diff options
| author |   Eric Dumazet <edumazet@google.com> | 2018-11-08 17:34:27 -0800 |
|---|---|---|
| committer |   David S. Miller <davem@davemloft.net> | 2018-11-08 18:40:30 -0800 |
| commit | 0d5b9311baf27bb545f187f12ecfd558220c607d (patch) | |
| tree | 73bc70b75089cf84c4d473506bfa59db36a08baa /tools/perf/scripts/python/export-to-sqlite.py | |
| parent | net: hns3: bugfix for not checking return value (diff) | |
| download | linux-dev-0d5b9311baf27bb545f187f12ecfd558220c607d.tar.xz linux-dev-0d5b9311baf27bb545f187f12ecfd558220c607d.zip | |
inet: frags: better deal with smp races
Multiple cpus might attempt to insert a new fragment in rhashtable,
if for example RPS is buggy, as reported by 배석진 in
https://patchwork.ozlabs.org/patch/994601/
We use rhashtable_lookup_get_insert_key() instead of
rhashtable_insert_fast() to let cpus losing the race
free their own inet_frag_queue and use the one that
was inserted by another cpu.
Fixes: 648700f76b03 ("inet: frags: use rhashtables for reassembly units")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: 배석진 <soukjin.bae@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'tools/perf/scripts/python/export-to-sqlite.py')
0 files changed, 0 insertions, 0 deletions