Christoph, >> Nothing prevents future improvements in that direction. It just seems >> out of scope for what Kanchan is trying to enable for his customer use >> cases. This patch looks harmless. > > It's not really. Once we wire it up like this we mess up the ability > to use the feature in other ways. Additionally the per-I/O hints are > simply broken if you want a file system Here is my take: It is the kernel's job to manage the system's hardware resources and arbitrate and share these resources optimally and fairly between all running applications. What irks me is defining application interfaces which fundamentally tell the kernel that "these blocks are part of the same file". The kernel already knows this. It is the very entity which provides that abstraction. Why do we need an explicit interface to inform the kernel that concurrent writes to the same file should have the same "temperature" or need to go to the same "bin" on the storage device? Shouldn't that just happen automatically? Whether it's SCSI groups, streams, UFS hints, or NVMe FDP, it seems like we are consistently failing to deliver something that actually works for anything but a few specialized corner cases. I think that is a shame. -- Martin K. Petersen Oracle Linux Engineering
