3 ms·

I built a memory allocator for testing not too long ago that only allocated but no-op'd deallocate(using the C++ Allocator interface). It was quite amazing tha

I built a memory allocator for testing not too long ago that only allocated but no-op'd deallocate(using the C++ Allocator interface). It was quite amazing that it resulted in a 25-50% performance boost in many cases I was testing. It's not for real code as you only get destruction but not deallocation. It let me pull that part out of the tests to get a better view of the performance

fermentation 6y ago
Yes as it turns out malloc is much faster if free doesn’t exist.
- saagarjha 6y ago
  Not just that: malloc is faster if you don’t have to care about thread safety, or support interposing, or hooks, or if security is not a concern. It is easy to “beat malloc” in a specific usecase if you know what it will be beforehand. Doing this in general is what is hard.
troutwine 6y ago
In case you weren't aware, this is a common enough technique for it to actually have a name: bump allocation. It's very useful in short lived programs with bounded allocations, for the exact reason you outlined.
- MaxBarraclough 6y ago
  The dmd D compiler adopted this approach and got a significant performance boost. [0] Looks like it still uses a pointer-bump allocator today. [1] [0] https://web.archive.org/web/20190126213344/https://www.drdobbs.com/cpp/increasing-compiler-speed-by-over-75/240158941 https://web.archive.org/web/20190126213344/https://www.drdob... [1] https://github.com/dlang/dmd/blob/025caed79/src/dmd/root/rmem.d#L153 https://github.com/dlang/dmd/blob/025caed79/src/dmd/root/rme...
dnautics 6y ago
this is why composable allocators are fantastic. You can request a dynamic memory slate for a given context, then allocate as you wish, no-op delete, when you're done with the context, you delete all of them in one go.