-
Notifications
You must be signed in to change notification settings - Fork 7
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
upgrade to polkadot-v0.9.40 #350
Conversation
@ebma the weights have increased. You can check each extrinsic's comments from the benchmark. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for adding all these changes @b-yap 🙏
I see that the weights changed indeed. Sometimes the execution time varies just a little and sometimes the weights doubled (which might look confusing at first since the weight comments now state everything in pico seconds, but since pico is 1e-3 smaller than nano and the values mostly have three additional 0s now, they are pretty similar.) But it could be that the fact that you were now using --execution=wasm
also amplified this variation. It's great though that you managed to use that parameter as it's preferred and should be more accurate AFAIK. I only did not use it before because it did not work for some reason.
Maybe we can fix some of the CI warnings but besides that it's good to go 👍
// --extrinsic=* | ||
// --chain | ||
// dev | ||
// --execution=wasm |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Wow, since when does it work with wasm
? That's great. I remember when I created all the benchmarks a while ago this parameter was not working and unfortunately I was not able to figure out why. Great that it works now
pallets/issue/src/benchmarking.rs
Outdated
// this amount is when running the benchmark with the testchain | ||
// let amount = 1000_0000u32.into(); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why this comment? I need to change the amount when running benchmarks with the testchain but use the 1000
otherwise? Can we change it such that it always uses 100_00000u32
then so that it works in both cases?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
changed amount to the fixed 1000_0000u32.into()
…obs/9118542926?pr=350#step:9:1563 and remove extra comments of grandpa
407d32b
to
365d3f4
Compare
partially closes pendulum issue 219.
I have updated the weights as well, since a lot of the pallets are
#![deny(warnings)]
.summary of changes:
from_ref...
, changed tofrom_parts()
functionSpecific to clients directory:
Specific to testchain directory:
warp_sync
towarp_sync_params
sc-network
WeightToFee
andLengthToFee
impls to transaction-payment Runtime APIKeyOwnerProofSystem
of grandpa removedextra:
test_check_slot_position
test case after updating theMAX_SLOTS_TO_REMEMBER
value.