From 9d0907b07606ab4da16ae67aac9523fce7ff9525 Mon Sep 17 00:00:00 2001
From: Thiago Macieira <thiago.macieira@intel.com>
Date: Wed, 9 May 2018 19:56:30 -0700
Subject: QUrl: use the SIMD ASCII-checking code from the last commit

Improves performance a little. This is just because I can and the
function is right there for the taking, as this qt_urlRecodeByteArray
function is only used in deprecated QUrl code.

Change-Id: I5d0ee9389a794d80983efffd152d290e570af387
Reviewed-by: Allan Sandfeld Jensen <allan.jensen@qt.io>
---
 src/corelib/io/qurlrecode.cpp | 10 ++++------
 src/corelib/tools/qstring.cpp | 25 +++++++++++++++++++------
 2 files changed, 23 insertions(+), 12 deletions(-)

diff --git a/src/corelib/io/qurlrecode.cpp b/src/corelib/io/qurlrecode.cpp
index a9b23babc0..91d7ac554a 100644
--- a/src/corelib/io/qurlrecode.cpp
+++ b/src/corelib/io/qurlrecode.cpp
@@ -603,6 +603,9 @@ qt_urlRecode(QString &appendTo, const QChar *begin, const QChar *end,
                   encoding, actionTable, false);
 }
 
+// qstring.cpp
+bool qt_is_ascii(const char *&ptr, const char *end) Q_DECL_NOTHROW;
+
 /*!
     \internal
     \since 5.0
@@ -623,12 +626,7 @@ QString qt_urlRecodeByteArray(const QByteArray &ba)
     // control points below 0x20 are fine in QString
     const char *in = ba.constData();
     const char *const end = ba.constEnd();
-    for ( ; in < end; ++in) {
-        if (*in & 0x80)
-            break;
-    }
-
-    if (in == end) {
+    if (qt_is_ascii(in, end)) {
         // no non-ASCII found, we're safe to convert to QString
         return QString::fromLatin1(ba, ba.size());
     }
diff --git a/src/corelib/tools/qstring.cpp b/src/corelib/tools/qstring.cpp
index 773b15032d..719c685f69 100644
--- a/src/corelib/tools/qstring.cpp
+++ b/src/corelib/tools/qstring.cpp
@@ -333,11 +333,10 @@ static bool simdTestMask(const char *&ptr, const char *end, quint32 maskval)
 }
 #endif
 
-bool QtPrivate::isAscii(QLatin1String s) Q_DECL_NOTHROW
+// Note: ptr on output may be off by one and point to a preceding US-ASCII
+// character. Usually harmless.
+bool qt_is_ascii(const char *&ptr, const char *end) Q_DECL_NOTHROW
 {
-    const char *ptr = s.begin();
-    const char *end = s.end();
-
 #if defined(__AVX2__)
     if (!simdTestMask(ptr, end, 0x80808080))
         return false;
@@ -346,16 +345,22 @@ bool QtPrivate::isAscii(QLatin1String s) Q_DECL_NOTHROW
     while (ptr + 16 < end) {
         __m128i data = _mm_loadu_si128(reinterpret_cast<const __m128i *>(ptr));
         quint32 mask = _mm_movemask_epi8(data);
-        if (mask)
+        if (mask) {
+            uint idx = qCountTrailingZeroBits(mask);
+            ptr += idx;
             return false;
+        }
         ptr += 16;
     }
 #endif
 
     while (ptr + 4 <= end) {
         quint32 data = qFromUnaligned<quint32>(ptr);
-        if (data & 0x80808080U)
+        if (data &= 0x80808080U) {
+            uint idx = qCountTrailingZeroBits(data);
+            ptr += idx / 8;
             return false;
+        }
         ptr += 4;
     }
 
@@ -367,6 +372,14 @@ bool QtPrivate::isAscii(QLatin1String s) Q_DECL_NOTHROW
     return true;
 }
 
+bool QtPrivate::isAscii(QLatin1String s) Q_DECL_NOTHROW
+{
+    const char *ptr = s.begin();
+    const char *end = s.end();
+
+    return qt_is_ascii(ptr, end);
+}
+
 static bool isAscii(const QChar *&ptr, const QChar *end)
 {
 #ifdef __SSE2__
-- 
cgit v1.2.3