Switch 64-bit sparcv9 platforms from bn(64,64) to bn(64,32). This doesn't
have impact on performance, because amount of multiplications does not
increase with this switch, not on sparcv9 that is. On the contrary, it
actually improves performance, because it spares a load of instructions
used to chase carries. Not to mention that BN assembler modules can be
shared more freely between 32- and 64-bit builts.

TABLE

diff --git a/TABLE b/TABLE
index 1c59c9c..f2acd14 100644
--- a/TABLE
+++ b/TABLE
@@ -142,7 +142,7 @@
 $thread_cflag = -pthread -D_THREAD_SAFE -D_REENTRANT
 $sys_id       = 
 $lflags       = 
-$bn_ops       = SIXTY_FOUR_BIT_LONG RC2_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC2 BF_PTR
+$bn_ops       = BN_LLONG RC2_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC2 BF_PTR
 $cpuid_obj    = 
 $bn_obj       = bn_asm.o sparcv9a-mont.o
 $des_obj      = des_enc-sparc.o fcrypt_b.o
@@ -2923,7 +2923,7 @@
 $thread_cflag = -D_REENTRANT
 $sys_id       = ULTRASPARC
 $lflags       = -ldl
-$bn_ops       = SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR
+$bn_ops       = BN_LLONG RC4_CHAR RC4_CHUNK DES_UNROLL BF_PTR
 $cpuid_obj    = 
 $bn_obj       = bn_asm.o sparcv9a-mont.o
 $des_obj      = des_enc-sparc.o fcrypt_b.o
@@ -3625,7 +3625,7 @@
 $thread_cflag = -D_REENTRANT
 $sys_id       = ULTRASPARC
 $lflags       = -lsocket -lnsl -ldl
-$bn_ops       = SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL BF_PTR
+$bn_ops       = BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL BF_PTR
 $cpuid_obj    = 
 $bn_obj       = bn_asm.o sparcv9a-mont.o
 $des_obj      = des_enc-sparc.o fcrypt_b.o
@@ -3652,7 +3652,7 @@
 $thread_cflag = -D_REENTRANT
 $sys_id       = ULTRASPARC
 $lflags       = -lsocket -lnsl -ldl
-$bn_ops       = SIXTY_FOUR_BIT_LONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL BF_PTR
+$bn_ops       = BN_LLONG RC4_CHAR RC4_CHUNK DES_INT DES_PTR DES_RISC1 DES_UNROLL BF_PTR
 $cpuid_obj    = 
 $bn_obj       = bn_asm.o sparcv9a-mont.o
 $des_obj      = des_enc-sparc.o fcrypt_b.o