https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
--- Comment #5 from Alan Somers ---
I decided that the problem with copy_file_range was actually distinct from the
problem with o_direct. I fixed it in 1c909c300b92 . But the o_direct problem
remains.
--
You are receiving this mail beca
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
Alan Somers changed:
What|Removed |Added
Summary|fusefs: data corruption |[fusefs]: data corruption
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
--- Comment #4 from Konstantin Belousov ---
(In reply to Alan Somers from comment #3)
I cannot guarantee anything, but also I do not see a reason to try.
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
--- Comment #3 from Alan Somers ---
kib, do you think that I could fix this bug by doing a vm_object_page_clean
followed by vinvalbuf just prior to FUSE_COPY_FILE_RANGE, FUSE_FALLOCATE, or a
FUSE_WRITE with O_DIRECT ?
--
You are receiving
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
--- Comment #2 from Alan Somers ---
Update: I can also produce the same corruption by mixing mmap'd writes and
copy_file_range, using fsx-rs 0.2.0 with a config like this:
flen = 524288
nomsyncafterwrite = true
[opsize]
max = 131072
min =
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
Graham Perrin changed:
What|Removed |Added
CC||f...@freebsd.org,
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
Konstantin Belousov changed:
What|Removed |Added
CC||k...@freebsd.org
--- Comment
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=269330
Bug ID: 269330
Summary: fusefs: data corruption with mmap and either o_direct
or fspacectl
Product: Base System
Version: CURRENT
Hardware: Any
OS: Any
