Towards heap walk #569
Towards heap walk #569
Conversation
|
Will we finally reach precise heap allocation statistics? |
This is not aimed at better statistics. We could add that, but it is not on my current plan. My rough thought would be to track per sizeclass sizeclass | total allocations | curr slabs | max slabsThe total allocations would be bumped everytime we grab a new free list, and curr slabs and max slabs would be increased on |
|
This last commit 9042b40 is my attempt to get the fix passed the CHERI compiler. We won't really be using this value, except to look up in the pagemap, but changing all the types in |
This adds an additional SeqSet that is used to track all the fully used slabs and large allocations. This gives more chances to detect memory leaks, and additionally catch some more UAF failures where the object is not recycled.
Use the head of the free list builder to track an interior pointer to the slab. This is unused unless the list contains something. Hence, we can use this to represent an interior pointer to the slab and report more accurate leaks.
Add an operation to get a tag free pointer from an address_t, and use it
This includes a two changes that are on the way to implementing heap walk. The changes performed are:
ladenranges of memory: the full and almost full slabs and large allocations.