Stagger Application of `stepSize` in ZSTD_fast

This replicates the behavior of @terrelln's `ZSTD_fast` implementation. That
is, it always looks at adjacent pairs of positions, and only applies the
acceleration every other position. This produces a more fine-grained
acceleration.

diff --git a/lib/compress/zstd_fast.c b/lib/compress/zstd_fast.c
index 87de17f81c140f80a58335019e6e01d58b5a7885..8e26cc70313a4eb7a2a0b24ecffb53c217b83316 100644 (file)
--- a/lib/compress/zstd_fast.c
+++ b/lib/compress/zstd_fast.c
@@ -99,7 +99,7 @@ ZSTD_compressBlock_fast_noDict_generic(
     U32* const hashTable = ms->hashTable;
     U32 const hlog = cParams->hashLog;
     /* support stepSize of 0 */
-    size_t const stepSize = cParams->targetLength + !(cParams->targetLength);
+    size_t const stepSize = cParams->targetLength + !(cParams->targetLength) - 1;
     const BYTE* const base = ms->window.base;
     const BYTE* const istart = (const BYTE*)src;
     const U32   endIndex = (U32)((size_t)(istart - base) + srcSize);
@@ -144,13 +144,13 @@ ZSTD_compressBlock_fast_noDict_generic(
     /* start each op */
 _start: /* Requires: ip0 */
 
-    step = stepSize;
+    step = 1;
     nextStep = ip0 + kStepIncr;
 
     /* calculate positions, ip0 - anchor == 0, so we skip step calc */
-    ip1 = ip0 + stepSize;
-    ip2 = ip1 + stepSize;
-    ip3 = ip2 + stepSize;
+    ip1 = ip0 + step;
+    ip2 = ip1 + step + stepSize;
+    ip3 = ip2 + step;
 
     if (ip3 >= ilimit) {
         goto _cleanup;
@@ -241,7 +241,7 @@ _start: /* Requires: ip0 */
         /* advance to next positions */
         ip0 = ip1;
         ip1 = ip2;
-        ip2 = ip2 + step;
+        ip2 = ip2 + step + stepSize;
         ip3 = ip2 + step;
     } while (ip3 < ilimit);