Even Linus Torvalds can have trouble with autocomplete • The Register

Next time autocomplete takes over and you accidentally send an email to the wrong person or group, perhaps it will be a little solace to know that one of the world’s most accomplished technologists – Linux kernel boss Linus Torvalds – just made that same mistake.

During most weeks of the kernel development cycle, Torvalds posts to the Linux Kernel Mailing list with an update on the previous week’s development activities. The posts usually announce new release candidates of the kernel, or entire new versions.

When Torvalds travels, takes a vacation or observes public holidays, he usually flags possible changes to his posting habits so kernelistas can understand why his regular writings won’t appear as usual.

Torvalds gave no such advice last week and his his usual weekly post did not appear on the list last Sunday.

Which was no cause for alarm, because Torvalds lives in the USA, which is busy celebrating Presidents Day. Perhaps the missing missive was therefore attributable to him partaking in the holiday?

On Monday, Torvalds revealed the real reason his weekly post didn’t appear.

“Bah,” he wrote. “I sent this out earlier, but it turns out that I sent it to the wrong list due to some messed-up autocompletion, so people didn’t see it.”

“So here’s the 6.14-rc3 announcement on the proper mailing list.”

His message confirmed that work on version 6.14 of the Linux Kernel is proceeding well.

“So the first few weeks of the 6.14 release development were smaller-than-usual, but rc3 is actually right in line with normal releases at this point,” Torvalds wrote. “Probably just timing of pull requests, and we’ll see how next week goes. But nothing looks worrisome.”

This release candidate adds something interesting in the form of a “faux device”.

As explained by Linux luminary, kernel developer and maintainer Greg Kroah-Hartman, this tech is designed “to allow platform devices from stop being abused.”

Linux kernel documentation defines platform devices as “devices that typically appear as autonomous entities in the system. This includes legacy port-based devices and host bridges to peripheral buses, and most controllers integrated into system-on-chip platforms.”

Kroah-Hartman explained the new structure is needed “to allow almost a straight or simpler conversion from platform devices that were not really a platform device.”

Interestingly, in the context of recent ructions about the use of the Rust language in the kernel, the faux bus “comes with a binding for rust, with an example driver in rust showing how it’s used.” ®