Javier Honduvilla Coto

@brancz I think it makes sense to remove these if the memory is completely unmapped. Semi-related issue: https://github.com/parca-dev/parca-agent/issues/359

Trying now with ```patch diff --git a/Makefile b/Makefile index e00bbc8..e2cf2d5 100644 --- a/Makefile +++ b/Makefile @@ -77,7 +77,7 @@ go_env := GOOS=linux GOARCH=$(ARCH:x86_64=amd64) CC=$(CMD_CLANG) CGO_CFLAGS="-I ifndef DOCKER $(OUT_BIN): $(LIBBPF_HEADERS) $(LIBBPF_OBJ)...

> // TODO(javierhonduco): Getting -ENOTSUPP, perhaps my kernel doesn't support it? Oh yes, I remember this, was quite odd. Can take a look in some weeks. Wondering how hot`DeleteKeyBatch` is,...

Curious to see if fixing the links addresses this! > One thing we need to clarify is where is the _start symbol stored for the given example. Seems like it's...

Love that we are thinking about this!! Should we maybe continue extracting the debug files for a little while and then attempt to upload them later on if the server...

I just wanted to update this issue as we have changed our kernel symbolizer in https://github.com/parca-dev/parca-agent/pull/708. The new implementation loads all the addresses and symbol names in memory, but there...

Sounds good! Let's do it after we land system-wide work if you don't mind. There's going to be a couple of concurrent changes in the same files

Thanks for doing this! The unknown frames seem odd, let's investigate what it is and why it can't symbolise

As we already have this internally, I'll close this issue

I've tried this PR on our internal Python wrapper and seems to work fine, It passes all our integration tests! 🎉 I was curious about performance and it seems to...