--- /dev/null
+@@\r
+@@ Copyright (C) 2012 Roman Pauer\r
+@@\r
+@@ Permission is hereby granted, free of charge, to any person obtaining a copy of\r
+@@ this software and associated documentation files (the "Software"), to deal in\r
+@@ the Software without restriction, including without limitation the rights to\r
+@@ use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies\r
+@@ of the Software, and to permit persons to whom the Software is furnished to do\r
+@@ so, subject to the following conditions:\r
+@@\r
+@@ The above copyright notice and this permission notice shall be included in all\r
+@@ copies or substantial portions of the Software.\r
+@@\r
+@@ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\r
+@@ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\r
+@@ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\r
+@@ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\r
+@@ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\r
+@@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE\r
+@@ SOFTWARE.\r
+@@\r
+\r
+\r
+@ S T U --\ E1 E2\r
+@ V C W --/ E3 E4\r
+@ X Y Z\r
+\r
+@ q0 = S1sl < S >\r
+@ q1 = S2sl < V >\r
+@ q2 = S3sl < X >\r
+@ q3 = S1sr < U >\r
+@ q4 = S2sr < W >\r
+@ q5 = S3sr < Z >\r
+@ q6 = E3\r
+@ q7 = E4\r
+@ q8 = S1\r
+@ q9 = S2\r
+@ q10 = S3\r
+@ q11 = S1prev < T >\r
+@ q12 = S2prev < C >\r
+@ q13 = S3prev < Y >\r
+@ q14 = E1\r
+@ q15 = E2\r
+\r
+\r
+.macro __neon_eagle2x_8_8_line src1, src2, src3, counter, dst1, dst2, reg1, qT, qY, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vld1.8 {d23[7]}, [\src1] @ S1prev[15] = src[-srcstride]\r
+ .endif\r
+ vld1.8 {d25[7]}, [\src2] @ S2prev[15] = src[0]\r
+ .ifeqs "\qY", "q13"\r
+ vld1.8 {d27[7]}, [\src3] @ S3prev[15] = src[srcstride]\r
+ .endif\r
+ andS \reg1, \counter, #15 @ reg1 = counter & 15\r
+\r
+ .ifnes "\qT", "q11"\r
+ add \src1, \src1, \counter @ src1 += counter\r
+ .endif\r
+ .ifnes "\qY", "q13"\r
+ add \src3, \src3, \counter @ src3 += counter\r
+ .endif\r
+ beq 1f\r
+\r
+ @ first 1-15 pixels - align counter to 16 bytes\r
+\r
+@ q0 = S1sl < S >\r
+@ q2 = S3sl < X >\r
+@ q7 = tmp2\r
+@ q15 = tmp1\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vld1.8 {q8}, [\src1], \reg1 @ S1 = [src - srcstride]; src1 += counter & 15\r
+ .endif\r
+\r
+ vld1.8 {q9}, [\src2], \reg1 @ S2 = [src ]; src2 += counter & 15\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vld1.8 {q10}, [\src3], \reg1 @ S3 = [src + srcstride]; src3 += counter & 15\r
+ .endif\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q0, \qT, q8, #15 @ S1sl = S1prev[15] | (S1 << 8) < S >\r
+\r
+ vmov \qT, q8 @ S1prev = S1 < T >\r
+ .endif\r
+ vext.8 q1, q12, q9, #15 @ S2sl = S2prev[15] | (S2 << 8) < V >\r
+\r
