To fix this problem, contact Microsoft Customer Support Services to get the hotfix. For a complete list of Microsoft Customer Support Services telephone numbers and information about support costs, visit the following Microsoft website:Note In special cases, charges that are ordinarily incurred for support calls may be canceled if a Microsoft Support Professional determines that a specific update will fix your problem. The usual support costs will apply to additional support questions and issues that don't qualify for the specific update in question.
PrerequisitesTo apply this hotfix, you must have the .NET Framework 4.5.2, the .NET Framework 4.5.1, or the .NET Framework 4.5 installed.
Restart requirementYou have to restart the computer after you apply this hotfix if any affected files are being used. We recommend that you close all .NET Framework-based applications before you apply this hotfix.
Hotfix replacement informationThis hotfix package doesn't replace a previously released hotfix package.
Issues that this hotfix rollup fixesIssue 1
Assume that you use a Windows Presentation Foundation (WPF) application. A WPF TreeView can spontaneously scroll to an unpredictable position after you add a new item. This issue occurs when the following conditions are true:
- The new item is the last of its parent's children.
- The parent item is the last of its parent's children.
- The new item is visible at the current scroll position.
In a WPF application, removing an editing control from the visual tree can leak an ImmComposition object. Doing this repeatedly can eventually cause an out-of-memory exception.
Note An "editing control" is any control that creates and uses a TextEditor object. The main examples are TextBox, RichTextBox, and FlowDocumentScrollViewer.
A WPF application can become deadlocked when the following conditions are true:
- The application raises an event whose delivery is routed through the weak-event system.
- Another thread is concurrently using the weak-event system. This causes the first thread to wait for a lock.
- When it waits, the thread's message pump receives a message that causes re-entrancy into the application's code.
- The re-entered code tries to use the weak-event system.
This is fixed in the weak-event system by waiting for locks in a way that doesn't run the thread's message pump.
When you try to run a command on the server, it times out on the client side, but finishes on the server. In this situation, the client sends an attention packet to the server and sets itself into a timed-out state. However, it doesn't reset it as soon as it recovers from the timed-out state after the server processed the statement that the client timed out on.
This hotfix resets the state of the client when it recovers from the timed-out state.
In some circumstances, some files in the ASP.NET shadow cache can become corrupted during computer shutdown. This may cause unpredictable behavior or exceptions being raised by applications that use the cache.
This hotfix makes sure that the cache contents are more proactively flushed to disk.
Computers that have 80 gigabytes (GB) or more of memory can experience decrease in performance in low memory situations because of the Garbage Collector running back to back Gen2 collections.
This hotfix includes changes to the Garbage Collector, adjusting the conditions in which Gen2 collections occur on computers that have 80 GB or more of memory.
Known issuesA fix that was previously released for touch-enabled VSTO add-ins in WPF was found to have a regression in scenarios in which partial-trust applications are used on touch-enabled devices. In these scenarios, customers see an application crash and cannot load a partial trust application. Therefore, the fix for touch-enabled VSTO add-ins has been removed.
Article ID: 3095334 - Last Review: Oct 28, 2015 - Revision: 1