All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev,
	syzbot+1fa663a2100308ab6eab@syzkaller.appspotmail.com,
	Alexei Starovoitov <ast@kernel.org>,
	Andrii Nakryiko <andrii@kernel.org>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.6 41/67] bpf: Avoid kfree_rcu() under lock in bpf_lpm_trie.
Date: Thu, 15 Aug 2024 15:25:55 +0200	[thread overview]
Message-ID: <20240815131839.895503593@linuxfoundation.org> (raw)
In-Reply-To: <20240815131838.311442229@linuxfoundation.org>

6.6-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Alexei Starovoitov <ast@kernel.org>

[ Upstream commit 59f2f841179aa6a0899cb9cf53659149a35749b7 ]

syzbot reported the following lock sequence:
cpu 2:
  grabs timer_base lock
    spins on bpf_lpm lock

cpu 1:
  grab rcu krcp lock
    spins on timer_base lock

cpu 0:
  grab bpf_lpm lock
    spins on rcu krcp lock

bpf_lpm lock can be the same.
timer_base lock can also be the same due to timer migration.
but rcu krcp lock is always per-cpu, so it cannot be the same lock.
Hence it's a false positive.
To avoid lockdep complaining move kfree_rcu() after spin_unlock.

Reported-by: syzbot+1fa663a2100308ab6eab@syzkaller.appspotmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Link: https://lore.kernel.org/bpf/20240329171439.37813-1-alexei.starovoitov@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 kernel/bpf/lpm_trie.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c
index 050fe1ebf0f7d..d0febf07051ed 100644
--- a/kernel/bpf/lpm_trie.c
+++ b/kernel/bpf/lpm_trie.c
@@ -308,6 +308,7 @@ static long trie_update_elem(struct bpf_map *map,
 {
 	struct lpm_trie *trie = container_of(map, struct lpm_trie, map);
 	struct lpm_trie_node *node, *im_node = NULL, *new_node = NULL;
+	struct lpm_trie_node *free_node = NULL;
 	struct lpm_trie_node __rcu **slot;
 	struct bpf_lpm_trie_key_u8 *key = _key;
 	unsigned long irq_flags;
@@ -382,7 +383,7 @@ static long trie_update_elem(struct bpf_map *map,
 			trie->n_entries--;
 
 		rcu_assign_pointer(*slot, new_node);
-		kfree_rcu(node, rcu);
+		free_node = node;
 
 		goto out;
 	}
@@ -429,6 +430,7 @@ static long trie_update_elem(struct bpf_map *map,
 	}
 
 	spin_unlock_irqrestore(&trie->lock, irq_flags);
+	kfree_rcu(free_node, rcu);
 
 	return ret;
 }
@@ -437,6 +439,7 @@ static long trie_update_elem(struct bpf_map *map,
 static long trie_delete_elem(struct bpf_map *map, void *_key)
 {
 	struct lpm_trie *trie = container_of(map, struct lpm_trie, map);
+	struct lpm_trie_node *free_node = NULL, *free_parent = NULL;
 	struct bpf_lpm_trie_key_u8 *key = _key;
 	struct lpm_trie_node __rcu **trim, **trim2;
 	struct lpm_trie_node *node, *parent;
@@ -506,8 +509,8 @@ static long trie_delete_elem(struct bpf_map *map, void *_key)
 		else
 			rcu_assign_pointer(
 				*trim2, rcu_access_pointer(parent->child[0]));
-		kfree_rcu(parent, rcu);
-		kfree_rcu(node, rcu);
+		free_parent = parent;
+		free_node = node;
 		goto out;
 	}
 
@@ -521,10 +524,12 @@ static long trie_delete_elem(struct bpf_map *map, void *_key)
 		rcu_assign_pointer(*trim, rcu_access_pointer(node->child[1]));
 	else
 		RCU_INIT_POINTER(*trim, NULL);
-	kfree_rcu(node, rcu);
+	free_node = node;
 
 out:
 	spin_unlock_irqrestore(&trie->lock, irq_flags);
+	kfree_rcu(free_parent, rcu);
+	kfree_rcu(free_node, rcu);
 
