introduce all the necessary features for reusing contract data memories #11144
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #11144 +/- ##
==========================================
+ Coverage 71.10% 71.11% +0.01%
==========================================
Files 773 773
Lines 154176 154214 +38
Branches 154176 154214 +38
==========================================
+ Hits 109625 109671 +46
+ Misses 40100 40093 -7
+ Partials 4451 4450 -1
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
|
The untested code lines should become tested once the actual implementation using these new helper functions will land :) |
| pub fn into_preallocated(self) -> Result<PreallocatedMemory, String> { | ||
| Ok(PreallocatedMemory( | ||
| Arc::into_inner(self.0) | ||
| .ok_or("Multiple references to NearVmMemory prevent its reuse")? |
There was a problem hiding this comment.
Might be worthwhile to tracing::debug this too, given that we're unlikely to act on this error.
| )?))) | ||
| } | ||
|
|
||
| #[cfg(unused)] // TODO: this will be used once we reuse the memories | ||
| pub fn into_preallocated(self) -> Result<PreallocatedMemory, String> { |
There was a problem hiding this comment.
| pub fn into_preallocated(self) -> Result<PreallocatedMemory, String> { | |
| pub fn into_preallocated(self) -> Result<PreallocatedMemory, &'static str> { |
|
No worries and thanks for the review! I haven't managed to finish the subsequent PR anyway. Once the next one is closer to completion, I'll handle your comments and land this one :) |
|
Closing, as per the results written down in #11033: it turns out the "optimization" is actually slower than the baseline, so there’s no longer any reason to land this code |
With this, all that’s missing should be the contract data memory queue, and checking that the
Arc::into_inneractually works properly at least most of the time. I’m… not that hopeful about this second precondition, but let’s see what happens with a bit of testing.Part of #11033