Update TODO

This commit is contained in:
Joakim Frostegård 2020-04-08 12:12:05 +02:00
parent 209072aa4a
commit c75e36c105

View file

@ -14,9 +14,13 @@
* Run multiple threads to test performance when contested? * Run multiple threads to test performance when contested?
* Iterate over whole returned buffer and run e.g. xor on it (.iter().fold()) * Iterate over whole returned buffer and run e.g. xor on it (.iter().fold())
* Generic bench function since current functions are almost identical * Generic bench function since current functions are almost identical
* State is now accumulated over runs. Is this good?
* Performance * Performance
* cpu-target=native good? * cpu-target=native good?
* mialloc good? * mialloc good?
* TorrentMap: mutable access only to insert peer, then drop reference
and get read-only access to gather peers. This could speed up
multi-threaded performance a lot
* Use less bytes from PeerId for hashing? Would need to implement * Use less bytes from PeerId for hashing? Would need to implement
"faulty" PartialEq too (on PeerMapKey, which would be OK) "faulty" PartialEq too (on PeerMapKey, which would be OK)
* bittorrent_udp * bittorrent_udp
@ -41,4 +45,4 @@
* Use `bytes` crate for bittorrent_udp: seems to worsen performance somewhat * Use `bytes` crate for bittorrent_udp: seems to worsen performance somewhat
* Zerocopy (https://docs.rs/zerocopy/0.3.0/zerocopy/index.html) for requests * Zerocopy (https://docs.rs/zerocopy/0.3.0/zerocopy/index.html) for requests
and responses? Doesn't work on Vec etc and responses? Doesn't work on Vec etc
* New array buffer each time in response_to_bytes: doesn't help performance * New array buffer each time in response_to_bytes: doesn't help performance