[LuaJIT 1/n] Add initial no-op plumbing for LuaJIT unwinder#1397
Conversation
fabled
left a comment
fabled
left a comment
There was a problem hiding this comment.
Wondering also if metrics/errors addition makes sense here or not. But if the implementation is fairly stable and those values are sufficient then they are probly good.
I'd still leave the introspection related file for later PR.
| @@ -0,0 +1,30 @@ | |||
| // Copyright The OpenTelemetry Authors | |||
There was a problem hiding this comment.
Could this file be left out?
These seems Lua introspection data and is unrelated to the internal identifier allocation
|
@umanwizard is out for a bit so I made the changes, should be ready to go |
florianl
left a comment
florianl
left a comment
There was a problem hiding this comment.
Errors/metrics, like ERR_LUAJIT_NO_PROC_INFO or UnwindLuaJITErrNoProcInfo, seem to be implementation specific and might fit better as part of the implementation PR than this scaffolding.
|
|
||
| // number of attempts to unwind LuaJIT | ||
| metricID_UnwindLuaJITAttempts, | ||
|
|
There was a problem hiding this comment.
To follow the other unwinders, metricID_UnwindLuaJITFrames is missing here and in metrics/metrics.json.
gnurizen
force-pushed
the
lua1
branch
3 times, most recently
from
007b16a to
5c7194f
Compare
|
Re-worked to not include errors/metrics. RFAL |
| Dotnet: "dotnet", | ||
| BEAM: "beam", | ||
| APMInt: "apm-integration", | ||
| LuaJIT: "luajit", |
There was a problem hiding this comment.
FYI: opened open-telemetry/semantic-conventions#3707 to land support for this frame type.
|
Thank you @gnurizen for shepherding this while I was out on PTO. |
This reserves various constants, metric IDs, error codes, and so on, but should result in no behavior changes.
|
Rebased to fix unwinder conflicts and pushed. |
5 participants
This reserves various constants, metric IDs, error codes, and so on, but should result in no behavior changes.