This fixes two problems in the handling of negative times:
• rem is signed, but the rem * c->sb.nsec_per_time_unit operation produced a bogus unsigned result, because s32 * u32 = u32.
• The timespec was not normalized (it could contain more than a billion nanoseconds).
For example, { .tv_sec = -14245441, .tv_nsec = 750000000 }, after being round tripped through timespec_to_bch2_time and then bch2_time_to_timespec would come back as { .tv_sec = -14245440, .tv_nsec = 4044967296 } (more than 4 billion nanoseconds).
Cc: stable@vger.kernel.org Fixes: 595c1e9bab7f ("bcachefs: Fix time handling") Closes: https://github.com/koverstreet/bcachefs/issues/743 Co-developed-by: Erin Shepherd erin.shepherd@e43.eu Signed-off-by: Erin Shepherd erin.shepherd@e43.eu Co-developed-by: Ryan Lahfa ryan@lahfa.xyz Signed-off-by: Ryan Lahfa ryan@lahfa.xyz Signed-off-by: Alyssa Ross hi@alyssa.is --- I've submitted an RFC to fstests to add a regression test for this: https://lore.kernel.org/fstests/20240907154527.604864-2-hi@alyssa.is/
fs/bcachefs/bcachefs.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/fs/bcachefs/bcachefs.h b/fs/bcachefs/bcachefs.h index 0c7086e00d18..81c4d935cca8 100644 --- a/fs/bcachefs/bcachefs.h +++ b/fs/bcachefs/bcachefs.h @@ -1195,12 +1195,15 @@ static inline bool btree_id_cached(const struct bch_fs *c, enum btree_id btree) static inline struct timespec64 bch2_time_to_timespec(const struct bch_fs *c, s64 time) { struct timespec64 t; + s64 sec; s32 rem;
time += c->sb.time_base_lo;
- t.tv_sec = div_s64_rem(time, c->sb.time_units_per_sec, &rem); - t.tv_nsec = rem * c->sb.nsec_per_time_unit; + sec = div_s64_rem(time, c->sb.time_units_per_sec, &rem); + + set_normalized_timespec64(&t, sec, rem * (s64)c->sb.nsec_per_time_unit); + return t; }
base-commit: 53f6619554fb1edf8d7599b560d44dbea085c730