 	return ret;
 }
-- 
2.43.0




  parent reply	other threads:[~2024-08-15 14:09 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-15 13:25 [PATCH 6.6 00/67] 6.6.47-rc1 review Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 01/67] exec: Fix ToCToU between perm check and set-uid/gid usage Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 02/67] ASoC: topology: Clean up route loading Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 03/67] ASoC: topology: Fix route memory corruption Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 04/67] LoongArch: Define __ARCH_WANT_NEW_STAT in unistd.h Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 05/67] NFSD: Rewrite synopsis of nfsd_percpu_counters_init() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 06/67] NFSD: Fix frame size warning in svc_export_parse() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 07/67] sunrpc: dont change ->sv_stats if it doesnt exist Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 08/67] nfsd: stop setting ->pg_stats for unused stats Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 09/67] sunrpc: pass in the sv_stats struct through svc_create_pooled Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 10/67] sunrpc: remove ->pg_stats from svc_program Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 11/67] sunrpc: use the struct net as the svc proc private Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 12/67] nfsd: rename NFSD_NET_* to NFSD_STATS_* Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 13/67] nfsd: expose /proc/net/sunrpc/nfsd in net namespaces Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 14/67] nfsd: make all of the nfsd stats per-network namespace Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 15/67] nfsd: remove nfsd_stats, make th_cnt a global counter Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 16/67] nfsd: make svc_stat per-network namespace instead of global Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 17/67] mm: gup: stop abusing try_grab_folio Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 18/67] nvme/pci: Add APST quirk for Lenovo N60z laptop Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 19/67] genirq/cpuhotplug: Skip suspended interrupts when restoring affinity Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 20/67] genirq/cpuhotplug: Retry with cpu_online_mask when migration fails Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 21/67] cgroup: Make operations on the cgroup root_list RCU safe Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 22/67] tcp_metrics: optimize tcp_metrics_flush_all() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 23/67] wifi: mac80211: take wiphy lock for MAC addr change Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 24/67] wifi: mac80211: fix change_address deadlock during unregister Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 25/67] fs: Convert to bdev_open_by_dev() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 26/67] jfs: " Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 27/67] jfs: fix log->bdev_handle null ptr deref in lbmStartIO Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 28/67] net: dont dump stack on queue timeout Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 29/67] jfs: fix shift-out-of-bounds in dbJoin Greg Kroah-Hartman
2024-08-15 14:13   ` Dave Kleikamp
2024-08-15 14:19     ` Greg Kroah-Hartman
2024-08-15 16:24       ` Dave Kleikamp
2024-08-15 13:25 ` [PATCH 6.6 30/67] squashfs: squashfs_read_data need to check if the length is 0 Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 31/67] Squashfs: fix variable overflow triggered by sysbot Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 32/67] reiserfs: fix uninit-value in comp_keys Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 33/67] erofs: avoid debugging output for (de)compressed data Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 34/67] net: tls, add test to capture error on large splice Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 35/67] Input: bcm5974 - check endpoint type before starting traffic Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 36/67] quota: Detect loops in quota tree Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 37/67] net:rds: Fix possible deadlock in rds_message_put Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 38/67] net: sctp: fix skb leak in sctp_inq_free() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 39/67] pppoe: Fix memory leak in pppoe_sendmsg() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 40/67] bpf: Replace bpf_lpm_trie_key 0-length array with flexible array Greg Kroah-Hartman
2024-08-15 13:25 ` Greg Kroah-Hartman [this message]
2024-08-15 13:25 ` [PATCH 6.6 42/67] fs: Annotate struct file_handle with __counted_by() and use struct_size() Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 43/67] mISDN: fix MISDN_TIME_STAMP handling Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 44/67] net: add copy_safe_from_sockptr() helper Greg Kroah-Hartman
2024-08-15 13:25 ` [PATCH 6.6 45/67] nfc: llcp: fix nfc_llcp_setsockopt() unsafe copies Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 46/67] Bluetooth: RFCOMM: Fix not validating setsockopt user input Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 47/67] ext4: fold quota accounting into ext4_xattr_inode_lookup_create() Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 48/67] ext4: do not create EA inode under buffer lock Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 49/67] mm/page_table_check: support userfault wr-protect entries Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 50/67] wifi: cfg80211: restrict NL80211_ATTR_TXQ_QUANTUM values Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 51/67] ext4: convert ext4_da_do_write_end() to take a folio Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 52/67] ext4: sanity check for NULL pointer after ext4_force_shutdown Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 53/67] bpf, net: Use DEV_STAT_INC() Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 54/67] f2fs: fix to do sanity check on F2FS_INLINE_DATA flag in inode during GC Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 55/67] f2fs: fix to cover read extent cache access with lock Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 56/67] fou: remove warn in gue_gro_receive on unsupported protocol Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 57/67] jfs: fix null ptr deref in dtInsertEntry Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 58/67] jfs: Fix shift-out-of-bounds in dbDiscardAG Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 59/67] fs/ntfs3: Do copy_to_user out of run_lock Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 60/67] ALSA: usb: Fix UBSAN warning in parse_audio_unit() Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 61/67] binfmt_flat: Fix corruption when not offsetting data start Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 62/67] Revert "jfs: fix shift-out-of-bounds in dbJoin" Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 63/67] Revert "Input: bcm5974 - check endpoint type before starting traffic" Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 64/67] mm/debug_vm_pgtable: drop RANDOM_ORVALUE trick Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 65/67] cgroup: Move rcu_head up near the top of cgroup_root Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 66/67] KVM: arm64: Dont defer TLB invalidation when zapping table entries Greg Kroah-Hartman
2024-08-15 13:26 ` [PATCH 6.6 67/67] KVM: arm64: Dont pass a TLBI level hint " Greg Kroah-Hartman
2024-08-15 19:35 ` [PATCH 6.6 00/67] 6.6.47-rc1 review ChromeOS Kernel Stable Merge
2024-08-15 19:46 ` Peter Schneider
2024-08-15 21:59 ` Florian Fainelli
2024-08-16  8:47 ` Anders Roxell
2024-08-16 11:24 ` Mark Brown
2024-08-16 11:56 ` Takeshi Ogasawara
2024-08-16 19:47 ` Jon Hunter
2024-08-16 20:40 ` Ron Economos

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240815131839.895503593@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=patches@lists.linux.dev \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=syzbot+1fa663a2100308ab6eab@syzkaller.appspotmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.