bsaes-armv7.pl: take it into build loop.

Configure

@@ -136,7 +136,7 @@ my $alpha_asm="alphacpuid.o:bn_asm.o alpha-mont.o:::::sha1-alpha.o:::::::ghash-a
 my $mips64_asm=":bn-mips.o mips-mont.o::aes_cbc.o aes-mips.o:::sha1-mips.o sha256-mips.o sha512-mips.o::::::::";
 my $mips32_asm=$mips64_asm; $mips32_asm =~ s/\s*sha512\-mips\.o//;
 my $s390x_asm="s390xcap.o s390xcpuid.o:bn-s390x.o s390x-mont.o s390x-gf2m.o::aes-s390x.o aes-ctr.o aes-xts.o:::sha1-s390x.o sha256-s390x.o sha512-s390x.o::rc4-s390x.o:::::ghash-s390x.o:";
-my $armv4_asm="armcap.o armv4cpuid.o:bn_asm.o armv4-mont.o armv4-gf2m.o::aes_cbc.o aes-armv4.o:::sha1-armv4-large.o sha256-armv4.o sha512-armv4.o:::::::ghash-armv4.o::void";
+my $armv4_asm="armcap.o armv4cpuid.o:bn_asm.o armv4-mont.o armv4-gf2m.o::aes_cbc.o aes-armv4.o bsaes-armv7.o:::sha1-armv4-large.o sha256-armv4.o sha512-armv4.o:::::::ghash-armv4.o::void";
 my $parisc11_asm="pariscid.o:bn_asm.o parisc-mont.o::aes_core.o aes_cbc.o aes-parisc.o:::sha1-parisc.o sha256-parisc.o sha512-parisc.o::rc4-parisc.o:::::ghash-parisc.o::32";
 my $parisc20_asm="pariscid.o:pa-risc2W.o parisc-mont.o::aes_core.o aes_cbc.o aes-parisc.o:::sha1-parisc.o sha256-parisc.o sha512-parisc.o::rc4-parisc.o:::::ghash-parisc.o::64";
 my $ppc64_asm="ppccpuid.o ppccap.o:bn-ppc.o ppc-mont.o ppc64-mont.o::aes_core.o aes_cbc.o aes-ppc.o:::sha1-ppc.o sha256-ppc.o sha512-ppc.o::::::::";

TABLE

@@ -1067,7 +1067,7 @@ $bn_ops       = BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR
 $cpuid_obj    = armcap.o armv4cpuid.o
 $bn_obj       = bn_asm.o armv4-mont.o armv4-gf2m.o
 $des_obj      = 
-$aes_obj      = aes_cbc.o aes-armv4.o
+$aes_obj      = aes_cbc.o aes-armv4.o bsaes-armv7.o
 $bf_obj       = 
 $md5_obj      = 
 $sha1_obj     = sha1-armv4-large.o sha256-armv4.o sha512-armv4.o
@@ -4004,7 +4004,7 @@ $bn_ops       = BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_UNROLL BF_PTR
 $cpuid_obj    = armcap.o armv4cpuid.o
 $bn_obj       = bn_asm.o armv4-mont.o armv4-gf2m.o
 $des_obj      = 
-$aes_obj      = aes_cbc.o aes-armv4.o
+$aes_obj      = aes_cbc.o aes-armv4.o bsaes-armv7.o
 $bf_obj       = 
 $md5_obj      = 
 $sha1_obj     = sha1-armv4-large.o sha256-armv4.o sha512-armv4.o

crypto/aes/Makefile

@@ -83,6 +83,8 @@ aes-mips.S:	asm/aes-mips.pl
 # GNU make "catch all"
 aes-%.S:	asm/aes-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
 aes-armv4.o:	aes-armv4.S
+bsaes-%.S:	asm/bsaes-%.pl;	$(PERL) $< $(PERLASM_SCHEME) $@
+bsaes-armv7.o:	bsaes-armv7.S
 
 files:
 	$(PERL) $(TOP)/util/files.pl "AES_ENC=$(AES_ENC)" Makefile >> $(TOP)/MINFO

crypto/aes/asm/bsaes-armv7.pl

@@ -26,6 +26,9 @@
 # which is [much] worse than anticipated (for further details see
 # http://www.openssl.org/~appro/Snapdragon-S4.html).
 #
+# Cortex-A15 manages in 14.2/19.6 cycles [when integer-only code
+# manages in 20.0 cycles].
+#
 # When comparing to x86_64 results keep in mind that NEON unit is
 # [mostly] single-issue and thus can't [fully] benefit from
 # instruction-level parallelism. And when comparing to aes-armv4
@@ -617,6 +620,9 @@ ___
 }
 
 $code.=<<___;
+#include "arm_arch.h"
+
+#if __ARM_ARCH__>=7
 .text
 .code	32
 .fpu	neon
@@ -975,6 +981,9 @@ bsaes_decrypt_128:
 .size	bsaes_decrypt_128,.-bsaes_decrypt_128
 ___
 }
+$code.=<<___;
+#endif
+___
 
 $code =~ s/\`([^\`]*)\`/eval($1)/gem;