From 3fcbb06cab8264310f2c971709925ca66e3b5b2c Mon Sep 17 00:00:00 2001
From: Alexey Milovidov <milovidov@yandex-team.ru>
Date: Mon, 11 May 2020 08:04:59 +0300
Subject: [PATCH] Fix MSan report in MergeTreeIndexFullText

---
 src/Storages/MergeTree/MergeTreeIndexFullText.cpp | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/src/Storages/MergeTree/MergeTreeIndexFullText.cpp b/src/Storages/MergeTree/MergeTreeIndexFullText.cpp
index 4d1106e7663..61ea5987e76 100644
--- a/src/Storages/MergeTree/MergeTreeIndexFullText.cpp
+++ b/src/Storages/MergeTree/MergeTreeIndexFullText.cpp
@@ -14,6 +14,7 @@
 #include <Parsers/ASTIdentifier.h>
 #include <Parsers/ASTLiteral.h>
 #include <Parsers/ASTSubquery.h>
+#include <Core/Defines.h>
 
 #include <Poco/Logger.h>
 
@@ -617,7 +618,7 @@ bool SplitTokenExtractor::next(const char * data, size_t len, size_t * pos, size
 
     while (*pos < len)
     {
-#if defined(__SSE2__)
+#if defined(__SSE2__) && !defined(MEMORY_SANITIZER) /// We read uninitialized bytes and decide on the calcualted mask
         // NOTE: we assume that `data` string is padded from the right with 15 bytes.
         const __m128i haystack = _mm_loadu_si128(reinterpret_cast<const __m128i *>(data + *pos));
         const size_t haystack_length = 16;
@@ -691,7 +692,7 @@ bool SplitTokenExtractor::next(const char * data, size_t len, size_t * pos, size
 #endif
     }
 
-#if defined(__SSE2__)
+#if defined(__SSE2__) && !defined(MEMORY_SANITIZER)
     // Could happen only if string is not padded with zeroes, and we accidentally hopped over end of data.
     if (*token_start > len)
         return false;