Hello, I have been developing my application on AIFM for some time. Recently I encountered a bug with gc_cache() and CircularBuffer().
Describe the bug
In gc_cache() a GC worker first checks is_free_cache_high(), which is the size of free_regions_ over the capacity of free_regions, then tries to pick up regions from used_regions_. There is a case where the is_free_cache_high() is not satisfied by "default".
This is because the free_regions_ is constructed with free_regions_count as the argument.
free_regions_ = std::move(CircularBuffer<Region, false>(free_regions_count));
The free_regions_count includes the number of real free regions that can be allocated and the number of per-core pre-allocated free regions. For example, if the kNumSocket1CPUs is set to be 30 and the free_regions_count is 76 (16 + 30 * 2), then only 16 free regions will be in the free_regions_ circular buffer and the is_free_cache_high() is not satisfied by default.
So if GC is triggered, the GC will run into an indefinite loop because the is_free_cache_high() will never be satisfied.
Priority
I think this can be fixed. I can change this line which initialized the free_regions_ from using free_regions_count as the size (capacity) argument to using free_regions_coun - 2 * helpers::kNumSocket1CPUs.
free_regions_ = std::move(CircularBuffer<Region, false>(free_regions_count - 2* helpers::kNumSocket1CPUs));
I can create a PR for this if other things are fine after the change because there may be some places where I miss.
Screenshots
The gc_cache() runs into an indefinite loop.
server
I think the environment does not matter much in this case.
g++ version: g++ 9.2.0
kernel version: 5.14.0
Hello, I have been developing my application on AIFM for some time. Recently I encountered a bug with
gc_cache()andCircularBuffer().Describe the bug
In
gc_cache()a GC worker first checksis_free_cache_high(), which is the size offree_regions_over the capacity offree_regions, then tries to pick up regions fromused_regions_. There is a case where theis_free_cache_high()is not satisfied by "default".This is because the
free_regions_is constructed withfree_regions_countas the argument.free_regions_ = std::move(CircularBuffer<Region, false>(free_regions_count));The
free_regions_countincludes the number of real free regions that can be allocated and the number of per-core pre-allocated free regions. For example, if thekNumSocket1CPUsis set to be 30 and thefree_regions_countis 76 (16 + 30 * 2), then only 16 free regions will be in thefree_regions_circular buffer and theis_free_cache_high()is not satisfied by default.So if GC is triggered, the GC will run into an indefinite loop because the
is_free_cache_high()will never be satisfied.Priority
I think this can be fixed. I can change this line which initialized the
free_regions_from usingfree_regions_countas the size (capacity) argument to usingfree_regions_coun - 2 * helpers::kNumSocket1CPUs.I can create a PR for this if other things are fine after the change because there may be some places where I miss.
Screenshots
The
gc_cache()runs into an indefinite loop.server
I think the environment does not matter much in this case.
g++ version: g++ 9.2.0
kernel version: 5.14.0