From c2d336140a1016b133da3416ba166b2b7e60d058 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 21 Nov 2023 12:41:26 +0100 Subject: [PATCH] lockdep: Fix block chain corruption [ Upstream commit bca4104b00fec60be330cd32818dd5c70db3d469 ] Kent reported an occasional KASAN splat in lockdep. Mark then noted: > I suspect the dodgy access is to chain_block_buckets[-1], which hits the last 4 > bytes of the redzone and gets (incorrectly/misleadingly) attributed to > nr_large_chain_blocks. That would mean @size == 0, at which point size_to_bucket() returns -1 and the above happens. alloc_chain_hlocks() has 'size - req', for the first with the precondition 'size >= rq', which allows the 0. This code is trying to split a block, del_chain_block() takes what we need, and add_chain_block() puts back the remainder, except in the above case the remainder is 0 sized and things go sideways. Fixes: 810507fe6fd5 ("locking/lockdep: Reuse freed chain_hlocks entries") Reported-by: Kent Overstreet Signed-off-by: Peter Zijlstra (Intel) Tested-by: Kent Overstreet Link: https://lkml.kernel.org/r/20231121114126.GH8262@noisy.programming.kicks-ass.net Signed-off-by: Sasha Levin --- kernel/locking/lockdep.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 0224b03..3b38303 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -3453,7 +3453,8 @@ static int alloc_chain_hlocks(int req) size = chain_block_size(curr); if (likely(size >= req)) { del_chain_block(0, size, chain_block_next(curr)); - add_chain_block(curr + req, size - req); + if (size > req) + add_chain_block(curr + req, size - req); return curr; } } -- 2.7.4