[PATCH AUTOSEL 5.16 051/217] libbpf: Accommodate DWARF/compiler bug with duplicated structs

public inbox for netdev@vger.kernel.org 
 help / color / mirror / Atom feed

From: Sasha Levin <sashal@kernel•org>
To: linux-kernel@vger•kernel.org, stable@vger•kernel.org
Cc: Andrii Nakryiko <andrii@kernel•org>, Jiri Olsa <jolsa@kernel•org>,
	Daniel Borkmann <daniel@iogearbox•net>,
	Sasha Levin <sashal@kernel•org>,
	ast@kernel•org, netdev@vger•kernel.org, bpf@vger•kernel.org
Subject: [PATCH AUTOSEL 5.16 051/217] libbpf: Accommodate DWARF/compiler bug with duplicated structs
Date: Mon, 17 Jan 2022 21:16:54 -0500	[thread overview]
Message-ID: <20220118021940.1942199-51-sashal@kernel.org> (raw)
In-Reply-To: <20220118021940.1942199-1-sashal@kernel.org>

From: Andrii Nakryiko <andrii@kernel•org>

[ Upstream commit efdd3eb8015e7447095f02a26eaabd164cd18004 ]

According to [0], compilers sometimes might produce duplicate DWARF
definitions for exactly the same struct/union within the same
compilation unit (CU). We've had similar issues with identical arrays
and handled them with a similar workaround in 6b6e6b1d09aa ("libbpf:
Accomodate DWARF/compiler bug with duplicated identical arrays"). Do the
same for struct/union by ensuring that two structs/unions are exactly
the same, down to the integer values of field referenced type IDs.

Solving this more generically (allowing referenced types to be
equivalent, but using different type IDs, all within a single CU)
requires a huge complexity increase to handle many-to-many mappings
between canonidal and candidate type graphs. Before we invest in that,
let's see if this approach handles all the instances of this issue in
practice. Thankfully it's pretty rare, it seems.

  [0] https://lore.kernel.org/bpf/YXr2NFlJTAhHdZqq@krava/

Reported-by: Jiri Olsa <jolsa@kernel•org>
Signed-off-by: Andrii Nakryiko <andrii@kernel•org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox•net>
Link: https://lore.kernel.org/bpf/20211117194114.347675-1-andrii@kernel.org
Signed-off-by: Sasha Levin <sashal@kernel•org>
---
 tools/lib/bpf/btf.c | 45 +++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 41 insertions(+), 4 deletions(-)

diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 7e4c5586bd877..6ab66bd0ecdfb 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -3443,8 +3443,8 @@ static long btf_hash_struct(struct btf_type *t)
 }
 
 /*
- * Check structural compatibility of two FUNC_PROTOs, ignoring referenced type
- * IDs. This check is performed during type graph equivalence check and
+ * Check structural compatibility of two STRUCTs/UNIONs, ignoring referenced
+ * type IDs. This check is performed during type graph equivalence check and
  * referenced types equivalence is checked separately.
  */
 static bool btf_shallow_equal_struct(struct btf_type *t1, struct btf_type *t2)
@@ -3817,6 +3817,31 @@ static int btf_dedup_identical_arrays(struct btf_dedup *d, __u32 id1, __u32 id2)
 	return btf_equal_array(t1, t2);
 }
 
+/* Check if given two types are identical STRUCT/UNION definitions */
+static bool btf_dedup_identical_structs(struct btf_dedup *d, __u32 id1, __u32 id2)
+{
+	const struct btf_member *m1, *m2;
+	struct btf_type *t1, *t2;
+	int n, i;
+
+	t1 = btf_type_by_id(d->btf, id1);
+	t2 = btf_type_by_id(d->btf, id2);
+
+	if (!btf_is_composite(t1) || btf_kind(t1) != btf_kind(t2))
+		return false;
+
+	if (!btf_shallow_equal_struct(t1, t2))
+		return false;
+
+	m1 = btf_members(t1);
+	m2 = btf_members(t2);
+	for (i = 0, n = btf_vlen(t1); i < n; i++, m1++, m2++) {
+		if (m1->type != m2->type)
+			return false;
+	}
+	return true;
+}
+
 /*
  * Check equivalence of BTF type graph formed by candidate struct/union (we'll
  * call it "candidate graph" in this description for brevity) to a type graph
@@ -3928,6 +3953,8 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 
 	hypot_type_id = d->hypot_map[canon_id];
 	if (hypot_type_id <= BTF_MAX_NR_TYPES) {
+		if (hypot_type_id == cand_id)
+			return 1;
 		/* In some cases compiler will generate different DWARF types
 		 * for *identical* array type definitions and use them for
 		 * different fields within the *same* struct. This breaks type
@@ -3936,8 +3963,18 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 		 * types within a single CU. So work around that by explicitly
 		 * allowing identical array types here.
 		 */
-		return hypot_type_id == cand_id ||
-		       btf_dedup_identical_arrays(d, hypot_type_id, cand_id);
+		if (btf_dedup_identical_arrays(d, hypot_type_id, cand_id))
+			return 1;
+		/* It turns out that similar situation can happen with
+		 * struct/union sometimes, sigh... Handle the case where
+		 * structs/unions are exactly the same, down to the referenced
+		 * type IDs. Anything more complicated (e.g., if referenced
+		 * types are different, but equivalent) is *way more*
+		 * complicated and requires a many-to-many equivalence mapping.
+		 */
+		if (btf_dedup_identical_structs(d, hypot_type_id, cand_id))
+			return 1;
+		return 0;
 	}
 
