[Bug 276002] nfscl: data corruption using both copy_file_range and mmap'd I/O

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=276002

--- Comment #44 from Konstantin Belousov <kib@FreeBSD.org> ---
(In reply to Rick Macklem from comment #42)
If the buffer size is 1M (shrug), then to map any page from this buffer,
kernel would validate the whole buffer.  At least this is the default setup
where vfs.nfs.use_buf_pager is true.

In other words, it should be not possible to have a buffer consisting of
a single dirty page and rest of invalid pages.

-- 
You are receiving this mail because:
You are the assignee for the bug.