Add PowerPC vpmsum implementation of CRC

* cipher/Makefile.am: Add 'crc-ppc.c'.
* cipher/crc-armv8-ce.c: Remove 'USE_INTEL_PCLMUL' comment.
* cipher/crc-ppc.c: New.
* cipher/crc.c (USE_PPC_VPMSUM): New.
(CRC_CONTEXT): Add 'use_vpmsum'.
(_gcry_crc32_ppc8_vpmsum, _gcry_crc24rfc2440_ppc8_vpmsum): New.
(crc32_init, crc24rfc2440_init): Add HWF check for 'use_vpmsum'.
(crc32_write, crc24rfc2440_write): Add 'use_vpmsum' code-path.
* configure.ac: Add 'vpmsumd' instruction to PowerPC VSX inline
assembly check; Add 'crc-ppc.lo'.
--

Benchmark on POWER8 (ppc64le, ~3.8Ghz):
Before:
                |  nanosecs/byte   mebibytes/sec   cycles/byte
 CRC32          |     0.978 ns/B     975.0 MiB/s      3.72 c/B
 CRC24RFC2440   |     0.974 ns/B     978.8 MiB/s      3.70 c/B
After(~22x faster):
                |  nanosecs/byte   mebibytes/sec   cycles/byte
 CRC32          |     0.044 ns/B     21878 MiB/s     0.166 c/B
 CRC24RFC2440   |     0.043 ns/B     22077 MiB/s     0.164 c/B

Benchmark on POWER9 (ppc64le, ~3.8Ghz):
Before:
                |  nanosecs/byte   mebibytes/sec   cycles/byte
 CRC32          |      1.01 ns/B     943.7 MiB/s      3.84 c/B
 CRC24RFC2440   |     0.993 ns/B     960.6 MiB/s      3.77 c/B
After (~20x faster):
                |  nanosecs/byte   mebibytes/sec   cycles/byte
 CRC32          |     0.046 ns/B     20675 MiB/s     0.175 c/B
 CRC24RFC2440   |     0.048 ns/B     19691 MiB/s     0.184 c/B

GnuPG-bug-id: 4460
Signed-off-by: Jussi Kivilinna <jussi.kivilinna@iki.fi>

1 files changed, 46 insertions, 0 deletions

diff --git a/cipher/crc.c b/cipher/crc.c
index 2abbab28..6d70f644 100644
--- a/cipher/crc.c
+++ b/cipher/crc.c
@@ -52,6 +52,19 @@
 # endif
 #endif /* USE_ARM_PMULL */
 
+/* USE_PPC_VPMSUM indicates whether to enable PowerPC vector
+ * accelerated code. */
+#undef USE_PPC_VPMSUM
+#ifdef ENABLE_PPC_CRYPTO_SUPPORT
+# if defined(HAVE_COMPATIBLE_CC_PPC_ALTIVEC) && \
+     defined(HAVE_GCC_INLINE_ASM_PPC_ALTIVEC)
+#  if __GNUC__ >= 4
+#   define USE_PPC_VPMSUM 1
+#  endif
+# endif
+#endif /* USE_PPC_VPMSUM */
+
+
 typedef struct
 {
   u32 CRC;
@@ -61,6 +74,9 @@ typedef struct
 #ifdef USE_ARM_PMULL
   unsigned int use_pmull:1;            /* ARMv8 PMULL shall be used. */
 #endif
+#ifdef USE_PPC_VPMSUM
+  unsigned int use_vpmsum:1;           /* POWER vpmsum shall be used. */
+#endif
   byte buf[4];
 }
 CRC_CONTEXT;
@@ -80,6 +96,13 @@ void _gcry_crc24rfc2440_armv8_ce_pmull (u32 *pcrc, const byte *inbuf,
 					size_t inlen);
 #endif
 
+#ifdef USE_PPC_VPMSUM
+/*-- crc-ppc.c --*/
+void _gcry_crc32_ppc8_vpmsum (u32 *pcrc, const byte *inbuf, size_t inlen);
+void _gcry_crc24rfc2440_ppc8_vpmsum (u32 *pcrc, const byte *inbuf,
+				     size_t inlen);
+#endif
+
 
 /*
  * Code generated by universal_crc by Danjel McGougan
@@ -388,6 +411,9 @@ crc32_init (void *context, unsigned int flags)
 #ifdef USE_ARM_PMULL
   ctx->use_pmull = (hwf & HWF_ARM_NEON) && (hwf & HWF_ARM_PMULL);
 #endif
+#ifdef USE_PPC_VPMSUM
+  ctx->use_vpmsum = !!(hwf & HWF_PPC_ARCH_2_07);
+#endif
 
   (void)flags;
   (void)hwf;
@@ -416,6 +442,13 @@ crc32_write (void *context, const void *inbuf_arg, size_t inlen)
       return;
     }
 #endif
+#ifdef USE_PPC_VPMSUM
+  if (ctx->use_vpmsum)
+    {
+      _gcry_crc32_ppc8_vpmsum(&ctx->CRC, inbuf, inlen);
+      return;
+    }
+#endif
 
   if (!inbuf || !inlen)
     return;
@@ -477,6 +510,9 @@ crc32rfc1510_init (void *context, unsigned int flags)
 #ifdef USE_ARM_PMULL
   ctx->use_pmull = (hwf & HWF_ARM_NEON) && (hwf & HWF_ARM_PMULL);
 #endif
+#ifdef USE_PPC_VPMSUM
+  ctx->use_vpmsum = !!(hwf & HWF_PPC_ARCH_2_07);
+#endif
 
   (void)flags;
   (void)hwf;
@@ -811,6 +847,9 @@ crc24rfc2440_init (void *context, unsigned int flags)
 #ifdef USE_ARM_PMULL
   ctx->use_pmull = (hwf & HWF_ARM_NEON) && (hwf & HWF_ARM_PMULL);
 #endif
+#ifdef USE_PPC_VPMSUM
+  ctx->use_vpmsum = !!(hwf & HWF_PPC_ARCH_2_07);
+#endif
 
   (void)hwf;
   (void)flags;
@@ -839,6 +878,13 @@ crc24rfc2440_write (void *context, const void *inbuf_arg, size_t inlen)
       return;
     }
 #endif
+#ifdef USE_PPC_VPMSUM
+  if (ctx->use_vpmsum)
+    {
+      _gcry_crc24rfc2440_ppc8_vpmsum(&ctx->CRC, inbuf, inlen);
+      return;
+    }
+#endif
 
   if (!inbuf || !inlen)
     return;