 	if (btf_dedup_hypot_map_add(d, canon_id, cand_id))
-- 
2.34.1

next prev parent reply	other threads:[~2022-01-18  2:23 UTC|newest]

Thread overview: 95+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-18  2:16 [PATCH AUTOSEL 5.16 001/217] Bluetooth: hci_sock: purge socket queues in the destruct() callback Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 002/217] Bluetooth: Fix debugfs entry leak in hci_register_dev() Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 003/217] Bluetooth: Fix memory leak of hci device Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 005/217] Bluetooth: Fix removing adv when processing cmd complete Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 008/217] libbpf: Detect corrupted ELF symbols section Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 009/217] libbpf: Improve sanity checking during BTF fix up Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 010/217] libbpf: Validate that .BTF and .BTF.ext sections contain data Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 012/217] selftests/bpf: Fix memory leaks in btf_type_c_dump() helper Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 013/217] selftests/bpf: Destroy XDP link correctly Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 014/217] selftests/bpf: Fix bpf_object leak in skb_ctx selftest Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 015/217] ar5523: Fix null-ptr-deref with unexpected WDCMSG_TARGET_START reply Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 024/217] ath11k: enable IEEE80211_VHT_EXT_NSS_BW_CAPABLE if NSS ratio enabled Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 025/217] batman-adv: allow netlink usage in unprivileged containers Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 026/217] bpf: Change value of MAX_TAIL_CALL_CNT from 32 to 33 Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 028/217] ath11k: Fix crash caused by uninitialized TX ring Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 034/217] amd: lance: use eth_hw_addr_set() Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 035/217] amd: ni65: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 036/217] amd: a2065/ariadne: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 037/217] amd: hplance: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 038/217] amd: atarilance: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 039/217] amd: mvme147: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 040/217] 8390: smc-ultra: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 041/217] 8390: hydra: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 042/217] 8390: mac8390: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 043/217] 8390: wd: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 044/217] smc9194: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 045/217] lasi_82594: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 046/217] apple: macmace: " Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 047/217] cirrus: mac89x0: " Sasha Levin
2022-01-18 16:56   ` Jakub Kicinski
2022-01-22 18:51     ` Sasha Levin
2022-01-18  2:16 ` Sasha Levin [this message]
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 052/217] ethernet: renesas: Use div64_ul instead of do_div Sasha Levin
2022-01-18  2:16 ` [PATCH AUTOSEL 5.16 053/217] 82596: use eth_hw_addr_set() Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 059/217] ath11k: Fix mon status ring rx tlv processing Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 063/217] mlxsw: pci: Add shutdown method in PCI driver Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 074/217] rtw89: fix potentially access out of range of RF register array Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 075/217] rtw88: add quirk to disable pci caps on HP 250 G7 Notebook PC Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 076/217] mwifiex: Fix skb_over_panic in mwifiex_usb_recv() Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 077/217] rsi: Fix use-after-free in rsi_rx_done_handler() Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 078/217] rsi: Fix out-of-bounds read in rsi_read_pkt() Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 079/217] ath11k: Avoid NULL ptr access during mgmt tx cleanup Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 104/217] iwlwifi: mvm: synchronize with FW after multicast commands Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 105/217] iwlwifi: mvm: avoid clearing a just saved session protection id Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 106/217] iwlwifi: acpi: fix wgds rev 3 size Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 108/217] net: dsa: ocelot: felix: Remove requirement for PCS in felix devices Sasha Levin
2022-01-18 12:23   ` Vladimir Oltean
2022-01-22 19:02     ` Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 109/217] ath11k: avoid deadlock by change ieee80211_queue_work for regd_update_work Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 110/217] ath10k: Fix tx hanging Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 111/217] rtw89: don't kick off TX DMA if failed to write skb Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 112/217] net-sysfs: update the queue counts in the unregistration path Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 113/217] ath10k: drop beacon and probe response which leak from other channel Sasha Levin
