Nick Ripley
Nick Ripley
Fair question! Here are some pros and cons for doing the ELF/Mach-O stuff in C: Pros: * I need to call into C at least a little bit to find...
I'm going to close this for now, we can re-open if/when we want to spend more time on C allocation profiling.
Nice optimization! This change unfortunately requires Go 1.17 for `atomic.(*Value).Swap`. Even if we adopt our [new Go version support policy](https://github.com/DataDog/dd-trace-go/pull/1239), we're still likely going to support Go 1.16 until Go...
I'll add that to the profiler to match the tracer 👍
Thank you for the report! I'll look into this ASAP.
So I haven't yet reproduced the crash, and I suspect it may be difficult on your end since it happens infrequently. However, one thing jumps out to me in the...
Possibly, yes, though the address is _slightly_ different in each crash (see below). If you're able to get the crash under a debugger and look at the arguments going into...
Good idea, please report back if you find anything. It might also be worth compiling the program with the address sanitizer (Go 1.18 added the `-asan` flag), though I'm not...
Ah, very interesting! Thank you for running more tests. This looks like a promising lead. Have you tried compiling and running your program with the `-race` flag to enable the...
Do you have a similar log starting with `Profiler configuration:`?