+ vmov q12, q9 @ S2prev = S2 < C >\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q2, \qY, q10, #15 @ S3sl = S3prev[15] | (S3 << 8) < X >\r
+\r
+ vmov \qY, q10 @ S3prev = S3 < Y >\r
+ .endif\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q3, \qT, q8, #1 @ S1sr = (S1prev >> 8) | ... < U >\r
+ .endif\r
+\r
+ vext.8 q4, q12, q9, #1 @ S2sr = (S2prev >> 8) | ... < W >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q5, \qY, q10, #1 @ S3sr = (S3prev >> 8) | ... < Z >\r
+ .else\r
+ vmov q2, q1 @ S3sl = S2sl < X >\r
+\r
+ vmov q5, q4 @ S3sr = S2sr < Z >\r
+ .endif\r
+\r
+ .ifnes "\qT", "q11"\r
+ vmov q0, q1 @ S1sl = S2sl < S >\r
+\r
+ vmov q3, q4 @ S1sr = S2sr < U >\r
+ .endif\r
+\r
+ vceq.i8 q14, q0, \qT @ E1 = < S == T >\r
+\r
+ vceq.i8 q15, q0, q1 @ tmp1 = < S == V >\r
+\r
+ vceq.i8 q6, q2, \qY @ E3 = < X == Y >\r
+\r
+ vceq.i8 q7, q2, q1 @ tmp2 = < X == V >\r
+\r
+ vand q14, q14, q15 @ E1 = < S == T && S == V >\r
+\r
+@ q0 = tmp3\r
+@ q15 = E2\r
+\r
+ vceq.i8 q15, q3, \qT @ E2 = < U == T >\r
+\r
+ vceq.i8 q0, q3, q4 @ tmp3 = < U == W >\r
+\r
+ vand q6, q6, q7 @ E3 = < X == Y && X == V >\r
+\r
+@ q2 = tmp4\r
+@ q7 = E4\r
+ vceq.i8 q7, q5, \qY @ E4 = < Z == Y >\r
+\r
+ vceq.i8 q2, q5, q4 @ tmp4 = < Z == W >\r
+\r
+ vand q15, q15, q0 @ E2 = < U == T && U == W >\r
+\r
+ vbsl q14, \qT, q12 @ E1 = < (S == T && S == V) ? T : C >\r
+\r
+ vbsl q15, \qT, q12 @ E2 = < (U == T && U == W) ? T : C >\r
+\r
+ vand q7, q7, q2 @ E4 = < Z == Y && Z == W >\r
+\r
+ vbsl q6, \qY, q12 @ E3 = < (X == Y && X == V) ? Y : C >\r
+\r
+ .ifeqs "\qT", "q11"\r
+ sub \reg1, \src1, #1\r
+ .else\r
+ sub \reg1, \src2, #1\r
+ .endif\r
+\r
+ vbsl q7, \qY, q12 @ E4 = < (Z == Y && Z == W) ? Y : C >\r
+ .ifeqs "\qT", "q11"\r
+ vld1.8 {d23[7]}, [\reg1] @ S1prev[15] = src[counter & 15 - 1 - srcstride]\r
+\r
+ sub \reg1, \src2, #1\r
+ .endif\r
+\r
+ vld1.8 {d25[7]}, [\reg1] @ S2prev[15] = src[counter & 15 - 1]\r
+\r
+ .ifeqs "\qY", "q13"\r
+ sub \reg1, \src3, #1\r
+\r
+ vld1.8 {d27[7]}, [\reg1] @ S3prev[15] = src[counter & 15 - 1 + srcstride]\r
+ .endif\r
+\r
+ ubfx \reg1, \counter, #0, #4 @ reg1 = counter & 15\r
+\r
+ lsl \reg1, #1\r
+\r
+ vst2.8 {q14-q15}, [\dst1],\reg1 @ [dst] = E1,E2; dst1 += reg1\r
+\r
+ bic \counter, \counter, #15\r
+\r
+ vst2.8 {q6-q7}, [\dst2], \reg1 @ [dst + dststride] = E3,E4; dst2 += reg1\r
+\r
+ @ counter is aligned to 16 bytes\r
+\r
+ 1:\r
+ .ifeqs "\qT", "q11"\r
+ vld1.8 {q8}, [\alsrc1]! @ S1 = [src - srcstride]; src1 += 16\r
+ .endif\r
+ vld1.8 {q9}, [\alsrc2]! @ S2 = [src ]; src2 += 16\r
+ .ifeqs "\qY", "q13"\r
+ vld1.8 {q10}, [\alsrc3]! @ S3 = [src + srcstride]; src3 += 16\r
+ .endif\r
+\r
+ @ inner loop (16 pixels per iteration)\r
+ 2:\r
+\r
+@ q0 = S1sl < S >\r
+@ q2 = S3sl < X >\r
+@ q7 = tmp2\r
+@ q15 = tmp1\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q0, \qT, q8, #15 @ S1sl = S1prev[15] | (S1 << 8) < S >\r
