150     __m128i aa = _mm_slli_epi64(xx, 63);          // x1<<63:x0<<63 = stuff:a  in gcm_reduce()
151     __m128i bb = _mm_slli_epi64(xx, 62);          // x1<<62:x0<<62 = stuff:b  in gcm_reduce()
152     __m128i cc = _mm_slli_epi64(xx, 57);          // x1<<57:x0<<57 = stuff:c  in gcm_reduce()
166     __m128i eh = _mm_slli_epi64(dx, 63);          // d<<63:stuff  in gcm_mix()
167     __m128i fh = _mm_slli_epi64(dx, 62);          // d<<62:stuff  in gcm_mix()
168     __m128i gh = _mm_slli_epi64(dx, 57);          // d<<57:stuff  in gcm_mix()
229      * On entry state is r3:r2:r1:r0 and xword is X:stuff:stuff:stuff  in aesni_set_rk_128()
275      * On entry state0 is r3:r2:r1:r0, state1 is stuff:stuff:r5:r4  in aesni_set_rk_192()
276      * and xword is stuff:stuff:X:stuff with X = rot( sub( r3 ) ) ^ RCON  in aesni_set_rk_192()
279      * On exit, state0 is r9:r8:r7:r6 and state1 is stuff:stuff:r11:r10  in aesni_set_rk_192()
293     xword = _mm_xor_si128(xword, *state1);    // stuff:stuff:r9+r5:r9+r4  in aesni_set_rk_192()
294     *state1 = _mm_slli_si128(*state1, 4);     // stuff:stuff:r4:0  in aesni_set_rk_192()
295     xword = _mm_xor_si128(xword, *state1);    // stuff:stuff:r9+r5+r4:r9+r4  in aesni_set_rk_192()
296     *state1 = xword;                          // = stuff:stuff:r11:r10  in aesni_set_rk_192()
336      * xword is X:stuff:stuff:stuff with X = rot( sub( r7 )) ^ RCON  in aesni_set_rk_256()
351     /* Set xword to stuff:Y:stuff:stuff with Y = subword( r11 )  in aesni_set_rk_256()
532          "psllq $63, %%xmm3                 \n\t" // x1<<63:x0<<63 = stuff:a  in mbedtls_aesni_gcm_mult()
533          "psllq $62, %%xmm4                 \n\t" // x1<<62:x0<<62 = stuff:b  in mbedtls_aesni_gcm_mult()
534          "psllq $57, %%xmm5                 \n\t" // x1<<57:x0<<57 = stuff:c  in mbedtls_aesni_gcm_mult()
537          "pxor %%xmm4, %%xmm3               \n\t" // stuff:a+b  in mbedtls_aesni_gcm_mult()
538          "pxor %%xmm5, %%xmm3               \n\t" // stuff:a+b+c  in mbedtls_aesni_gcm_mult()
556          "psllq $63, %%xmm3                 \n\t" // d<<63:stuff  in mbedtls_aesni_gcm_mult()
557          "psllq $62, %%xmm4                 \n\t" // d<<62:stuff  in mbedtls_aesni_gcm_mult()
558          "psllq $57, %%xmm5                 \n\t" // d<<57:stuff  in mbedtls_aesni_gcm_mult()
559          "pxor %%xmm4, %%xmm3               \n\t" // d<<63+d<<62:stuff  in mbedtls_aesni_gcm_mult()
560          "pxor %%xmm5, %%xmm3               \n\t" // missing bits of d:stuff  in mbedtls_aesni_gcm_mult()
615           * On entry xmm0 is r3:r2:r1:r0 and xmm1 is X:stuff:stuff:stuff  in aesni_setkey_enc_128()
670           * On entry xmm0 is r3:r2:r1:r0, xmm1 is stuff:stuff:r5:r4  in aesni_setkey_enc_192()
671           * and xmm2 is stuff:stuff:X:stuff with X = rot( sub( r3 ) ) ^ RCON.  in aesni_setkey_enc_192()
673           * On exit, xmm0 is r9:r8:r7:r6 and xmm1 is stuff:stuff:r11:r10  in aesni_setkey_enc_192()
688          "pxor %%xmm1, %%xmm2           \n\t" // stuff:stuff:r9+r5:r10  in aesni_setkey_enc_192()
690          "pxor %%xmm2, %%xmm1           \n\t" // xmm1 = stuff:stuff:r11:r10  in aesni_setkey_enc_192()
729           * xmm2 is X:stuff:stuff:stuff with X = rot( sub( r7 )) ^ RCON  in aesni_setkey_enc_256()
746          /* Set xmm2 to stuff:Y:stuff:stuff with Y = subword( r11 )  in aesni_setkey_enc_256()