{"id":43090,"date":"2024-05-06T16:24:41","date_gmt":"2024-05-06T20:24:41","guid":{"rendered":"https:\/\/mjtsai.com\/blog\/?p=43090"},"modified":"2024-05-06T16:24:41","modified_gmt":"2024-05-06T20:24:41","slug":"swifts-native-clocks-are-very-inefficient","status":"publish","type":"post","link":"https:\/\/mjtsai.com\/blog\/2024\/05\/06\/swifts-native-clocks-are-very-inefficient\/","title":{"rendered":"Swift’s Native Clocks Are Very Inefficient"},"content":{"rendered":"

Wade Tregaskis<\/a> (Hacker News<\/a>):<\/p>\n

In a nutshell, the problem is that Swift’s Clock<\/a><\/code> protocol has significant overheads by design. If you look at a time profile of code like this, you’ll see things like[…]<\/p>

That’s a lot of time wasted in function calls and struct initialisation and type conversion and protocol witnesses and all that guff. The only part that’s actually<\/em> retrieving the time is the swift_get_time<\/a><\/code> call (which is just a wrapper over clock_gettime<\/a><\/code>, which is just a wrapper over clock_gettime_nsec_np<\/a>(CLOCK_UPTIME_RAW)<\/code>, which is just a wrapper over mach_absolute_time<\/a><\/code>).<\/p>

[…]<\/p>

The downside to calling mach_absolute_time<\/code>directly<\/em>, though, is that it’s on Apple’s “naughty” list<\/a> – apparently it’s been abused for device fingerprinting, so Apple require you to beg for special permission if you want to use it (even though it’s used by all these other APIs anyway, as the basis for their implementations, and there’s nothing you can get from mach_absolute_time<\/code> that you can’t get from them too 🤨).<\/p><\/blockquote>\n

This matches my experience that intuition is often wrong regarding Swift performance. Sometimes what seems like it would be a simple virtual call has more overhead than an Objective-C message send. Various dynamic stuff involving checking types\/protocols can also be much slower than with Objective-C. The good news is that Date<\/code> is fast, not even calling down to NSDate<\/code>, and that there’s a pull request<\/a> to inline some of this.<\/p>\n\n

Previously:<\/p>\n