+ vmov \qT, q8 @ S1prev = S1 < T >\r
+ .endif\r
+\r
+ vext.8 q1, q12, q9, #15 @ S2sl = S2prev[15] | (S2 << 8) < V >\r
+ vmov q12, q9 @ S2prev = S2 < C >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q2, \qY, q10, #15 @ S3sl = S3prev[15] | (S3 << 8) < X >\r
+ vmov \qY, q10 @ S3prev = S3 < Y >\r
+ .endif\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vld1.8 {q8}, [\alsrc1]! @ S1 = [src - srcstride]; src1 += 16\r
+ vext.8 q3, \qT, q8, #1 @ S1sr = (S1prev >> 8) | S1[0] < U >\r
+ .endif\r
+\r
+ vld1.8 {q9}, [\alsrc2]! @ S2 = [src ]; src2 += 16\r
+ vext.8 q4, q12, q9, #1 @ S2sr = (S2prev >> 8) | S2[0] < W >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vld1.8 {q10}, [\alsrc3]! @ S3 = [src + srcstride]; src3 += 16\r
+ vext.8 q5, \qY, q10, #1 @ S3sr = (S3prev >> 8) | S3[0] < Z >\r
+ .else\r
+ vmov q2, q1 @ S3sl = S2sl < X >\r
+\r
+ vmov q5, q4 @ S3sr = S2sr < Z >\r
+ .endif\r
+\r
+ .ifnes "\qT", "q11"\r
+ vmov q0, q1 @ S1sl = S2sl < S >\r
+\r
+ vmov q3, q4 @ S1sr = S2sr < U >\r
+ .endif\r
+\r
+ sub \counter, \counter, #16 @ counter -= 16\r
+ vceq.i8 q14, q0, \qT @ E1 = < S == T >\r
+\r
+ vceq.i8 q15, q0, q1 @ tmp1 = < S == V >\r
+\r
+ vceq.i8 q6, q2, \qY @ E3 = < X == Y >\r
+\r
+ vceq.i8 q7, q2, q1 @ tmp2 = < X == V >\r
+\r
+ vand q14, q14, q15 @ E1 = < S == T && S == V >\r
+\r
+@ q0 = tmp3\r
+@ q15 = E2\r
+\r
+ vceq.i8 q15, q3, \qT @ E2 = < U == T >\r
+\r
+ vceq.i8 q0, q3, q4 @ tmp3 = < U == W >\r
+\r
+ vand q6, q6, q7 @ E3 = < X == Y && X == V >\r
+\r
+@ q2 = tmp4\r
+@ q7 = E4\r
+ vceq.i8 q7, q5, \qY @ E4 = < Z == Y >\r
+\r
+ vceq.i8 q2, q5, q4 @ tmp4 = < Z == W >\r
+\r
+ vand q15, q15, q0 @ E2 = < U == T && U == W >\r
+\r
+ vbsl q14, \qT, q12 @ E1 = < (S == T && S == V) ? T : C >\r
+\r
+ vbsl q15, \qT, q12 @ E2 = < (U == T && U == W) ? T : C >\r
+\r
+ vand q7, q7, q2 @ E4 = < Z == Y && Z == W >\r
+\r
+ vbsl q6, \qY, q12 @ E3 = < (X == Y && X == V) ? Y : C >\r
+\r
+ vbsl q7, \qY, q12 @ E4 = < (Z == Y && Z == W) ? Y : C >\r
+ vst2.8 {q14-q15}, [\aldst1]! @ [dst] = E1,E2; dst1 += 2*16\r
+\r
+ cmp \counter, #16\r
+\r
+ vst2.8 {q6-q7}, [\aldst2]! @ [dst + dststride] = E3,E4; dst2 += 2*16\r
+ bhi 2b\r
+\r
+ @ last 16 pixels\r
+\r
+@ q0 = S1sl < S >\r
+@ q2 = S3sl < X >\r
+@ q7 = tmp2\r
+@ q15 = tmp1\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q0, \qT, q8, #15 @ S1sl = S1prev[15] | (S1 << 8) < S >\r
+ vmov \qT, q8 @ S1prev = S1 < T >\r
+ .endif\r
+\r
+ vext.8 q1, q12, q9, #15 @ S2sl = S2prev[15] | (S2 << 8) < V >\r
+ vmov q12, q9 @ S2prev = S2 < C >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q2, \qY, q10, #15 @ S3sl = S3prev[15] | (S3 << 8) < X >\r
+ vmov \qY, q10 @ S3prev = S3 < Y >\r
+ .endif\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vshr.u64 d16, d17, #(64-8) @ S1[0] = S1[15] | ...\r
