Unified - adapt the generation of modes assembler to use GENERATE

[openssl.git] / crypto / modes / asm / aesni-gcm-x86_64.pl

diff --git a/crypto/modes/asm/aesni-gcm-x86_64.pl b/crypto/modes/asm/aesni-gcm-x86_64.pl
index 1ac6b5b8458a8a5b164661683fa4e03bfc2aa65c..8f5026b3a40fb9b41c7c8cc905aa1b62fca9f6de 100644 (file)
--- a/crypto/modes/asm/aesni-gcm-x86_64.pl
+++ b/crypto/modes/asm/aesni-gcm-x86_64.pl
@@ -22,7 +22,11 @@
 # [1] and [2], with MOVBE twist suggested by Ilya Albrekht and Max
 # Locktyukhin of Intel Corp. who verified that it reduces shuffles
 # pressure with notable relative improvement, achieving 1.0 cycle per
-# byte processed with 128-bit key on Haswell processor.
+# byte processed with 128-bit key on Haswell processor, 0.74 - on
+# Broadwell, 0.63 - on Skylake... [Mentioned results are raw profiled
+# measurements for favourable packet size, one divisible by 96.
+# Applications using the EVP interface will observe a few percent
+# worse performance.]
 #
 # [1] http://rt.openssl.org/Ticket/Display.html?id=2900&user=guest&pass=guest
 # [2] http://www.intel.com/content/dam/www/public/us/en/documents/software-support/enabling-high-performance-gcm.pdf
@@ -40,7 +44,7 @@ die "can't locate x86_64-xlate.pl";
 
 if (`$ENV{CC} -Wa,-v -c -o /dev/null -x assembler /dev/null 2>&1`
                =~ /GNU assembler version ([2-9]\.[0-9]+)/) {
-       $avx = ($1>=2.19) + ($1>=2.22);
+       $avx = ($1>=2.20) + ($1>=2.22);
 }
 
 if (!$avx && $win64 && ($flavour =~ /nasm/ || $ENV{ASM} =~ /nasm/) &&
@@ -53,8 +57,8 @@ if (!$avx && $win64 && ($flavour =~ /masm/ || $ENV{ASM} =~ /ml64/) &&
        $avx = ($1>=10) + ($1>=11);
 }
 
-if (!$avx && `$ENV{CC} -v` =~ /LLVM ([3-9]\.[0-9]+)/) {
-       $avx = ($1>=3.0) + ($1>=3.1);
+if (!$avx && `$ENV{CC} -v 2>&1` =~ /((?:^clang|LLVM) version|.*based on LLVM) ([3-9]\.[0-9]+)/) {
+       $avx = ($2>=3.0) + ($2>3.0);
 }
 
 open OUT,"| \"$^X\" $xlate $flavour $output";
@@ -92,7 +96,7 @@ _aesni_ctr32_ghash_6x:
 
 .align 32
 .Loop6x:
-       add             \$6<<24,$counter
+       add             \$`6<<24`,$counter
        jc              .Lhandle_ctr32          # discard $inout[1-5]?
        vmovdqu         0x00-0x20($Xip),$Hkey   # $Hkey^1
          vpaddb        $T2,$inout5,$T1         # next counter value
@@ -486,7 +490,7 @@ $code.=<<___;
 ___
 $code.=<<___ if ($win64);
        movaps  -0xd8(%rax),%xmm6
-       movaps  -0xd8(%rax),%xmm7
+       movaps  -0xc8(%rax),%xmm7
        movaps  -0xb8(%rax),%xmm8
        movaps  -0xa8(%rax),%xmm9
        movaps  -0x98(%rax),%xmm10
@@ -520,7 +524,7 @@ _aesni_ctr32_6x:
        vmovups         0x10-0x80($key),$rndkey
        lea             0x20-0x80($key),%r12
        vpxor           $Z0,$T1,$inout0
-       add             \$6<<24,$counter
+       add             \$`6<<24`,$counter
        jc              .Lhandle_ctr32_2
        vpaddb          $T2,$T1,$inout1
        vpaddb          $T2,$inout1,$inout2