Francesco Nigro
Francesco Nigro
It contains 3 improvements (unfinished): - [O(1) collation lookup table](https://github.com/eclipse-vertx/vertx-sql-client/commit/e94a0288c9b20c4373e7f3b5bd4b3af3c4945360) - [branchless/batchy parse digits](https://github.com/eclipse-vertx/vertx-sql-client/commit/cf176f47c270cffb04df4e47639da440450d4a0e) - NOT YET STARTED: https://github.com/eclipse-vertx/vertx-sql-client/blob/9073da6171b7688a7491398522dcb6c454bd5960/vertx-mysql-client/src/main/java/io/vertx/mysqlclient/impl/datatype/DataTypeCodec.java#L61 exceed `MinJumpTableSize` (see https://stackoverflow.com/questions/44988841/force-tableswitch-instead-of-lookupswitch for more info): we should find a...
I see that 0cd7eca06ef17e09a8c87221d9b9f08d353614f7 from @tsegismont has introduced updaters, but I still don't understand why is using such heavy weight atomic operations there. Given that is a pretty hot path...
While looking at #2442 I've read some of the historic changes on this part and it's not clear to me yet what the logic and ownership of indexes changes will...
This is to keep up with the Vertx API changes for https://github.com/eclipse-vertx/vert.x/commit/88cf77bfe4f44c3dc5ee89d65b8891674088de46, although I'm not very happy with the current state (that's why is in draft): 1. `VertxLocalsHelper.` is "stealthy"...
One of the key behaviour of a coordinated omission free load generator is to always consider as start time the so-called "intended" one, in order to properly account for coordination...
Performing an heap dump of wrk2 while in the `test` phase reveal that "calibaration" allocated `Sessions` and objects are still retained in the heap, bloating the old generation. It would...
In wrk/wrk2 (but not only) both requests and sequence instances uses LimitedPool with just 1 element capacity. It would be great to optimize this case, given that it can be...
This is a suggested direction for #386 NOT to shrink the existing resources (although it does it - see the TODO for the missing pieces there), but more to have...