+ .endif\r
+\r
+ vshr.u64 d18, d19, #(64-8) @ S2[0] = S2[15] | ...\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vshr.u64 d20, d21, #(64-8) @ S3[0] = S3[15] | ...\r
+ .endif\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q3, \qT, q8, #1 @ S1sr = (S1prev >> 8) | S1[0] < U >\r
+ .endif\r
+\r
+ vext.8 q4, q12, q9, #1 @ S2sr = (S2prev >> 8) | S2[0] < W >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q5, \qY, q10, #1 @ S3sr = (S3prev >> 8) | S3[0] < Z >\r
+ .else\r
+ vmov q2, q1 @ S3sl = S2sl < X >\r
+\r
+ vmov q5, q4 @ S3sr = S2sr < Z >\r
+ .endif\r
+\r
+ .ifnes "\qT", "q11"\r
+ vmov q0, q1 @ S1sl = S2sl < S >\r
+\r
+ vmov q3, q4 @ S1sr = S2sr < U >\r
+ .endif\r
+\r
+ vceq.i8 q14, q0, \qT @ E1 = < S == T >\r
+\r
+ vceq.i8 q15, q0, q1 @ tmp1 = < S == V >\r
+\r
+ vceq.i8 q6, q2, \qY @ E3 = < X == Y >\r
+\r
+ vceq.i8 q7, q2, q1 @ tmp2 = < X == V >\r
+\r
+ vand q14, q14, q15 @ E1 = < S == T && S == V >\r
+\r
+@ q0 = tmp3\r
+@ q15 = E2\r
+\r
+ vceq.i8 q15, q3, \qT @ E2 = < U == T >\r
+\r
+ vceq.i8 q0, q3, q4 @ tmp3 = < U == W >\r
+\r
+ vand q6, q6, q7 @ E3 = < X == Y && X == V >\r
+\r
+@ q2 = tmp4\r
+@ q7 = E4\r
+ vceq.i8 q7, q5, \qY @ E4 = < Z == Y >\r
+\r
+ vceq.i8 q2, q5, q4 @ tmp4 = < Z == W >\r
+\r
+ vand q15, q15, q0 @ E2 = < U == T && U == W >\r
+\r
+ vbsl q14, \qT, q12 @ E1 = < (S == T && S == V) ? T : C >\r
+\r
+ vbsl q15, \qT, q12 @ E2 = < (U == T && U == W) ? T : C >\r
+\r
+ vand q7, q7, q2 @ E4 = < Z == Y && Z == W >\r
+\r
+ vbsl q6, \qY, q12 @ E3 = < (X == Y && X == V) ? Y : C >\r
+\r
+ vbsl q7, \qY, q12 @ E4 = < (Z == Y && Z == W) ? Y : C >\r
+ vst2.8 {q14-q15}, [\aldst1]! @ [dst] = E1,E2; dst1 += 2*16\r
+\r
+ vst2.8 {q6-q7}, [\aldst2]! @ [dst + dststride] = E3,E4; dst2 += 2*16\r
+\r
+.endm\r
+\r
+.macro _neon_eagle2x_8_8_line_first src1, src2, src3, counter, dst1, dst2, reg1, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+ __neon_eagle2x_8_8_line \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, q12, q13, \alsrc1, \alsrc2, \alsrc3, \aldst1, \aldst2\r
+.endm\r
+\r
+.macro _neon_eagle2x_8_8_line_middle src1, src2, src3, counter, dst1, dst2, reg1, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+ __neon_eagle2x_8_8_line \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, q11, q13, \alsrc1, \alsrc2, \alsrc3, \aldst1, \aldst2\r
+.endm\r
+\r
+.macro _neon_eagle2x_8_8_line_last src1, src2, src3, counter, dst1, dst2, reg1, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+ __neon_eagle2x_8_8_line \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, q11, q12, \alsrc1, \alsrc2, \alsrc3, \aldst1, \aldst2\r
+.endm\r
+\r
+.macro neon_eagle2x_8_8_line part, src1, src2, src3, counter, dst1, dst2, reg1, srcalign16, dstalign32\r
+ .ifeq \srcalign16\r
