From 01258b62c62710297dab4e2b72f46e01be392cc6 Mon Sep 17 00:00:00 2001 From: Lukas Bulwahn Date: Fri, 9 Dec 2022 10:59:37 +0100 Subject: [PATCH 001/112] wifi: ti: remove obsolete lines in the Makefile Commit 06463f6e98df ("wifi: wl1251: drop support for platform data") removes TI WiLink platform data, but leaves some dead lines in the Makefile. Remove these obsolete lines in the Makefile. Signed-off-by: Lukas Bulwahn Reviewed-by: Dmitry Torokhov Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221209095937.17773-1-lukas.bulwahn@gmail.com --- drivers/net/wireless/ti/Makefile | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/wireless/ti/Makefile b/drivers/net/wireless/ti/Makefile index 0530dd744275..05ee016594f8 100644 --- a/drivers/net/wireless/ti/Makefile +++ b/drivers/net/wireless/ti/Makefile @@ -3,6 +3,3 @@ obj-$(CONFIG_WLCORE) += wlcore/ obj-$(CONFIG_WL12XX) += wl12xx/ obj-$(CONFIG_WL1251) += wl1251/ obj-$(CONFIG_WL18XX) += wl18xx/ - -# small builtin driver bit -obj-$(CONFIG_WILINK_PLATFORM_DATA) += wilink_platform_data.o From cc074822465d18a2d39e0b3e2b48b6766a568db2 Mon Sep 17 00:00:00 2001 From: Hou Tao Date: Sat, 17 Dec 2022 14:21:44 +0800 Subject: [PATCH 002/112] bpf: Define sock security related BTF IDs under CONFIG_SECURITY_NETWORK There are warnings reported from resolve_btfids when building vmlinux with CONFIG_SECURITY_NETWORK disabled: WARN: resolve_btfids: unresolved symbol bpf_lsm_sk_free_security WARN: resolve_btfids: unresolved symbol bpf_lsm_sk_alloc_security So only define BTF IDs for these LSM hooks when CONFIG_SECURITY_NETWORK is enabled. Fixes: c0c852dd1876 ("bpf: Do not mark certain LSM hook arguments as trusted") Signed-off-by: Hou Tao Signed-off-by: Daniel Borkmann Acked-by: Yonghong Song Link: https://lore.kernel.org/bpf/20221217062144.2507222-1-houtao@huaweicloud.com --- kernel/bpf/bpf_lsm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/bpf/bpf_lsm.c b/kernel/bpf/bpf_lsm.c index 9ea42a45da47..a4a41ee3e80b 100644 --- a/kernel/bpf/bpf_lsm.c +++ b/kernel/bpf/bpf_lsm.c @@ -351,8 +351,10 @@ BTF_ID(func, bpf_lsm_bpf_prog_alloc_security) BTF_ID(func, bpf_lsm_bpf_prog_free_security) BTF_ID(func, bpf_lsm_file_alloc_security) BTF_ID(func, bpf_lsm_file_free_security) +#ifdef CONFIG_SECURITY_NETWORK BTF_ID(func, bpf_lsm_sk_alloc_security) BTF_ID(func, bpf_lsm_sk_free_security) +#endif /* CONFIG_SECURITY_NETWORK */ BTF_ID(func, bpf_lsm_task_free) BTF_SET_END(untrusted_lsm_hooks) From 1c4c0b28b517d778d37900deedfe91088839f07a Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 13 Dec 2022 23:15:04 +0200 Subject: [PATCH 003/112] wifi: iwlwifi: fw: skip PPAG for JF MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit For JF RFs we don't support PPAG, but many firmware images lie about it. Always skip support for JF to avoid firmware errors when sending the command. Reported-and-tested-by: Íñigo Huguet Link: https://lore.kernel.org/linux-wireless/CACT4oufQsqHGp6bah2c4+jPn2wG1oZqY=UKa_TmPx=F6Lxng8Q@mail.gmail.com Signed-off-by: Johannes Berg Signed-off-by: Gregory Greenman Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221213225723.2a43415d8990.I9ac210740a45b41f1b2e15274e1daf4284f2808a@changeid --- drivers/net/wireless/intel/iwlwifi/fw/acpi.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c index e6d64152c81a..a02e5a67b706 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c @@ -1106,6 +1106,11 @@ int iwl_read_ppag_table(struct iwl_fw_runtime *fwrt, union iwl_ppag_table_cmd *c int i, j, num_sub_bands; s8 *gain; + /* many firmware images for JF lie about this */ + if (CSR_HW_RFID_TYPE(fwrt->trans->hw_rf_id) == + CSR_HW_RFID_TYPE(CSR_HW_RF_ID_TYPE_JF)) + return -EOPNOTSUPP; + if (!fw_has_capa(&fwrt->fw->ucode_capa, IWL_UCODE_TLV_CAPA_SET_PPAG)) { IWL_DEBUG_RADIO(fwrt, "PPAG capability not supported by FW, command not sent.\n"); From 37fc9ad1617a303bbfd28870eb25aaa4766e79ab Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 15 Dec 2022 17:31:10 +0100 Subject: [PATCH 004/112] wifi: mt76: mt7996: select CONFIG_RELAY Without CONFIG_RELAY, the driver fails to link: ERROR: modpost: "relay_flush" [drivers/net/wireless/mediatek/mt76/mt7996/mt7996e.ko] undefined! ERROR: modpost: "relay_switch_subbuf" [drivers/net/wireless/mediatek/mt76/mt7996/mt7996e.ko] undefined! ERROR: modpost: "relay_open" [drivers/net/wireless/mediatek/mt76/mt7996/mt7996e.ko] undefined! ERROR: modpost: "relay_reset" [drivers/net/wireless/mediatek/mt76/mt7996/mt7996e.ko] undefined! ERROR: modpost: "relay_file_operations" [drivers/net/wireless/mediatek/mt76/mt7996/mt7996e.ko] undefined! The same change was done in mt7915 for the corresponding copy of the code. Fixes: 98686cd21624 ("wifi: mt76: mt7996: add driver for MediaTek Wi-Fi 7 (802.11be) devices") See-also: 988845c9361a ("mt76: mt7915: add support for passing chip/firmware debug data to user space") Signed-off-by: Arnd Bergmann Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221215163133.4152299-1-arnd@kernel.org --- drivers/net/wireless/mediatek/mt76/mt7996/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig index 5c5fc569e6d5..79fb47a73c91 100644 --- a/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig +++ b/drivers/net/wireless/mediatek/mt76/mt7996/Kconfig @@ -2,6 +2,7 @@ config MT7996E tristate "MediaTek MT7996 (PCIe) support" select MT76_CONNAC_LIB + select RELAY depends on MAC80211 depends on PCI help From b7dc753fe33a707379e2254317794a4dad6c0fe2 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 15 Dec 2022 17:55:42 +0100 Subject: [PATCH 005/112] wifi: ath9k: use proper statements in conditionals MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A previous cleanup patch accidentally broke some conditional expressions by replacing the safe "do {} while (0)" constructs with empty macros. gcc points this out when extra warnings are enabled: drivers/net/wireless/ath/ath9k/hif_usb.c: In function 'ath9k_skb_queue_complete': drivers/net/wireless/ath/ath9k/hif_usb.c:251:57: error: suggest braces around empty body in an 'else' statement [-Werror=empty-body] 251 | TX_STAT_INC(hif_dev, skb_failed); Make both sets of macros proper expressions again. Fixes: d7fc76039b74 ("ath9k: htc: clean up statistics macros") Signed-off-by: Arnd Bergmann Acked-by: Toke Høiland-Jørgensen Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221215165553.1950307-1-arnd@kernel.org --- drivers/net/wireless/ath/ath9k/htc.h | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/net/wireless/ath/ath9k/htc.h b/drivers/net/wireless/ath/ath9k/htc.h index 30f0765fb9fd..237f4ec2cffd 100644 --- a/drivers/net/wireless/ath/ath9k/htc.h +++ b/drivers/net/wireless/ath/ath9k/htc.h @@ -327,9 +327,9 @@ static inline struct ath9k_htc_tx_ctl *HTC_SKB_CB(struct sk_buff *skb) } #ifdef CONFIG_ATH9K_HTC_DEBUGFS -#define __STAT_SAFE(hif_dev, expr) ((hif_dev)->htc_handle->drv_priv ? (expr) : 0) -#define CAB_STAT_INC(priv) ((priv)->debug.tx_stats.cab_queued++) -#define TX_QSTAT_INC(priv, q) ((priv)->debug.tx_stats.queue_stats[q]++) +#define __STAT_SAFE(hif_dev, expr) do { ((hif_dev)->htc_handle->drv_priv ? (expr) : 0); } while (0) +#define CAB_STAT_INC(priv) do { ((priv)->debug.tx_stats.cab_queued++); } while (0) +#define TX_QSTAT_INC(priv, q) do { ((priv)->debug.tx_stats.queue_stats[q]++); } while (0) #define TX_STAT_INC(hif_dev, c) \ __STAT_SAFE((hif_dev), (hif_dev)->htc_handle->drv_priv->debug.tx_stats.c++) @@ -378,10 +378,10 @@ void ath9k_htc_get_et_stats(struct ieee80211_hw *hw, struct ethtool_stats *stats, u64 *data); #else -#define TX_STAT_INC(hif_dev, c) -#define TX_STAT_ADD(hif_dev, c, a) -#define RX_STAT_INC(hif_dev, c) -#define RX_STAT_ADD(hif_dev, c, a) +#define TX_STAT_INC(hif_dev, c) do { } while (0) +#define TX_STAT_ADD(hif_dev, c, a) do { } while (0) +#define RX_STAT_INC(hif_dev, c) do { } while (0) +#define RX_STAT_ADD(hif_dev, c, a) do { } while (0) #define CAB_STAT_INC(priv) #define TX_QSTAT_INC(priv, c) From 54c3f1a81421f85e60ae2eaae7be3727a09916ee Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Mon, 19 Dec 2022 16:47:00 -0800 Subject: [PATCH 006/112] bpf: pull before calling skb_postpull_rcsum() Anand hit a BUG() when pulling off headers on egress to a SW tunnel. We get to skb_checksum_help() with an invalid checksum offset (commit d7ea0d9df2a6 ("net: remove two BUG() from skb_checksum_help()") converted those BUGs to WARN_ONs()). He points out oddness in how skb_postpull_rcsum() gets used. Indeed looks like we should pull before "postpull", otherwise the CHECKSUM_PARTIAL fixup from skb_postpull_rcsum() will not be able to do its job: if (skb->ip_summed == CHECKSUM_PARTIAL && skb_checksum_start_offset(skb) < 0) skb->ip_summed = CHECKSUM_NONE; Reported-by: Anand Parthasarathy Fixes: 6578171a7ff0 ("bpf: add bpf_skb_change_proto helper") Signed-off-by: Jakub Kicinski Acked-by: Stanislav Fomichev Link: https://lore.kernel.org/r/20221220004701.402165-1-kuba@kernel.org Signed-off-by: Martin KaFai Lau --- net/core/filter.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/core/filter.c b/net/core/filter.c index 929358677183..43cc1fe58a2c 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -3180,15 +3180,18 @@ static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len) static int bpf_skb_generic_pop(struct sk_buff *skb, u32 off, u32 len) { + void *old_data; + /* skb_ensure_writable() is not needed here, as we're * already working on an uncloned skb. */ if (unlikely(!pskb_may_pull(skb, off + len))) return -ENOMEM; - skb_postpull_rcsum(skb, skb->data + off, len); - memmove(skb->data + len, skb->data, off); + old_data = skb->data; __skb_pull(skb, len); + skb_postpull_rcsum(skb, old_data + off, len); + memmove(skb->data, old_data, off); return 0; } From 5eb119da94ac5d67a31eaa869621dc6e25eb125e Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 15 Dec 2022 15:16:33 +0100 Subject: [PATCH 007/112] netfilter: conntrack: fix ipv6 exthdr error check smatch warnings: net/netfilter/nf_conntrack_proto.c:167 nf_confirm() warn: unsigned 'protoff' is never less than zero. We need to check if ipv6_skip_exthdr() returned an error, but protoff is unsigned. Use a signed integer for this. Fixes: a70e483460d5 ("netfilter: conntrack: merge ipv4+ipv6 confirm functions") Reported-by: kernel test robot Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_proto.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 99323fb12d0f..ccef340be575 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -141,6 +141,7 @@ unsigned int nf_confirm(void *priv, struct nf_conn *ct; bool seqadj_needed; __be16 frag_off; + int start; u8 pnum; ct = nf_ct_get(skb, &ctinfo); @@ -163,9 +164,11 @@ unsigned int nf_confirm(void *priv, break; case NFPROTO_IPV6: pnum = ipv6_hdr(skb)->nexthdr; - protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &pnum, &frag_off); - if (protoff < 0 || (frag_off & htons(~0x7)) != 0) + start = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &pnum, &frag_off); + if (start < 0 || (frag_off & htons(~0x7)) != 0) return nf_conntrack_confirm(skb); + + protoff = start; break; default: return nf_conntrack_confirm(skb); From bed4a63ea4ae77cfe5aae004ef87379f0655260a Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 19 Dec 2022 20:07:52 +0100 Subject: [PATCH 008/112] netfilter: nf_tables: consolidate set description Add the following fields to the set description: - key type - data type - object type - policy - gc_int: garbage collection interval) - timeout: element timeout This prepares for stricter set type checks on updates in a follow up patch. Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 12 +++++++ net/netfilter/nf_tables_api.c | 58 +++++++++++++++---------------- 2 files changed, 40 insertions(+), 30 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index e69ce23566ea..4957b4775757 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -312,17 +312,29 @@ struct nft_set_iter { /** * struct nft_set_desc - description of set elements * + * @ktype: key type * @klen: key length + * @dtype: data type * @dlen: data length + * @objtype: object type + * @flags: flags * @size: number of set elements + * @policy: set policy + * @gc_int: garbage collector interval * @field_len: length of each field in concatenation, bytes * @field_count: number of concatenated fields in element * @expr: set must support for expressions */ struct nft_set_desc { + u32 ktype; unsigned int klen; + u32 dtype; unsigned int dlen; + u32 objtype; unsigned int size; + u32 policy; + u32 gc_int; + u64 timeout; u8 field_len[NFT_REG32_COUNT]; u8 field_count; bool expr; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 832b881f7c17..1deecc1a6c00 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -3780,8 +3780,7 @@ static bool nft_set_ops_candidate(const struct nft_set_type *type, u32 flags) static const struct nft_set_ops * nft_select_set_ops(const struct nft_ctx *ctx, const struct nlattr * const nla[], - const struct nft_set_desc *desc, - enum nft_set_policies policy) + const struct nft_set_desc *desc) { struct nftables_pernet *nft_net = nft_pernet(ctx->net); const struct nft_set_ops *ops, *bops; @@ -3810,7 +3809,7 @@ nft_select_set_ops(const struct nft_ctx *ctx, if (!ops->estimate(desc, flags, &est)) continue; - switch (policy) { + switch (desc->policy) { case NFT_SET_POL_PERFORMANCE: if (est.lookup < best.lookup) break; @@ -4392,7 +4391,6 @@ static int nf_tables_set_desc_parse(struct nft_set_desc *desc, static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, const struct nlattr * const nla[]) { - u32 ktype, dtype, flags, policy, gc_int, objtype; struct netlink_ext_ack *extack = info->extack; u8 genmask = nft_genmask_next(info->net); u8 family = info->nfmsg->nfgen_family; @@ -4405,10 +4403,10 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, struct nft_set *set; struct nft_ctx ctx; size_t alloc_size; - u64 timeout; char *name; int err, i; u16 udlen; + u32 flags; u64 size; if (nla[NFTA_SET_TABLE] == NULL || @@ -4419,10 +4417,10 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, memset(&desc, 0, sizeof(desc)); - ktype = NFT_DATA_VALUE; + desc.ktype = NFT_DATA_VALUE; if (nla[NFTA_SET_KEY_TYPE] != NULL) { - ktype = ntohl(nla_get_be32(nla[NFTA_SET_KEY_TYPE])); - if ((ktype & NFT_DATA_RESERVED_MASK) == NFT_DATA_RESERVED_MASK) + desc.ktype = ntohl(nla_get_be32(nla[NFTA_SET_KEY_TYPE])); + if ((desc.ktype & NFT_DATA_RESERVED_MASK) == NFT_DATA_RESERVED_MASK) return -EINVAL; } @@ -4447,17 +4445,17 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, return -EOPNOTSUPP; } - dtype = 0; + desc.dtype = 0; if (nla[NFTA_SET_DATA_TYPE] != NULL) { if (!(flags & NFT_SET_MAP)) return -EINVAL; - dtype = ntohl(nla_get_be32(nla[NFTA_SET_DATA_TYPE])); - if ((dtype & NFT_DATA_RESERVED_MASK) == NFT_DATA_RESERVED_MASK && - dtype != NFT_DATA_VERDICT) + desc.dtype = ntohl(nla_get_be32(nla[NFTA_SET_DATA_TYPE])); + if ((desc.dtype & NFT_DATA_RESERVED_MASK) == NFT_DATA_RESERVED_MASK && + desc.dtype != NFT_DATA_VERDICT) return -EINVAL; - if (dtype != NFT_DATA_VERDICT) { + if (desc.dtype != NFT_DATA_VERDICT) { if (nla[NFTA_SET_DATA_LEN] == NULL) return -EINVAL; desc.dlen = ntohl(nla_get_be32(nla[NFTA_SET_DATA_LEN])); @@ -4472,34 +4470,34 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, if (!(flags & NFT_SET_OBJECT)) return -EINVAL; - objtype = ntohl(nla_get_be32(nla[NFTA_SET_OBJ_TYPE])); - if (objtype == NFT_OBJECT_UNSPEC || - objtype > NFT_OBJECT_MAX) + desc.objtype = ntohl(nla_get_be32(nla[NFTA_SET_OBJ_TYPE])); + if (desc.objtype == NFT_OBJECT_UNSPEC || + desc.objtype > NFT_OBJECT_MAX) return -EOPNOTSUPP; } else if (flags & NFT_SET_OBJECT) return -EINVAL; else - objtype = NFT_OBJECT_UNSPEC; + desc.objtype = NFT_OBJECT_UNSPEC; - timeout = 0; + desc.timeout = 0; if (nla[NFTA_SET_TIMEOUT] != NULL) { if (!(flags & NFT_SET_TIMEOUT)) return -EINVAL; - err = nf_msecs_to_jiffies64(nla[NFTA_SET_TIMEOUT], &timeout); + err = nf_msecs_to_jiffies64(nla[NFTA_SET_TIMEOUT], &desc.timeout); if (err) return err; } - gc_int = 0; + desc.gc_int = 0; if (nla[NFTA_SET_GC_INTERVAL] != NULL) { if (!(flags & NFT_SET_TIMEOUT)) return -EINVAL; - gc_int = ntohl(nla_get_be32(nla[NFTA_SET_GC_INTERVAL])); + desc.gc_int = ntohl(nla_get_be32(nla[NFTA_SET_GC_INTERVAL])); } - policy = NFT_SET_POL_PERFORMANCE; + desc.policy = NFT_SET_POL_PERFORMANCE; if (nla[NFTA_SET_POLICY] != NULL) - policy = ntohl(nla_get_be32(nla[NFTA_SET_POLICY])); + desc.policy = ntohl(nla_get_be32(nla[NFTA_SET_POLICY])); if (nla[NFTA_SET_DESC] != NULL) { err = nf_tables_set_desc_parse(&desc, nla[NFTA_SET_DESC]); @@ -4544,7 +4542,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, if (!(info->nlh->nlmsg_flags & NLM_F_CREATE)) return -ENOENT; - ops = nft_select_set_ops(&ctx, nla, &desc, policy); + ops = nft_select_set_ops(&ctx, nla, &desc); if (IS_ERR(ops)) return PTR_ERR(ops); @@ -4584,18 +4582,18 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, set->table = table; write_pnet(&set->net, net); set->ops = ops; - set->ktype = ktype; + set->ktype = desc.ktype; set->klen = desc.klen; - set->dtype = dtype; - set->objtype = objtype; + set->dtype = desc.dtype; + set->objtype = desc.objtype; set->dlen = desc.dlen; set->flags = flags; set->size = desc.size; - set->policy = policy; + set->policy = desc.policy; set->udlen = udlen; set->udata = udata; - set->timeout = timeout; - set->gc_int = gc_int; + set->timeout = desc.timeout; + set->gc_int = desc.gc_int; set->field_count = desc.field_count; for (i = 0; i < desc.field_count; i++) From a8fe4154fa5a1bae590b243ed60f871e5a5e1378 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 19 Dec 2022 18:00:10 +0100 Subject: [PATCH 009/112] netfilter: nf_tables: add function to create set stateful expressions Add a helper function to allocate and initialize the stateful expressions that are defined in a set. This patch allows to reuse this code from the set update path, to check that type of the update matches the existing set in the kernel. Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 106 ++++++++++++++++++++++------------ 1 file changed, 68 insertions(+), 38 deletions(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 1deecc1a6c00..b9b0ae29f5f6 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -4388,6 +4388,59 @@ static int nf_tables_set_desc_parse(struct nft_set_desc *desc, return err; } +static int nft_set_expr_alloc(struct nft_ctx *ctx, struct nft_set *set, + const struct nlattr * const *nla, + struct nft_expr **exprs, int *num_exprs, + u32 flags) +{ + struct nft_expr *expr; + int err, i; + + if (nla[NFTA_SET_EXPR]) { + expr = nft_set_elem_expr_alloc(ctx, set, nla[NFTA_SET_EXPR]); + if (IS_ERR(expr)) { + err = PTR_ERR(expr); + goto err_set_expr_alloc; + } + exprs[0] = expr; + (*num_exprs)++; + } else if (nla[NFTA_SET_EXPRESSIONS]) { + struct nlattr *tmp; + int left; + + if (!