PR: 2432
Submitted by: Michael Heyman

for more details.

PR: 2335

collected with 'apps/openssl speed ghash'.

ghash-x86.pl.

Modest improvement coefficients mean that code already had some
parallelism and there was not very much room for improvement. Special
thanks to Ted Krovetz for benchmarking the code with such patience.

add ghash benchmark to apps/speed.c.

ARM assembler modules.

opcodes detected in executable segment" error.

improvement.

in gcm_ghash_4bit. Taking the idea a step further I've added extra
256+16 bytes of per-key storage, so that one can speak about 3rd variant
in addition to "256B" and "4KB": "528B" one. Commonly it should be
~50% faster than "256B" implementation or ~25% slower than "4KB" one.

- gcm128.c: support for Intel PCLMULQDQ, readability improvements;
- asm/ghash-x86.pl: splitted vanilla, MMX, PCLMULQDQ subroutines;
- asm/ghash-x86_64.pl: add PCLMULQDQ implementations.

- introduce common modes_lcl.h;
- ctr128.c: implement additional CRYPTO_ctr128_encrypt_ctr32 interface;
- gcm128.c: add omitted ARM initialization, remove ctx.ctr;

ghash-x86*.pl: fix performance numbers for Core2, as it turned out
previous ones were "tainted" by variable clock frequency.