URL
https://opencores.org/ocsvn/openrisc/openrisc/trunk
Subversion Repositories openrisc
[/] [openrisc/] [trunk/] [gnu-old/] [binutils-2.18.50/] [gas/] [testsuite/] [gas/] [bfin/] [arithmetic.s] - Rev 816
Compare with Previous | Blame | View Log
.text .global abs abs: a0 = abs a0; A0 = ABS A1; A1 = Abs a0; a1 = aBs A1; A1 = abs a1, a0 = ABS A0; r0 = abs r2; .text .global add add: sp = sp + P0; SP = SP + P2; FP = p1 + fp; R7 = r7 + r2 (NS); r6 = r6 + r0 (s); r4.L = R2.h + r0.L (s); r5.H = R1.H + R1.L (S); r6.L = R6.L + r5.l (NS); .text .global add_sub_prescale_down add_sub_prescale_down: r4.l = r0 + r1 (RND20); R3.H = r5 + r0 (rnd20); r1.L = r7 - R5 (rND20); .text .global add_sub_prescale_up add_sub_prescale_up: r2.L = R0 + R1 (rnd12); r7.H = r7 + r6 (RND12); r5.l = r3 - R2 (rNd12); r2.h = R1 - R2 (Rnd12); .text .global add_immediate add_immediate: R5 += -64; r2 += 63; i0 += 2; I3 += 2; I2 += 4; i1 += 4; P0 += 4; sp += 16; FP += -32; .text .global divide_primitive divide_primitive: divs (r3, r5); divq (R3, R5); .text .global expadj expadj: r6.L = EXPADJ (r5, r4.l); R5.l = ExpAdj (r0.h, r1.l); R4.L = expadj (R3, R5.L) (V); .text .global max max: R6 = MAX (r5, R2); r0 = max (r1, r3); .text .global min min: r5 = mIn (r2, R3); R4 = Min (r7, R0); .text .global modify_decrement modify_decrement: A0 -= A1; a0 -= a1 (w32); fp -= p2; SP -= P0; I3 -= M0; i1 -= m1; .text .global modify_increment modify_increment: a0 += a1; A0 += A1 (w32); Sp += P1 (Brev); P5 += Fp (BREV); i2 += M2; I0 += m0 (brev); r7 = ( a0 += a1); r6.l = (A0 += a1); R0.H = (a0 += A1); .text .global multiply16 multiply16: R0.l = r1.h * r2.l; r1.L = r5.H * r0.H (s2rnd); r7.l = r3.l * r3.H (FU); r4 = r2.H * r5.H (iSS2); r0 = r1.l * r3.l (is); r6 = R5.H * r0.l; r2.h = r7.l * r6.H (M, iu); r3.H = r5.H * r0.L; R0.H = r1.L * r1.H (M); r1 = r7.H * r6.L (M, is); R5 = r0.l * r2.h; r3 = r6.H * r0.H (m); .text .global multiply32 multiply32: R4 *= r0; r7 *= R2; .text .global multiply_accumulate multiply_accumulate: a0 = r5.l * R7.H (w32); a0 = r0.h * r0.l; A0 += R2.L * r3.H (FU); A0 += r4.h * r1.L; a0 -= r7.l * r6.H (Is); A0 -= R5.H * r2.H; a1 = r1.L * r0.H (M); A1 = r2.h * r0.L; A1 = R7.H * R6.L (M, W32); a1 += r3.l * r2.l (fu); a1 += R6.H * r1.L; A1 -= r0.L * R3.H (is); a1 -= r2.l * r7.h; .text .global multiply_accumulate_half multiply_accumulate_half: r7.l = (a0 = r6.H * r5.L); r0.L = (A0 = r1.h * R2.l) (tfu); R2.L = (a0 += r5.L * r4.L); r3.l = (A0 += r7.H * r6.h) (T); r0.l = (a0 -= r3.h * r2.h); r1.l = (a0 -= r5.L * r4.L) (iH); r1.H = (a1 = r1.l * R0.H); r2.h = (A1 = r0.H * r3.L) (M, Iss2); R6.H = (a1 += r7.l * r7.H); r7.h = (a1 += R2.L * R3.L) (S2rnd); r6.H = (A1 -= R4.h * r2.h); r5.h = (a1 -= r3.H * r7.L) (M, tFu); .text .global multiply_accumulate_data_reg multiply_accumulate_data_reg: R0 = (A0 = R1.L * R2.L); R2 = (A0 = r1.l * r2.l) (is); r4 = (a0 += r7.h * r6.L); r6 = (A0 += R5.L * r3.h) (s2RND); R6 = (a0 -= r2.h * r7.l); r4 = (A0 -= R0.L * r6.H) (FU); r7 = (a1 = r0.h * r1.l); R5 = (A1 = r2.H * r3.H) (M, fu); R3 = (A1 += r7.l * r5.l); r1 = (a1 += r2.h * r7.h) (iss2); r3 = (A1 -= r0.l * R0.H); R5 = (a1 -= R2.l * R7.h) (m, is); r7 = (a1+=r1.h*r3.h); .text .global negate negate: R5 = - r0; r7 = -R2(s); R7 = -r2(Ns); A0 = -A0; a0 = -a1; A1 = -A0; a1 = -A1; a1 = -a1, a0 = -a0; .text .global round_half round_half: R5.L = r3 (rnd); r6.H = r0 (RND); .text .global saturate saturate: A0 = A0 (S); a1 = a1 (s); A1 = a1 (S), a0 = A0 (s); .text .global signbits signbits: R5.l = signbits r0; r0.L = SIGNbits r7.H; r3.l = signBits A0; r7.L = SIGNBITS a1; .text .global subtract subtract: R5 = R3 - R0; R7 = R7 - r0 (S); r3 = r2 - r1 (ns); r5.l = R6.H - R7.h (s); r0.H = r3.l - r3.h (NS); .text .global subtract_immediate subtract_immediate: I2 -= 2; i0 -= 4;