(flags & NFT_SET_EXPR)) { + err = -EINVAL; + goto err_set_expr_alloc; + } + i = 0; + nla_for_each_nested(tmp, nla[NFTA_SET_EXPRESSIONS], left) { + if (i == NFT_SET_EXPR_MAX) { + err = -E2BIG; + goto err_set_expr_alloc; + } + if (nla_type(tmp) != NFTA_LIST_ELEM) { + err = -EINVAL; + goto err_set_expr_alloc; + } + expr = nft_set_elem_expr_alloc(ctx, set, tmp); + if (IS_ERR(expr)) { + err = PTR_ERR(expr); + goto err_set_expr_alloc; + } + exprs[i++] = expr; + (*num_exprs)++; + } + } + + return 0; + +err_set_expr_alloc: + for (i = 0; i < *num_exprs; i++) + nft_expr_destroy(ctx, exprs[i]); + + return err; +} + static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, const struct nlattr * const nla[]) { @@ -4395,7 +4448,6 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, u8 genmask = nft_genmask_next(info->net); u8 family = info->nfmsg->nfgen_family; const struct nft_set_ops *ops; - struct nft_expr *expr = NULL; struct net *net = info->net; struct nft_set_desc desc; struct nft_table *table; @@ -4403,6 +4455,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, struct nft_set *set; struct nft_ctx ctx; size_t alloc_size; + int num_exprs = 0; char *name; int err, i; u16 udlen; @@ -4529,6 +4582,8 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, return PTR_ERR(set); } } else { + struct nft_expr *exprs[NFT_SET_EXPR_MAX] = {}; + if (info->nlh->nlmsg_flags & NLM_F_EXCL) { NL_SET_BAD_ATTR(extack, nla[NFTA_SET_NAME]); return -EEXIST; @@ -4536,6 +4591,13 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, if (info->nlh->nlmsg_flags & NLM_F_REPLACE) return -EOPNOTSUPP; + err = nft_set_expr_alloc(&ctx, set, nla, exprs, &num_exprs, flags); + if (err < 0) + return err; + + for (i = 0; i < num_exprs; i++) + nft_expr_destroy(&ctx, exprs[i]); + return 0; } @@ -4603,43 +4665,11 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, if (err < 0) goto err_set_init; - if (nla[NFTA_SET_EXPR]) { - expr = nft_set_elem_expr_alloc(&ctx, set, nla[NFTA_SET_EXPR]); - if (IS_ERR(expr)) { - err = PTR_ERR(expr); - goto err_set_expr_alloc; - } - set->exprs[0] = expr; - set->num_exprs++; - } else if (nla[NFTA_SET_EXPRESSIONS]) { - struct nft_expr *expr; - struct nlattr *tmp; - int left; - - if (!(flags & NFT_SET_EXPR)) { - err = -EINVAL; - goto err_set_expr_alloc; - } - i = 0; - nla_for_each_nested(tmp, nla[NFTA_SET_EXPRESSIONS], left) { - if (i == NFT_SET_EXPR_MAX) { - err = -E2BIG; - goto err_set_expr_alloc; - } - if (nla_type(tmp) != NFTA_LIST_ELEM) { - err = -EINVAL; - goto err_set_expr_alloc; - } - expr = nft_set_elem_expr_alloc(&ctx, set, tmp); - if (IS_ERR(expr)) { - err = PTR_ERR(expr); - goto err_set_expr_alloc; - } - set->exprs[i++] = expr; - set->num_exprs++; - } - } + err = nft_set_expr_alloc(&ctx, set, nla, set->exprs, &num_exprs, flags); + if (err < 0) + goto err_set_destroy; + set->num_exprs = num_exprs; set->handle = nf_tables_alloc_handle(table); err = nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set); @@ -4653,7 +4683,7 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, err_set_expr_alloc: for (i = 0; i < set->num_exprs; i++) nft_expr_destroy(&ctx, set->exprs[i]); - +err_set_destroy: ops->destroy(set); err_set_init: kfree(set->name); From f6594c372afd5cec8b1e9ee9ea8f8819d59c6fb1 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 19 Dec 2022 20:09:00 +0100 Subject: [PATCH 010/112] netfilter: nf_tables: perform type checking for existing sets If a ruleset declares a set name that matches an existing set in the kernel, then validate that this declaration really refers to the same set, otherwise bail out with EEXIST. Currently, the kernel reports success when adding a set that already exists in the kernel. This usually results in EINVAL errors at a later stage, when the user adds elements to the set, if the set declaration mismatches the existing set representation in the kernel. Add a new function to check that the set declaration really refers to the same existing set in the kernel. Fixes: 96518518cc41 ("netfilter: add nftables") Reported-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_tables_api.c | 36 ++++++++++++++++++++++++++++++++++- 1 file changed, 35 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index b9b0ae29f5f6..319887f4d3ef 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -4441,6 +4441,34 @@ err_set_expr_alloc: return err; } +static bool nft_set_is_same(const struct nft_set *set, + const struct nft_set_desc *desc, + struct nft_expr *exprs[], u32 num_exprs, u32 flags) +{ + int i; + + if (set->ktype != desc->ktype || + set->dtype != desc->dtype || + set->flags != flags || + set->klen != desc->klen || + set->dlen != desc->dlen || + set->field_count != desc->field_count || + set->num_exprs != num_exprs) + return false; + + for (i = 0; i < desc->field_count; i++) { + if (set->field_len[i] != desc->field_len[i]) + return false; + } + + for (i = 0; i < num_exprs; i++) { + if (set->exprs[i]->ops != exprs[i]->ops) + return false; + } + + return true; +} + static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, const struct nlattr * const nla[]) { @@ -4595,10 +4623,16 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, if (err < 0) return err; + err = 0; + if (!nft_set_is_same(set, &desc, exprs, num_exprs, flags)) { + NL_SET_BAD_ATTR(extack, nla[NFTA_SET_NAME]); + err = -EEXIST; + } + for (i = 0; i < num_exprs; i++) nft_expr_destroy(&ctx, exprs[i]); - return 0; + return err; } if (!(info->nlh->nlmsg_flags & NLM_F_CREATE)) From 70a00e2f1dbae11dc3444444c6bd7555763d8421 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Wed, 21 Dec 2022 10:56:53 -0800 Subject: [PATCH 011/112] selftests/bpf: Test bpf_skb_adjust_room on CHECKSUM_PARTIAL When the bpf_skb_adjust_room() shrinks the skb such that its csum_start is invalid, the skb->ip_summed should be reset from CHECKSUM_PARTIAL to CHECKSUM_NONE. The commit 54c3f1a81421 ("bpf: pull before calling skb_postpull_rcsum()") fixed it. This patch adds a test to ensure the skb->ip_summed changed from CHECKSUM_PARTIAL to CHECKSUM_NONE after bpf_skb_adjust_room(). Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann Acked-by: Stanislav Fomichev Link: https://lore.kernel.org/bpf/20221221185653.1589961-1-martin.lau@linux.dev --- tools/testing/selftests/bpf/DENYLIST.s390x | 1 + .../selftests/bpf/prog_tests/decap_sanity.c | 85 +++++++++++++++++++ .../selftests/bpf/progs/bpf_tracing_net.h | 6 ++ .../selftests/bpf/progs/decap_sanity.c | 68 +++++++++++++++ 4 files changed, 160 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/decap_sanity.c create mode 100644 tools/testing/selftests/bpf/progs/decap_sanity.c diff --git a/tools/testing/selftests/bpf/DENYLIST.s390x b/tools/testing/selftests/bpf/DENYLIST.s390x index 585fcf73c731..3fc3e54b19aa 100644 --- a/tools/testing/selftests/bpf/DENYLIST.s390x +++ b/tools/testing/selftests/bpf/DENYLIST.s390x @@ -14,6 +14,7 @@ cgrp_kfunc # JIT does not support calling kernel f cgrp_local_storage # prog_attach unexpected error: -524 (trampoline) core_read_macros # unknown func bpf_probe_read#4 (overlapping) d_path # failed to auto-attach program 'prog_stat': -524 (trampoline) +decap_sanity # JIT does not support calling kernel function (kfunc) deny_namespace # failed to attach: ERROR: strerror_r(-524)=22 (trampoline) dummy_st_ops # test_run unexpected error: -524 (errno 524) (trampoline) fentry_fexit # fentry attach failed: -524 (trampoline) diff --git a/tools/testing/selftests/bpf/prog_tests/decap_sanity.c b/tools/testing/selftests/bpf/prog_tests/decap_sanity.c new file mode 100644 index 000000000000..0b2f73b88c53 --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/decap_sanity.c @@ -0,0 +1,85 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2022 Meta Platforms, Inc. and affiliates. */ + +#include +#include +#include +#include + +#include "test_progs.h" +#include "network_helpers.h" +#include "decap_sanity.skel.h" + +#define SYS(fmt, ...) \ + ({ \ + char cmd[1024]; \ + snprintf(cmd, sizeof(cmd), fmt, ##__VA_ARGS__); \ + if (!ASSERT_OK(system(cmd), cmd)) \ + goto fail; \ + }) + +#define NS_TEST "decap_sanity_ns" +#define IPV6_IFACE_ADDR "face::1" +#define UDP_TEST_PORT 7777 + +void test_decap_sanity(void) +{ + LIBBPF_OPTS(bpf_tc_hook, qdisc_hook, .attach_point = BPF_TC_EGRESS); + LIBBPF_OPTS(bpf_tc_opts, tc_attach); + struct nstoken *nstoken = NULL; + struct decap_sanity *skel; + struct sockaddr_in6 addr; + socklen_t addrlen; + char buf[128] = {}; + int sockfd, err; + + skel = decap_sanity__open_and_load(); + if (!ASSERT_OK_PTR(skel, "skel open_and_load")) + return; + + SYS("ip netns add %s", NS_TEST); + SYS("ip -net %s -6 addr add %s/128 dev lo nodad", NS_TEST, IPV6_IFACE_ADDR); + SYS("ip -net %s link set dev lo up", NS_TEST); + + nstoken = open_netns(NS_TEST); + if (!ASSERT_OK_PTR(nstoken, "open_netns")) + goto fail; + + qdisc_hook.ifindex = if_nametoindex("lo"); + if (!ASSERT_GT(qdisc_hook.ifindex, 0, "if_nametoindex lo")) + goto fail; + + err = bpf_tc_hook_create(&qdisc_hook); + if (!ASSERT_OK(err, "create qdisc hook")) + goto fail; + + tc_attach.prog_fd = bpf_program__fd(skel->progs.decap_sanity); + err = bpf_tc_attach(&qdisc_hook, &tc_attach); + if (!ASSERT_OK(err, "attach filter")) + goto fail; + + addrlen = sizeof(addr); + err = make_sockaddr(AF_INET6, IPV6_IFACE_ADDR, UDP_TEST_PORT, + (void *)&addr, &addrlen); + if (!ASSERT_OK(err, "make_sockaddr")) + goto fail; + sockfd = socket(AF_INET6, SOCK_DGRAM, 0); + if (!ASSERT_NEQ(sockfd, -1, "socket")) + goto fail; + err = sendto(sockfd, buf, sizeof(buf), 0, (void *)&addr, addrlen); + close(sockfd); + if (!ASSERT_EQ(err, sizeof(buf), "send")) + goto fail; + + ASSERT_TRUE(skel->bss->init_csum_partial, "init_csum_partial"); + ASSERT_TRUE(skel->bss->final_csum_none, "final_csum_none"); + ASSERT_FALSE(skel->bss->broken_csum_start, "broken_csum_start"); + +fail: + if (nstoken) { + bpf_tc_hook_destroy(&qdisc_hook); + close_netns(nstoken); + } + system("ip netns del " NS_TEST " >& /dev/null"); + decap_sanity__destroy(skel); +} diff --git a/tools/testing/selftests/bpf/progs/bpf_tracing_net.h b/tools/testing/selftests/bpf/progs/bpf_tracing_net.h index b394817126cf..cfed4df490f3 100644 --- a/tools/testing/selftests/bpf/progs/bpf_tracing_net.h +++ b/tools/testing/selftests/bpf/progs/bpf_tracing_net.h @@ -50,6 +50,12 @@ #define ICSK_TIME_LOSS_PROBE 5 #define ICSK_TIME_REO_TIMEOUT 6 +#define ETH_HLEN 14 +#define ETH_P_IPV6 0x86DD + +#define CHECKSUM_NONE 0 +#define CHECKSUM_PARTIAL 3 + #define IFNAMSIZ 16 #define RTF_GATEWAY 0x0002 diff --git a/tools/testing/selftests/bpf/progs/decap_sanity.c b/tools/testing/selftests/bpf/progs/decap_sanity.c new file mode 100644 index 000000000000..bd3c657c58a7 --- /dev/null +++ b/tools/testing/selftests/bpf/progs/decap_sanity.c @@ -0,0 +1,68 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2022 Meta Platforms, Inc. and affiliates. */ + +#include "vmlinux.h" +#include "bpf_tracing_net.h" +#include +#include + +#define UDP_TEST_PORT 7777 + +void *bpf_cast_to_kern_ctx(void *) __ksym; +bool init_csum_partial = false; +bool final_csum_none = false; +bool broken_csum_start = false; + +static unsigned int skb_headlen(const struct sk_buff *skb) +{ + return skb->len - skb->data_len; +} + +static unsigned int skb_headroom(const struct sk_buff *skb) +{ + return skb->data - skb->head; +} + +static int skb_checksum_start_offset(const struct sk_buff *skb) +{ + return skb->csum_start - skb_headroom(skb); +} + +SEC("tc") +int decap_sanity(struct __sk_buff *skb) +{ + struct sk_buff *kskb; + struct ipv6hdr ip6h; + struct udphdr udph; + int err; + + if (skb->protocol != __bpf_constant_htons(ETH_P_IPV6)) + return TC_ACT_SHOT; + + if (bpf_skb_load_bytes(skb, ETH_HLEN, &ip6h, sizeof(ip6h))) + return TC_ACT_SHOT; + + if (ip6h.nexthdr != IPPROTO_UDP) + return TC_ACT_SHOT; + + if (bpf_skb_load_bytes(skb, ETH_HLEN + sizeof(ip6h), &udph, sizeof(udph))) + return TC_ACT_SHOT; + + if (udph.dest != __bpf_constant_htons(UDP_TEST_PORT)) + return TC_ACT_SHOT; + + kskb = bpf_cast_to_kern_ctx(skb); + init_csum_partial = (kskb->ip_summed == CHECKSUM_PARTIAL); + err = bpf_skb_adjust_room(skb, -(s32)(ETH_HLEN + sizeof(ip6h) + sizeof(udph)), + 1, BPF_F_ADJ_ROOM_FIXED_GSO); + if (err) + return TC_ACT_SHOT; + final_csum_none = (kskb->ip_summed == CHECKSUM_NONE); + if (kskb->ip_summed == CHECKSUM_PARTIAL && + (unsigned int)skb_checksum_start_offset(kskb) >= skb_headlen(kskb)) + broken_csum_start = true; + + return TC_ACT_SHOT; +} + +char __license[] SEC("license") = "GPL"; From 53fc61be273a1e76dd5e356f91805dce00ff2d2c Mon Sep 17 00:00:00 2001 From: Maciej Fijalkowski Date: Tue, 20 Dec 2022 09:54:48 -0800 Subject: [PATCH 012/112] ice: xsk: do not use xdp_return_frame() on tx_buf->raw_buf Previously ice XDP xmit routine was changed in a way that it avoids xdp_buff->xdp_frame conversion as it is simply not needed for handling XDP_TX action and what is more it saves us CPU cycles. This routine is re-used on ZC driver to handle XDP_TX action. Although for XDP_TX on Rx ZC xdp_buff that comes from xsk_buff_pool is converted to xdp_frame, xdp_frame itself is not stored inside ice_tx_buf, we only store raw data pointer. Casting this pointer to xdp_frame and calling against it xdp_return_frame in ice_clean_xdp_tx_buf() results in undefined behavior. To fix this, simply call page_frag_free() on tx_buf->raw_buf. Later intention is to remove the buff->frame conversion in order to simplify the codebase and improve XDP_TX performance on ZC. Fixes: 126cdfe1007a ("ice: xsk: Improve AF_XDP ZC Tx and use batching API") Reported-and-tested-by: Robin Cowley Signed-off-by: Maciej Fijalkowski Tested-by: Chandan Kumar Rout (A Contingent Worker at Intel) Signed-off-by: Tony Nguyen Reviewed-by: Piotr Raczynski Link: https://lore.kernel.org/r/20221220175448.693999-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/intel/ice/ice_xsk.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 907055b77af0..7105de6fb344 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -783,7 +783,7 @@ construct_skb: static void ice_clean_xdp_tx_buf(struct ice_tx_ring *xdp_ring, struct ice_tx_buf *tx_buf) { - xdp_return_frame((struct xdp_frame *)tx_buf->raw_buf); + page_frag_free(tx_buf->raw_buf); xdp_ring->xdp_tx_active--; dma_unmap_single(xdp_ring->dev, dma_unmap_addr(tx_buf, dma), dma_unmap_len(tx_buf, len), DMA_TO_DEVICE); From f2575c8f404911da83f25b688e12afcf4273e640 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Tue, 20 Dec 2022 18:18:25 +0100 Subject: [PATCH 013/112] net: vrf: determine the dst using the original ifindex for multicast Multicast packets received on an interface bound to a VRF are marked as belonging to the VRF and the skb device is updated to point to the VRF device itself. This was fine even when a route was associated to a device as when performing a fib table lookup 'oif' in fib6_table_lookup (coming from 'skb->dev->ifindex' in ip6_route_input) was set to 0 when FLOWI_FLAG_SKIP_NH_OIF was set. With commit 40867d74c374 ("net: Add l3mdev index to flow struct and avoid oif reset for port devices") this is not longer true and multicast traffic is not received on the original interface. Instead of adding back a similar check in fib6_table_lookup determine the dst using the original ifindex for multicast VRF traffic. To make things consistent across the function do the above for all strict packets, which was the logic before commit 6f12fa775530 ("vrf: mark skb for multicast or link-local as enslaved to VRF"). Note that reverting to this behavior should be fine as the change was about marking packets belonging to the VRF, not about their dst. Fixes: 40867d74c374 ("net: Add l3mdev index to flow struct and avoid oif reset for port devices") Reported-by: Jianlin Shi Signed-off-by: Antoine Tenart Reviewed-by: David Ahern Link: https://lore.kernel.org/r/20221220171825.1172237-1-atenart@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/vrf.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 6b5a4d036d15..bdb3a76a352e 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -1385,8 +1385,8 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, /* loopback, multicast & non-ND link-local traffic; do not push through * packet taps again. Reset pkt_type for upper layers to process skb. - * For strict packets with a source LLA, determine the dst using the - * original ifindex. + * For non-loopback strict packets, determine the dst using the original + * ifindex. */ if (skb->pkt_type == PACKET_LOOPBACK || (need_strict && !is_ndisc)) { skb->dev = vrf_dev; @@ -1395,7 +1395,7 @@ static struct sk_buff *vrf_ip6_rcv(struct net_device *vrf_dev, if (skb->pkt_type == PACKET_LOOPBACK) skb->pkt_type = PACKET_HOST; - else if (ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL) + else vrf_ip6_input_dst(skb, vrf_dev, orig_iif); goto out; From 95637d91fefdb94d6e7389222ba9ddab0e9f5abe Mon Sep 17 00:00:00 2001 From: Aaron Conole Date: Tue, 20 Dec 2022 16:27:17 -0500 Subject: [PATCH 014/112] net: openvswitch: release vport resources on failure A recent commit introducing upcall packet accounting failed to properly release the vport object when the per-cpu stats struct couldn't be allocated. This can cause dangling pointers to dp objects long after they've been released. Cc: wangchuanlei Fixes: 1933ea365aa7 ("net: openvswitch: Add support to count upcall packets") Reported-by: syzbot+8f4e2dcfcb3209ac35f9@syzkaller.appspotmail.com Signed-off-by: Aaron Conole Acked-by: Eelco Chaudron Reviewed-by: Michal Swiatkowski Link: https://lore.kernel.org/r/20221220212717.526780-1-aconole@redhat.com Signed-off-by: Jakub Kicinski --- net/openvswitch/datapath.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index 9ca721c9fa71..a71795355aec 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -1861,7 +1861,7 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) vport->upcall_stats = netdev_alloc_pcpu_stats(struct vport_upcall_stats_percpu); if (!vport->upcall_stats) { err = -ENOMEM; - goto err_destroy_portids; + goto err_destroy_vport; } err = ovs_dp_cmd_fill_info(dp, reply, info->snd_portid, @@ -1876,6 +1876,8 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info) ovs_notify(&dp_datapath_genl_family, reply, info); return 0; +err_destroy_vport: + ovs_dp_detach_port(vport); err_destroy_portids: kfree(rcu_dereference_raw(dp->upcall_portids)); err_unlock_and_destroy_meters: @@ -2323,7 +2325,7 @@ restart: vport->upcall_stats = netdev_alloc_pcpu_stats(struct vport_upcall_stats_percpu); if (!vport->upcall_stats) { err = -ENOMEM; - goto exit_unlock_free; + goto exit_unlock_free_vport; } err = ovs_vport_cmd_fill_info(vport, reply, genl_info_net(info), @@ -2343,6 +2345,8 @@ restart: ovs_notify(&dp_vport_genl_family, reply, info); return 0; +exit_unlock_free_vport: + ovs_dp_detach_port(vport); exit_unlock_free: ovs_unlock(); kfree_skb(reply); From 3d8f2c4269d08f8793e946279dbdf5e972cc4911 Mon Sep 17 00:00:00 2001 From: Ronak Doshi Date: Tue, 20 Dec 2022 12:25:55 -0800 Subject: [PATCH 015/112] vmxnet3: correctly report csum_level for encapsulated packet Commit dacce2be3312 ("vmxnet3: add geneve and vxlan tunnel offload support") added support for encapsulation offload. However, the pathc did not report correctly the csum_level for encapsulated packet. This patch fixes this issue by reporting correct csum level for the encapsulated packet. Fixes: dacce2be3312 ("vmxnet3: add geneve and vxlan tunnel offload support") Signed-off-by: Ronak Doshi Acked-by: Peng Li Link: https://lore.kernel.org/r/20221220202556.24421-1-doshir@vmware.com Signed-off-by: Jakub Kicinski --- drivers/net/vmxnet3/vmxnet3_drv.