Closed kevingosse closed 2 months ago
Execution-time results for samples comparing the following branches/commits:
Execution-time benchmarks measure the whole time it takes to execute a program. And are intended to measure the one-off costs. Cases where the execution time results for the PR are worse than latest master results are shown in red. The following thresholds were used for comparing the execution times:
Note that these results are based on a single point-in-time result for each branch. For full results, see the dashboard.
Graphs show the p99 interval based on the mean and StdDev of the test run, as well as the mean value of the run (shown as a diamond below the graph).
Branch report: kevin/error_loader
Commit report: 4b29f0a
Test service: dd-trace-dotnet
:white_check_mark: 0 Failed, 306654 Passed, 2349 Skipped, 22h 2m 18.67s Total Time
Benchmarks for #5929 compared to master:
The following thresholds were used for comparing the benchmark speeds:
Allocation changes below 0.5% are ignored.
Summary of changes
Wrap the call to the managed loader in a try/catch, and use a native logger to log the exception if any.
Reason for change
There's already a try/catch in the managed loader, but in some situations we can fail before even getting there (if jitting the method failed, for instance because of a trimmed runtime). Also, the managed loader logger can fail.
Implementation details
This PR adds a try/catch in the code emitted to load the managed loader. In the catch clause, we call a native logging function, with a direct unmanaged
calli
call to its address.Because we have now this outer catch block, I removed the
IsInSsi
logic which isn't needed anymore.