diff options
author | Marcel Ziswiler <marcel.ziswiler@toradex.com> | 2020-05-21 00:54:36 +0200 |
---|---|---|
committer | Marcel Ziswiler <marcel.ziswiler@toradex.com> | 2020-05-21 00:54:36 +0200 |
commit | 135d39dcedcedd1f44ea0bba52f15ac5922c114f (patch) | |
tree | 12379fc8ef7489eaca1c7245f8bd6af74c619a8f /arch/x86/crypto/poly1305-avx2-x86_64.S | |
parent | 187764bd111b27783b6d68ffb3b3dbb3a9bafd38 (diff) | |
parent | 1279cd128bba968ebe0a2df7f7ae38bae90250ef (diff) |
Merge remote-tracking branch 'remotes/fslc/4.9-2.3.x-imx' into toradex_4.9-2.3.x-imx-next
Conflicts:
sound/soc/codecs/sgtl5000.c
sound/soc/fsl/imx-sgtl5000.c
Diffstat (limited to 'arch/x86/crypto/poly1305-avx2-x86_64.S')
-rw-r--r-- | arch/x86/crypto/poly1305-avx2-x86_64.S | 14 |
1 files changed, 10 insertions, 4 deletions
diff --git a/arch/x86/crypto/poly1305-avx2-x86_64.S b/arch/x86/crypto/poly1305-avx2-x86_64.S index eff2f414e22b..ec234c43b3f4 100644 --- a/arch/x86/crypto/poly1305-avx2-x86_64.S +++ b/arch/x86/crypto/poly1305-avx2-x86_64.S @@ -321,6 +321,12 @@ ENTRY(poly1305_4block_avx2) vpaddq t2,t1,t1 vmovq t1x,d4 + # Now do a partial reduction mod (2^130)-5, carrying h0 -> h1 -> h2 -> + # h3 -> h4 -> h0 -> h1 to get h0,h2,h3,h4 < 2^26 and h1 < 2^26 + a small + # amount. Careful: we must not assume the carry bits 'd0 >> 26', + # 'd1 >> 26', 'd2 >> 26', 'd3 >> 26', and '(d4 >> 26) * 5' fit in 32-bit + # integers. It's true in a single-block implementation, but not here. + # d1 += d0 >> 26 mov d0,%rax shr $26,%rax @@ -359,16 +365,16 @@ ENTRY(poly1305_4block_avx2) # h0 += (d4 >> 26) * 5 mov d4,%rax shr $26,%rax - lea (%eax,%eax,4),%eax - add %eax,%ebx + lea (%rax,%rax,4),%rax + add %rax,%rbx # h4 = d4 & 0x3ffffff mov d4,%rax and $0x3ffffff,%eax mov %eax,h4 # h1 += h0 >> 26 - mov %ebx,%eax - shr $26,%eax + mov %rbx,%rax + shr $26,%rax add %eax,h1 # h0 = h0 & 0x3ffffff andl $0x3ffffff,%ebx |