From 4c72a1a9c2dd2a1e5268289c21a8df07428eda0d Mon Sep 17 00:00:00 2001
From: Bimba Shrestha <bimbashrestha@fb.com>
Date: Thu, 5 Mar 2020 09:55:38 -0800
Subject: [PATCH 1/2] adding vector to main loop

---
 lib/compress/zstd_fast.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/lib/compress/zstd_fast.c b/lib/compress/zstd_fast.c
index 6dbefee6b..8d4c82b1f 100644
--- a/lib/compress/zstd_fast.c
+++ b/lib/compress/zstd_fast.c
@@ -80,6 +80,10 @@ ZSTD_compressBlock_fast_generic(
     }
 
     /* Main Search Loop */
+#ifdef __INTEL_COMPILER
+    #pragma vector always
+    #pragma ivdep
+#endif
     while (ip1 < ilimit) {   /* < instead of <=, because check at ip0+2 */
         size_t mLength;
         BYTE const* ip2 = ip0 + 2;

From 43fc88f44358d659c2d07c58710d131d370545e6 Mon Sep 17 00:00:00 2001
From: Bimba Shrestha <bimbashrestha@fb.com>
Date: Tue, 10 Mar 2020 14:57:27 -0500
Subject: [PATCH 2/2] Adding comment and remvoing ivdep

---
 lib/compress/zstd_fast.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/lib/compress/zstd_fast.c b/lib/compress/zstd_fast.c
index 8d4c82b1f..09e220b3a 100644
--- a/lib/compress/zstd_fast.c
+++ b/lib/compress/zstd_fast.c
@@ -81,8 +81,11 @@ ZSTD_compressBlock_fast_generic(
 
     /* Main Search Loop */
 #ifdef __INTEL_COMPILER
+    /* From intel 'The vector pragma indicates that the loop should be 
+     * vectorized if it is legal to do so'. Can be used together with 
+     * #pragma ivdep (but have opted to exclude that because intel 
+     * warns against using it).*/
     #pragma vector always
-    #pragma ivdep
 #endif
     while (ip1 < ilimit) {   /* < instead of <=, because check at ip0+2 */
         size_t mLength;