[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <i3vf5e63aqbfvoywuftt7h3qd4abdzomdeuvp3ygmzgza4xjdt@qx36kyyxxkfi>
Date: Sat, 23 Nov 2024 12:21:21 -0500
From: Kent Overstreet <kent.overstreet@...ux.dev>
To: Thomas Graf <tgraf@...g.ch>, Herbert Xu <herbert@...dor.apana.org.au>
Cc: netdev@...r.kernel.org
Subject: rhashtable issue - -EBUSY
I'm seeing an issue where rhashtable inserts sporadically return -EBUSY,
thrown from rhashtable_insert_rehash() when there's already a rehash in
progress, i.e. rehashes aren't keeping up.
This looks like a clear bug, the -EBUSY isn't handled at all, nor is it
documented, and inserts shouldn't randomly fail.
Looking through the history it's hard to tell what the intention of the
code is...
Powered by blists - more mailing lists