July 2018 | The Old New Thing

Server names: One of the remaining places where IT managers can be a little silly

Raymond ChenJuly 31, 2018Jul 31, 201807/31/18

Some time ago, ComputerWorld ran a story on whimsical names for servers. I recall that at my university, one academic department had machines named up and down. I'm sure this led to some strange Who's on First?-style messages.¹ The early email servers at Microsoft were named after muppets. I'm guessing that this came from the term ...

Dark Pattern: Opt-in above the fold, but opt-out below the fold

Raymond ChenJuly 30, 2018Jul 30, 201807/30/18

One of my colleagues reminded me of a Dark Pattern employed by the installer for the media player for that internet protocol that got the rug pulled out from under it. When you got to the shovelware portion of the installer, it showed you this: We have partnered with other companies to bring you these valuable offers. Select the ones you...

TerminateThread function synchronous?"/>

Is the `TerminateThread` function synchronous?

Raymond ChenJuly 27, 2018Jul 27, 201807/27/18

A customer wanted to know whether the TerminateThread function was synchronous. In other words, does the TerminateThread function wait until the target thread has definitely terminated before returning? No. The TerminateThread function queues a termination to the target thread but does not wait for confirmation that the ...

What happens to custom unhandled exception filters if a debugger is not running?

Raymond ChenJuly 26, 2018Jul 26, 201807/26/18

A customer wanted to know what happens to custom exception filters if a debugger is not running. They included this sample code fragment: The customer reported that the results of the Test function are highly context-dependent. If the Test function is called from main, then it works, provided that a debugger is not attached. But if they call ...

Why are my file write operations synchronous, even though I opened the file as <CODE data-lazy-src=

FILE_FLAG_OVERLAPPED?"/>

Why are my file write operations synchronous, even though I opened the file as `FILE_FLAG_OVERLAPPED`?

Raymond ChenJuly 25, 2018Jul 25, 201807/25/18

A customer found that even though they opened a file with the FILE_FLAG_OVERLAPPED, their calls to WriteFile were nevertheless synchronous. Why are their overlapped writes completing synchronously? As currently implemented, there are at least three things which will cause a write to be forced synchronous: The first two are easy to ...

Pushing the boundaries of cryptography in a security vulnerability report

Raymond ChenJuly 24, 2018Jul 24, 201807/24/18

A security vulnerability report arrived that implied that the finder had made an earth-shattering breakthrough in cryptography. Specifically, the finder claimed to have found an efficient way to factor the large numbers used in RSA cryptography. This would be a remarkable breakthrough if true, but the description of the algorithm, while ...

The heavy metal umlaut reaches baked goods

Raymond ChenJuly 23, 2018Jul 23, 201807/23/18

For that extra Germanic oomph.

How do I trigger an EN_UPDATE notification for all of my edit controls when the user’s locale information changes?

Raymond ChenJuly 23, 2018Jul 23, 201807/23/18

A customer had a dialog box with a bunch of edit controls. Some of these edit controls contained floating-point values, and the code parses them by calling a locale-sensitive parser, so that users in the United States (say) can use the period as the decimal marker, whereas users in Germany can use the comma. The user might change the locale ...

How can I get the actual window procedure address and not a thunk?

Raymond ChenJuly 20, 2018Jul 20, 201807/20/18

We saw some time ago that the GetWindowLongPtr function returns a magic cookie if it needs to thunk the message. The CallWindowProc function understands this magic cookie and calls the original window procedure after converting the message to the appropriate character set. But what if you want to get the actual window ...

When is it appropriate to use the current processor number as an optimization hint?

Raymond ChenJuly 19, 2018Jul 19, 201807/19/18

Some time ago, on the topic of avoiding heap contention, I left an exercise that asked whether it would be appropriate to use the current processor number (as reported by GetCurrentProcessorNumber) to select which heap to use. In this case, the answer is no. While using the current processor would avoid contention at allocation...

The Old New Thing