2 * Buffer submit code for multi buffer SHA1 algorithm
4 * This file is provided under a dual BSD/GPLv2 license. When using or
5 * redistributing this file, you may do so under either license.
9 * Copyright(c) 2014 Intel Corporation.
11 * This program is free software; you can redistribute it and/or modify
12 * it under the terms of version 2 of the GNU General Public License as
13 * published by the Free Software Foundation.
15 * This program is distributed in the hope that it will be useful, but
16 * WITHOUT ANY WARRANTY; without even the implied warranty of
17 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
18 * General Public License for more details.
20 * Contact Information:
21 * James Guilford <james.guilford@intel.com>
22 * Tim Chen <tim.c.chen@linux.intel.com>
26 * Copyright(c) 2014 Intel Corporation.
28 * Redistribution and use in source and binary forms, with or without
29 * modification, are permitted provided that the following conditions
32 * * Redistributions of source code must retain the above copyright
33 * notice, this list of conditions and the following disclaimer.
34 * * Redistributions in binary form must reproduce the above copyright
35 * notice, this list of conditions and the following disclaimer in
36 * the documentation and/or other materials provided with the
38 * * Neither the name of Intel Corporation nor the names of its
39 * contributors may be used to endorse or promote products derived
40 * from this software without specific prior written permission.
42 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
43 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
44 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
45 * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
46 * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
47 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
48 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
49 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
50 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
51 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
52 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
55 #include <linux/linkage.h>
56 #include <asm/frame.h>
57 #include "sha1_mb_mgr_datastruct.S"
62 # LINUX register definitions
75 # idx must be a register not clobberred by sha1_x8_avx2
84 BYTE_unused_lanes = %bl
98 # JOB* submit_mb_mgr_submit_avx2(MB_MGR *state, job_sha1 *job)
101 ENTRY(sha1_mb_mgr_submit_avx2)
106 mov _unused_lanes(state), unused_lanes
107 mov unused_lanes, lane
110 imul $_LANE_DATA_size, lane, lane_data
111 movl $STS_BEING_PROCESSED, _status(job)
112 lea _ldata(state, lane_data), lane_data
113 mov unused_lanes, _unused_lanes(state)
114 movl _len(job), DWORD_len
116 mov job, _job_in_lane(lane_data)
120 movl DWORD_len, _lens(state , lane, 4)
122 # Load digest words from result_digest
123 vmovdqu _result_digest(job), %xmm0
124 mov _result_digest+1*16(job), DWORD_tmp
125 vmovd %xmm0, _args_digest(state, lane, 4)
126 vpextrd $1, %xmm0, _args_digest+1*32(state , lane, 4)
127 vpextrd $2, %xmm0, _args_digest+2*32(state , lane, 4)
128 vpextrd $3, %xmm0, _args_digest+3*32(state , lane, 4)
129 movl DWORD_tmp, _args_digest+4*32(state , lane, 4)
132 mov p, _args_data_ptr(state, lane, 8)
134 cmp $0xF, unused_lanes
139 vmovdqa _lens(state), %xmm0
140 vmovdqa _lens+1*16(state), %xmm1
142 vpminud %xmm1, %xmm0, %xmm2 # xmm2 has {D,C,B,A}
143 vpalignr $8, %xmm2, %xmm3, %xmm3 # xmm3 has {x,x,D,C}
144 vpminud %xmm3, %xmm2, %xmm2 # xmm2 has {x,x,E,F}
145 vpalignr $4, %xmm2, %xmm3, %xmm3 # xmm3 has {x,x,x,E}
146 vpminud %xmm3, %xmm2, %xmm2 # xmm2 has min value in low dword
148 vmovd %xmm2, DWORD_idx
154 vpand clear_low_nibble(%rip), %xmm2, %xmm2
155 vpshufd $0, %xmm2, %xmm2
157 vpsubd %xmm2, %xmm0, %xmm0
158 vpsubd %xmm2, %xmm1, %xmm1
160 vmovdqa %xmm0, _lens + 0*16(state)
161 vmovdqa %xmm1, _lens + 1*16(state)
164 # "state" and "args" are the same address, arg1
168 # state and idx are intact
171 # process completed job "idx"
172 imul $_LANE_DATA_size, idx, lane_data
173 lea _ldata(state, lane_data), lane_data
175 mov _job_in_lane(lane_data), job_rax
176 mov _unused_lanes(state), unused_lanes
177 movq $0, _job_in_lane(lane_data)
178 movl $STS_COMPLETED, _status(job_rax)
181 mov unused_lanes, _unused_lanes(state)
183 movl $0xFFFFFFFF, _lens(state, idx, 4)
185 vmovd _args_digest(state, idx, 4), %xmm0
186 vpinsrd $1, _args_digest+1*32(state , idx, 4), %xmm0, %xmm0
187 vpinsrd $2, _args_digest+2*32(state , idx, 4), %xmm0, %xmm0
188 vpinsrd $3, _args_digest+3*32(state , idx, 4), %xmm0, %xmm0
189 movl _args_digest+4*32(state, idx, 4), DWORD_tmp
191 vmovdqu %xmm0, _result_digest(job_rax)
192 movl DWORD_tmp, _result_digest+1*16(job_rax)
204 ENDPROC(sha1_mb_mgr_submit_avx2)
206 .section .rodata.cst16.clear_low_nibble, "aM", @progbits, 16
209 .octa 0x000000000000000000000000FFFFFFF0