Hello Team, I have been experimenting with the FUSE passthrough(https://docs.kernel.org/filesystems/fuse-passthrough.html) feature. I have a question regarding its interaction with the kernel page cache. During my evaluation, I observed a performance degradation when using passthrough. My understanding is that the I/O on the passthrough backing file descriptor bypasses the kernel page cache and goes directly to disk. This is significantly less performant for my use case compared to cached I/O. While looking into the kernel source, I found a couple of references that seem to confirm that bypassing the page cache is intentional. 1. The FUSE driver appears to explicitly disallow the FOPEN_KEEP_CACHE flag for passthrough: https://github.com/torvalds/linux/blob/66701750d5565c574af42bef0b789ce0203e3071/fs/fuse/iomode.c#L162 2. The passthrough feature uses the fuse_file_uncached_io_open function. This function returns an ETXTBSY error if any cached file handles for the inode are already open: https://github.com/torvalds/linux/blob/66701750d5565c574af42bef0b789ce0203e3071/fs/fuse/iomode.c#L97 Could you please share the reasoning behind this design choice? Additionally, would it be feasible to implement support for using the page cache with passthrough in the future? Thank you for your time and any insights you can provide. Best regards, Ashmeen
