kcsan: Fix misreporting if concurrent races on same address
authorMarco Elver <elver@google.com>
Mon, 10 Feb 2020 14:56:39 +0000 (15:56 +0100)
committerIngo Molnar <mingo@kernel.org>
Sat, 21 Mar 2020 08:43:41 +0000 (09:43 +0100)
commit3a5b45e5031fa395ab6e53545fb726b2d5b104f0
tree311b9c0dc636cd1f33a077372e14ed9153014bd1
parent80d4c4775216602ccdc9e761ce251c8451d0c6ca
kcsan: Fix misreporting if concurrent races on same address

If there are at least 4 threads racing on the same address, it can
happen that one of the readers may observe another matching reader in
other_info. To avoid locking up, we have to consume 'other_info'
regardless, but skip the report. See the added comment for more details.

Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/kcsan/report.c