On Thu, Apr 16, 2026 at 08:56:46AM -0700, Joanne Koong wrote: > On Tue, Apr 14, 2026 at 5:10 PM John Groves <[email protected]> wrote: > > > > There is a FUSE_DAX_FMAP capability that the kernel may advertise or not > > at init time; this capability "is" the famfs GET_FMAP AND GET_DAXDEV > > commands. In the future, if we find a way to use BPF (or some other > > mechanism) to avoid needing those fuse messages, the kernel could be updated > > to NEVER advertise the FUSE_DAX_FMAP capability. All of the famfs-specific > > code could be taken out of kernels that never advertise that capability. > > I’m not sure the capability bit can be used like that (though I am > hoping it can!). As I understand it, once the kernel advertises a > capability, it must continue supporting it in future kernels else > userspace programs that rely on it will break. >
FUSE_DAX_FMAP is already conditional on CONFIG_FUSE_DAX, the kernel is not required to continue advertising FUSE_DAX_FMAP in perpetuity. Setting CONFIG_FUSE_DAX=n does not mean userland "is broken", this would only be the case if FUSE_DAX_FMAP was advertised but not actually supported. If DAX were removed from the kernel (unlikely, but stick with me) this would be equivalent to permanently changing CONFIG_FUSE_DAX to always off, and there would be no squabbles over whether that particular change broke userland (there would be much strife over removing dax). While not a deprecation method, this is what capability bits are designed for. Same as cpuid capability bits - just because the bit is there doesn't mean a processor is required to support it in perpetuity. They're only required to support it if the bit is turned on. --- I think the focus here needs to be on whether this interface ACTUALLY needs to be more generic - and whether that is actually FEASIBLE. It's not like this is a new problem - and there are real design reasons why John chose this route. The additional overhead is not trivial for FAMFS - FAMFS is not doing i/o. He already has data showing fuse caused a performance hit due to overhead on open - his concern of overhead on fault being catastrophic is grounded in data. For others it's an age old problem of self-describing protocols (parsing vs giant inflexible binary blobs, pick your poison). It's extremely unlikely we will find a one-size-fits-all solution that doesn't eventually run right back into this same problem. I worry that this discussion is going to turn towards implementing a solution grounded in parsing arbitrary formats and how to store them, and that is completely detached from why FAMFS went this route in the first place. I question whether the actual issue here lies in the interface APPEARING more general purpose than it actually is - and therefore inviting attempts to over-genericize it. Is there a world here where this is solved by a name change and a capability bit? I think so. ~Gregory