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c index 6f1e560fb15c..56267c327f0b 100644 --- a/drivers/net/vmxnet3/vmxnet3_drv.c +++ b/drivers/net/vmxnet3/vmxnet3_drv.c @@ -1288,6 +1288,10 @@ vmxnet3_rx_csum(struct vmxnet3_adapter *adapter, (le32_to_cpu(gdesc->dword[3]) & VMXNET3_RCD_CSUM_OK) == VMXNET3_RCD_CSUM_OK) { skb->ip_summed = CHECKSUM_UNNECESSARY; + if ((le32_to_cpu(gdesc->dword[0]) & + (1UL << VMXNET3_RCD_HDR_INNER_SHIFT))) { + skb->csum_level = 1; + } WARN_ON_ONCE(!(gdesc->rcd.tcp || gdesc->rcd.udp) && !(le32_to_cpu(gdesc->dword[0]) & (1UL << VMXNET3_RCD_HDR_INNER_SHIFT))); @@ -1297,6 +1301,10 @@ vmxnet3_rx_csum(struct vmxnet3_adapter *adapter, } else if (gdesc->rcd.v6 && (le32_to_cpu(gdesc->dword[3]) & (1 << VMXNET3_RCD_TUC_SHIFT))) { skb->ip_summed = CHECKSUM_UNNECESSARY; + if ((le32_to_cpu(gdesc->dword[0]) & + (1UL << VMXNET3_RCD_HDR_INNER_SHIFT))) { + skb->csum_level = 1; + } WARN_ON_ONCE(!(gdesc->rcd.tcp || gdesc->rcd.udp) && !(le32_to_cpu(gdesc->dword[0]) & (1UL << VMXNET3_RCD_HDR_INNER_SHIFT))); From e20aa071cd955aabc15be0ec1e914283592ddef4 Mon Sep 17 00:00:00 2001 From: Yinjun Zhang Date: Tue, 20 Dec 2022 16:21:00 +0100 Subject: [PATCH 016/112] nfp: fix schedule in atomic context when sync mc address The callback `.ndo_set_rx_mode` is called in atomic context, sleep is not allowed in the implementation. Now use workqueue mechanism to avoid this issue. Fixes: de6248644966 ("nfp: add support for multicast filter") Signed-off-by: Yinjun Zhang Reviewed-by: Louis Peens Signed-off-by: Simon Horman Link: https://lore.kernel.org/r/20221220152100.1042774-1-simon.horman@corigine.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/netronome/nfp/nfp_net.h | 7 +++ .../ethernet/netronome/nfp/nfp_net_common.c | 61 +++++++++++++++++-- 2 files changed, 64 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net.h b/drivers/net/ethernet/netronome/nfp/nfp_net.h index da33f09facb9..432d79d691c2 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net.h +++ b/drivers/net/ethernet/netronome/nfp/nfp_net.h @@ -617,6 +617,9 @@ struct nfp_net_dp { * @vnic_no_name: For non-port PF vNIC make ndo_get_phys_port_name return * -EOPNOTSUPP to keep backwards compatibility (set by app) * @port: Pointer to nfp_port structure if vNIC is a port + * @mc_lock: Protect mc_addrs list + * @mc_addrs: List of mc addrs to add/del to HW + * @mc_work: Work to update mc addrs * @app_priv: APP private data for this vNIC */ struct nfp_net { @@ -718,6 +721,10 @@ struct nfp_net { struct nfp_port *port; + spinlock_t mc_lock; + struct list_head mc_addrs; + struct work_struct mc_work; + void *app_priv; }; diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 09053373288f..18fc9971f1c8 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1334,9 +1334,14 @@ err_unlock: return err; } -static int nfp_net_mc_cfg(struct net_device *netdev, const unsigned char *addr, const u32 cmd) +struct nfp_mc_addr_entry { + u8 addr[ETH_ALEN]; + u32 cmd; + struct list_head list; +}; + +static int nfp_net_mc_cfg(struct nfp_net *nn, const unsigned char *addr, const u32 cmd) { - struct nfp_net *nn = netdev_priv(netdev); int ret; ret = nfp_net_mbox_lock(nn, NFP_NET_CFG_MULTICAST_SZ); @@ -1351,6 +1356,25 @@ static int nfp_net_mc_cfg(struct net_device *netdev, const unsigned char *addr, return nfp_net_mbox_reconfig_and_unlock(nn, cmd); } +static int nfp_net_mc_prep(struct nfp_net *nn, const unsigned char *addr, const u32 cmd) +{ + struct nfp_mc_addr_entry *entry; + + entry = kmalloc(sizeof(*entry), GFP_ATOMIC); + if (!entry) + return -ENOMEM; + + ether_addr_copy(entry->addr, addr); + entry->cmd = cmd; + spin_lock_bh(&nn->mc_lock); + list_add_tail(&entry->list, &nn->mc_addrs); + spin_unlock_bh(&nn->mc_lock); + + schedule_work(&nn->mc_work); + + return 0; +} + static int nfp_net_mc_sync(struct net_device *netdev, const unsigned char *addr) { struct nfp_net *nn = netdev_priv(netdev); @@ -1361,12 +1385,35 @@ static int nfp_net_mc_sync(struct net_device *netdev, const unsigned char *addr) return -EINVAL; } - return nfp_net_mc_cfg(netdev, addr, NFP_NET_CFG_MBOX_CMD_MULTICAST_ADD); + return nfp_net_mc_prep(nn, addr, NFP_NET_CFG_MBOX_CMD_MULTICAST_ADD); } static int nfp_net_mc_unsync(struct net_device *netdev, const unsigned char *addr) { - return nfp_net_mc_cfg(netdev, addr, NFP_NET_CFG_MBOX_CMD_MULTICAST_DEL); + struct nfp_net *nn = netdev_priv(netdev); + + return nfp_net_mc_prep(nn, addr, NFP_NET_CFG_MBOX_CMD_MULTICAST_DEL); +} + +static void nfp_net_mc_addr_config(struct work_struct *work) +{ + struct nfp_net *nn = container_of(work, struct nfp_net, mc_work); + struct nfp_mc_addr_entry *entry, *tmp; + struct list_head tmp_list; + + INIT_LIST_HEAD(&tmp_list); + + spin_lock_bh(&nn->mc_lock); + list_splice_init(&nn->mc_addrs, &tmp_list); + spin_unlock_bh(&nn->mc_lock); + + list_for_each_entry_safe(entry, tmp, &tmp_list, list) { + if (nfp_net_mc_cfg(nn, entry->addr, entry->cmd)) + nn_err(nn, "Config mc address to HW failed.\n"); + + list_del(&entry->list); + kfree(entry); + } } static void nfp_net_set_rx_mode(struct net_device *netdev) @@ -2633,6 +2680,11 @@ int nfp_net_init(struct nfp_net *nn) if (!nn->dp.netdev) return 0; + + spin_lock_init(&nn->mc_lock); + INIT_LIST_HEAD(&nn->mc_addrs); + INIT_WORK(&nn->mc_work, nfp_net_mc_addr_config); + return register_netdev(nn->dp.netdev); err_clean_mbox: @@ -2652,5 +2704,6 @@ void nfp_net_clean(struct nfp_net *nn) unregister_netdev(nn->dp.netdev); nfp_net_ipsec_clean(nn); nfp_ccm_mbox_clean(nn); + flush_work(&nn->mc_work); nfp_net_reconfig_wait_posted(nn); } From 7d803344fdc3e38079fabcf38b1e4cb6f8faa655 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 20 Dec 2022 11:52:14 -0800 Subject: [PATCH 017/112] mptcp: fix deadlock in fastopen error path MatM reported a deadlock at fastopening time: INFO: task syz-executor.0:11454 blocked for more than 143 seconds. Tainted: G S 6.1.0-rc5-03226-gdb0157db5153 #1 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. task:syz-executor.0 state:D stack:25104 pid:11454 ppid:424 flags:0x00004006 Call Trace: context_switch kernel/sched/core.c:5191 [inline] __schedule+0x5c2/0x1550 kernel/sched/core.c:6503 schedule+0xe8/0x1c0 kernel/sched/core.c:6579 __lock_sock+0x142/0x260 net/core/sock.c:2896 lock_sock_nested+0xdb/0x100 net/core/sock.c:3466 __mptcp_close_ssk+0x1a3/0x790 net/mptcp/protocol.c:2328 mptcp_destroy_common+0x16a/0x650 net/mptcp/protocol.c:3171 mptcp_disconnect+0xb8/0x450 net/mptcp/protocol.c:3019 __inet_stream_connect+0x897/0xa40 net/ipv4/af_inet.c:720 tcp_sendmsg_fastopen+0x3dd/0x740 net/ipv4/tcp.c:1200 mptcp_sendmsg_fastopen net/mptcp/protocol.c:1682 [inline] mptcp_sendmsg+0x128a/0x1a50 net/mptcp/protocol.c:1721 inet6_sendmsg+0x11f/0x150 net/ipv6/af_inet6.c:663 sock_sendmsg_nosec net/socket.c:714 [inline] sock_sendmsg+0xf7/0x190 net/socket.c:734 ____sys_sendmsg+0x336/0x970 net/socket.c:2476 ___sys_sendmsg+0x122/0x1c0 net/socket.c:2530 __sys_sendmmsg+0x18d/0x460 net/socket.c:2616 __do_sys_sendmmsg net/socket.c:2645 [inline] __se_sys_sendmmsg net/socket.c:2642 [inline] __x64_sys_sendmmsg+0x9d/0x110 net/socket.c:2642 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x38/0x90 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7f5920a75e7d RSP: 002b:00007f59201e8028 EFLAGS: 00000246 ORIG_RAX: 0000000000000133 RAX: ffffffffffffffda RBX: 00007f5920bb4f80 RCX: 00007f5920a75e7d RDX: 0000000000000001 RSI: 0000000020002940 RDI: 0000000000000005 RBP: 00007f5920ae7593 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000020004050 R11: 0000000000000246 R12: 0000000000000000 R13: 000000000000000b R14: 00007f5920bb4f80 R15: 00007f59201c8000 In the error path, tcp_sendmsg_fastopen() ends-up calling mptcp_disconnect(), and the latter tries to close each subflow, acquiring the socket lock on each of them. At fastopen time, we have a single subflow, and such subflow socket lock is already held by the called, causing the deadlock. We already track the 'fastopen in progress' status inside the msk socket. Use it to address the issue, making mptcp_disconnect() a no op when invoked from the fastopen (error) path and doing the relevant cleanup after releasing the subflow socket lock. While at the above, rename the fastopen status bit to something more meaningful. Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/321 Fixes: fa9e57468aa1 ("mptcp: fix abba deadlock on fastopen") Reported-by: Mat Martineau Reviewed-by: Mat Martineau Signed-off-by: Paolo Abeni Signed-off-by: Mat Martineau Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 18 +++++++++++++++--- net/mptcp/protocol.h | 2 +- 2 files changed, 16 insertions(+), 4 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index f6f93957275b..907b435e2984 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1662,6 +1662,8 @@ static void mptcp_set_nospace(struct sock *sk) set_bit(MPTCP_NOSPACE, &mptcp_sk(sk)->flags); } +static int mptcp_disconnect(struct sock *sk, int flags); + static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msghdr *msg, size_t len, int *copied_syn) { @@ -1672,9 +1674,9 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msgh lock_sock(ssk); msg->msg_flags |= MSG_DONTWAIT; msk->connect_flags = O_NONBLOCK; - msk->is_sendmsg = 1; + msk->fastopening = 1; ret = tcp_sendmsg_fastopen(ssk, msg, copied_syn, len, NULL); - msk->is_sendmsg = 0; + msk->fastopening = 0; msg->msg_flags = saved_flags; release_sock(ssk); @@ -1688,6 +1690,8 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct sock *ssk, struct msgh */ if (ret && ret != -EINPROGRESS && ret != -ERESTARTSYS && ret != -EINTR) *copied_syn = 0; + } else if (ret && ret != -EINPROGRESS) { + mptcp_disconnect(sk, 0); } return ret; @@ -2989,6 +2993,14 @@ static int mptcp_disconnect(struct sock *sk, int flags) { struct mptcp_sock *msk = mptcp_sk(sk); + /* We are on the fastopen error path. We can't call straight into the + * subflows cleanup code due to lock nesting (we are already under + * msk->firstsocket lock). Do nothing and leave the cleanup to the + * caller. + */ + if (msk->fastopening) + return 0; + inet_sk_state_store(sk, TCP_CLOSE); mptcp_stop_timer(sk); @@ -3532,7 +3544,7 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) /* if reaching here via the fastopen/sendmsg path, the caller already * acquired the subflow socket lock, too. */ - if (msk->is_sendmsg) + if (msk->fastopening) err = __inet_stream_connect(ssock, uaddr, addr_len, msk->connect_flags, 1); else err = inet_stream_connect(ssock, uaddr, addr_len, msk->connect_flags); diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index 955fb3d88eb3..f47d3e4018b5 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -295,7 +295,7 @@ struct mptcp_sock { u8 recvmsg_inq:1, cork:1, nodelay:1, - is_sendmsg:1; + fastopening:1; int connect_flags; struct work_struct work; struct sk_buff *ooo_last_skb; From fec3adfd754ccc99a7230e8ab9f105b65fb07bcc Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 20 Dec 2022 11:52:15 -0800 Subject: [PATCH 018/112] mptcp: fix lockdep false positive MattB reported a lockdep splat in the mptcp listener code cleanup: WARNING: possible circular locking dependency detected packetdrill/14278 is trying to acquire lock: ffff888017d868f0 ((work_completion)(&msk->work)){+.+.}-{0:0}, at: __flush_work (kernel/workqueue.c:3069) but task is already holding lock: ffff888017d84130 (sk_lock-AF_INET){+.+.}-{0:0}, at: mptcp_close (net/mptcp/protocol.c:2973) which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (sk_lock-AF_INET){+.+.}-{0:0}: __lock_acquire (kernel/locking/lockdep.c:5055) lock_acquire (kernel/locking/lockdep.c:466) lock_sock_nested (net/core/sock.c:3463) mptcp_worker (net/mptcp/protocol.c:2614) process_one_work (kernel/workqueue.c:2294) worker_thread (include/linux/list.h:292) kthread (kernel/kthread.c:376) ret_from_fork (arch/x86/entry/entry_64.S:312) -> #0 ((work_completion)(&msk->work)){+.+.}-{0:0}: check_prev_add (kernel/locking/lockdep.c:3098) validate_chain (kernel/locking/lockdep.c:3217) __lock_acquire (kernel/locking/lockdep.c:5055) lock_acquire (kernel/locking/lockdep.c:466) __flush_work (kernel/workqueue.c:3070) __cancel_work_timer (kernel/workqueue.c:3160) mptcp_cancel_work (net/mptcp/protocol.c:2758) mptcp_subflow_queue_clean (net/mptcp/subflow.c:1817) __mptcp_close_ssk (net/mptcp/protocol.c:2363) mptcp_destroy_common (net/mptcp/protocol.c:3170) mptcp_destroy (include/net/sock.h:1495) __mptcp_destroy_sock (net/mptcp/protocol.c:2886) __mptcp_close (net/mptcp/protocol.c:2959) mptcp_close (net/mptcp/protocol.c:2974) inet_release (net/ipv4/af_inet.c:432) __sock_release (net/socket.c:651) sock_close (net/socket.c:1367) __fput (fs/file_table.c:320) task_work_run (kernel/task_work.c:181 (discriminator 1)) exit_to_user_mode_prepare (include/linux/resume_user_mode.h:49) syscall_exit_to_user_mode (kernel/entry/common.c:130) do_syscall_64 (arch/x86/entry/common.c:87) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120) other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(sk_lock-AF_INET); lock((work_completion)(&msk->work)); lock(sk_lock-AF_INET); lock((work_completion)(&msk->work)); *** DEADLOCK *** The report is actually a false positive, since the only existing lock nesting is the msk socket lock acquired by the mptcp work. cancel_work_sync() is invoked without the relevant socket lock being held, but under a different (the msk listener) socket lock. We could silence the splat adding a per workqueue dynamic lockdep key, but that looks overkill. Instead just tell lockdep the msk socket lock is not held around cancel_work_sync(). Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/322 Fixes: 30e51b923e43 ("mptcp: fix unreleased socket in accept queue") Reported-by: Matthieu Baerts Reviewed-by: Mat Martineau Signed-off-by: Paolo Abeni Signed-off-by: Mat Martineau Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 2 +- net/mptcp/protocol.h | 2 +- net/mptcp/subflow.c | 19 +++++++++++++++++-- 3 files changed, 19 insertions(+), 4 deletions(-) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 907b435e2984..b7ad030dfe89 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2357,7 +2357,7 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, /* otherwise tcp will dispose of the ssk and subflow ctx */ if (ssk->sk_state == TCP_LISTEN) { tcp_set_state(ssk, TCP_CLOSE); - mptcp_subflow_queue_clean(ssk); + mptcp_subflow_queue_clean(sk, ssk); inet_csk_listen_stop(ssk); mptcp_event_pm_listener(ssk, MPTCP_EVENT_LISTENER_CLOSED); } diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index f47d3e4018b5..a0d1658ce59e 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -628,7 +628,7 @@ void mptcp_close_ssk(struct sock *sk, struct sock *ssk, struct mptcp_subflow_context *subflow); void __mptcp_subflow_send_ack(struct sock *ssk); void mptcp_subflow_reset(struct sock *ssk); -void mptcp_subflow_queue_clean(struct sock *ssk); +void mptcp_subflow_queue_clean(struct sock *sk, struct sock *ssk); void mptcp_sock_graft(struct sock *sk, struct socket *parent); struct socket *__mptcp_nmpc_socket(const struct mptcp_sock *msk); bool __mptcp_close(struct sock *sk, long timeout); diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c index d1d32a66ae3f..bd387d4b5a38 100644 --- a/net/mptcp/subflow.c +++ b/net/mptcp/subflow.c @@ -1791,7 +1791,7 @@ static void subflow_state_change(struct sock *sk) } } -void mptcp_subflow_queue_clean(struct sock *listener_ssk) +void mptcp_subflow_queue_clean(struct sock *listener_sk, struct sock *listener_ssk) { struct request_sock_queue *queue = &inet_csk(listener_ssk)->icsk_accept_queue; struct mptcp_sock *msk, *next, *head = NULL; @@ -1840,8 +1840,23 @@ void mptcp_subflow_queue_clean(struct sock *listener_ssk) do_cancel_work = __mptcp_close(sk, 0); release_sock(sk); - if (do_cancel_work) + if (do_cancel_work) { + /* lockdep will report a false positive ABBA deadlock + * between cancel_work_sync and the listener socket. + * The involved locks belong to different sockets WRT + * the existing AB chain. + * Using a per socket key is problematic as key + * deregistration requires process context and must be + * performed at socket disposal time, in atomic + * context. + * Just tell lockdep to consider the listener socket + * released here. + */ + mutex_release(&listener_sk->sk_lock.dep_map, _RET_IP_); mptcp_cancel_work(sk); + mutex_acquire(&listener_sk->sk_lock.dep_map, + SINGLE_DEPTH_NESTING, 0, _RET_IP_); + } sock_put(sk); } From 123b99619cca94bdca0bf7bde9abe28f0a0dfe06 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 19 Dec 2022 20:10:12 +0100 Subject: [PATCH 019/112] netfilter: nf_tables: honor set timeout and garbage collection updates Set timeout and garbage collection interval updates are ignored on updates. Add transaction to update global set element timeout and garbage collection interval. Fixes: 96518518cc41 ("netfilter: add nftables") Suggested-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 13 ++++++- net/netfilter/nf_tables_api.c | 63 ++++++++++++++++++++++--------- 2 files changed, 57 insertions(+), 19 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index 4957b4775757..9430128aae99 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -597,7 +597,9 @@ void *nft_set_catchall_gc(const struct nft_set *set); static inline unsigned long nft_set_gc_interval(const struct nft_set *set) { - return set->gc_int ? msecs_to_jiffies(set->gc_int) : HZ; + u32 gc_int = READ_ONCE(set->gc_int); + + return gc_int ? msecs_to_jiffies(gc_int) : HZ; } /** @@ -1570,6 +1572,9 @@ struct nft_trans_rule { struct nft_trans_set { struct nft_set *set; u32 set_id; + u32 gc_int; + u64 timeout; + bool update; bool bound; }; @@ -1579,6 +1584,12 @@ struct nft_trans_set { (((struct nft_trans_set *)trans->data)->set_id) #define nft_trans_set_bound(trans) \ (((struct nft_trans_set *)trans->data)->bound) +#define nft_trans_set_update(trans) \ + (((struct nft_trans_set *)trans->data)->update) +#define nft_trans_set_timeout(trans) \ + (((struct nft_trans_set *)trans->data)->timeout) +#define nft_trans_set_gc_int(trans) \ + (((struct nft_trans_set *)trans->data)->gc_int) struct nft_trans_chain { bool update; diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 319887f4d3ef..8c09e4d12ac1 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -465,8 +465,9 @@ static int nft_delrule_by_chain(struct nft_ctx *ctx) return 0; } -static int nft_trans_set_add(const struct nft_ctx *ctx, int msg_type, - struct nft_set *set) +static int __nft_trans_set_add(const struct nft_ctx *ctx, int msg_type, + struct nft_set *set, + const struct nft_set_desc *desc) { struct nft_trans *trans; @@ -474,17 +475,28 @@ static int nft_trans_set_add(const struct nft_ctx *ctx, int msg_type, if (trans == NULL) return -ENOMEM; - if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] != NULL) { + if (msg_type == NFT_MSG_NEWSET && ctx->nla[NFTA_SET_ID] && !desc) { nft_trans_set_id(trans) = ntohl(nla_get_be32(ctx->nla[NFTA_SET_ID])); nft_activate_next(ctx->net, set); } nft_trans_set(trans) = set; + if (desc) { + nft_trans_set_update(trans) = true; + nft_trans_set_gc_int(trans) = desc->gc_int; + nft_trans_set_timeout(trans) = desc->timeout; + } nft_trans_commit_list_add_tail(ctx->net, trans); return 0; } +static int nft_trans_set_add(const struct nft_ctx *ctx, int msg_type, + struct nft_set *set) +{ + return __nft_trans_set_add(ctx, msg_type, set, NULL); +} + static int nft_delset(const struct nft_ctx *ctx, struct nft_set *set) { int err; @@ -4044,8 +4056,10 @@ static int nf_tables_fill_set_concat(struct sk_buff *skb, static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, const struct nft_set *set, u16 event, u16 flags) { - struct nlmsghdr *nlh; + u64 timeout = READ_ONCE(set->timeout); + u32 gc_int = READ_ONCE(set->gc_int); u32 portid = ctx->portid; + struct nlmsghdr *nlh; struct nlattr *nest; u32 seq = ctx->seq; int i; @@ -4081,13 +4095,13 @@ static int nf_tables_fill_set(struct sk_buff *skb, const struct nft_ctx *ctx, nla_put_be32(skb, NFTA_SET_OBJ_TYPE, htonl(set->objtype))) goto nla_put_failure; - if (set->timeout && + if (timeout && nla_put_be64(skb, NFTA_SET_TIMEOUT, - nf_jiffies64_to_msecs(set->timeout), + nf_jiffies64_to_msecs(timeout), NFTA_SET_PAD)) goto nla_put_failure; - if (set->gc_int && - nla_put_be32(skb, NFTA_SET_GC_INTERVAL, htonl(set->gc_int))) + if (gc_int && + nla_put_be32(skb, NFTA_SET_GC_INTERVAL, htonl(gc_int))) goto nla_put_failure; if (set->policy != NFT_SET_POL_PERFORMANCE) { @@ -4632,7 +4646,10 @@ static int nf_tables_newset(struct sk_buff *skb, const struct nfnl_info *info, for (i = 0; i < num_exprs; i++) nft_expr_destroy(&ctx, exprs[i]); - return err; + if (err < 0) + return err; + + return __nft_trans_set_add(&ctx, NFT_MSG_NEWSET, set, &desc); } if (!