Lines Matching refs:NOW
749 # in fp1. now, simply jump to _real_ovfl()!
924 # now, what's left that's not dyadic is fsincos. we can distinguish it
999 # in fp1 (don't forget to save fp0). what to do now?
1361 # No exceptions occurred that were also enabled. Now:
1385 # * this is the case where we must call _real_inex() now or else
1612 # now, copy the result to the proper place on the stack
1786 # now, copy the result to the proper place on the stack
1939 # No exceptions occurred that were also enabled. Now:
1954 bra.w fu_in_exc_ovfl_p # do _real_inex() now
1963 # * this is the case where we must call _real_inex() now or else
2157 # now, copy the result to the proper place on the stack
2213 # now, we copy the default result to it's proper location
2253 # now, we copy the default result to it's proper location
2293 # now, we copy the default result to it's proper location
2509 # so, now it's immediate data extended precision AND PACKED FORMAT!
2560 # FP_SRC now holds the src operand.
2593 # now, we're left with ftst and fcmp. so, first let's tag them so that they don't
2632 # now, we save the result, unless, of course, the operation was ftst or fcmp.
2662 # no exception occurred. now, did a disabled, exact overflow occur with inexact
2773 # right now, d0 = the size.
2813 # right now, d1 = size and d0 = the strg.
5062 # 9. Overwrite X by X := X rem 2Pi. Now that |X| <= Pi, #
5089 # 8. Overwrite X by X := X rem 2Pi. Now that |X| <= Pi, #
5212 eor.l %d1,X(%a6) # X IS NOW R'= SGN*R
5269 eor.l %d1,X(%a6) # X IS NOW S'= SGN*S
5579 #--THE REMAINDER INSTRUCTION WHICH IS NOW IN SOFTWARE.
5650 #--WE MUST NOW FIND INT(FP2). SINCE WE NEED THIS VALUE IN
5679 #--We are now ready to perform (R+r) - N*P1 - N*P2, P1 = 2**(L) * Piby2_1 and
5698 #--Now we need to normalize (A,a) to "new (R,r)" where R+r = A+a but
5763 # 8. Overwrite X by X := X rem 2Pi. Now that |X| <= Pi, go back #
6010 #--THE REMAINDER INSTRUCTION WHICH IS NOW IN SOFTWARE.
6081 #--WE MUST NOW FIND INT(FP2). SINCE WE NEED THIS VALUE IN
6110 #--We are now ready to perform (R+r) - N*P1 - N*P2, P1 = 2**(L) * Piby2_1 and
6129 #--Now we need to normalize (A,a) to "new (R,r)" where R+r = A+a but
6378 #--TRUE THAT A DIVIDE IS NOW NEEDED, BUT THE APPROXIMATION FOR
6387 #--NOW WE SEE X AS +-2^K * 1.BBBBBBB....B <- 1. + 63 BITS
6398 mov.l &0x00000000,XFRACLO(%a6) # LOCATION OF X IS NOW F
6422 mov.l (%a1)+,ATANFLO(%a6) # ATANF IS NOW ATAN(|F|)
6425 or.l %d1,ATANF(%a6) # ATANF IS NOW SIGN(F)*ATAN(|F|)
6428 #--THAT'S ALL I HAVE TO DO FOR NOW,
6431 #--U IN FP0, WE ARE NOW READY TO COMPUTE ATAN(U) AS
6649 # not be caught until gen_except(). Now, with the exceptions being
6893 # Now |X| <= 16446 log2, thus #
7266 #--WE NOW COMPUTE EXP(R)-1 BY A POLYNOMIAL
7429 #--WE NOW COMPUTE EXP(R)-1 BY A POLYNOMIAL
8372 mov.l &0x3FFF0000,X(%a6) # X IS NOW Y, I.E. 2^(-K)*X
8454 #--REGISTERS SAVED ARE NOW FPCR,FP1,FP2,FP3
9590 mov.l %d0,-(%sp) # store off ctrl bits for now
9619 mov.l %d0,-(%sp) # save src for now
11001 # has been decoded. Flow continues here where we now decode
11628 # normalized then we really don't need to go through this scaling. but for now,
11684 # save setting this until now because this is where fmul_may_ovfl may jump in
12285 # now, round to extended(and don't alter the FPSR).
13065 # now, round to extended(and don't alter the FPSR).
13680 # now, round to extended(and don't alter the FPSR).
14044 # save setting this until now because this is where fsglmul_may_ovfl may jump in
14685 fbeq.w fadd_zero_exit # if result is zero, end now
14868 # ok, so now the result has a exponent equal to the smallest normalized
14872 # now, we must determine whether the pre-rounded result was an underflow
15138 fbeq.w fsub_zero_exit # if result zero, end now
15321 # ok, so now the result has a exponent equal to the smallest normalized
15325 # now, we must determine whether the pre-rounded result was an underflow
15494 fbge.w fsub_inf_done # sign is now positive
15762 # now, round to extended(and don't alter the FPSR).
18056 # the bsun exception bit was set. now, check to see is BSUN
18067 # now, check to see if the result is to be stored in the data register
19630 # we're not using op011 for now...
19802 # if (packed) then punt for now #
20391 # the stack frame. so, leave it in FP_SRC for now and deal with it later...
20607 # if exp now equals one, then it overflowed so call ovf_res.
20818 # if exp now equals one, then it overflowed so call ovf_res.
20885 swap %d0 # d0 now in upper word
20992 # we'll have to change this, but for now, tough luck!!!
21014 # the question now is whether the exponents sign bit is allowed to be non-zero
21879 subi.w &0x20, %d1 # %d1 now between 0 and 32
22359 clr.l FTEMP_LO(%a0) # lo(man) is now zero
22520 # must distinguish now "Unnormalized zeroes" which we
22739 # result is now rounded properly. convert back to normal format
22749 tst.l FTEMP_HI(%a0) # is value now a zero?
22799 # result is now rounded properly. convert back to normal format
22809 tst.l FTEMP_HI(%a0) # is value now a zero?
22952 # FP_SRC(a6) = packed operand now as a binary FP number #
23127 neg.l %d1 # now negative, make pos and set SE
23277 neg.l %d1 # now its neg; get abs
23706 fmov.x FP_SCR1(%a6),%fp0 # now fp0 has 1.f
23910 # multiply by 10^(d2), which is now only allowed to be 24,
Indexes created Tue Oct 28 12:10:06 GMT 2025