Snaipe

To update this, I'm not actively looking at replacing the malloc calls in the assertion calls. It's hard to not rely on vital functions like that without massively increasing the...

Alternatively, since the string building facilities are subpar, we could alter the prototype to provide a printf-like behaviour: ```c void cr_stream_open(struct cr_stream *, const char *url_fmt, ...); void cr_mem_open(struct cr_mem...

This is definitely something that we would want. I wrote a while ago [tizzy](https://github.com/Snaipe/tizzy) with the intent of eventually providing an API for Criterion to use, but got side-tracked since...

Sure, that sounds reasonable. Having a strong foundation for backtrace should allow us to extend the feature with crash backtraces when we get to that point.

Last time I tried something remotely similar with libunwind (which one, by the way? I know there's at least two of them, since LLVM ships their own), I could not...

Looking through this, this seems fairly nice, especially the builtin text processing where normally one would have to use sed or something else in the cram test to manipulate the...

This is troubling. Can you check in criterion's process tree that gdbserver is indeed started?

BoxFort essentially forks then re-execs itself (or CreateProcess-es itself on Windows). A program compiled with BoxFort installs an init hook that checks if its parent has setup a worker context...

I don't use Bazel personally, and I'm not currently interested in maintaining multiple build system definitions. If there is enough desire for it, I would expect someone to step up...

> Also, do you really do execve()? I thought criterion only fork()s Yes, because just fork()ing had some problems. The first issue was that the network part of criterion was...