News for hackers. Stuff that matters.
Re:It’s Time for a Modern Synthesis Kernel by obl on Tuesday July 02, 2019 @03:35PM (#20338639)
Nice article.

Of course if you get some of the benefits of run time specialization, you also get some of the downside, i.e., overspecialization.

The heuristics to know when to stop specializing is a really hard part of those kind of systems.

You could have a program that does a single write() on 100k different files. You could have thousands of processes each writing to a single file and bloating the I$ with useless copies of write.

Re:It’s Time for a Modern Synthesis Kernel by trurl on Tuesday July 02, 2019 @03:54PM (#20338818)
This isn't exactly a new issue in runtime code generation, for example, the HotSpot Java JIT was introduced over 20 years ago.
Re:It’s Time for a Modern Synthesis Kernel by vidarh on Tuesday July 02, 2019 @11:09PM (#20341295)
The typical approach to this (and yes it can of course still fail), is to use counters and first optimize when you see a given pattern enough times.