9379 # 3.1 If R = Y, go to Step 9. #
9381 # 3.3 If j = 0, go to Step 4. #
9383 # Step 3.1. #
9385 # Step 4. At this point, R = X - QY = MOD(X,Y). Set #
9386 # Last_Subtract := false (used in Step 7 below). If #
9387 # MOD is requested, go to Step 6. #
9389 # Step 5. R = MOD(X,Y), but REM(X,Y) is requested. #
9391 # Step 6. #
9393 # Q := Q + 1, Y := signY*Y }. Go to Step 6. #
9397 # Step 6. R := signX*R. #
9399 # Step 7. If Last_Subtract = true, R := R - Y. #
9401 # Step 8. Return signQ, last 7 bits of Q, and R as required. #
9403 # Step 9. At this point, R = 2^(-j)*X - Q Y = Y. Thus, #
Indexes created Fri Oct 17 03:10:13 GMT 2025