(info->nlh->nlmsg_flags & NLM_F_CREATE)) @@ -6070,7 +6087,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, return err; } else if (set->flags & NFT_SET_TIMEOUT && !(flags & NFT_SET_ELEM_INTERVAL_END)) { - timeout = set->timeout; + timeout = READ_ONCE(set->timeout); } expiration = 0; @@ -6171,7 +6188,7 @@ static int nft_add_set_elem(struct nft_ctx *ctx, struct nft_set *set, if (err < 0) goto err_parse_key_end; - if (timeout != set->timeout) { + if (timeout != READ_ONCE(set->timeout)) { err = nft_set_ext_add(&tmpl, NFT_SET_EXT_TIMEOUT); if (err < 0) goto err_parse_key_end; @@ -9093,14 +9110,20 @@ static int nf_tables_commit(struct net *net, struct sk_buff *skb) nft_flow_rule_destroy(nft_trans_flow_rule(trans)); break; case NFT_MSG_NEWSET: - nft_clear(net, nft_trans_set(trans)); - /* This avoids hitting -EBUSY when deleting the table - * from the transaction. - */ - if (nft_set_is_anonymous(nft_trans_set(trans)) && - !list_empty(&nft_trans_set(trans)->bindings)) - trans->ctx.table->use--; + if (nft_trans_set_update(trans)) { + struct nft_set *set = nft_trans_set(trans); + WRITE_ONCE(set->timeout, nft_trans_set_timeout(trans)); + WRITE_ONCE(set->gc_int, nft_trans_set_gc_int(trans)); + } else { + nft_clear(net, nft_trans_set(trans)); + /* This avoids hitting -EBUSY when deleting the table + * from the transaction. + */ + if (nft_set_is_anonymous(nft_trans_set(trans)) && + !list_empty(&nft_trans_set(trans)->bindings)) + trans->ctx.table->use--; + } nf_tables_set_notify(&trans->ctx, nft_trans_set(trans), NFT_MSG_NEWSET, GFP_KERNEL); nft_trans_destroy(trans); @@ -9322,6 +9345,10 @@ static int __nf_tables_abort(struct net *net, enum nfnl_abort_action action) nft_trans_destroy(trans); break; case NFT_MSG_NEWSET: + if (nft_trans_set_update(trans)) { + nft_trans_destroy(trans); + break; + } trans->ctx.table->use--; if (nft_trans_set_bound(trans)) { nft_trans_destroy(trans); From 42c7ded0eeacd2ba5db599205c71c279dc715de7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 20 Dec 2022 13:08:31 +0000 Subject: [PATCH 020/112] bonding: fix lockdep splat in bond_miimon_commit() bond_miimon_commit() is run while RTNL is held, not RCU. WARNING: suspicious RCU usage 6.1.0-syzkaller-09671-g89529367293c #0 Not tainted ----------------------------- drivers/net/bonding/bond_main.c:2704 suspicious rcu_dereference_check() usage! Fixes: e95cc44763a4 ("bonding: do failover when high prio link up") Signed-off-by: Eric Dumazet Reported-by: syzbot Cc: Hangbin Liu Cc: Jay Vosburgh Cc: Veaceslav Falico Cc: Andy Gospodarek Link: https://lore.kernel.org/r/20221220130831.1480888-1-edumazet@google.com Signed-off-by: Paolo Abeni --- drivers/net/bonding/bond_main.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index b4c65783960a..0363ce597661 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2654,10 +2654,12 @@ static void bond_miimon_link_change(struct bonding *bond, static void bond_miimon_commit(struct bonding *bond) { - struct slave *slave, *primary; + struct slave *slave, *primary, *active; bool do_failover = false; struct list_head *iter; + ASSERT_RTNL(); + bond_for_each_slave(bond, slave, iter) { switch (slave->link_new_state) { case BOND_LINK_NOCHANGE: @@ -2700,8 +2702,8 @@ static void bond_miimon_commit(struct bonding *bond) bond_miimon_link_change(bond, slave, BOND_LINK_UP); - if (!rcu_access_pointer(bond->curr_active_slave) || slave == primary || - slave->prio > rcu_dereference(bond->curr_active_slave)->prio) + active = rtnl_dereference(bond->curr_active_slave); + if (!active || slave == primary || slave->prio > active->prio) do_failover = true; continue; From d717f9474e3fb7e6bd3e43ca16e131f04320ed6f Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Wed, 21 Dec 2022 10:33:15 +0100 Subject: [PATCH 021/112] net: lan966x: Fix configuration of the PCS When the PCS was taken out of reset, we were changing by mistake also the speed to 100 Mbit. But in case the link was going down, the link up routine was setting correctly the link speed. If the link was not getting down then the speed was forced to run at 100 even if the speed was something else. On lan966x, to set the speed link to 1G or 2.5G a value of 1 needs to be written in DEV_CLOCK_CFG_LINK_SPEED. This is similar to the procedure in lan966x_port_init. The issue was reproduced using 1000base-x sfp module using the commands: ip link set dev eth2 up ip link addr add 10.97.10.2/24 dev eth2 ethtool -s eth2 speed 1000 autoneg off Fixes: d28d6d2e37d1 ("net: lan966x: add port module support") Signed-off-by: Horatiu Vultur Reviewed-by: Piotr Raczynski Link: https://lore.kernel.org/r/20221221093315.939133-1-horatiu.vultur@microchip.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/microchip/lan966x/lan966x_port.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_port.c b/drivers/net/ethernet/microchip/lan966x/lan966x_port.c index 1a61c6cdb077..0050fcb988b7 100644 --- a/drivers/net/ethernet/microchip/lan966x/lan966x_port.c +++ b/drivers/net/ethernet/microchip/lan966x/lan966x_port.c @@ -381,7 +381,7 @@ int lan966x_port_pcs_set(struct lan966x_port *port, } /* Take PCS out of reset */ - lan_rmw(DEV_CLOCK_CFG_LINK_SPEED_SET(2) | + lan_rmw(DEV_CLOCK_CFG_LINK_SPEED_SET(LAN966X_SPEED_1000) | DEV_CLOCK_CFG_PCS_RX_RST_SET(0) | DEV_CLOCK_CFG_PCS_TX_RST_SET(0), DEV_CLOCK_CFG_LINK_SPEED | From fa349e396e4886d742fd6501c599ec627ef1353b Mon Sep 17 00:00:00 2001 From: Shawn Bohrer Date: Tue, 20 Dec 2022 12:59:03 -0600 Subject: [PATCH 022/112] veth: Fix race with AF_XDP exposing old or uninitialized descriptors When AF_XDP is used on on a veth interface the RX ring is updated in two steps. veth_xdp_rcv() removes packet descriptors from the FILL ring fills them and places them in the RX ring updating the cached_prod pointer. Later xdp_do_flush() syncs the RX ring prod pointer with the cached_prod pointer allowing user-space to see the recently filled in descriptors. The rings are intended to be SPSC, however the existing order in veth_poll allows the xdp_do_flush() to run concurrently with another CPU creating a race condition that allows user-space to see old or uninitialized descriptors in the RX ring. This bug has been observed in production systems. To summarize, we are expecting this ordering: CPU 0 __xsk_rcv_zc() CPU 0 __xsk_map_flush() CPU 2 __xsk_rcv_zc() CPU 2 __xsk_map_flush() But we are seeing this order: CPU 0 __xsk_rcv_zc() CPU 2 __xsk_rcv_zc() CPU 0 __xsk_map_flush() CPU 2 __xsk_map_flush() This occurs because we rely on NAPI to ensure that only one napi_poll handler is running at a time for the given veth receive queue. napi_schedule_prep() will prevent multiple instances from getting scheduled. However calling napi_complete_done() signals that this napi_poll is complete and allows subsequent calls to napi_schedule_prep() and __napi_schedule() to succeed in scheduling a concurrent napi_poll before the xdp_do_flush() has been called. For the veth driver a concurrent call to napi_schedule_prep() and __napi_schedule() can occur on a different CPU because the veth xmit path can additionally schedule a napi_poll creating the race. The fix as suggested by Magnus Karlsson, is to simply move the xdp_do_flush() call before napi_complete_done(). This syncs the producer ring pointers before another instance of napi_poll can be scheduled on another CPU. It will also slightly improve performance by moving the flush closer to when the descriptors were placed in the RX ring. Fixes: d1396004dd86 ("veth: Add XDP TX and REDIRECT") Suggested-by: Magnus Karlsson Signed-off-by: Shawn Bohrer Link: https://lore.kernel.org/r/20221220185903.1105011-1-sbohrer@cloudflare.com Signed-off-by: Paolo Abeni --- drivers/net/veth.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index ac7c0653695f..dfc7d87fad59 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -974,6 +974,9 @@ static int veth_poll(struct napi_struct *napi, int budget) xdp_set_return_frame_no_direct(); done = veth_xdp_rcv(rq, budget, &bq, &stats); + if (stats.xdp_redirect > 0) + xdp_do_flush(); + if (done < budget && napi_complete_done(napi, done)) { /* Write rx_notify_masked before reading ptr_ring */ smp_store_mb(rq->rx_notify_masked, false); @@ -987,8 +990,6 @@ static int veth_poll(struct napi_struct *napi, int budget) if (stats.xdp_tx > 0) veth_xdp_flush(rq, &bq); - if (stats.xdp_redirect > 0) - xdp_do_flush(); xdp_clear_return_frame_no_direct(); return done; From 55171f2930be98c8a49991435cdf3a8b574353b6 Mon Sep 17 00:00:00 2001 From: Anton Protopopov Date: Thu, 22 Dec 2022 10:26:27 +0000 Subject: [PATCH 023/112] bpftool: Fix linkage with statically built libllvm Since the commit eb9d1acf634b ("bpftool: Add LLVM as default library for disassembling JIT-ed programs") we might link the bpftool program with the libllvm library. This works fine when a shared libllvm library is available, but fails if we want to link bpftool with a statically built LLVM: [...] /usr/bin/ld: /usr/local/lib/libLLVMSupport.a(CrashRecoveryContext.cpp.o): in function `llvm::CrashRecoveryContextCleanup::~CrashRecoveryContextCleanup()': CrashRecoveryContext.cpp:(.text._ZN4llvm27CrashRecoveryContextCleanupD0Ev+0x17): undefined reference to `operator delete(void*, unsigned long)' /usr/bin/ld: /usr/local/lib/libLLVMSupport.a(CrashRecoveryContext.cpp.o): in function `llvm::CrashRecoveryContext::~CrashRecoveryContext()': CrashRecoveryContext.cpp:(.text._ZN4llvm20CrashRecoveryContextD2Ev+0xc8): undefined reference to `operator delete(void*, unsigned long)' [...] So in the case of static libllvm we need to explicitly link bpftool with required libraries, namely, libstdc++ and those provided by the `llvm-config --system-libs` command. We can distinguish between the shared and static cases by using the `llvm-config --shared-mode` command. Fixes: eb9d1acf634b ("bpftool: Add LLVM as default library for disassembling JIT-ed programs") Signed-off-by: Anton Protopopov Signed-off-by: Daniel Borkmann Acked-by: Stanislav Fomichev Link: https://lore.kernel.org/bpf/20221222102627.1643709-1-aspsk@isovalent.com --- tools/bpf/bpftool/Makefile | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/bpf/bpftool/Makefile b/tools/bpf/bpftool/Makefile index 787b857d3fb5..f610e184ce02 100644 --- a/tools/bpf/bpftool/Makefile +++ b/tools/bpf/bpftool/Makefile @@ -145,6 +145,10 @@ ifeq ($(feature-llvm),1) LLVM_CONFIG_LIB_COMPONENTS := mcdisassembler all-targets CFLAGS += $(shell $(LLVM_CONFIG) --cflags --libs $(LLVM_CONFIG_LIB_COMPONENTS)) LIBS += $(shell $(LLVM_CONFIG) --libs $(LLVM_CONFIG_LIB_COMPONENTS)) + ifeq ($(shell $(LLVM_CONFIG) --shared-mode),static) + LIBS += $(shell $(LLVM_CONFIG) --system-libs $(LLVM_CONFIG_LIB_COMPONENTS)) + LIBS += -lstdc++ + endif LDFLAGS += $(shell $(LLVM_CONFIG) --ldflags) else # Fall back on libbfd From 8374bfd5a3c90a5b250f7c087c4d2b8ac467b12e Mon Sep 17 00:00:00 2001 From: Hao Sun Date: Thu, 22 Dec 2022 10:44:13 +0800 Subject: [PATCH 024/112] bpf: fix nullness propagation for reg to reg comparisons After befae75856ab, the verifier would propagate null information after JEQ/JNE, e.g., if two pointers, one is maybe_null and the other is not, the former would be marked as non-null in eq path. However, as comment "PTR_TO_BTF_ID points to a kernel struct that does not need to be null checked by the BPF program ... The verifier must keep this in mind and can make no assumptions about null or non-null when doing branch ...". If one pointer is maybe_null and the other is PTR_TO_BTF, the former is incorrectly marked non-null. The following BPF prog can trigger a null-ptr-deref, also see this report for more details[1]: 0: (18) r1 = map_fd ; R1_w=map_ptr(ks=4, vs=4) 2: (79) r6 = *(u64 *)(r1 +8) ; R6_w=bpf_map->inner_map_data ; R6 is PTR_TO_BTF_ID ; equals to null at runtime 3: (bf) r2 = r10 4: (07) r2 += -4 5: (62) *(u32 *)(r2 +0) = 0 6: (85) call bpf_map_lookup_elem#1 ; R0_w=map_value_or_null 7: (1d) if r6 == r0 goto pc+1 8: (95) exit ; from 7 to 9: R0=map_value R6=ptr_bpf_map 9: (61) r0 = *(u32 *)(r0 +0) ; null-ptr-deref 10: (95) exit So, make the verifier propagate nullness information for reg to reg comparisons only if neither reg is PTR_TO_BTF_ID. [1] https://lore.kernel.org/bpf/CACkBjsaFJwjC5oiw-1KXvcazywodwXo4zGYsRHwbr2gSG9WcSw@mail.gmail.com/T/#u Fixes: befae75856ab ("bpf: propagate nullness information for reg to reg comparisons") Signed-off-by: Hao Sun Acked-by: Yonghong Song Link: https://lore.kernel.org/r/20221222024414.29539-1-sunhao.th@gmail.com Signed-off-by: Martin KaFai Lau --- kernel/bpf/verifier.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index a5255a0dcbb6..243d06ce6842 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -11822,10 +11822,17 @@ static int check_cond_jmp_op(struct bpf_verifier_env *env, * register B - not null * for JNE A, B, ... - A is not null in the false branch; * for JEQ A, B, ... - A is not null in the true branch. + * + * Since PTR_TO_BTF_ID points to a kernel struct that does + * not need to be null checked by the BPF program, i.e., + * could be null even without PTR_MAYBE_NULL marking, so + * only propagate nullness when neither reg is that type. */ if (!is_jmp32 && BPF_SRC(insn->code) == BPF_X && __is_pointer_value(false, src_reg) && __is_pointer_value(false, dst_reg) && - type_may_be_null(src_reg->type) != type_may_be_null(dst_reg->type)) { + type_may_be_null(src_reg->type) != type_may_be_null(dst_reg->type) && + base_type(src_reg->type) != PTR_TO_BTF_ID && + base_type(dst_reg->type) != PTR_TO_BTF_ID) { eq_branch_regs = NULL; switch (opcode) { case BPF_JEQ: From cedebd74cf3883f0384af9ec26b4e6f8f1964dd4 Mon Sep 17 00:00:00 2001 From: Hao Sun Date: Thu, 22 Dec 2022 10:44:14 +0800 Subject: [PATCH 025/112] selftests/bpf: check null propagation only neither reg is PTR_TO_BTF_ID Verify that nullness information is not porpagated in the branches of register to register JEQ and JNE operations if one of them is PTR_TO_BTF_ID. Implement this in C level so we can use CO-RE. Signed-off-by: Hao Sun Suggested-by: Martin KaFai Lau Link: https://lore.kernel.org/r/20221222024414.29539-2-sunhao.th@gmail.com Signed-off-by: Martin KaFai Lau --- .../bpf/prog_tests/jeq_infer_not_null.c | 9 ++++ .../bpf/progs/jeq_infer_not_null_fail.c | 42 +++++++++++++++++++ 2 files changed, 51 insertions(+) create mode 100644 tools/testing/selftests/bpf/prog_tests/jeq_infer_not_null.c create mode 100644 tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c diff --git a/tools/testing/selftests/bpf/prog_tests/jeq_infer_not_null.c b/tools/testing/selftests/bpf/prog_tests/jeq_infer_not_null.c new file mode 100644 index 000000000000..3add34df5767 --- /dev/null +++ b/tools/testing/selftests/bpf/prog_tests/jeq_infer_not_null.c @@ -0,0 +1,9 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include +#include "jeq_infer_not_null_fail.skel.h" + +void test_jeq_infer_not_null(void) +{ + RUN_TESTS(jeq_infer_not_null_fail); +} diff --git a/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c b/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c new file mode 100644 index 000000000000..f46965053acb --- /dev/null +++ b/tools/testing/selftests/bpf/progs/jeq_infer_not_null_fail.c @@ -0,0 +1,42 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "vmlinux.h" +#include +#include "bpf_misc.h" + +char _license[] SEC("license") = "GPL"; + +struct { + __uint(type, BPF_MAP_TYPE_HASH); + __uint(max_entries, 1); + __type(key, u64); + __type(value, u64); +} m_hash SEC(".maps"); + +SEC("?raw_tp") +__failure __msg("R8 invalid mem access 'map_value_or_null") +int jeq_infer_not_null_ptr_to_btfid(void *ctx) +{ + struct bpf_map *map = (struct bpf_map *)&m_hash; + struct bpf_map *inner_map = map->inner_map_meta; + u64 key = 0, ret = 0, *val; + + val = bpf_map_lookup_elem(map, &key); + /* Do not mark ptr as non-null if one of them is + * PTR_TO_BTF_ID (R9), reject because of invalid + * access to map value (R8). + * + * Here, we need to inline those insns to access + * R8 directly, since compiler may use other reg + * once it figures out val==inner_map. + */ + asm volatile("r8 = %[val];\n" + "r9 = %[inner_map];\n" + "if r8 != r9 goto +1;\n" + "%[ret] = *(u64 *)(r8 +0);\n" + : [ret] "+r"(ret) + : [inner_map] "r"(inner_map), [val] "r"(val) + : "r8", "r9"); + + return ret; +} From 8d8bee13ae9e316443c6666286360126a19c8d94 Mon Sep 17 00:00:00 2001 From: Sean Anderson Date: Fri, 16 Dec 2022 12:29:37 -0500 Subject: [PATCH 026/112] powerpc: dts: t208x: Disable 10G on MAC1 and MAC2 There aren't enough resources to run these ports at 10G speeds. Disable 10G for these ports, reverting to the previous speed. Fixes: 36926a7d70c2 ("powerpc: dts: t208x: Mark MAC1 and MAC2 as 10G") Reported-by: Camelia Alexandra Groza Signed-off-by: Sean Anderson Reviewed-by: Camelia Groza Tested-by: Camelia Groza Link: https://lore.kernel.org/r/20221216172937.2960054-1-sean.anderson@seco.com Signed-off-by: Jakub Kicinski --- arch/powerpc/boot/dts/fsl/t2081si-post.dtsi | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/arch/powerpc/boot/dts/fsl/t2081si-post.dtsi b/arch/powerpc/boot/dts/fsl/t2081si-post.dtsi index 74e17e134387..27714dc2f04a 100644 --- a/arch/powerpc/boot/dts/fsl/t2081si-post.dtsi +++ b/arch/powerpc/boot/dts/fsl/t2081si-post.dtsi @@ -659,3 +659,19 @@ interrupts = <16 2 1 9>; }; }; + +&fman0_rx_0x08 { + /delete-property/ fsl,fman-10g-port; +}; + +&fman0_tx_0x28 { + /delete-property/ fsl,fman-10g-port; +}; + +&fman0_rx_0x09 { + /delete-property/ fsl,fman-10g-port; +}; + +&fman0_tx_0x29 { + /delete-property/ fsl,fman-10g-port; +}; From 7fac54b93ad13e5e7ac237af33eb2a0940eaeea0 Mon Sep 17 00:00:00 2001 From: Rong Tao Date: Wed, 21 Dec 2022 20:36:27 +0800 Subject: [PATCH 027/112] atm: uapi: fix spelling typos in comments Fix the typo of 'Unsuported' in atmbr2684.h Signed-off-by: Rong Tao Link: https://lore.kernel.org/r/tencent_F1354BEC925C65EA357E741E91DF2044E805@qq.com Signed-off-by: Jakub Kicinski --- include/uapi/linux/atmbr2684.