Get rid of blocking behavior for regenerate response.

gpt4all-chat/chat.cpp

@@ -63,10 +63,10 @@ void Chat::connectLLM()
     connect(this, &Chat::loadDefaultModelRequested, m_llmodel, &ChatLLM::loadDefaultModel, Qt::QueuedConnection);
     connect(this, &Chat::loadModelRequested, m_llmodel, &ChatLLM::loadModel, Qt::QueuedConnection);
     connect(this, &Chat::generateNameRequested, m_llmodel, &ChatLLM::generateName, Qt::QueuedConnection);
+    connect(this, &Chat::regenerateResponseRequested, m_llmodel, &ChatLLM::regenerateResponse, Qt::QueuedConnection);
 
     // The following are blocking operations and will block the gui thread, therefore must be fast
     // to respond to
-    connect(this, &Chat::regenerateResponseRequested, m_llmodel, &ChatLLM::regenerateResponse, Qt::BlockingQueuedConnection);
     connect(this, &Chat::resetResponseRequested, m_llmodel, &ChatLLM::resetResponse, Qt::BlockingQueuedConnection);
     connect(this, &Chat::resetContextRequested, m_llmodel, &ChatLLM::resetContext, Qt::BlockingQueuedConnection);
 }
@@ -151,7 +151,7 @@ void Chat::handleLocalDocsRetrieved(const QString &uid, const QList<ResultInfo>
 
 void Chat::regenerateResponse()
 {
-    emit regenerateResponseRequested(); // blocking queued connection
+    emit regenerateResponseRequested();
 }
 
 void Chat::stopGenerating()

gpt4all-chat/chatllm.cpp

@@ -146,7 +146,7 @@ bool ChatLLM::loadModel(const QString &modelName)
     // We have a live model, but it isn't the one we want
     bool alreadyAcquired = isModelLoaded();
     if (alreadyAcquired) {
-        resetContextProtected();
+        resetContext();
 #if defined(DEBUG_MODEL_LOADING)
         qDebug() << "already acquired model deleted" << m_chat->id() << m_modelInfo.model;
 #endif
@@ -301,12 +301,6 @@ void ChatLLM::resetResponse()
 }
 
 void ChatLLM::resetContext()
-{
-    resetContextProtected();
-    emit sendResetContext();
-}
-
-void ChatLLM::resetContextProtected()
 {
     regenerateResponse();
     m_ctx = LLModel::PromptContext();

gpt4all-chat/chatllm.h

@@ -81,14 +81,12 @@ Q_SIGNALS:
     void recalcChanged();
     void sendStartup();
     void sendModelLoaded();
-    void sendResetContext();
     void generatedNameChanged();
     void stateChanged();
     void threadStarted();
     void shouldBeLoadedChanged();
 
 protected:
-    void resetContextProtected();
     bool handlePrompt(int32_t token);
     bool handleResponse(int32_t token, const std::string &response);
     bool handleRecalculate(bool isRecalc);

gpt4all-chat/server.cpp

@@ -286,7 +286,7 @@ QHttpServerResponse Server::handleCompletionRequest(const QHttpServerRequest &re
     }
 
     // don't remember any context
-    resetContextProtected();
+    resetContext();
 
     QSettings settings;
     settings.sync();