Weekly update 2022-05-25 #6900

akhi3030 · 2022-05-27T17:06:57Z

akhi3030
May 27, 2022
Maintainer

Most of the team was attending the protocol off-site the week before and hence was primarily occupied by that.

@matklad: Broke the build for someone for two different unrelated reasons within one week!
@matklad: Assembled a huge standing desk for my new home office. Coupled with the new laptop from last week, I am almost fully set (the only thing missing is a web camera so that I don't have to present my profile in the calls)

@nagisa: Get two stripes on a test
@Ekleog: Finished up the fuzzer, if it's stable until EOW we'll invite everyone with credentials to #pagoda/fuzzer/private to have access to the findings
@Ekleog: Added debug logs, to more easily know which part of our system refused deserialization of a contract
@matklad: Minor logging cleanups fix: reduce spammy logging #6880, refactor: use consistent field name for error #6882
@matklad: Start using strip = true profile option when building various bits of wasm fix: ensure test contract compile without warnings #6878, refactor: use modern way to strip symbols from WASM #6852
@matklad: Fixed some annoyingly failing tests refactor: simplify client_primitives::Error #6865, fix: don't type convert errors to strings unnecessary #6864, fix: remove test that fails locally #6867, test: more helpful assertion messages #6863 particular, fix: race in NetworkRecepient #6828 is the bottom of a pretty deep rabbit hole involving rewriting the core of once_cell implementation.
@matklad: Made the core block data-structure cheaply-clonable via Arc refactor: make blocks and headers cheaply clonable #6866. This is useful to enable caches on top of DB
@matklad: Assorted cleanups refactor: use ? #6877, refactor: use idiomatic rust signature #6872

@nagisa: Locals refactor in wasmer
@Ekleog: Actually instrument all interesting wasmer functions with tracing
@matklad: Still stabilizing alt_bn128. Lesson for the future -- assume stabilization takes at least a week of work feat: stabilize alt_bn128 familiy of host functions #6824
@matklad: Still refactoring the chain store, uncovering various curiosities in process: refactor: use &access for chain store #6811

@matklad: Maybe look at refactoring vm-errors again? This came up once more this week.
@Ekleog: Actually instrument all interesting wasmer functions with tracing