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/atmbr2684.h b/include/uapi/linux/atmbr2684.h index a9e2250cd720..d47c47d06f11 100644 --- a/include/uapi/linux/atmbr2684.h +++ b/include/uapi/linux/atmbr2684.h @@ -38,7 +38,7 @@ */ #define BR2684_ENCAPS_VC (0) /* VC-mux */ #define BR2684_ENCAPS_LLC (1) -#define BR2684_ENCAPS_AUTODETECT (2) /* Unsuported */ +#define BR2684_ENCAPS_AUTODETECT (2) /* Unsupported */ /* * Is this VC bridged or routed? From 09e6b30eeb254f1818a008cace3547159e908dfd Mon Sep 17 00:00:00 2001 From: Jie Wang Date: Thu, 22 Dec 2022 14:43:41 +0800 Subject: [PATCH 028/112] net: hns3: add interrupts re-initialization while doing VF FLR Currently keep alive message between PF and VF may be lost and the VF is unalive in PF. So the VF will not do reset during PF FLR reset process. This would make the allocated interrupt resources of VF invalid and VF would't receive or respond to PF any more. So this patch adds VF interrupts re-initialization during VF FLR for VF recovery in above cases. Fixes: 862d969a3a4d ("net: hns3: do VF's pci re-initialization while PF doing FLR") Signed-off-by: Jie Wang Signed-off-by: Hao Lan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index db6f7cdba958..081bd2c3f289 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -2767,7 +2767,8 @@ static int hclgevf_pci_reset(struct hclgevf_dev *hdev) struct pci_dev *pdev = hdev->pdev; int ret = 0; - if (hdev->reset_type == HNAE3_VF_FULL_RESET && + if ((hdev->reset_type == HNAE3_VF_FULL_RESET || + hdev->reset_type == HNAE3_FLR_RESET) && test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) { hclgevf_misc_irq_uninit(hdev); hclgevf_uninit_msi(hdev); From 7d89b53cea1a702f97117fb4361523519bb1e52c Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Thu, 22 Dec 2022 14:43:42 +0800 Subject: [PATCH 029/112] net: hns3: fix miss L3E checking for rx packet For device supports RXD advanced layout, the driver will return directly if the hardware finish the checksum calculate. It cause missing L3E checking for ip packets. Fixes it. Fixes: 1ddc028ac849 ("net: hns3: refactor out RX completion checksum") Signed-off-by: Jian Shen Signed-off-by: Hao Lan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index 0ec5730b1788..b4c4fb873568 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -3855,18 +3855,16 @@ static int hns3_gro_complete(struct sk_buff *skb, u32 l234info) return 0; } -static bool hns3_checksum_complete(struct hns3_enet_ring *ring, +static void hns3_checksum_complete(struct hns3_enet_ring *ring, struct sk_buff *skb, u32 ptype, u16 csum) { if (ptype == HNS3_INVALID_PTYPE || hns3_rx_ptype_tbl[ptype].ip_summed != CHECKSUM_COMPLETE) - return false; + return; hns3_ring_stats_update(ring, csum_complete); skb->ip_summed = CHECKSUM_COMPLETE; skb->csum = csum_unfold((__force __sum16)csum); - - return true; } static void hns3_rx_handle_csum(struct sk_buff *skb, u32 l234info, @@ -3926,8 +3924,7 @@ static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb, ptype = hnae3_get_field(ol_info, HNS3_RXD_PTYPE_M, HNS3_RXD_PTYPE_S); - if (hns3_checksum_complete(ring, skb, ptype, csum)) - return; + hns3_checksum_complete(ring, skb, ptype, csum); /* check if hardware has done checksum */ if (!(bd_base_info & BIT(HNS3_RXD_L3L4P_B))) @@ -3936,6 +3933,7 @@ static void hns3_rx_checksum(struct hns3_enet_ring *ring, struct sk_buff *skb, if (unlikely(l234info & (BIT(HNS3_RXD_L3E_B) | BIT(HNS3_RXD_L4E_B) | BIT(HNS3_RXD_OL3E_B) | BIT(HNS3_RXD_OL4E_B)))) { + skb->ip_summed = CHECKSUM_NONE; hns3_ring_stats_update(ring, l3l4_csum_err); return; From 8ee57c7b8406c7aa8ca31e014440c87c6383f429 Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Thu, 22 Dec 2022 14:43:43 +0800 Subject: [PATCH 030/112] net: hns3: fix VF promisc mode not update when mac table full Currently, it missed set HCLGE_VPORT_STATE_PROMISC_CHANGE flag for VF when vport->overflow_promisc_flags changed. So the VF won't check whether to update promisc mode in this case. So add it. Fixes: 1e6e76101fd9 ("net: hns3: configure promisc mode for VF asynchronously") Signed-off-by: Jian Shen Signed-off-by: Hao Lan Signed-off-by: Jakub Kicinski --- .../hisilicon/hns3/hns3pf/hclge_main.c | 83 +++++++++++-------- 1 file changed, 47 insertions(+), 36 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 4e54f91f7a6c..6c2742f59c77 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -12754,60 +12754,71 @@ static int hclge_gro_en(struct hnae3_handle *handle, bool enable) return ret; } -static void hclge_sync_promisc_mode(struct hclge_dev *hdev) +static int hclge_sync_vport_promisc_mode(struct hclge_vport *vport) { - struct hclge_vport *vport = &hdev->vport[0]; struct hnae3_handle *handle = &vport->nic; + struct hclge_dev *hdev = vport->back; + bool uc_en = false; + bool mc_en = false; u8 tmp_flags; + bool bc_en; int ret; - u16 i; if (vport->last_promisc_flags != vport->overflow_promisc_flags) { set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state); vport->last_promisc_flags = vport->overflow_promisc_flags; } - if (test_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state)) { + if (!test_and_clear_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, + &vport->state)) + return 0; + + /* for PF */ + if (!vport->vport_id) { tmp_flags = handle->netdev_flags | vport->last_promisc_flags; ret = hclge_set_promisc_mode(handle, tmp_flags & HNAE3_UPE, tmp_flags & HNAE3_MPE); - if (!ret) { - clear_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, - &vport->state); + if (!ret) set_bit(HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE, &vport->state); - } - } - - for (i = 1; i < hdev->num_alloc_vport; i++) { - bool uc_en = false; - bool mc_en = false; - bool bc_en; - - vport = &hdev->vport[i]; - - if (!test_and_clear_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, - &vport->state)) - continue; - - if (vport->vf_info.trusted) { - uc_en = vport->vf_info.request_uc_en > 0 || - vport->overflow_promisc_flags & - HNAE3_OVERFLOW_UPE; - mc_en = vport->vf_info.request_mc_en > 0 || - vport->overflow_promisc_flags & - HNAE3_OVERFLOW_MPE; - } - bc_en = vport->vf_info.request_bc_en > 0; - - ret = hclge_cmd_set_promisc_mode(hdev, vport->vport_id, uc_en, - mc_en, bc_en); - if (ret) { + else set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state); + return ret; + } + + /* for VF */ + if (vport->vf_info.trusted) { + uc_en = vport->vf_info.request_uc_en > 0 || + vport->overflow_promisc_flags & HNAE3_OVERFLOW_UPE; + mc_en = vport->vf_info.request_mc_en > 0 || + vport->overflow_promisc_flags & HNAE3_OVERFLOW_MPE; + } + bc_en = vport->vf_info.request_bc_en > 0; + + ret = hclge_cmd_set_promisc_mode(hdev, vport->vport_id, uc_en, + mc_en, bc_en); + if (ret) { + set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state); + return ret; + } + hclge_set_vport_vlan_fltr_change(vport); + + return 0; +} + +static void hclge_sync_promisc_mode(struct hclge_dev *hdev) +{ + struct hclge_vport *vport; + int ret; + u16 i; + + for (i = 0; i < hdev->num_alloc_vport; i++) { + vport = &hdev->vport[i]; + + ret = hclge_sync_vport_promisc_mode(vport); + if (ret) return; - } - hclge_set_vport_vlan_fltr_change(vport); } } From fcbb408a1aaf426f88d8fb3b4c14e3625745b02f Mon Sep 17 00:00:00 2001 From: Stanislav Fomichev Date: Thu, 22 Dec 2022 13:39:58 -0800 Subject: [PATCH 031/112] selftests/bpf: Add host-tools to gitignore Shows up when cross-compiling: HOST_SCRATCH_DIR := $(OUTPUT)/host-tools vs SCRATCH_DIR := $(OUTPUT)/tools HOST_SCRATCH_DIR := $(SCRATCH_DIR) Reported-by: John Sperbeck Signed-off-by: Stanislav Fomichev Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20221222213958.2302320-1-sdf@google.com --- tools/testing/selftests/bpf/.gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore index 07d2d0a8c5cb..401a75844cc0 100644 --- a/tools/testing/selftests/bpf/.gitignore +++ b/tools/testing/selftests/bpf/.gitignore @@ -36,6 +36,7 @@ test_cpp *.lskel.h /no_alu32 /bpf_gcc +/host-tools /tools /runqslower /bench From 399ab7fe0fa0d846881685fd4e57e9a8ef7559f7 Mon Sep 17 00:00:00 2001 From: Hawkins Jiawei Date: Thu, 22 Dec 2022 11:51:19 +0800 Subject: [PATCH 032/112] net: sched: fix memory leak in tcindex_set_parms Syzkaller reports a memory leak as follows: ==================================== BUG: memory leak unreferenced object 0xffff88810c287f00 (size 256): comm "syz-executor105", pid 3600, jiffies 4294943292 (age 12.990s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmalloc_trace+0x20/0x90 mm/slab_common.c:1046 [] kmalloc include/linux/slab.h:576 [inline] [] kmalloc_array include/linux/slab.h:627 [inline] [] kcalloc include/linux/slab.h:659 [inline] [] tcf_exts_init include/net/pkt_cls.h:250 [inline] [] tcindex_set_parms+0xa7/0xbe0 net/sched/cls_tcindex.c:342 [] tcindex_change+0xdf/0x120 net/sched/cls_tcindex.c:553 [] tc_new_tfilter+0x4f2/0x1100 net/sched/cls_api.c:2147 [] rtnetlink_rcv_msg+0x4dc/0x5d0 net/core/rtnetlink.c:6082 [] netlink_rcv_skb+0x87/0x1d0 net/netlink/af_netlink.c:2540 [] netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline] [] netlink_unicast+0x397/0x4c0 net/netlink/af_netlink.c:1345 [] netlink_sendmsg+0x396/0x710 net/netlink/af_netlink.c:1921 [] sock_sendmsg_nosec net/socket.c:714 [inline] [] sock_sendmsg+0x56/0x80 net/socket.c:734 [] ____sys_sendmsg+0x178/0x410 net/socket.c:2482 [] ___sys_sendmsg+0xa8/0x110 net/socket.c:2536 [] __sys_sendmmsg+0x105/0x330 net/socket.c:2622 [] __do_sys_sendmmsg net/socket.c:2651 [inline] [] __se_sys_sendmmsg net/socket.c:2648 [inline] [] __x64_sys_sendmmsg+0x24/0x30 net/socket.c:2648 [] do_syscall_x64 arch/x86/entry/common.c:50 [inline] [] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 [] entry_SYSCALL_64_after_hwframe+0x63/0xcd ==================================== Kernel uses tcindex_change() to change an existing filter properties. Yet the problem is that, during the process of changing, if `old_r` is retrieved from `p->perfect`, then kernel uses tcindex_alloc_perfect_hash() to newly allocate filter results, uses tcindex_filter_result_init() to clear the old filter result, without destroying its tcf_exts structure, which triggers the above memory leak. To be more specific, there are only two source for the `old_r`, according to the tcindex_lookup(). `old_r` is retrieved from `p->perfect`, or `old_r` is retrieved from `p->h`. * If `old_r` is retrieved from `p->perfect`, kernel uses tcindex_alloc_perfect_hash() to newly allocate the filter results. Then `r` is assigned with `cp->perfect + handle`, which is newly allocated. So condition `old_r && old_r != r` is true in this situation, and kernel uses tcindex_filter_result_init() to clear the old filter result, without destroying its tcf_exts structure * If `old_r` is retrieved from `p->h`, then `p->perfect` is NULL according to the tcindex_lookup(). Considering that `cp->h` is directly copied from `p->h` and `p->perfect` is NULL, `r` is assigned with `tcindex_lookup(cp, handle)`, whose value should be the same as `old_r`, so condition `old_r && old_r != r` is false in this situation, kernel ignores using tcindex_filter_result_init() to clear the old filter result. So only when `old_r` is retrieved from `p->perfect` does kernel use tcindex_filter_result_init() to clear the old filter result, which triggers the above memory leak. Considering that there already exists a tc_filter_wq workqueue to destroy the old tcindex_data by tcindex_partial_destroy_work() at the end of tcindex_set_parms(), this patch solves this memory leak bug by removing this old filter result clearing part and delegating it to the tc_filter_wq workqueue. Note that this patch doesn't introduce any other issues. If `old_r` is retrieved from `p->perfect`, this patch just delegates old filter result clearing part to the tc_filter_wq workqueue; If `old_r` is retrieved from `p->h`, kernel doesn't reach the old filter result clearing part, so removing this part has no effect. [Thanks to the suggestion from Jakub Kicinski, Cong Wang, Paolo Abeni and Dmitry Vyukov] Fixes: b9a24bb76bf6 ("net_sched: properly handle failure case of tcf_exts_init()") Link: https://lore.kernel.org/all/0000000000001de5c505ebc9ec59@google.com/ Reported-by: syzbot+232ebdbd36706c965ebf@syzkaller.appspotmail.com Tested-by: syzbot+232ebdbd36706c965ebf@syzkaller.appspotmail.com Cc: Cong Wang Cc: Jakub Kicinski Cc: Paolo Abeni Cc: Dmitry Vyukov Acked-by: Paolo Abeni Signed-off-by: Hawkins Jiawei Signed-off-by: David S. Miller --- net/sched/cls_tcindex.c | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c index eb0e9458e722..ee2a050c887b 100644 --- a/net/sched/cls_tcindex.c +++ b/net/sched/cls_tcindex.c @@ -333,7 +333,7 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, struct tcindex_filter_result *r, struct nlattr **tb, struct nlattr *est, u32 flags, struct netlink_ext_ack *extack) { - struct tcindex_filter_result new_filter_result, *old_r = r; + struct tcindex_filter_result new_filter_result; struct tcindex_data *cp = NULL, *oldp; struct tcindex_filter *f = NULL; /* make gcc behave */ struct tcf_result cr = {}; @@ -402,7 +402,7 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, err = tcindex_filter_result_init(&new_filter_result, cp, net); if (err < 0) goto errout_alloc; - if (old_r) + if (r) cr = r->res; err = -EBUSY; @@ -479,14 +479,6 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, tcf_bind_filter(tp, &cr, base); } - if (old_r && old_r != r) { - err = tcindex_filter_result_init(old_r, cp, net); - if (err < 0) { - kfree(f); - goto errout_alloc; - } - } - oldp = p; r->res = cr; tcf_exts_change(&r->exts, &e); From 13a7c8964afcd8ca43c0b6001ebb0127baa95362 Mon Sep 17 00:00:00 2001 From: Daniil Tatianin Date: Thu, 22 Dec 2022 14:52:28 +0300 Subject: [PATCH 033/112] qlcnic: prevent ->dcb use-after-free on qlcnic_dcb_enable() failure adapter->dcb would get silently freed inside qlcnic_dcb_enable() in case qlcnic_dcb_attach() would return an error, which always happens under OOM conditions. This would lead to use-after-free because both of the existing callers invoke qlcnic_dcb_get_info() on the obtained pointer, which is potentially freed at that point. Propagate errors from qlcnic_dcb_enable(), and instead free the dcb pointer at callsite using qlcnic_dcb_free(). This also removes the now unused qlcnic_clear_dcb_ops() helper, which was a simple wrapper around kfree() also causing memory leaks for partially initialized dcb. Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Fixes: 3c44bba1d270 ("qlcnic: Disable DCB operations from SR-IOV VFs") Reviewed-by: Michal Swiatkowski Signed-off-by: Daniil Tatianin Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c | 8 +++++++- drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h | 10 ++-------- drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c | 8 +++++++- 3 files changed, 16 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c index dbb800769cb6..c95d56e56c59 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c @@ -2505,7 +2505,13 @@ int qlcnic_83xx_init(struct qlcnic_adapter *adapter) goto disable_mbx_intr; qlcnic_83xx_clear_function_resources(adapter); - qlcnic_dcb_enable(adapter->dcb); + + err = qlcnic_dcb_enable(adapter->dcb); + if (err) { + qlcnic_dcb_free(adapter->dcb); + goto disable_mbx_intr; + } + qlcnic_83xx_initialize_nic(adapter, 1); qlcnic_dcb_get_info(adapter->dcb); diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h index 7519773eaca6..22afa2be85fd 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h @@ -41,11 +41,6 @@ struct qlcnic_dcb { unsigned long state; }; -static inline void qlcnic_clear_dcb_ops(struct qlcnic_dcb *dcb) -{ - kfree(dcb); -} - static inline int qlcnic_dcb_get_hw_capability(struct qlcnic_dcb *dcb) { if (dcb && dcb->ops->get_hw_capability) @@ -112,9 +107,8 @@ static inline void qlcnic_dcb_init_dcbnl_ops(struct qlcnic_dcb *dcb) dcb->ops->init_dcbnl_ops(dcb); } -static inline void qlcnic_dcb_enable(struct qlcnic_dcb *dcb) +static inline int qlcnic_dcb_enable(struct qlcnic_dcb *dcb) { - if (dcb && qlcnic_dcb_attach(dcb)) - qlcnic_clear_dcb_ops(dcb); + return dcb ? qlcnic_dcb_attach(dcb) : 0; } #endif diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c index 28476b982bab..44dac3c0908e 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c @@ -2599,7 +2599,13 @@ qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) "Device does not support MSI interrupts\n"); if (qlcnic_82xx_check(adapter)) { - qlcnic_dcb_enable(adapter->dcb); + err = qlcnic_dcb_enable(adapter->dcb); + if (err) { + qlcnic_dcb_free(adapter->dcb); + dev_err(&pdev->dev, "Failed to enable DCB\n"); + goto err_out_free_hw; + } + qlcnic_dcb_get_info(adapter->dcb); err = qlcnic_setup_intr(adapter); From 30e725537546248bddc12eaac2fe0a258917f190 Mon Sep 17 00:00:00 2001 From: "Johnny S. Lee" Date: Thu, 22 Dec 2022 22:34:05 +0800 Subject: [PATCH 034/112] net: dsa: mv88e6xxx: depend on PTP conditionally PTP hardware timestamping related objects are not linked when PTP support for MV88E6xxx (NET_DSA_MV88E6XXX_PTP) is disabled, therefore NET_DSA_MV88E6XXX should not depend on PTP_1588_CLOCK_OPTIONAL regardless of NET_DSA_MV88E6XXX_PTP. Instead, condition more strictly on how NET_DSA_MV88E6XXX_PTP's dependencies are met, making sure that it cannot be enabled when NET_DSA_MV88E6XXX=y and PTP_1588_CLOCK=m. In other words, this commit allows NET_DSA_MV88E6XXX to be built-in while PTP_1588_CLOCK is a module, as long as NET_DSA_MV88E6XXX_PTP is prevented from being enabled. Fixes: e5f31552674e ("ethernet: fix PTP_1588_CLOCK dependencies") Signed-off-by: Johnny S. Lee Signed-off-by: David S. Miller --- drivers/net/dsa/mv88e6xxx/Kconfig | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/dsa/mv88e6xxx/Kconfig b/drivers/net/dsa/mv88e6xxx/Kconfig index 7a2445a34eb7..e3181d5471df 100644 --- a/drivers/net/dsa/mv88e6xxx/Kconfig +++ b/drivers/net/dsa/mv88e6xxx/Kconfig @@ -2,7 +2,6 @@ config NET_DSA_MV88E6XXX tristate "Marvell 88E6xxx Ethernet switch fabric support" depends on NET_DSA - depends on PTP_1588_CLOCK_OPTIONAL select IRQ_DOMAIN select NET_DSA_TAG_EDSA select NET_DSA_TAG_DSA @@ -13,7 +12,8 @@ config NET_DSA_MV88E6XXX config NET_DSA_MV88E6XXX_PTP bool "PTP support for Marvell 88E6xxx" default n - depends on NET_DSA_MV88E6XXX && PTP_1588_CLOCK + depends on (NET_DSA_MV88E6XXX = y && PTP_1588_CLOCK = y) || \ + (NET_DSA_MV88E6XXX = m && PTP_1588_CLOCK) help Say Y to enable PTP hardware timestamping on Marvell 88E6xxx switch chips that support it. From df49908f3c52d211aea5e2a14a93bbe67a2cb3af Mon Sep 17 00:00:00 2001 From: Miaoqian Lin Date: Fri, 23 Dec 2022 11:37:18 +0400 Subject: [PATCH 035/112] nfc: Fix potential resource leaks nfc_get_device() take reference for the device, add missing nfc_put_device() to release it when not need anymore. Also fix the style warnning by use error EOPNOTSUPP instead of ENOTSUPP. Fixes: 5ce3f32b5264 ("NFC: netlink: SE API implementation") Fixes: 29e76924cf08 ("nfc: netlink: Add capability to reply to vendor_cmd with data") Signed-off-by: Miaoqian Lin Signed-off-by: David S. Miller --- net/nfc/netlink.c | 52 ++++++++++++++++++++++++++++++++++------------- 1 file changed, 38 insertions(+), 14 deletions(-) diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c index 9d91087b9399..1fc339084d89 100644 --- a/net/nfc/netlink.c +++ b/net/nfc/netlink.c @@ -1497,6 +1497,7 @@ static int nfc_genl_se_io(struct sk_buff *skb, struct genl_info *info) u32 dev_idx, se_idx; u8 *apdu; size_t apdu_len; + int rc; if (!info->attrs[NFC_ATTR_DEVICE_INDEX] || !info->attrs[NFC_ATTR_SE_INDEX] || @@ -1510,25 +1511,37 @@ static int nfc_genl_se_io(struct sk_buff *skb, struct genl_info *info) if (!dev) return -ENODEV; - if (!dev->ops || !dev->ops->se_io) - return -ENOTSUPP; + if (!dev->ops || !dev->ops->se_io) { + rc = -EOPNOTSUPP; + goto put_dev; + } apdu_len = nla_len(info->attrs[NFC_ATTR_SE_APDU]); - if (apdu_len == 0) - return -EINVAL; + if (apdu_len == 0) { + rc = -EINVAL; + goto put_dev; + } apdu = nla_data(info->attrs[NFC_ATTR_SE_APDU]); - if (!apdu) - return -EINVAL; + if (!apdu) { + rc = -EINVAL; + goto put_dev; + } ctx = kzalloc(sizeof(struct se_io_ctx), GFP_KERNEL); - if (!ctx) - return -ENOMEM; + if (!ctx) { + rc = -ENOMEM; + goto put_dev; + } ctx->dev_idx = dev_idx; ctx->se_idx = se_idx; - return nfc_se_io(dev, se_idx, apdu, apdu_len, se_io_cb, ctx); + rc = nfc_se_io(dev, se_idx, apdu, apdu_len, se_io_cb, ctx); + +put_dev: + nfc_put_device(dev); + return rc; } static int nfc_genl_vendor_cmd(struct sk_buff *skb, @@ -1551,14 +1564,21 @@ static int nfc_genl_vendor_cmd(struct sk_buff *skb, subcmd = nla_get_u32(info->attrs[NFC_ATTR_VENDOR_SUBCMD]); dev = nfc_get_device(dev_idx); - if (!dev || !dev->vendor_cmds || !dev->n_vendor_cmds) + if (!dev) return -ENODEV; + if (!dev->vendor_cmds || !dev->n_vendor_cmds) { + err = -ENODEV; + goto put_dev; + } + if (info->attrs[NFC_ATTR_VENDOR_DATA]) { data = nla_data(info->attrs[NFC_ATTR_VENDOR_DATA]); data_len = nla_len(info->attrs[NFC_ATTR_VENDOR_DATA]); - if (data_len == 0) - return -EINVAL; + if (data_len == 0) { + err = -EINVAL; + goto put_dev; + } } else { data = NULL; data_len = 0; @@ -1573,10 +1593,14 @@ static int nfc_genl_vendor_cmd(struct sk_buff *skb, dev->cur_cmd_info = info; err = cmd->doit(dev, data, data_len); dev->cur_cmd_info = NULL; - return err; + goto put_dev; } - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + +put_dev: + nfc_put_device(dev); + return err; } /* message building helper */ From d3805695fe1e7383517903715cefc9bbdcffdc90 Mon Sep 17 00:00:00 2001 From: Anuradha Weeraman Date: Sun, 25 Dec 2022 23:12:22 +0530 Subject: [PATCH 036/112] net: ethernet: marvell: octeontx2: Fix uninitialized variable warning Fix for uninitialized variable warning. Addresses-Coverity: ("Uninitialized scalar variable") Signed-off-by: Anuradha Weeraman Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/octeontx2/af/mcs_rvu_if.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mcs_rvu_if.c b/drivers/net/ethernet/marvell/octeontx2/af/mcs_rvu_if.c index fa8029a94068..eb25e458266c 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/mcs_rvu_if.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/mcs_rvu_if.c @@ -589,7 +589,7 @@ int rvu_mbox_handler_mcs_free_resources(struct rvu *rvu, u16 pcifunc = req->hdr.pcifunc; struct mcs_rsrc_map *map; struct mcs *mcs; - int rc; + int rc = 0; if (req->mcs_id >= rvu->mcs_blk_cnt) return MCS_AF_ERR_INVALID_MCSID; From 0e50d999903c009b6a9cd2277c82d6798d982e31 Mon Sep 17 00:00:00 2001 From: David Howells Date: Sat, 24 Dec 2022 14:49:00 +0000 Subject: [PATCH 037/112] rxrpc: Fix a couple of potential use-after-frees At the end of rxrpc_recvmsg(), if a call is found, the call is put and then a trace line is emitted referencing that call in a couple of places - but the call may have been deallocated by the time those traces happen. Fix this by stashing the call debug_id in a variable and passing that to the tracepoint rather than the call pointer. Fixes: 849979051cbc ("rxrpc: Add a tracepoint to follow what recvmsg does") Signed-off-by: David Howells cc: Marc Dionne cc: linux-afs@lists.infradead.org Signed-off-by: David S. Miller --- include/trace/events/rxrpc.h | 6 +++--- net/rxrpc/recvmsg.c | 14 ++++++++------ 2 files changed, 11 insertions(+), 9 deletions(-) diff --git a/include/trace/events/rxrpc.h b/include/trace/events/rxrpc.h index c6cfed00d0c6..5f9dd7389536 100644 --- a/include/trace/events/rxrpc.h +++ b/include/trace/events/rxrpc.h @@ -1062,10 +1062,10 @@ TRACE_EVENT(rxrpc_receive, ); TRACE_EVENT(rxrpc_recvmsg, - TP_PROTO(struct rxrpc_call *call, enum rxrpc_recvmsg_trace why, + TP_PROTO(unsigned int call_debug_id, enum rxrpc_recvmsg_trace why, int ret), - TP_ARGS(call, why, ret), + TP_ARGS(call_debug_id, why, ret), TP_STRUCT__entry( __field(unsigned int, call ) @@ -1074,7 +1074,7 @@ TRACE_EVENT(rxrpc_recvmsg, ), TP_fast_assign( - __entry->call = call ? call->debug_id : 0; + __entry->call = call_debug_id; __entry->why = why; __entry->ret = ret; ), diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c index 36b25d003cf0..6ebd6440a2b7 100644 --- a/net/rxrpc/recvmsg.c +++ b/net/rxrpc/recvmsg.c @@ -388,13 +388,14 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, struct rxrpc_call *call; struct rxrpc_sock *rx = rxrpc_sk(sock->sk); struct list_head *l; + unsigned int call_debug_id = 0; size_t copied = 0; long timeo; int ret; DEFINE_WAIT(wait); - trace_rxrpc_recvmsg(NULL, rxrpc_recvmsg_enter, 0); + trace_rxrpc_recvmsg(0, rxrpc_recvmsg_enter, 0); if (flags & (MSG_OOB | MSG_TRUNC)) return -EOPNOTSUPP; @@ -431,7 +432,7 @@ try_again: if (list_empty(&rx->recvmsg_q)) { if (signal_pending(current)) goto wait_interrupted; - trace_rxrpc_recvmsg(NULL, rxrpc_recvmsg_wait, 0); + trace_rxrpc_recvmsg(0, rxrpc_recvmsg_wait, 0); timeo = schedule_timeout(timeo); } finish_wait(sk_sleep(&rx->sk), &wait); @@ -450,7 +451,8 @@ try_again: rxrpc_get_call(call, rxrpc_call_get_recvmsg); write_unlock(&rx->recvmsg_lock); - trace_rxrpc_recvmsg(call, rxrpc_recvmsg_dequeue, 0); + call_debug_id = call->debug_id; + trace_rxrpc_recvmsg(call_debug_id, rxrpc_recvmsg_dequeue, 0); /* We're going to drop the socket lock, so we need to lock the call * against interference by sendmsg. @@ -531,7 +533,7 @@ try_again: error_unlock_call: mutex_unlock(&call->user_mutex); rxrpc_put_call(call, rxrpc_call_put_recvmsg); - trace_rxrpc_recvmsg(call, rxrpc_recvmsg_return, ret); + trace_rxrpc_recvmsg(call_debug_id, rxrpc_recvmsg_return, ret); return ret; error_requeue_call: @@ -539,14 +541,14 @@ error_requeue_call: write_lock(&rx->recvmsg_lock); list_add(&call->recvmsg_link, &rx->recvmsg_q); write_unlock(&rx->recvmsg_lock); - trace_rxrpc_recvmsg(call, rxrpc_recvmsg_requeue, 0); + trace_rxrpc_recvmsg(call_debug_id, rxrpc_recvmsg_requeue, 0); } else { rxrpc_put_call(call, rxrpc_call_put_recvmsg); } error_no_call: release_sock(&rx->sk); error_trace: - trace_rxrpc_recvmsg(call, rxrpc_recvmsg_return, ret); + trace_rxrpc_recvmsg(call_debug_id, rxrpc_recvmsg_return, ret); return ret; wait_interrupted: From f4ef681115f822daf7f36f8b1892d9f1e1a26fbf Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 22 Dec 2022 11:22:47 -0800 Subject: [PATCH 038/112] docs: netdev: reshuffle sections in prep for de-FAQization Subsequent changes will reformat the doc away from FAQ. To make that more readable perform the pure section moves now. Reviewed-by: Randy Dunlap Reviewed-by: Andrew Lunn Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- Documentation/process/maintainer-netdev.rst | 186 ++++++++++---------- 1 file changed, 93 insertions(+), 93 deletions(-) diff --git a/Documentation/process/maintainer-netdev.rst b/Documentation/process/maintainer-netdev.rst index 1fa5ab8754d3..8f22f8a3dcd1 100644 --- a/Documentation/process/maintainer-netdev.rst +++ b/Documentation/process/maintainer-netdev.rst @@ -44,17 +44,6 @@ for the future release. You can find the trees here: - https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git - https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git -How do I indicate which tree (net vs. net-next) my patch should be in? ----------------------------------------------------------------------- -To help maintainers and CI bots you should explicitly mark which tree -your patch is targeting. Assuming that you use git, use the prefix -flag:: - - git format-patch --subject-prefix='PATCH net-next' start..finish - -Use ``net`` instead of ``net-next`` (always lower case) in the above for -bug-fix ``net`` content. - How often do changes from these trees make it to the mainline Linus tree? ------------------------------------------------------------------------- To understand this, you need to know a bit of background information on @@ -127,15 +116,6 @@ patch. Patches are indexed by the ``Message-ID`` header of the emails which carried them so if you have trouble finding your patch append the value of ``Message-ID`` to the URL above. -How long before my patch is accepted? -------------------------------------- -Generally speaking, the patches get triaged quickly (in less than -48h). But be patient, if your patch is active in patchwork (i.e. it's -listed on the project's patch list) the chances it was missed are close to zero. -Asking the maintainer for status updates on your -patch is a good way to ensure your patch is ignored or pushed to the -bottom of the priority list. - Should I directly update patchwork state of my own patches? ----------------------------------------------------------- It may be tempting to help the maintainers and update the state of your @@ -145,19 +125,14 @@ it to the maintainer to figure out what is the most recent and current version that should be applied. If there is any doubt, the maintainer will reply and ask what should be done. -How do I divide my work into patches? +How long before my patch is accepted? ------------------------------------- - -Put yourself in the shoes of the reviewer. Each patch is read separately -and therefore should constitute a comprehensible step towards your stated -goal. - -Avoid sending series longer than 15 patches. Larger series takes longer -to review as reviewers will defer looking at it until they find a large -chunk of time. A small series can be reviewed in a short time, so Maintainers -just do it. As a result, a sequence of smaller series gets merged quicker and -with better review coverage. Re-posting large series also increases the mailing -list traffic. +Generally speaking, the patches get triaged quickly (in less than +48h). But be patient, if your patch is active in patchwork (i.e. it's +listed on the project's patch list) the chances it was missed are close to zero. +Asking the maintainer for status updates on your +patch is a good way to ensure your patch is ignored or pushed to the +bottom of the priority list. I made changes to only a few patches in a patch series should I resend only those changed? ------------------------------------------------------------------------------------------ @@ -165,17 +140,6 @@ No, please resend the entire patch series and make sure you do number your patches such that it is clear this is the latest and greatest set of patches that can be applied. -I have received review feedback, when should I post a revised version of the patches? -------------------------------------------------------------------------------------- -Allow at least 24 hours to pass between postings. This will ensure reviewers -from all geographical locations have a chance to chime in. Do not wait -too long (weeks) between postings either as it will make it harder for reviewers -to recall all the context. - -Make sure you address all the feedback in your new posting. Do not post a new -version of the code if the discussion about the previous version is still -ongoing, unless directly instructed by a reviewer. - I submitted multiple versions of a patch series and it looks like a version other than the last one has been accepted, what should I do? ---------------------------------------------------------------------------------------------------------------------------------------- There is no revert possible, once it is pushed out, it stays like that. @@ -191,6 +155,82 @@ the case today. Please follow the standard stable rules in :ref:`Documentation/process/stable-kernel-rules.rst `, and make sure you include appropriate Fixes tags! +I found a bug that might have possible security implications or similar. Should I mail the main netdev maintainer off-list? +--------------------------------------------------------------------------------------------------------------------------- +No. The current netdev maintainer has consistently requested that +people use the mailing lists and not reach out directly. If you aren't +OK with that, then perhaps consider mailing security@kernel.org or +reading about http://oss-security.openwall.org/wiki/mailing-lists/distros +as possible alternative mechanisms. + +How do I post corresponding changes to user space components? +------------------------------------------------------------- +User space code exercising kernel features should be posted +alongside kernel patches. This gives reviewers a chance to see +how any new interface is used and how well it works. + +When user space tools reside in the kernel repo itself all changes +should generally come as one series. If series becomes too large +or the user space project is not reviewed on netdev include a link +to a public repo where user space patches can be seen. + +In case user space tooling lives in a separate repository but is +reviewed on netdev (e.g. patches to ``iproute2`` tools) kernel and +user space patches should form separate series (threads) when posted +to the mailing list, e.g.:: + + [PATCH net-next 0/3] net: some feature cover letter + └─ [PATCH net-next 1/3] net: some feature prep + └─ [PATCH net-next 2/3] net: some feature do it + └─ [PATCH net-next 3/3] selftest: net: some feature + + [PATCH iproute2-next] ip: add support for some feature + +Posting as one thread is discouraged because it confuses patchwork +(as of patchwork 2.2.2). + +Any other tips to help ensure my net/net-next patch gets OK'd? +-------------------------------------------------------------- +Attention to detail. Re-read your own work as if you were the +reviewer. You can start with using ``checkpatch.pl``, perhaps even with +the ``--strict`` flag. But do not be mindlessly robotic in doing so. +If your change is a bug fix, make sure your commit log indicates the +end-user visible symptom, the underlying reason as to why it happens, +and then if necessary, explain why the fix proposed is the best way to +get things done. Don't mangle whitespace, and as is common, don't +mis-indent function arguments that span multiple lines. If it is your +first patch, mail it to yourself so you can test apply it to an +unpatched tree to confirm infrastructure didn't mangle it. + +Finally, go back and read +:ref:`Documentation/process/submitting-patches.rst ` +to be sure you are not repeating some common mistake documented there. + +How do I indicate which tree (net vs. net-next) my patch should be in? +---------------------------------------------------------------------- +To help maintainers and CI bots you should explicitly mark which tree +your patch is targeting. Assuming that you use git, use the prefix +flag:: + + git format-patch --subject-prefix='PATCH net-next' start..finish + +Use ``net`` instead of ``net-next`` (always lower case) in the above for +bug-fix ``net`` content. + +How do I divide my work into patches? +------------------------------------- + +Put yourself in the shoes of the reviewer. Each patch is read separately +and therefore should constitute a comprehensible step towards your stated +goal. + +Avoid sending series longer than 15 patches. Larger series takes longer +to review as reviewers will defer looking at it until they find a large +chunk of time. A small series can be reviewed in a short time, so Maintainers +just do it. As a result, a sequence of smaller series gets merged quicker and +with better review coverage. Re-posting large series also increases the mailing +list traffic. + Is the comment style convention different for the networking content? --------------------------------------------------------------------- Yes, in a largely trivial way. Instead of this:: @@ -224,13 +264,16 @@ I am working in existing code which uses non-standard formatting. Which formatti Make your code follow the most recent guidelines, so that eventually all code in the domain of netdev is in the preferred format. -I found a bug that might have possible security implications or similar. Should I mail the main netdev maintainer off-list? ---------------------------------------------------------------------------------------------------------------------------- -No. The current netdev maintainer has consistently requested that -people use the mailing lists and not reach out directly. If you aren't -OK with that, then perhaps consider mailing security@kernel.org or -reading about http://oss-security.openwall.org/wiki/mailing-lists/distros -as possible alternative mechanisms. +I have received review feedback, when should I post a revised version of the patches? +------------------------------------------------------------------------------------- +Allow at least 24 hours to pass between postings. This will ensure reviewers +from all geographical locations have a chance to chime in. Do not wait +too long (weeks) between postings either as it will make it harder for reviewers +to recall all the context. + +Make sure you address all the feedback in your new posting. Do not post a new +version of the code if the discussion about the previous version is still +ongoing, unless directly instructed by a reviewer. What level of testing is expected before I submit my change? ------------------------------------------------------------ @@ -244,32 +287,6 @@ and the patch series contains a set of kernel selftest for You are expected to test your changes on top of the relevant networking tree (``net`` or ``net-next``) and