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, 8 insertions, 0 deletions

diff --git a/inflate.c b/inflate.c
index 84ae6e7..8484bc3 100644
--- a/inflate.c
+++ b/inflate.c
@@ -371,9 +371,17 @@ static int updatewindow(z_stream *strm, const unsigned char *end, uint32_t copy)
 
     /* if it hasn't been done already, allocate space for the window */
     if (state->window == NULL) {
+#ifdef INFFAST_CHUNKSIZE
+        unsigned wsize = 1U << state->wbits;
+        state->window = (unsigned char *) ZALLOC(strm, wsize + INFFAST_CHUNKSIZE, sizeof(unsigned char));
+        if (state->window == Z_NULL)
+            return 1;
+        memset(state->window + wsize, 0, INFFAST_CHUNKSIZE);
+#else
         state->window = (unsigned char *) ZALLOC(strm, 1U << state->wbits, sizeof(unsigned char));
         if (state->window == NULL)
             return 1;
+#endif
     }
 
     /* if window not in use yet, initialize */