A message for Linux.com registered users: We are in the process of making changes to the Linux forums. Starting Monday, 8/13/18 at 6:00 PM PT, you will be unable to access the forums. They will re-launch as soon as possible on Wednesday, 8/15/18 with new features and improved functionality. Thank you for your patience and stay tuned for the new improved forums.
Dev contribution to hugetlbfs
So I'm posting here because I'm not entirely sure who to get in touch with to ask, so I'll just cross my fingers that this is roughly the right place.
Basically, in the course of my work I've used/read up on hugetlbfs a fair amount, and think it's a very useful capability. But in the course of use, I've also had some difficulties with it. Specifically, the reliability of memory allocation after boot time is poor if the system is busy.
Therefore, I am considering attempting to write an enhancement to the system to increase the reliability of changes made to the number of allocated hugepages (i.e., trying to get more).
I'm aware of the transparent huge page technology, but I also know that they have not been a silver bullet, and in any case they would not strictly speaking help with my specific use case.
I'm an experienced C developer and I've written (personal) kernel mods before, and I've also extensively browsed the hugetlbfs code itself but to sum it up, I have two questions:
1) Does anyone have any specific resources they would recommend I read/utilize? People to get in touch with if the enhancement seems to turn out well?
2) Is this something for which there is broad interest?