not e.g. a stable tree or ``linux-next``. -How do I post corresponding changes to user space components? -------------------------------------------------------------- -User space code exercising kernel features should be posted -alongside kernel patches. This gives reviewers a chance to see -how any new interface is used and how well it works. - -When user space tools reside in the kernel repo itself all changes -should generally come as one series. If series becomes too large -or the user space project is not reviewed on netdev include a link -to a public repo where user space patches can be seen. - -In case user space tooling lives in a separate repository but is -reviewed on netdev (e.g. patches to ``iproute2`` tools) kernel and -user space patches should form separate series (threads) when posted -to the mailing list, e.g.:: - - [PATCH net-next 0/3] net: some feature cover letter - └─ [PATCH net-next 1/3] net: some feature prep - └─ [PATCH net-next 2/3] net: some feature do it - └─ [PATCH net-next 3/3] selftest: net: some feature - - [PATCH iproute2-next] ip: add support for some feature - -Posting as one thread is discouraged because it confuses patchwork -(as of patchwork 2.2.2). - Can I reproduce the checks from patchwork on my local machine? -------------------------------------------------------------- @@ -303,23 +320,6 @@ it has a real, in-tree user. Mock-ups and tests based on ``netdevsim`` are strongly encouraged when adding new APIs, but ``netdevsim`` in itself is **not** considered a use case/user. -Any other tips to help ensure my net/net-next patch gets OK'd? --------------------------------------------------------------- -Attention to detail. Re-read your own work as if you were the -reviewer. You can start with using ``checkpatch.pl``, perhaps even with -the ``--strict`` flag. But do not be mindlessly robotic in doing so. -If your change is a bug fix, make sure your commit log indicates the -end-user visible symptom, the underlying reason as to why it happens, -and then if necessary, explain why the fix proposed is the best way to -get things done. Don't mangle whitespace, and as is common, don't -mis-indent function arguments that span multiple lines. If it is your -first patch, mail it to yourself so you can test apply it to an -unpatched tree to confirm infrastructure didn't mangle it. - -Finally, go back and read -:ref:`Documentation/process/submitting-patches.rst ` -to be sure you are not repeating some common mistake documented there. - My company uses peer feedback in employee performance reviews. Can I ask netdev maintainers for feedback? --------------------------------------------------------------------------------------------------------- From ff249be5cca9f982e58936847ba6c30104abbcad Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 22 Dec 2022 11:22:48 -0800 Subject: [PATCH 039/112] docs: netdev: convert to a non-FAQ document The netdev-FAQ document has grown over the years to the point where finding information in it is somewhat challenging. The length of the questions prevents readers from locating content that's relevant at a glance. Convert to a more standard documentation format with sections and sub-sections rather than questions and answers. The content edits are limited to what's necessary to change the format, and very minor clarifications. Reviewed-by: Randy Dunlap Reviewed-by: Andrew Lunn Signed-off-by: Jakub Kicinski Signed-off-by: David S. Miller --- Documentation/process/maintainer-netdev.rst | 223 +++++++++++--------- 1 file changed, 126 insertions(+), 97 deletions(-) diff --git a/Documentation/process/maintainer-netdev.rst b/Documentation/process/maintainer-netdev.rst index 8f22f8a3dcd1..4a75686d35ab 100644 --- a/Documentation/process/maintainer-netdev.rst +++ b/Documentation/process/maintainer-netdev.rst @@ -2,9 +2,9 @@ .. _netdev-FAQ: -========== -netdev FAQ -========== +============================= +Networking subsystem (netdev) +============================= tl;dr ----- @@ -15,14 +15,15 @@ tl;dr - don't repost your patches within one 24h period - reverse xmas tree -What is netdev? ---------------- -It is a mailing list for all network-related Linux stuff. This +netdev +------ + +netdev is a mailing list for all network-related Linux stuff. This includes anything found under net/ (i.e. core code like IPv6) and drivers/net (i.e. hardware specific drivers) in the Linux source tree. Note that some subsystems (e.g. wireless drivers) which have a high -volume of traffic have their own specific mailing lists. +volume of traffic have their own specific mailing lists and trees. The netdev list is managed (like many other Linux mailing lists) through VGER (http://vger.kernel.org/) with archives available at @@ -32,21 +33,10 @@ Aside from subsystems like those mentioned above, all network-related Linux development (i.e. RFC, review, comments, etc.) takes place on netdev. -How do the changes posted to netdev make their way into Linux? --------------------------------------------------------------- -There are always two trees (git repositories) in play. Both are -driven by David Miller, the main network maintainer. There is the -``net`` tree, and the ``net-next`` tree. As you can probably guess from -the names, the ``net`` tree is for fixes to existing code already in the -mainline tree from Linus, and ``net-next`` is where the new code goes -for the future release. You can find the trees here: +Development cycle +----------------- -- https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git -- https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git - -How often do changes from these trees make it to the mainline Linus tree? -------------------------------------------------------------------------- -To understand this, you need to know a bit of background information on +Here is a bit of background information on the cadence of Linux development. Each new release starts off with a two week "merge window" where the main maintainers feed their new stuff to Linus for merging into the mainline tree. After the two weeks, the @@ -58,9 +48,33 @@ rc2 is released. This repeats on a roughly weekly basis until rc7 state of churn), and a week after the last vX.Y-rcN was done, the official vX.Y is released. -Relating that to netdev: At the beginning of the 2-week merge window, -the ``net-next`` tree will be closed - no new changes/features. The -accumulated new content of the past ~10 weeks will be passed onto +To find out where we are now in the cycle - load the mainline (Linus) +page here: + + https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git + +and note the top of the "tags" section. If it is rc1, it is early in +the dev cycle. If it was tagged rc7 a week ago, then a release is +probably imminent. If the most recent tag is a final release tag +(without an ``-rcN`` suffix) - we are most likely in a merge window +and ``net-next`` is closed. + +git trees and patch flow +------------------------ + +There are two networking trees (git repositories) in play. Both are +driven by David Miller, the main network maintainer. There is the +``net`` tree, and the ``net-next`` tree. As you can probably guess from +the names, the ``net`` tree is for fixes to existing code already in the +mainline tree from Linus, and ``net-next`` is where the new code goes +for the future release. You can find the trees here: + +- https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git +- https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git + +Relating that to kernel development: At the beginning of the 2-week +merge window, the ``net-next`` tree will be closed - no new changes/features. +The accumulated new content of the past ~10 weeks will be passed onto mainline/Linus via a pull request for vX.Y -- at the same time, the ``net`` tree will start accumulating fixes for this pulled content relating to vX.Y @@ -92,22 +106,14 @@ focus for ``net`` is on stabilization and bug fixes. Finally, the vX.Y gets released, and the whole cycle starts over. -So where are we now in this cycle? ----------------------------------- +netdev patch review +------------------- -Load the mainline (Linus) page here: +Patch status +~~~~~~~~~~~~ - https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git - -and note the top of the "tags" section. If it is rc1, it is early in -the dev cycle. If it was tagged rc7 a week ago, then a release is -probably imminent. If the most recent tag is a final release tag -(without an ``-rcN`` suffix) - we are most likely in a merge window -and ``net-next`` is closed. - -How can I tell the status of a patch I've sent? ------------------------------------------------ -Start by looking at the main patchworks queue for netdev: +Status of a patch can be checked by looking at the main patchwork +queue for netdev: https://patchwork.kernel.org/project/netdevbpf/list/ @@ -116,17 +122,20 @@ patch. Patches are indexed by the ``Message-ID`` header of the emails which carried them so if you have trouble finding your patch append the value of ``Message-ID`` to the URL above. -Should I directly update patchwork state of my own patches? ------------------------------------------------------------ +Updating patch status +~~~~~~~~~~~~~~~~~~~~~ + It may be tempting to help the maintainers and update the state of your -own patches when you post a new version or spot a bug. Please do not do that. +own patches when you post a new version or spot a bug. Please **do not** +do that. Interfering with the patch status on patchwork will only cause confusion. Leave it to the maintainer to figure out what is the most recent and current version that should be applied. If there is any doubt, the maintainer will reply and ask what should be done. -How long before my patch is accepted? -------------------------------------- +Review timelines +~~~~~~~~~~~~~~~~ + Generally speaking, the patches get triaged quickly (in less than 48h). But be patient, if your patch is active in patchwork (i.e. it's listed on the project's patch list) the chances it was missed are close to zero. @@ -134,37 +143,47 @@ Asking the maintainer for status updates on your patch is a good way to ensure your patch is ignored or pushed to the bottom of the priority list. -I made changes to only a few patches in a patch series should I resend only those changed? ------------------------------------------------------------------------------------------- -No, please resend the entire patch series and make sure you do number your -patches such that it is clear this is the latest and greatest set of patches -that can be applied. +Partial resends +~~~~~~~~~~~~~~~ -I submitted multiple versions of a patch series and it looks like a version other than the last one has been accepted, what should I do? ----------------------------------------------------------------------------------------------------------------------------------------- +Please always resend the entire patch series and make sure you do number your +patches such that it is clear this is the latest and greatest set of patches +that can be applied. Do not try to resend just the patches which changed. + +Handling misapplied patches +~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +Occasionally a patch series gets applied before receiving critical feedback, +or the wrong version of a series gets applied. There is no revert possible, once it is pushed out, it stays like that. Please send incremental versions on top of what has been merged in order to fix the patches the way they would look like if your latest patch series was to be merged. -Are there special rules regarding stable submissions on netdev? ---------------------------------------------------------------- +Stable tree +~~~~~~~~~~~ + While it used to be the case that netdev submissions were not supposed to carry explicit ``CC: stable@vger.kernel.org`` tags that is no longer the case today. Please follow the standard stable rules in :ref:`Documentation/process/stable-kernel-rules.rst `, and make sure you include appropriate Fixes tags! -I found a bug that might have possible security implications or similar. Should I mail the main netdev maintainer off-list? ---------------------------------------------------------------------------------------------------------------------------- -No. The current netdev maintainer has consistently requested that +Security fixes +~~~~~~~~~~~~~~ + +Do not email netdev maintainers directly if you think you discovered +a bug that might have possible security implications. +The current netdev maintainer has consistently requested that people use the mailing lists and not reach out directly. If you aren't OK with that, then perhaps consider mailing security@kernel.org or reading about http://oss-security.openwall.org/wiki/mailing-lists/distros as possible alternative mechanisms. -How do I post corresponding changes to user space components? -------------------------------------------------------------- + +Co-posting changes to user space components +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + User space code exercising kernel features should be posted alongside kernel patches. This gives reviewers a chance to see how any new interface is used and how well it works. @@ -189,9 +208,10 @@ to the mailing list, e.g.:: Posting as one thread is discouraged because it confuses patchwork (as of patchwork 2.2.2). -Any other tips to help ensure my net/net-next patch gets OK'd? --------------------------------------------------------------- -Attention to detail. Re-read your own work as if you were the +Preparing changes +----------------- + +Attention to detail is important. Re-read your own work as if you were the reviewer. You can start with using ``checkpatch.pl``, perhaps even with the ``--strict`` flag. But do not be mindlessly robotic in doing so. If your change is a bug fix, make sure your commit log indicates the @@ -206,8 +226,9 @@ Finally, go back and read :ref:`Documentation/process/submitting-patches.rst ` to be sure you are not repeating some common mistake documented there. -How do I indicate which tree (net vs. net-next) my patch should be in? ----------------------------------------------------------------------- +Indicating target tree +~~~~~~~~~~~~~~~~~~~~~~ + To help maintainers and CI bots you should explicitly mark which tree your patch is targeting. Assuming that you use git, use the prefix flag:: @@ -217,8 +238,8 @@ flag:: Use ``net`` instead of ``net-next`` (always lower case) in the above for bug-fix ``net`` content. -How do I divide my work into patches? -------------------------------------- +Dividing work into patches +~~~~~~~~~~~~~~~~~~~~~~~~~~ Put yourself in the shoes of the reviewer. Each patch is read separately and therefore should constitute a comprehensible step towards your stated @@ -231,9 +252,11 @@ just do it. As a result, a sequence of smaller series gets merged quicker and with better review coverage. Re-posting large series also increases the mailing list traffic. -Is the comment style convention different for the networking content? ---------------------------------------------------------------------- -Yes, in a largely trivial way. Instead of this:: +Multi-line comments +~~~~~~~~~~~~~~~~~~~ + +Comment style convention is slightly different for networking and most of +the tree. Instead of this:: /* * foobar blah blah blah @@ -246,8 +269,8 @@ it is requested that you make it look like this:: * another line of text */ -What is "reverse xmas tree"? ----------------------------- +Local variable ordering ("reverse xmas tree", "RCS") +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Netdev has a convention for ordering local variables in functions. Order the variable declaration lines longest to shortest, e.g.:: @@ -259,13 +282,16 @@ Order the variable declaration lines longest to shortest, e.g.:: If there are dependencies between the variables preventing the ordering move the initialization out of line. -I am working in existing code which uses non-standard formatting. Which formatting should I use? ------------------------------------------------------------------------------------------------- -Make your code follow the most recent guidelines, so that eventually all code +Format precedence +~~~~~~~~~~~~~~~~~ + +When working in existing code which uses nonstandard formatting make +your code follow the most recent guidelines, so that eventually all code in the domain of netdev is in the preferred format. -I have received review feedback, when should I post a revised version of the patches? -------------------------------------------------------------------------------------- +Resending after review +~~~~~~~~~~~~~~~~~~~~~~ + Allow at least 24 hours to pass between postings. This will ensure reviewers from all geographical locations have a chance to chime in. Do not wait too long (weeks) between postings either as it will make it harder for reviewers @@ -275,8 +301,12 @@ Make sure you address all the feedback in your new posting. Do not post a new version of the code if the discussion about the previous version is still ongoing, unless directly instructed by a reviewer. -What level of testing is expected before I submit my change? ------------------------------------------------------------- +Testing +------- + +Expected level of testing +~~~~~~~~~~~~~~~~~~~~~~~~~ + At the very minimum your changes must survive an ``allyesconfig`` and an ``allmodconfig`` build with ``W=1`` set without new warnings or failures. @@ -287,43 +317,42 @@ and the patch series contains a set of kernel selftest for You are expected to test your changes on top of the relevant networking tree (``net`` or ``net-next``) and not e.g. a stable tree or ``linux-next``. -Can I reproduce the checks from patchwork on my local machine? --------------------------------------------------------------- +patchwork checks +~~~~~~~~~~~~~~~~ Checks in patchwork are mostly simple wrappers around existing kernel scripts, the sources are available at: https://github.com/kuba-moo/nipa/tree/master/tests -Running all the builds and checks locally is a pain, can I post my patches and have the patchwork bot validate them? --------------------------------------------------------------------------------------------------------------------- - -No, you must ensure that your patches are ready by testing them locally +**Do not** post your patches just to run them through the checks. +You must ensure that your patches are ready by testing them locally before posting to the mailing list. The patchwork build bot instance gets overloaded very easily and netdev@vger really doesn't need more traffic if we can help it. -netdevsim is great, can I extend it for my out-of-tree tests? -------------------------------------------------------------- +netdevsim +~~~~~~~~~ -No, ``netdevsim`` is a test vehicle solely for upstream tests. -(Please add your tests under ``tools/testing/selftests/``.) +``netdevsim`` is a test driver which can be used to exercise driver +configuration APIs without requiring capable hardware. +Mock-ups and tests based on ``netdevsim`` are strongly encouraged when +adding new APIs, but ``netdevsim`` in itself is **not** considered +a use case/user. You must also implement the new APIs in a real driver. -We also give no guarantees that ``netdevsim`` won't change in the future +We give no guarantees that ``netdevsim`` won't change in the future in a way which would break what would normally be considered uAPI. -Is netdevsim considered a "user" of an API? -------------------------------------------- +``netdevsim`` is reserved for use by upstream tests only, so any +new ``netdevsim`` features must be accompanied by selftests under +``tools/testing/selftests/``. -Linux kernel has a long standing rule that no API should be added unless -it has a real, in-tree user. Mock-ups and tests based on ``netdevsim`` are -strongly encouraged when adding new APIs, but ``netdevsim`` in itself -is **not** considered a use case/user. +Testimonials / feedback +----------------------- -My company uses peer feedback in employee performance reviews. Can I ask netdev maintainers for feedback? ---------------------------------------------------------------------------------------------------------- - -Yes, especially if you spend significant amount of time reviewing code +Some companies use peer feedback in employee performance reviews. +Please feel free to request feedback from netdev maintainers, +especially if you spend significant amount of time reviewing code and go out of your way to improve shared infrastructure. The feedback must be requested by you, the contributor, and will always From 8e6a8d7a3dd93e93645be061692cb4ee6702dff0 Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Mon, 26 Dec 2022 16:13:27 +0900 Subject: [PATCH 040/112] net: ethernet: renesas: rswitch: Fix error path in renesas_eth_sw_probe() If rswitch_init() returns non-zero and this driver is re-probed, the following error happens: renesas_eth_sw e6880000.ethernet: Unbalanced pm_runtime_enable! So, fix error path in renesas_eth_sw_probe(). Fixes: 3590918b5d07 ("net: ethernet: renesas: Add support for "Ethernet Switch"") Signed-off-by: Yoshihiro Shimoda Signed-off-by: David S. Miller --- drivers/net/ethernet/renesas/rswitch.