+\r
+ .ifeq \dstalign32\r
+ _neon_eagle2x_8_8_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1, \src2, \src3, \dst1, \dst2\r
+ .else\r
+ _neon_eagle2x_8_8_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1, \src2, \src3, \dst1:256, \dst2:256\r
+ .endif\r
+\r
+ .else\r
+\r
+ .ifeq \dstalign32\r
+ _neon_eagle2x_8_8_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1:128, \src2:128, \src3:128, \dst1, \dst2\r
+ .else\r
+ _neon_eagle2x_8_8_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1:128, \src2:128, \src3:128, \dst1:256, \dst2:256\r
+ .endif\r
+\r
+ .endif\r
+.endm\r
+\r
+\r
+.macro __neon_eagle2x_16_16_line src1, src2, src3, counter, dst1, dst2, reg1, qT, qY, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vld1.16 {d23[3]}, [\src1] @ S1prev[7] = src[-srcstride]\r
+ .endif\r
+ vld1.16 {d25[3]}, [\src2] @ S2prev[7] = src[0]\r
+ .ifeqs "\qY", "q13"\r
+ vld1.16 {d27[3]}, [\src3] @ S3prev[7] = src[srcstride]\r
+ .endif\r
+ andS \reg1, \counter, #7 @ reg1 = counter & 7\r
+\r
+ .ifnes "\qT", "q11"\r
+ add \src1, \src1, \counter, lsl #1 @ src1 += 2 * counter\r
+ .endif\r
+ .ifnes "\qY", "q13"\r
+ add \src3, \src3, \counter, lsl #1 @ src3 += 2 * counter\r
+ .endif\r
+ beq 1f\r
+\r
+ @ first 1-7 pixels - align counter to 16 bytes\r
+\r
+@ q0 = S1sl < S >\r
+@ q2 = S3sl < X >\r
+@ q7 = tmp2\r
+@ q15 = tmp1\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vld1.16 {q8}, [\src1] @ S1 = [src - srcstride]\r
+ add \src1, \src1, \reg1, lsl #1 @ src1 += 2 * (counter & 7)\r
+ .endif\r
+\r
+ vld1.16 {q9}, [\src2] @ S2 = [src ]\r
+ add \src2, \src2, \reg1, lsl #1 @ src2 += 2 * (counter & 7)\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vld1.16 {q10}, [\src3] @ S3 = [src + srcstride]\r
+ add \src3, \src3, \reg1, lsl #1 @ src3 += 2 * (counter & 7)\r
+ .endif\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q0, \qT, q8, #14 @ S1sl = S1prev[7] | (S1 << 16) < S >\r
+\r
+ vmov \qT, q8 @ S1prev = S1 < T >\r
+ .endif\r
+ vext.8 q1, q12, q9, #14 @ S2sl = S2prev[7] | (S2 << 16) < V >\r
+\r
+ vmov q12, q9 @ S2prev = S2 < C >\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q2, \qY, q10, #14 @ S3sl = S3prev[7] | (S3 << 16) < X >\r
+\r
+ vmov \qY, q10 @ S3prev = S3 < Y >\r
+ .endif\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q3, \qT, q8, #2 @ S1sr = (S1prev >> 16) | ... < U >\r
+ .endif\r
+\r
+ vext.8 q4, q12, q9, #2 @ S2sr = (S2prev >> 16) | ... < W >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q5, \qY, q10, #2 @ S3sr = (S3prev >> 16) | ... < Z >\r
+ .else\r
+ vmov q2, q1 @ S3sl = S2sl < X >\r
+\r
+ vmov q5, q4 @ S3sr = S2sr < Z >\r
+ .endif\r
+\r
+ .ifnes "\qT", "q11"\r
+ vmov q0, q1 @ S1sl = S2sl < S >\r
+\r
+ vmov q3, q4 @ S1sr = S2sr < U >\r
+ .endif\r
+\r
+ vceq.i16 q14, q0, \qT @ E1 = < S == T >\r
