XeLabs/go-profiler-notes
2
0
Fork 0
mirror of https://github.com/DataDog/go-profiler-notes.git synced 2026-06-20 16:45:26 +08:00
Code Issues Projects Releases Wiki Activity

Merge pull request #7 from ozerovandrei/fix-inuse-metric-name

Browse source 
Memory profiler: fix `inuse_` metric name
This commit is contained in:
Felix Geisendörfer 2022-01-11 10:30:31 +01:00 committed by GitHub
commit 8536a00d61
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
guide/README.md
View file

@ -273,7 +273,7 @@ Regardless of how you activate the Memory profiler, the resulting profile will e
A memory profile contains two major pieces of information:
- `alloc_*`: The amount of allocations that your program has made since the start of the process (or profiling period for delta profiles).
- `insue_*`: The amount of allocations that your program has made that were still reachable during the last GC.
- `inuse_*`: The amount of allocations that your program has made that were still reachable during the last GC.
You can use this information for various purposes. For example you can use the `alloc_*` data to determine which code paths might be producing a lot of garbage for the GC to deal with, and looking at the `inuse_*` data over time can help you with investigating memory leaks or high memory usage by your program.
@ -326,7 +326,7 @@ func sweep(object):
// deallocation magic
```
The `free_*` counters themselves are not included in the final memory profile. Instead they are used to calculate the `insue_*` counters in the profile via simple `allocs - frees` subtraction. Additionally the final output values are scaled by dividing them through their sampling probability.
The `free_*` counters themselves are not included in the final memory profile. Instead they are used to calculate the `inuse_*` counters in the profile via simple `allocs - frees` subtraction. Additionally the final output values are scaled by dividing them through their sampling probability.
### Memory Profiler Limitations