Add LIT_MEM define to use more memory for a small deflate speedup.

A bug fix in zlib 1.2.12 resulted in a slight slowdown (1-2%) of
deflate. This commit provides the option to #define LIT_MEM, which
uses more memory to reverse most of that slowdown. The memory for
the pending buffer and symbol buffers is increased by 25%, which
increases the total memory usage with the default parameters by
about 6%.

1 files changed, 21 insertions, 0 deletions

diff --git a/deflate.c b/deflate.c
index 5e72cd9..263bbc8 100644
--- a/deflate.c
+++ b/deflate.c
@@ -493,7 +493,11 @@ int ZEXPORT deflateInit2_(z_streamp strm, int level, int method,
      * symbols from which it is being constructed.
      */
 
+#ifdef LIT_MEM
+    s->pending_buf = (uchf *) ZALLOC(strm, s->lit_bufsize, 5);
+#else
     s->pending_buf = (uchf *) ZALLOC(strm, s->lit_bufsize, 4);
+#endif
     s->pending_buf_size = (ulg)s->lit_bufsize * 4;
 
     if (s->window == Z_NULL || s->prev == Z_NULL || s->head == Z_NULL ||
@@ -503,8 +507,14 @@ int ZEXPORT deflateInit2_(z_streamp strm, int level, int method,
         deflateEnd (strm);
         return Z_MEM_ERROR;
     }
+#ifdef LIT_MEM
+    s->d_buf = (ushf *)(s->pending_buf + (s->lit_bufsize << 1));
+    s->l_buf = s->pending_buf + (s->lit_bufsize << 2);
+    s->sym_end = s->lit_bufsize - 1;
+#else
     s->sym_buf = s->pending_buf + s->lit_bufsize;
     s->sym_end = (s->lit_bufsize - 1) * 3;
+#endif
     /* We avoid equality with lit_bufsize*3 because of wraparound at 64K
      * on 16 bit machines and because stored blocks are restricted to
      * 64K-1 bytes.
@@ -720,9 +730,15 @@ int ZEXPORT deflatePrime(z_streamp strm, int bits, int value) {
 
     if (deflateStateCheck(strm)) return Z_STREAM_ERROR;
     s = strm->state;
+#ifdef LIT_MEM
+    if (bits < 0 || bits > 16 ||
+        (uchf *)s->d_buf < s->pending_out + ((Buf_size + 7) >> 3))
+        return Z_BUF_ERROR;
+#else
     if (bits < 0 || bits > 16 ||
         s->sym_buf < s->pending_out + ((Buf_size + 7) >> 3))
         return Z_BUF_ERROR;
+#endif
     do {
         put = Buf_size - s->bi_valid;
         if (put > bits)
@@ -1308,7 +1324,12 @@ int ZEXPORT deflateCopy(z_streamp dest, z_streamp source) {
     zmemcpy(ds->pending_buf, ss->pending_buf, (uInt)ds->pending_buf_size);
 
     ds->pending_out = ds->pending_buf + (ss->pending_out - ss->pending_buf);
+#ifdef LIT_MEM
+    ds->d_buf = (ushf *)(ds->pending_buf + (ds->lit_bufsize << 1));
+    ds->l_buf = ds->pending_buf + (ds->lit_bufsize << 2);
+#else
     ds->sym_buf = ds->pending_buf + ds->lit_bufsize;
+#endif
 
     ds->l_desc.dyn_tree = ds->dyn_ltree;
     ds->d_desc.dyn_tree = ds->dyn_dtree;