From 35453446a511366e1250858b249e36c80b6ad044 Mon Sep 17 00:00:00 2001
From: Marc Mutz <marc.mutz@kdab.com>
Date: Sun, 4 Jul 2021 23:44:01 +0200
Subject: [PATCH] QCryptographicHash: don't present the same data over and over
 again

Need to decrement 'remaining' (check), but also increment data (meep).

Testing is a bit complicated, as most algorithms are just too slow to
fit into the 5min QTestLib timeout.  Picked the fast ones and Sha512
(which completes here in < 17s, with threads), at least.

Amends e12577b56396cca0df05f88f8787706a3a12c82d.

[ChangeLog][QtCore][QCryptographicHash] Fixed a bug where presenting
more than 4GiB in a single addData() call would calculate the wrong
result().

Pick-to: 6.1 6.2
Change-Id: Ic72916ebc33ba087d58225af6d8240e46e41f434
Reviewed-by: Lars Knoll <lars.knoll@qt.io>
---
 src/corelib/tools/qcryptographichash.cpp      |  4 +-
 .../tst_qcryptographichash.cpp                | 84 +++++++++++++++++++
 2 files changed, 85 insertions(+), 3 deletions(-)

diff --git a/src/corelib/tools/qcryptographichash.cpp b/src/corelib/tools/qcryptographichash.cpp
index 47b72038eb..f6db6b976c 100644
--- a/src/corelib/tools/qcryptographichash.cpp
+++ b/src/corelib/tools/qcryptographichash.cpp
@@ -363,10 +363,8 @@ void QCryptographicHash::addData(const char *data, qsizetype length)
     // feed the data UINT_MAX bytes at a time, as some of the methods below
     // take a uint (of course, feeding more than 4G of data into the hashing
     // functions will be pretty slow anyway)
-    qsizetype remaining = length;
-    while (remaining) {
+    for (auto remaining = length; remaining; remaining -= length, data += length) {
         length = qMin(qsizetype(std::numeric_limits<uint>::max()), remaining);
-        remaining -= length;
 #else
     {
 #endif
diff --git a/tests/auto/corelib/tools/qcryptographichash/tst_qcryptographichash.cpp b/tests/auto/corelib/tools/qcryptographichash/tst_qcryptographichash.cpp
index c819f02d70..92a2c3516c 100644
--- a/tests/auto/corelib/tools/qcryptographichash/tst_qcryptographichash.cpp
+++ b/tests/auto/corelib/tools/qcryptographichash/tst_qcryptographichash.cpp
@@ -29,9 +29,14 @@
 
 #include <QtCore/QCoreApplication>
 #include <QTest>
+#include <QScopeGuard>
 #include <QCryptographicHash>
 #include <QtCore/QMetaEnum>
 
+#if QT_CONFIG(cxx11_future)
+#  include <thread>
+#endif
+
 Q_DECLARE_METATYPE(QCryptographicHash::Algorithm)
 
 class tst_QCryptographicHash : public QObject
@@ -51,6 +56,11 @@ private slots:
     void files();
     void hashLength_data();
     void hashLength();
+    // keep last
+    void moreThan4GiBOfData_data();
+    void moreThan4GiBOfData();
+private:
+    std::vector<char> large;
 };
 
 void tst_QCryptographicHash::repeated_result_data()
@@ -419,5 +429,79 @@ void tst_QCryptographicHash::hashLength()
     QCOMPARE(QCryptographicHash::hashLength(algorithm), output.length());
 }
 
+void tst_QCryptographicHash::moreThan4GiBOfData_data()
+{
+#if QT_POINTER_SIZE > 4
+    QElapsedTimer timer;
+    timer.start();
+    const size_t GiB = 1024 * 1024 * 1024;
+    try {
+        large.resize(4 * GiB + 1, '\0');
+    } catch (const std::bad_alloc &) {
+        QSKIP("Could not allocate 4GiB plus one byte of RAM.");
+    }
+    QCOMPARE(large.size(), 4 * GiB + 1);
+    large.back() = '\1';
+    qDebug("created dataset in %lld ms", timer.elapsed());
+
+    QTest::addColumn<QCryptographicHash::Algorithm>("algorithm");
+    auto me = QMetaEnum::fromType<QCryptographicHash::Algorithm>();
+    auto row = [me] (QCryptographicHash::Algorithm algo) {
+        QTest::addRow("%s", me.valueToKey(int(algo))) << algo;
+    };
+    // these are reasonably fast (O(secs))
+    row(QCryptographicHash::Md4);
+    row(QCryptographicHash::Md5);
+    row(QCryptographicHash::Sha1);
+    // this is already significantly slower, but important (O(min)
+    row(QCryptographicHash::Sha512);
+    // the rest is just too slow
+#else
+    QSKIP("This test is 64-bit only.");
+#endif
+}
+
+void tst_QCryptographicHash::moreThan4GiBOfData()
+{
+    QFETCH(const QCryptographicHash::Algorithm, algorithm);
+
+# if QT_CONFIG(cxx11_future)
+    using MaybeThread = std::thread;
+# else
+    struct MaybeThread {
+        std::function<void()> func;
+        void join() { func(); }
+    };
+# endif
+
+    QElapsedTimer timer;
+    timer.start();
+    const auto sg = qScopeGuard([&] {
+        qDebug() << algorithm << "test finished in" << timer.restart() << "ms";
+    });
+
+    const auto begin = large.data();
+    const auto mid = begin + large.size() / 2;
+    const auto end = begin + large.size();
+
+    QByteArray single;
+    QByteArray chunked;
+
+    auto t = MaybeThread{[&] {
+        QCryptographicHash h(algorithm);
+        h.addData(begin, end - begin);
+        single = h.result();
+    }};
+    {
+        QCryptographicHash h(algorithm);
+        h.addData(begin, mid - begin);
+        h.addData(mid, end - mid);
+        chunked = h.result();
+    }
+    t.join();
+
+    QCOMPARE(single, chunked);
+}
+
 QTEST_MAIN(tst_QCryptographicHash)
 #include "tst_qcryptographichash.moc"