2022-01-18  2:17 ` [PATCH AUTOSEL 5.16 114/217] net: phy: prefer 1000baseT over 1000baseKX Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 118/217] net: Enable neighbor sysctls that is save for userns root Sasha Levin
2022-01-18 16:59   ` Jakub Kicinski
2022-01-19  3:46     ` David Ahern
2022-01-22 19:13       ` Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 120/217] ath11k: Avoid false DEADLOCK warning reported by lockdep Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 127/217] net: bonding: debug: avoid printing debug logs when bond is not notifying peers Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 129/217] bpf: Do not WARN in bpf_warn_invalid_xdp_action() Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 152/217] mt76: mt7615: fix possible deadlock while mt7615_register_ext_phy() Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 153/217] mt76: mt7915: fix SMPS operation fail Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 154/217] mt76: connac: fix a theoretical NULL pointer dereference in mt76_connac_get_phy_mode Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 155/217] mt76: do not pass the received frame with decryption error Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 156/217] mt76: mt7615: improve wmm index allocation Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 157/217] mt76: mt7921: fix network buffer leak by txs missing Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 158/217] ath9k_htc: fix NULL pointer dereference at ath9k_htc_rxep() Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 159/217] ath9k_htc: fix NULL pointer dereference at ath9k_htc_tx_get_packet() Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 160/217] ath9k: Fix out-of-bound memcpy in ath9k_hif_usb_rx_stream Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 161/217] rtw88: 8822c: update rx settings to prevent potential hw deadlock Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 163/217] iwlwifi: recognize missing PNVM data and then log filename Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 164/217] iwlwifi: fix leaks/bad data after failed firmware load Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 165/217] iwlwifi: remove module loading failure message Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 166/217] iwlwifi: mvm: Fix calculation of frame length Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 167/217] iwlwifi: mvm: fix AUX ROC removal Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 168/217] iwlwifi: pcie: make sure prph_info is set when treating wakeup IRQ Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 172/217] ath11k: Fix napi related hang Sasha Levin
2022-01-18  2:18 ` [PATCH AUTOSEL 5.16 176/217] xfrm: rate limit SA mapping change message to user space Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 191/217] net/mlx5: DR, Fix error flow in creating matcher Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 194/217] net: mdio: Demote probed message to debug print Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 195/217] mac80211: allow non-standard VHT MCS-10/11 Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 198/217] can: do not increase rx statistics when generating a CAN rx error message frame Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 199/217] bpf/selftests: Fix namespace mount setup in tc_redirect Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 200/217] mlxsw: pci: Avoid flow control for EMAD packets Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 201/217] net: phy: marvell: configure RGMII delays for 88E1118 Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 202/217] net: gemini: allow any RGMII interface mode Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 203/217] net: dsa: hold rtnl_mutex when calling dsa_master_{setup,teardown} Sasha Levin
2022-01-18 12:13   ` Vladimir Oltean
2022-01-22 19:17     ` Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 209/217] net/mlx5: Update log_max_qp value to FW max capability Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 210/217] net/mlx5e: Unblock setting vid 0 for VF in case PF isn't eswitch manager Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 213/217] can: flexcan: allow to change quirks at runtime Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 214/217] can: flexcan: rename RX modes Sasha Levin
2022-01-18  2:19 ` [PATCH AUTOSEL 5.16 215/217] can: flexcan: add more quirks to describe RX path capabilities Sasha Levin

find likely ancestor, descendant, or conflicting patches for this message:
( dfblob:7e4c5586bd87 dfblob:6ab66bd0ecdf )
 OR (
bs:"[PATCH AUTOSEL 5.16 051/217] libbpf: Accommodate DWARF/compiler bug with duplicated structs" )
	(help)

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=20220118021940.1942199-51-sashal@kernel.org \
    --to=sashal@kernel$(echo .)org \
    --cc=andrii@kernel$(echo .)org \
    --cc=ast@kernel$(echo .)org \
    --cc=bpf@vger$(echo .)kernel.org \
    --cc=daniel@iogearbox$(echo .)net \
    --cc=jolsa@kernel$(echo .)org \
    --cc=linux-kernel@vger$(echo .)kernel.org \
    --cc=netdev@vger$(echo .)kernel.org \
    --cc=stable@vger$(echo .)kernel.org \
    /path/to/YOUR_REPLY

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

Be sure your reply has a Subject: header at the top and a blank line before the message body.

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox