Lines Matching full:w
165 # Extract w[t-7]
166 MY_VPALIGNR YTMP0, Y_3, Y_2, 8 # YTMP0 = W[-7]
167 # Calculate w[t-16] + w[t-7]
168 vpaddq Y_0, YTMP0, YTMP0 # YTMP0 = W[-7] + W[-16]
169 # Extract w[t-15]
170 MY_VPALIGNR YTMP1, Y_1, Y_0, 8 # YTMP1 = W[-15]
174 # Calculate w[t-15] ror 1
177 vpor YTMP2, YTMP3, YTMP3 # YTMP3 = W[-15] ror 1
178 # Calculate w[t-15] shr 7
179 vpsrlq $7, YTMP1, YTMP4 # YTMP4 = W[-15] >> 7
184 add frame_XFER(%rsp),h # h = k + w + h # --
196 add h, d # d = k + w + h + d # --
209 add y1, h # h = k + w + h + S0 # --
211 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
213 add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
220 # Calculate w[t-15] ror 8
223 vpor YTMP2, YTMP1, YTMP1 # YTMP1 = W[-15] ror 8
225 vpxor YTMP4, YTMP3, YTMP3 # YTMP3 = W[-15] ror 1 ^ W[-15] >> 7
229 # Add three components, w[t-16], w[t-7] and sigma0
230 vpaddq YTMP1, YTMP0, YTMP0 # YTMP0 = W[-16] + W[-7] + s0
231 # Move to appropriate lanes for calculating w[16] and w[17]
232 vperm2f128 $0x0, YTMP0, YTMP0, Y_0 # Y_0 = W[-16] + W[-7] + s0 {BABA}
233 # Move to appropriate lanes for calculating w[18] and w[19]
234 vpand MASK_YMM_LO(%rip), YTMP0, YTMP0 # YTMP0 = W[-16] + W[-7] + s0 {DC00}
236 # Calculate w[16] and w[17] in both 128 bit lanes
238 # Calculate sigma1 for w[16] and w[17] on both 128 bit lanes
239 vperm2f128 $0x11, Y_3, Y_3, YTMP2 # YTMP2 = W[-2] {BABA}
240 vpsrlq $6, YTMP2, YTMP4 # YTMP4 = W[-2] >> 6 {BABA}
246 add 1*8+frame_XFER(%rsp), h # h = k + w + h # --
260 add h, d # d = k + w + h + d # --
274 add y1, h # h = k + w + h + S0 # --
276 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
277 add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
285 vpsrlq $19, YTMP2, YTMP3 # YTMP3 = W[-2] >> 19 {BABA}
286 vpsllq $(64-19), YTMP2, YTMP1 # YTMP1 = W[-2] << 19 {BABA}
287 vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 19 {BABA}
288 vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = W[-2] ror 19 ^ W[-2] >> 6 {BABA}
289 vpsrlq $61, YTMP2, YTMP3 # YTMP3 = W[-2] >> 61 {BABA}
290 vpsllq $(64-61), YTMP2, YTMP1 # YTMP1 = W[-2] << 61 {BABA}
291 vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 61 {BABA}
292 vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = s1 = (W[-2] ror 19) ^
293 # (W[-2] ror 61) ^ (W[-2] >> 6) {BABA}
295 # Add sigma1 to the other compunents to get w[16] and w[17]
296 vpaddq YTMP4, Y_0, Y_0 # Y_0 = {W[1], W[0], W[1], W[0]}
298 # Calculate sigma1 for w[18] and w[19] for upper 128 bit lane
299 vpsrlq $6, Y_0, YTMP4 # YTMP4 = W[-2] >> 6 {DC--}
303 add 2*8+frame_XFER(%rsp), h # h = k + w + h # --
315 add h, d # d = k + w + h + d # --
331 add y1, h # h = k + w + h + S0 # --
332 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
333 add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
341 vpsrlq $19, Y_0, YTMP3 # YTMP3 = W[-2] >> 19 {DC--}
342 vpsllq $(64-19), Y_0, YTMP1 # YTMP1 = W[-2] << 19 {DC--}
343 vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 19 {DC--}
344 vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = W[-2] ror 19 ^ W[-2] >> 6 {DC--}
345 vpsrlq $61, Y_0, YTMP3 # YTMP3 = W[-2] >> 61 {DC--}
346 vpsllq $(64-61), Y_0, YTMP1 # YTMP1 = W[-2] << 61 {DC--}
347 vpor YTMP1, YTMP3, YTMP3 # YTMP3 = W[-2] ror 61 {DC--}
348 vpxor YTMP3, YTMP4, YTMP4 # YTMP4 = s1 = (W[-2] ror 19) ^
349 # (W[-2] ror 61) ^ (W[-2] >> 6) {DC--}
351 # Add the sigma0 + w[t-7] + w[t-16] for w[18] and w[19]
352 # to newly calculated sigma1 to get w[18] and w[19]
353 vpaddq YTMP4, YTMP0, YTMP2 # YTMP2 = {W[3], W[2], --, --}
355 # Form w[19, w[18], w17], w[16]
356 vpblendd $0xF0, YTMP2, Y_0, Y_0 # Y_0 = {W[3], W[2], W[1], W[0]}
361 add 3*8+frame_XFER(%rsp), h # h = k + w + h # --
373 add h, d # d = k + w + h + d # --
383 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
392 add y1, h # h = k + w + h + S0 # --
393 add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
422 add frame_XFER(%rsp), h # h = k + w + h # --
431 add h, d # d = k + w + h + d # --
433 add y1, h # h = k + w + h + S0 # --
435 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
441 add y2, old_h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
460 add 8*1+frame_XFER(%rsp), h # h = k + w + h # --
469 add h, d # d = k + w + h + d # --
471 add y1, h # h = k + w + h + S0 # --
473 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
479 add y2, old_h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
498 add 8*2+frame_XFER(%rsp), h # h = k + w + h # --
507 add h, d # d = k + w + h + d # --
509 add y1, h # h = k + w + h + S0 # --
511 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
517 add y2, old_h # h = k + w + h + S0 + S1 + CH = t1 + S0# --
536 add 8*3+frame_XFER(%rsp), h # h = k + w + h # --
546 add h, d # d = k + w + h + d # --
548 add y1, h # h = k + w + h + S0 # --
550 add y2, d # d = k + w + h + d + S1 + CH = d + t1 # --
552 add y2, h # h = k + w + h + S0 + S1 + CH = t1 + S0# --