Refactor the exclude list for assembly instrumentation #3137
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
andrewlock
added
type:refactor
area:native-library
tonyredondo
reviewed
Aug 30, 2022
| { | ||
| Logger::Debug("ModuleLoadFinished matched module by pattern: ", module_id, " ", module_info.assembly.name, | ||
| "but assembly is explicitly included"); | ||
| goto inject; |
There was a problem hiding this comment.
You can leave it like this or or or use a break; to prevent my eyes from bleeding. 😂
There was a problem hiding this comment.
Can't use a break because we're two for-loops deep at this point 😛 We want to break out of both loops (because we've already matched the pattern, so don't want to check the others)
There was a problem hiding this comment.
There's ways of removing the break if it's too offensive 😉
We could check if the assembly is in the include_assembly list first. Only check the skip_assembly_prefixes if it's not in that list.
Given that most assemblies won't be on the list, the approach approach I used should have less overhead I think, by avoiding the extra loop for every assembly? It's a tiny difference though 😉 Happy to change it if you think that's preferable! 🙂
There was a problem hiding this comment.
or add the typical bool "found" and if (found) {break} once again... 🙄🙈
There was a problem hiding this comment.
or add the typical bool "found" and if (found) {break} once again...
Yeah, I think this is preferable. Too much of a coincidence that I add a goto and all our builds suddenly fail 🙈 . They say it's DNS, but I dunno, seems sus 🤫
NachoEchevarria
approved these changes
Aug 30, 2022
This comment has been minimized.
This comment has been minimized.
tonyredondo
approved these changes
Aug 30, 2022
lucaspimentel
approved these changes
Aug 30, 2022
andrewlock
force-pushed
the
andrew/simplify-assembly-skip-check
branch
from
692d69b
to
3c3c6b4
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
In #1663 we switched from excluding the Microsoft.Extensions prefix to excluding each of the non-logging assemblies individually. This is not future proof (as new libraries could be added), and increases the number of assemblies we need to match against. As a workaround, list adds an explicit "include" list for assemblies which match the "exclude" prefix, but which should be included anyway.
andrewlock
force-pushed
the
andrew/simplify-assembly-skip-check
branch
from
3c3c6b4
to
b5c00bd
Compare
Benchmarks Report 🐌Benchmarks for #3137 compared to master:
The following thresholds were used for comparing the benchmark speeds:
Allocation changes below 0.5% are ignored. Benchmark detailsBenchmarks.Trace.AgentWriterBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.AppSecBodyBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.AspNetCoreBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.DbCommandBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.ElasticsearchBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.GraphQLBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.HttpClientBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.ILoggerBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.Log4netBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.NLogBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.RedisBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.SerilogBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.SpanBenchmark - Same speed ✔️ Same allocations ✔️Raw results
Benchmarks.Trace.TraceAnnotationsBenchmark - Same speed ✔️ Same allocations ✔️Raw results
|
Code Coverage Report 📊✔️ Merging #3137 into master will not change line coverage
View the full report for further details: Datadog.Trace Breakdown ✔️
The following classes have significant coverage changes.
The following classes were added in #3137:
View the full reports for further details: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary of changes
Refactor the "exclude by pattern" list to allow specific inclusions.
Reason for change
In #1663 we switched from excluding the Microsoft.Extensions prefix to excluding each of the non-logging assemblies individually.
This is not future proof (as new Microsoft.Extensions libraries could (will) be added, that we don't want to instrument), and increases the number of assemblies we need to match against.
Implementation details
As a workaround, adds an explicit "include" list for assemblies which match the "exclude" prefix, but which should be included anyway.
I used a
goto. I understand if I'm not allowed to touch C++ again.Test coverage
Should be covered by existing
Other details
Should make it easier for @NachoEchevarria's work on
System.Diagnostics.Process.