+\r
+ vceq.i16 q15, q0, q1 @ tmp1 = < S == V >\r
+\r
+ vceq.i16 q6, q2, \qY @ E3 = < X == Y >\r
+\r
+ vceq.i16 q7, q2, q1 @ tmp2 = < X == V >\r
+\r
+ vand q14, q14, q15 @ E1 = < S == T && S == V >\r
+\r
+@ q0 = tmp3\r
+@ q15 = E2\r
+\r
+ vceq.i16 q15, q3, \qT @ E2 = < U == T >\r
+\r
+ vceq.i16 q0, q3, q4 @ tmp3 = < U == W >\r
+\r
+ vand q6, q6, q7 @ E3 = < X == Y && X == V >\r
+\r
+@ q2 = tmp4\r
+@ q7 = E4\r
+ vceq.i16 q7, q5, \qY @ E4 = < Z == Y >\r
+\r
+ vceq.i16 q2, q5, q4 @ tmp4 = < Z == W >\r
+\r
+ vand q15, q15, q0 @ E2 = < U == T && U == W >\r
+\r
+ vbsl q14, \qT, q12 @ E1 = < (S == T && S == V) ? T : C >\r
+\r
+ vbsl q15, \qT, q12 @ E2 = < (U == T && U == W) ? T : C >\r
+\r
+ vand q7, q7, q2 @ E4 = < Z == Y && Z == W >\r
+\r
+ vbsl q6, \qY, q12 @ E3 = < (X == Y && X == V) ? Y : C >\r
+\r
+ .ifeqs "\qT", "q11"\r
+ sub \reg1, \src1, #2\r
+ .else\r
+ sub \reg1, \src2, #2\r
+ .endif\r
+\r
+ vbsl q7, \qY, q12 @ E4 = < (Z == Y && Z == W) ? Y : C >\r
+ .ifeqs "\qT", "q11"\r
+ vld1.16 {d23[3]}, [\reg1] @ S1prev[7] = src[2 * (counter & 7) - 2 - srcstride]\r
+\r
+ sub \reg1, \src2, #2\r
+ .endif\r
+\r
+ vld1.16 {d25[3]}, [\reg1] @ S2prev[7] = src[2 * (counter & 7) - 2]\r
+\r
+ .ifeqs "\qY", "q13"\r
+ sub \reg1, \src3, #2\r
+\r
+ vld1.16 {d27[3]}, [\reg1] @ S3prev[7] = src[2 * (counter & 7) - 2 + srcstride]\r
+ .endif\r
+\r
+ ubfx \reg1, \counter, #0, #3 @ reg1 = counter & 7\r
+\r
+ lsl \reg1, #2\r
+\r
+ vst2.16 {q14-q15}, [\dst1], \reg1 @ [dst] = E1,E2; dst1 += reg1\r
+\r
+ bic \counter, \counter, #7\r
+\r
+ vst2.16 {q6-q7}, [\dst2], \reg1 @ [dst + dststride] = E3,E4; dst2 += reg1\r
+\r
+ @ counter is aligned to 16 bytes\r
+\r
+ 1:\r
+ .ifeqs "\qT", "q11"\r
+ vld1.16 {q8}, [\alsrc1]! @ S1 = [src - srcstride]; src1 += 2*8\r
+ .endif\r
+ vld1.16 {q9}, [\alsrc2]! @ S2 = [src ]; src2 += 2*8\r
+ .ifeqs "\qY", "q13"\r
+ vld1.16 {q10}, [\alsrc3]! @ S3 = [src + srcstride]; src3 += 2*8\r
+ .endif\r
+\r
+ @ inner loop (8 pixels per iteration)\r
+ 2:\r
+\r
+@ q0 = S1sl < S >\r
+@ q2 = S3sl < X >\r
+@ q7 = tmp2\r
+@ q15 = tmp1\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q0, \qT, q8, #14 @ S1sl = S1prev[7] | (S1 << 16) < S >\r
+ vmov \qT, q8 @ S1prev = S1 < T >\r
+ .endif\r
+\r
+ vext.8 q1, q12, q9, #14 @ S2sl = S2prev[7] | (S2 << 16) < V >\r
+ vmov q12, q9 @ S2prev = S2 < C >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q2, \qY, q10, #14 @ S3sl = S3prev[7] | (S3 << 16) < X >\r
+ vmov \qY, q10 @ S3prev = S3 < Y >\r
+ .endif\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vld1.16 {q8}, [\alsrc1]! @ S1 = [src - srcstride]; src1 += 2*8\r
+ vext.8 q3, \qT, q8, #2 @ S1sr = (S1prev >> 16) | S1[0] < U >\r
+ .endif\r
+\r
+ vld1.16 {q9}, [\alsrc2]! @ S2 = [src ]; src2 += 2*8\r
