Re: Algorithm Optimization

"Johann 'Myrkraverk' Oskarsson" <>
Wed, 21 Apr 2021 16:29:30 +0000

          From comp.compilers

Related articles
[9 earlier articles]
Re: Algorithm Optimization (Richard Harnden) (2020-09-16)
Re: Algorithm Optimization (Hans-Peter Diettrich) (2020-09-17)
Re: Algorithm Optimization (Thomas Koenig) (2020-09-17)
Re: Algorithm Optimization (A. K.) (2020-09-21)
Re: Algorithm Optimization (Hans-Peter Diettrich) (2020-12-13)
Re: Algorithm Optimization (gah4) (2020-12-20)
Re: Algorithm Optimization (Johann 'Myrkraverk' Oskarsson) (2021-04-21)
| List of all articles for this month |

From: "Johann 'Myrkraverk' Oskarsson" <>
Newsgroups: comp.compilers
Date: Wed, 21 Apr 2021 16:29:30 +0000
Organization: Compilers Central
References: 20-09-032 20-09-035 20-09-036
Injection-Info:; posting-host=""; logging-data="8258"; mail-complaints-to=""
Keywords: optimize
Posted-Date: 21 Apr 2021 12:40:15 EDT

On 16/09/2020 5:25 am, gah4 wrote:

> I think I remember this being discussed many years ago.

> One thought was that someone codes bubblesort, and the compiler
> generates quicksort. Small complication that bubblesort is stable, and
> quicksort isn't. (Add an array with the original position to break
> ties.)

Here are even more dragons.

What if you're sorting exactly three elements? In that situation, I'll
be very surprised if bubblesort doesn't outperform all other algorithms.

How does the /compiler/ decide that an algorithm is better, based
on data it never sees? Even if we account for profile guided optimi-
zations, we shouldn't allow the compiler to adjust for /what if this is
run with more data in production?/ kind of questions.

If you want to go down this rabbit hole, I'd much prefer warnings than
outright algorithm changes.


  I'm not from the internet, I just work there.

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.