Inflate using wider loads and stores and a minimum of branches. (#95)

* Inflate using wider loads and stores.

In inflate_fast() the output pointer always has plenty of room to write.  This
means that so long as the target is capable, wide un-aligned loads and stores
can be used to transfer several bytes at once.

When the reference distance is too short simply unroll the data a little to
increase the distance.

Change-Id: I59854eb25d2b1e43561c8a2afaf9175bf10cf674

1 files changed, 7 insertions, 0 deletions

diff --git a/inffast.h b/inffast.h
index 0d75614..fdf5a18 100644
--- a/inffast.h
+++ b/inffast.h
@@ -12,4 +12,11 @@
 
 void ZLIB_INTERNAL inflate_fast(z_stream *strm, unsigned long start);
 
+
+#if (defined(__GNUC__) || defined(__clang__)) && defined(__ARM_NEON__)
+#  include <arm_neon.h>
+typedef uint8x16_t inffast_chunk_t;
+#  define INFFAST_CHUNKSIZE sizeof(inffast_chunk_t)
+#endif
+
 #endif /* INFFAST_H_ */