+ vext.8 q4, q12, q9, #2 @ S2sr = (S2prev >> 16) | S2[0] < W >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vld1.16 {q10}, [\alsrc3]! @ S3 = [src + srcstride]; src3 += 2*8\r
+ vext.8 q5, \qY, q10, #2 @ S3sr = (S3prev >> 16) | S3[0] < Z >\r
+ .else\r
+ vmov q2, q1 @ S3sl = S2sl < X >\r
+\r
+ vmov q5, q4 @ S3sr = S2sr < Z >\r
+ .endif\r
+\r
+ .ifnes "\qT", "q11"\r
+ vmov q0, q1 @ S1sl = S2sl < S >\r
+\r
+ vmov q3, q4 @ S1sr = S2sr < U >\r
+ .endif\r
+\r
+ sub \counter, \counter, #8 @ counter -= 8\r
+ vceq.i16 q14, q0, \qT @ E1 = < S == T >\r
+\r
+ vceq.i16 q15, q0, q1 @ tmp1 = < S == V >\r
+\r
+ vceq.i16 q6, q2, \qY @ E3 = < X == Y >\r
+\r
+ vceq.i16 q7, q2, q1 @ tmp2 = < X == V >\r
+\r
+ vand q14, q14, q15 @ E1 = < S == T && S == V >\r
+\r
+@ q0 = tmp3\r
+@ q15 = E2\r
+\r
+ vceq.i16 q15, q3, \qT @ E2 = < U == T >\r
+\r
+ vceq.i16 q0, q3, q4 @ tmp3 = < U == W >\r
+\r
+ vand q6, q6, q7 @ E3 = < X == Y && X == V >\r
+\r
+@ q2 = tmp4\r
+@ q7 = E4\r
+ vceq.i16 q7, q5, \qY @ E4 = < Z == Y >\r
+\r
+ vceq.i16 q2, q5, q4 @ tmp4 = < Z == W >\r
+\r
+ vand q15, q15, q0 @ E2 = < U == T && U == W >\r
+\r
+ vbsl q14, \qT, q12 @ E1 = < (S == T && S == V) ? T : C >\r
+\r
+ vbsl q15, \qT, q12 @ E2 = < (U == T && U == W) ? T : C >\r
+\r
+ vand q7, q7, q2 @ E4 = < Z == Y && Z == W >\r
+\r
+ vbsl q6, \qY, q12 @ E3 = < (X == Y && X == V) ? Y : C >\r
+\r
+ vbsl q7, \qY, q12 @ E4 = < (Z == Y && Z == W) ? Y : C >\r
+ vst2.16 {q14-q15}, [\aldst1]! @ [dst] = E1,E2; dst1 += 2*2*8\r
+\r
+ cmp \counter, #8\r
+\r
+ vst2.16 {q6-q7}, [\aldst2]! @ [dst + dststride] = E3,E4; dst2 += 2*2*8\r
+ bhi 2b\r
+\r
+ @ last 8 pixels\r
+\r
+@ q0 = S1sl < S >\r
+@ q2 = S3sl < X >\r
+@ q7 = tmp2\r
+@ q15 = tmp1\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q0, \qT, q8, #14 @ S1sl = S1prev[7] | (S1 << 16) < S >\r
+ vmov \qT, q8 @ S1prev = S1 < T >\r
+ .endif\r
+\r
+ vext.8 q1, q12, q9, #14 @ S2sl = S2prev[7] | (S2 << 16) < V >\r
+ vmov q12, q9 @ S2prev = S2 < C >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q2, \qY, q10, #14 @ S3sl = S3prev[7] | (S3 << 16) < X >\r
+ vmov \qY, q10 @ S3prev = S3 < Y >\r
+ .endif\r
+\r
+ .ifeqs "\qT", "q11"\r
+ vshr.u64 d16, d17, #(64-16) @ S1[0] = S1[7] | ...\r
+ .endif\r
+\r
+ vshr.u64 d18, d19, #(64-16) @ S2[0] = S2[7] | ...\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vshr.u64 d20, d21, #(64-16) @ S3[0] = S3[7] | ...\r
+ .endif\r
+ .ifeqs "\qT", "q11"\r
+ vext.8 q3, \qT, q8, #2 @ S1sr = (S1prev >> 16) | S1[0] < U >\r
+ .endif\r
+\r
+ vext.8 q4, q12, q9, #2 @ S2sr = (S2prev >> 16) | S2[0] < W >\r
+\r
+ .ifeqs "\qY", "q13"\r
+ vext.8 q5, \qY, q10, #2 @ S3sr = (S3prev >> 16) | S3[0] < Z >\r
+ .else\r
+ vmov q2, q1 @ S3sl = S2sl < X >\r
+\r
+ vmov q5, q4 @ S3sr = S2sr < Z >\r
+ .endif\r
+\r
+ .ifnes "\qT", "q11"\r
+ vmov q0, q1 @ S1sl = S2sl < S >\r
