import 15.2.0 Octopus source

[ceph.git] / ceph / src / spdk / intel-ipsec-mb / avx / sha256_one_block_avx.asm

diff --git a/ceph/src/spdk/intel-ipsec-mb/avx/sha256_one_block_avx.asm b/ceph/src/spdk/intel-ipsec-mb/avx/sha256_one_block_avx.asm
index b92e1f6c46e1217c93e55232c064876dd623d745..040e66e817b844632e1874e73c1963e9237c702c 100644 (file)
--- a/ceph/src/spdk/intel-ipsec-mb/avx/sha256_one_block_avx.asm
+++ b/ceph/src/spdk/intel-ipsec-mb/avx/sha256_one_block_avx.asm
@@ -135,16 +135,8 @@ _XMM_SAVE: reso    7
 _XFER:         reso    1
 endstruc
 
-%ifndef H0
-%define H0 0x6a09e667
-%define H1 0xbb67ae85
-%define H2 0x3c6ef372
-%define H3 0xa54ff53a
-%define H4 0x510e527f
-%define H5 0x9b05688c
-%define H6 0x1f83d9ab
-%define H7 0x5be0cd19
-%define FUNC sha256_one_block_avx
+%ifndef FUNC
+%define FUNC sha256_block_avx
 %endif
 
 ; rotate_Xs
@@ -440,14 +432,14 @@ FUNC:
 %endif
 
        ;; load initial digest
-       mov     a,H0
-       mov     b,H1
-       mov     c,H2
-       mov     d,H3
-       mov     e,H4
-       mov     f,H5
-       mov     g,H6
-       mov     h,H7
+       mov     a, [4*0 + CTX]
+       mov     b, [4*1 + CTX]
+       mov     c, [4*2 + CTX]
+       mov     d, [4*3 + CTX]
+       mov     e, [4*4 + CTX]
+       mov     f, [4*5 + CTX]
+       mov     g, [4*6 + CTX]
+       mov     h, [4*7 + CTX]
 
        vmovdqa BYTE_FLIP_MASK, [rel PSHUFFLE_BYTE_FLIP_MASK]
        vmovdqa SHUF_00BA, [rel _SHUF_00BA]
@@ -508,22 +500,14 @@ loop2:
        sub     SRND, 1
        jne     loop2
 
-       add     a,H0
-       add     b,H1
-       add     c,H2
-       add     d,H3
-       add     e,H4
-       add     f,H5
-       add     g,H6
-       mov     [4*0 + CTX],a
-       mov     [4*1 + CTX],b
-       mov     [4*2 + CTX],c
-       mov     [4*3 + CTX],d
-       mov     [4*4 + CTX],e
-       mov     [4*5 + CTX],f
-       mov     [4*6 + CTX],g
-       add     h,H7
-       mov     [4*7 + CTX],h
+       add     [4*0 + CTX], a
+       add     [4*1 + CTX], b
+       add     [4*2 + CTX], c
+       add     [4*3 + CTX], d
+       add     [4*4 + CTX], e
+       add     [4*5 + CTX], f
+       add     [4*6 + CTX], g
+       add     [4*7 + CTX], h
 
 done_hash:
 %ifndef LINUX