Timeline for Why is processing a sorted array faster than processing an unsorted array?
Current License: CC BY-SA 3.0
12 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Jun 20, 2020 at 9:12 | history | edited | CommunityBot |
Commonmark migration
|
|
Feb 10, 2020 at 4:07 | comment | added | Peter Cordes | While flushing pipelines is super fast Not really. It's fast compared to a cache miss all the way to DRAM, but on a modern high-performance x86 (like Intel Sandybridge-family) it's about a dozen cycles. Although fast recovery does allow it to avoid waiting for all older independent instructions to reach retirement before starting recovery, you still lose a lot of front-end cycles on a mispredict. What exactly happens when a skylake CPU mispredicts a branch?. (And each cycle can be about 4 instructions of work.) Bad for high-throughput code. | |
Apr 2, 2018 at 11:29 | review | Suggested edits | |||
Apr 2, 2018 at 13:21 | |||||
Mar 16, 2018 at 12:30 | history | edited | Peter Mortensen | CC BY-SA 3.0 |
Active reading [<https://en.wiktionary.org/wiki/let%27s#Contraction> <http://en.wiktionary.org/wiki/nowadays>]
|
Aug 25, 2017 at 18:46 | audit | First posts | |||
Aug 25, 2017 at 19:18 | |||||
Aug 25, 2017 at 12:37 | audit | First posts | |||
Aug 25, 2017 at 13:06 | |||||
Aug 24, 2017 at 10:12 | audit | First posts | |||
Aug 24, 2017 at 10:13 | |||||
Aug 24, 2017 at 7:02 | audit | First posts | |||
Aug 24, 2017 at 7:03 | |||||
Aug 4, 2017 at 18:12 | history | edited | Tony Tannous | CC BY-SA 3.0 |
added 646 characters in body
|
Aug 4, 2017 at 10:32 | history | edited | Tony Tannous | CC BY-SA 3.0 |
added 9 characters in body
|
Aug 4, 2017 at 10:24 | history | edited | Tony Tannous | CC BY-SA 3.0 |
added 352 characters in body
|
Aug 4, 2017 at 10:07 | history | answered | Tony Tannous | CC BY-SA 3.0 |