+\r
+ vmov q3, q4 @ S1sr = S2sr < U >\r
+ .endif\r
+\r
+ vceq.i16 q14, q0, \qT @ E1 = < S == T >\r
+\r
+ vceq.i16 q15, q0, q1 @ tmp1 = < S == V >\r
+\r
+ vceq.i16 q6, q2, \qY @ E3 = < X == Y >\r
+\r
+ vceq.i16 q7, q2, q1 @ tmp2 = < X == V >\r
+\r
+ vand q14, q14, q15 @ E1 = < S == T && S == V >\r
+\r
+@ q0 = tmp3\r
+@ q15 = E2\r
+\r
+ vceq.i16 q15, q3, \qT @ E2 = < U == T >\r
+\r
+ vceq.i16 q0, q3, q4 @ tmp3 = < U == W >\r
+\r
+ vand q6, q6, q7 @ E3 = < X == Y && X == V >\r
+\r
+@ q2 = tmp4\r
+@ q7 = E4\r
+ vceq.i16 q7, q5, \qY @ E4 = < Z == Y >\r
+\r
+ vceq.i16 q2, q5, q4 @ tmp4 = < Z == W >\r
+\r
+ vand q15, q15, q0 @ E2 = < U == T && U == W >\r
+\r
+ vbsl q14, \qT, q12 @ E1 = < (S == T && S == V) ? T : C >\r
+\r
+ vbsl q15, \qT, q12 @ E2 = < (U == T && U == W) ? T : C >\r
+\r
+ vand q7, q7, q2 @ E4 = < Z == Y && Z == W >\r
+\r
+ vbsl q6, \qY, q12 @ E3 = < (X == Y && X == V) ? Y : C >\r
+\r
+ vbsl q7, \qY, q12 @ E4 = < (Z == Y && Z == W) ? Y : C >\r
+ vst2.16 {q14-q15}, [\aldst1]! @ [dst] = E1,E2; dst1 += 2*2*8\r
+\r
+ vst2.16 {q6-q7}, [\aldst2]! @ [dst + dststride] = E3,E4; dst2 += 2*2*8\r
+\r
+.endm\r
+\r
+.macro _neon_eagle2x_16_16_line_first src1, src2, src3, counter, dst1, dst2, reg1, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+ __neon_eagle2x_16_16_line \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, q12, q13, \alsrc1, \alsrc2, \alsrc3, \aldst1, \aldst2\r
+.endm\r
+\r
+.macro _neon_eagle2x_16_16_line_middle src1, src2, src3, counter, dst1, dst2, reg1, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+ __neon_eagle2x_16_16_line \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, q11, q13, \alsrc1, \alsrc2, \alsrc3, \aldst1, \aldst2\r
+.endm\r
+\r
+.macro _neon_eagle2x_16_16_line_last src1, src2, src3, counter, dst1, dst2, reg1, alsrc1, alsrc2, alsrc3, aldst1, aldst2\r
+ __neon_eagle2x_16_16_line \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, q11, q12, \alsrc1, \alsrc2, \alsrc3, \aldst1, \aldst2\r
+.endm\r
+\r
+.macro neon_eagle2x_16_16_line part, src1, src2, src3, counter, dst1, dst2, reg1, srcalign16, dstalign32\r
+ .ifeq \srcalign16\r
+\r
+ .ifeq \dstalign32\r
+ _neon_eagle2x_16_16_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1, \src2, \src3, \dst1, \dst2\r
+ .else\r
+ _neon_eagle2x_16_16_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1, \src2, \src3, \dst1:256, \dst2:256\r
+ .endif\r
+\r
+ .else\r
+\r
+ .ifeq \dstalign32\r
+ _neon_eagle2x_16_16_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1:128, \src2:128, \src3:128, \dst1, \dst2\r
+ .else\r
+ _neon_eagle2x_16_16_line_\part \src1, \src2, \src3, \counter, \dst1, \dst2, \reg1, \src1:128, \src2:128, \src3:128, \dst1:256, \dst2:256\r
+ .endif\r
+\r
+ .endif\r
+.endm\r
+\r
notaz.gp2x.de / gpsp.git / blobdiff