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/renesas/rswitch.c b/drivers/net/ethernet/renesas/rswitch.c index e42ceaa0099f..473d86bdf97d 100644 --- a/drivers/net/ethernet/renesas/rswitch.c +++ b/drivers/net/ethernet/renesas/rswitch.c @@ -1786,6 +1786,11 @@ static int renesas_eth_sw_probe(struct platform_device *pdev) pm_runtime_get_sync(&pdev->dev); ret = rswitch_init(priv); + if (ret < 0) { + pm_runtime_put(&pdev->dev); + pm_runtime_disable(&pdev->dev); + return ret; + } device_set_wakeup_capable(&pdev->dev, 1); From bd2adfe3b3b863c883309bcc915f13c831ca88da Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Mon, 26 Dec 2022 16:13:28 +0900 Subject: [PATCH 041/112] net: ethernet: renesas: rswitch: Fix getting mac address from device tree To get mac address from device tree which is from each ethernet-port, fix the first argument of of_get_ethdev_address(). Fixes: 3590918b5d07 ("net: ethernet: renesas: Add support for "Ethernet Switch"") Signed-off-by: Yoshihiro Shimoda Signed-off-by: David S. Miller --- drivers/net/ethernet/renesas/rswitch.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/renesas/rswitch.c b/drivers/net/ethernet/renesas/rswitch.c index 473d86bdf97d..6441892636db 100644 --- a/drivers/net/ethernet/renesas/rswitch.c +++ b/drivers/net/ethernet/renesas/rswitch.c @@ -1578,6 +1578,7 @@ static int rswitch_device_alloc(struct rswitch_private *priv, int index) { struct platform_device *pdev = priv->pdev; struct rswitch_device *rdev; + struct device_node *port; struct net_device *ndev; int err; @@ -1606,7 +1607,9 @@ static int rswitch_device_alloc(struct rswitch_private *priv, int index) netif_napi_add(ndev, &rdev->napi, rswitch_poll); - err = of_get_ethdev_address(pdev->dev.of_node, ndev); + port = rswitch_get_port_node(rdev); + err = of_get_ethdev_address(port, ndev); + of_node_put(port); if (err) { if (is_valid_ether_addr(rdev->etha->mac_addr)) eth_hw_addr_set(ndev, rdev->etha->mac_addr); From 0020ae2a4aa81becd182231bf48acd66c86c86dd Mon Sep 17 00:00:00 2001 From: Vikas Gupta Date: Mon, 26 Dec 2022 22:19:36 -0500 Subject: [PATCH 042/112] bnxt_en: fix devlink port registration to netdev We don't register a devlink port in case of a VF so avoid setting the devlink pointer to netdev. Also, SET_NETDEV_DEVLINK_PORT has to be moved so that we determine whether the device is PF/VF first. This fixes the NULL pointer dereference of devlink_port->devlink when creating VFs: BUG: kernel NULL pointer dereference, address: 0000000000000160 PGD 0 Oops: 0000 [#1] PREEMPT SMP NOPTI CPU: 14 PID: 388 Comm: kworker/14:1 Kdump: loaded Not tainted 6.1.0-rc8 #5 Hardware name: Dell Inc. PowerEdge R750/06V45N, BIOS 1.3.8 08/31/2021 Workqueue: events work_for_cpu_fn RIP: 0010:devlink_nl_port_handle_size+0xb/0x50 Code: 83 c4 10 5b 5d c3 cc cc cc cc b8 a6 ff ff ff eb de e8 c9 59 21 00 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 55 53 48 8b 47 20 <48> 8b a8 60 01 00 00 48 8b 45 60 48 8b 38 e8 92 90 1a 00 48 8b 7d RSP: 0018:ff4fe5394846fcd8 EFLAGS: 00010286 RAX: 0000000000000000 RBX: 0000000000000794 RCX: 0000000000000000 RDX: ff1f129683a30a40 RSI: 0000000000000008 RDI: ff1f1296bb496188 RBP: 0000000000000334 R08: 0000000000000cc0 R09: 0000000000000000 R10: ff1f1296bb494298 R11: ffffffffffffffc0 R12: 0000000000000000 R13: 0000000000000000 R14: ff1f1296bb494000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ff1f129e5fa00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000160 CR3: 000000131f610006 CR4: 0000000000771ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: if_nlmsg_size+0x14a/0x220 rtmsg_ifinfo_build_skb+0x3c/0x100 rtmsg_ifinfo+0x9c/0xc0 register_netdevice+0x59d/0x670 register_netdev+0x1c/0x40 bnxt_init_one+0x674/0xa60 [bnxt_en] local_pci_probe+0x42/0x80 work_for_cpu_fn+0x13/0x20 process_one_work+0x1e2/0x3b0 ? rescuer_thread+0x390/0x390 worker_thread+0x1c4/0x3a0 ? rescuer_thread+0x390/0x390 kthread+0xd6/0x100 ? kthread_complete_and_exit+0x20/0x20 Fixes: ac73d4bf2cda ("net: make drivers to use SET_NETDEV_DEVLINK_PORT to set devlink_port") Cc: Jiri Pirko Signed-off-by: Vikas Gupta Reviewed-by: Andy Gospodarek Reviewed-by: Kalesh Anakkur Purayil Reviewed-by: Damodharam Ammepalli Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 4c7d07c684c4..93d32b333007 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -13591,7 +13591,6 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) return -ENOMEM; bp = netdev_priv(dev); - SET_NETDEV_DEVLINK_PORT(dev, &bp->dl_port); bp->board_idx = ent->driver_data; bp->msg_enable = BNXT_DEF_MSG_ENABLE; bnxt_set_max_func_irqs(bp, max_irqs); @@ -13599,6 +13598,10 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (bnxt_vf_pciid(bp->board_idx)) bp->flags |= BNXT_FLAG_VF; + /* No devlink port registration in case of a VF */ + if (BNXT_PF(bp)) + SET_NETDEV_DEVLINK_PORT(dev, &bp->dl_port); + if (pdev->msix_cap) bp->flags |= BNXT_FLAG_MSIX_CAP; From bbfc17e50ba2ed18dfef46b1c433d50a58566bf1 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 26 Dec 2022 22:19:37 -0500 Subject: [PATCH 043/112] bnxt_en: Simplify bnxt_xdp_buff_init() bnxt_xdp_buff_init() does not modify the data_ptr or the len parameters, so no need to pass in the addresses of these parameters. Fixes: b231c3f3414c ("bnxt: refactor bnxt_rx_xdp to separate xdp_init_buff/xdp_prepare_buff") Reviewed-by: Andy Gospodarek Reviewed-by: Somnath Kotur Reviewed-by: Pavan Chebbi Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c | 6 +++--- drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 93d32b333007..b8639b7e6b2b 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1925,7 +1925,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, dma_addr = rx_buf->mapping; if (bnxt_xdp_attached(bp, rxr)) { - bnxt_xdp_buff_init(bp, rxr, cons, &data_ptr, &len, &xdp); + bnxt_xdp_buff_init(bp, rxr, cons, data_ptr, len, &xdp); if (agg_bufs) { u32 frag_len = bnxt_rx_agg_pages_xdp(bp, cpr, &xdp, cp_cons, agg_bufs, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c index c3065ec0a479..1847f191577d 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c @@ -177,7 +177,7 @@ bool bnxt_xdp_attached(struct bnxt *bp, struct bnxt_rx_ring_info *rxr) } void bnxt_xdp_buff_init(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, - u16 cons, u8 **data_ptr, unsigned int *len, + u16 cons, u8 *data_ptr, unsigned int len, struct xdp_buff *xdp) { struct bnxt_sw_rx_bd *rx_buf; @@ -191,13 +191,13 @@ void bnxt_xdp_buff_init(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, offset = bp->rx_offset; mapping = rx_buf->mapping - bp->rx_dma_offset; - dma_sync_single_for_cpu(&pdev->dev, mapping + offset, *len, bp->rx_dir); + dma_sync_single_for_cpu(&pdev->dev, mapping + offset, len, bp->rx_dir); if (bp->xdp_has_frags) buflen = BNXT_PAGE_MODE_BUF_SIZE + offset; xdp_init_buff(xdp, buflen, &rxr->xdp_rxq); - xdp_prepare_buff(xdp, *data_ptr - offset, offset, *len, false); + xdp_prepare_buff(xdp, data_ptr - offset, offset, len, false); } void bnxt_xdp_buff_frags_free(struct bnxt_rx_ring_info *rxr, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h index 505911ae095d..2bbdb8e7c506 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h @@ -27,7 +27,7 @@ int bnxt_xdp_xmit(struct net_device *dev, int num_frames, bool bnxt_xdp_attached(struct bnxt *bp, struct bnxt_rx_ring_info *rxr); void bnxt_xdp_buff_init(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, - u16 cons, u8 **data_ptr, unsigned int *len, + u16 cons, u8 *data_ptr, unsigned int len, struct xdp_buff *xdp); void bnxt_xdp_buff_frags_free(struct bnxt_rx_ring_info *rxr, struct xdp_buff *xdp); From 9b3e607871ea5ee90f10f5be3965fc07f2aa3ef7 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 26 Dec 2022 22:19:38 -0500 Subject: [PATCH 044/112] bnxt_en: Fix XDP RX path The XDP program can change the starting address of the RX data buffer and this information needs to be passed back from bnxt_rx_xdp() to bnxt_rx_pkt() for the XDP_PASS case so that the SKB can point correctly to the modified buffer address. Add back the data_ptr parameter to bnxt_rx_xdp() to make this work. Fixes: b231c3f3414c ("bnxt: refactor bnxt_rx_xdp to separate xdp_init_buff/xdp_prepare_buff") Reviewed-by: Andy Gospodarek Reviewed-by: Pavan Chebbi Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c | 7 +++++-- drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h | 4 ++-- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index b8639b7e6b2b..1acabfe26db1 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1940,7 +1940,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, } if (xdp_active) { - if (bnxt_rx_xdp(bp, rxr, cons, xdp, data, &len, event)) { + if (bnxt_rx_xdp(bp, rxr, cons, xdp, data, &data_ptr, &len, event)) { rc = 1; goto next_rx; } diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c index 1847f191577d..2ceeaa818c1c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c @@ -222,7 +222,8 @@ void bnxt_xdp_buff_frags_free(struct bnxt_rx_ring_info *rxr, * false - packet should be passed to the stack. */ bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons, - struct xdp_buff xdp, struct page *page, unsigned int *len, u8 *event) + struct xdp_buff xdp, struct page *page, u8 **data_ptr, + unsigned int *len, u8 *event) { struct bpf_prog *xdp_prog = READ_ONCE(rxr->xdp_prog); struct bnxt_tx_ring_info *txr; @@ -255,8 +256,10 @@ bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons, *event &= ~BNXT_RX_EVENT; *len = xdp.data_end - xdp.data; - if (orig_data != xdp.data) + if (orig_data != xdp.data) { offset = xdp.data - xdp.data_hard_start; + *data_ptr = xdp.data_hard_start + offset; + } switch (act) { case XDP_PASS: diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h index 2bbdb8e7c506..ea430d6961df 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.h @@ -18,8 +18,8 @@ struct bnxt_sw_tx_bd *bnxt_xmit_bd(struct bnxt *bp, struct xdp_buff *xdp); void bnxt_tx_int_xdp(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts); bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons, - struct xdp_buff xdp, struct page *page, unsigned int *len, - u8 *event); + struct xdp_buff xdp, struct page *page, u8 **data_ptr, + unsigned int *len, u8 *event); int bnxt_xdp(struct net_device *dev, struct netdev_bpf *xdp); int bnxt_xdp_xmit(struct net_device *dev, int num_frames, struct xdp_frame **frames, u32 flags); From 1abeacc1979fa4a756695f5030791d8f0fa934b9 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 26 Dec 2022 22:19:39 -0500 Subject: [PATCH 045/112] bnxt_en: Fix first buffer size calculations for XDP multi-buffer The size of the first buffer is always page size, and the useable space is the page size minus the offset and the skb_shared_info size. Make sure SKB and XDP buf sizes match so that the skb_shared_info is at the same offset seen from the SKB and XDP_BUF. build_skb() should be passed PAGE_SIZE. xdp_init_buff() should be passed PAGE_SIZE as well. xdp_get_shared_info_from_buff() will automatically deduct the skb_shared_info size if the XDP buffer has frags. There is no need to keep bp->xdp_has_frags. Change BNXT_PAGE_MODE_BUF_SIZE to BNXT_MAX_PAGE_MODE_MTU_SBUF since this constant is really the MTU with ethernet header size subtracted. Also fix the BNXT_MAX_PAGE_MODE_MTU macro with proper parentheses. Fixes: 32861236190b ("bnxt: change receive ring space parameters") Reviewed-by: Somnath Kotur Reviewed-by: Andy Gospodarek Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 +++++---- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 15 +++++++++++---- drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c | 7 +------ 3 files changed, 17 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 1acabfe26db1..a21c6829e301 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -991,8 +991,7 @@ static struct sk_buff *bnxt_rx_multi_page_skb(struct bnxt *bp, dma_addr -= bp->rx_dma_offset; dma_unmap_page_attrs(&bp->pdev->dev, dma_addr, PAGE_SIZE, bp->rx_dir, DMA_ATTR_WEAK_ORDERING); - skb = build_skb(page_address(page), BNXT_PAGE_MODE_BUF_SIZE + - bp->rx_dma_offset); + skb = build_skb(page_address(page), PAGE_SIZE); if (!skb) { __free_page(page); return NULL; @@ -3969,8 +3968,10 @@ void bnxt_set_ring_params(struct bnxt *bp) bp->rx_agg_ring_mask = (bp->rx_agg_nr_pages * RX_DESC_CNT) - 1; if (BNXT_RX_PAGE_MODE(bp)) { - rx_space = BNXT_PAGE_MODE_BUF_SIZE; - rx_size = BNXT_MAX_PAGE_MODE_MTU; + rx_space = PAGE_SIZE; + rx_size = PAGE_SIZE - + ALIGN(max(NET_SKB_PAD, XDP_PACKET_HEADROOM), 8) - + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); } else { rx_size = SKB_DATA_ALIGN(BNXT_RX_COPY_THRESH + NET_IP_ALIGN); rx_space = rx_size + NET_SKB_PAD + diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index 41c6dd0ae447..5163ef4a49ea 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -591,12 +591,20 @@ struct nqe_cn { #define BNXT_RX_PAGE_SIZE (1 << BNXT_RX_PAGE_SHIFT) #define BNXT_MAX_MTU 9500 -#define BNXT_PAGE_MODE_BUF_SIZE \ + +/* First RX buffer page in XDP multi-buf mode + * + * +-------------------------------------------------------------------------+ + * | XDP_PACKET_HEADROOM | bp->rx_buf_use_size | skb_shared_info| + * | (bp->rx_dma_offset) | | | + * +-------------------------------------------------------------------------+ + */ +#define BNXT_MAX_PAGE_MODE_MTU_SBUF \ ((unsigned int)PAGE_SIZE - VLAN_ETH_HLEN - NET_IP_ALIGN - \ XDP_PACKET_HEADROOM) #define BNXT_MAX_PAGE_MODE_MTU \ - BNXT_PAGE_MODE_BUF_SIZE - \ - SKB_DATA_ALIGN((unsigned int)sizeof(struct skb_shared_info)) + (BNXT_MAX_PAGE_MODE_MTU_SBUF - \ + SKB_DATA_ALIGN((unsigned int)sizeof(struct skb_shared_info))) #define BNXT_MIN_PKT_SIZE 52 @@ -2134,7 +2142,6 @@ struct bnxt { #define BNXT_DUMP_CRASH 1 struct bpf_prog *xdp_prog; - u8 xdp_has_frags; struct bnxt_ptp_cfg *ptp_cfg; u8 ptp_all_rx_tstamp; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c index 2ceeaa818c1c..36d5202c0aee 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c @@ -193,9 +193,6 @@ void bnxt_xdp_buff_init(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, mapping = rx_buf->mapping - bp->rx_dma_offset; dma_sync_single_for_cpu(&pdev->dev, mapping + offset, len, bp->rx_dir); - if (bp->xdp_has_frags) - buflen = BNXT_PAGE_MODE_BUF_SIZE + offset; - xdp_init_buff(xdp, buflen, &rxr->xdp_rxq); xdp_prepare_buff(xdp, data_ptr - offset, offset, len, false); } @@ -404,10 +401,8 @@ static int bnxt_xdp_set(struct bnxt *bp, struct bpf_prog *prog) netdev_warn(dev, "ethtool rx/tx channels must be combined to support XDP.\n"); return -EOPNOTSUPP; } - if (prog) { + if (prog) tx_xdp = bp->rx_nr_rings; - bp->xdp_has_frags = prog->aux->xdp_has_frags; - } tc = netdev_get_num_tc(dev); if (!tc) From a056ebcc30e2f78451d66f615d2f6bdada3e6438 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 26 Dec 2022 22:19:40 -0500 Subject: [PATCH 046/112] bnxt_en: Fix HDS and jumbo thresholds for RX packets The recent XDP multi-buffer feature has introduced regressions in the setting of HDS and jumbo thresholds. HDS was accidentally disabled in the nornmal mode without XDP. This patch restores jumbo HDS placement when not in XDP mode. In XDP multi-buffer mode, HDS should be disabled and the jumbo threshold should be set to the usable page size in the first page buffer. Fixes: 32861236190b ("bnxt: change receive ring space parameters") Reviewed-by: Mohammad Shuab Siddique Reviewed-by: Ajit Khaparde Reviewed-by: Andy Gospodarek Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index a21c6829e301..16ce7a90610c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -5399,15 +5399,16 @@ static int bnxt_hwrm_vnic_set_hds(struct bnxt *bp, u16 vnic_id) req->flags = cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_JUMBO_PLACEMENT); req->enables = cpu_to_le32(VNIC_PLCMODES_CFG_REQ_ENABLES_JUMBO_THRESH_VALID); - if (BNXT_RX_PAGE_MODE(bp) && !BNXT_RX_JUMBO_MODE(bp)) { + if (BNXT_RX_PAGE_MODE(bp)) { + req->jumbo_thresh = cpu_to_le16(bp->rx_buf_use_size); + } else { req->flags |= cpu_to_le32(VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV4 | VNIC_PLCMODES_CFG_REQ_FLAGS_HDS_IPV6); req->enables |= cpu_to_le32(VNIC_PLCMODES_CFG_REQ_ENABLES_HDS_THRESHOLD_VALID); + req->jumbo_thresh = cpu_to_le16(bp->rx_copy_thresh); + req->hds_threshold = cpu_to_le16(bp->rx_copy_thresh); } - /* thresholds not implemented in firmware yet */ - req->jumbo_thresh = cpu_to_le16(bp->rx_copy_thresh); - req->hds_threshold = cpu_to_le16(bp->rx_copy_thresh); req->vnic_id = cpu_to_le32(vnic->fw_vnic_id); return hwrm_req_send(bp, req); } From 9deb1e9fb88b1120a908676fa33bdf9e2eeaefce Mon Sep 17 00:00:00 2001 From: Daniil Tatianin Date: Mon, 26 Dec 2022 14:48:23 +0300 Subject: [PATCH 047/112] net/ethtool/ioctl: return -EOPNOTSUPP if we have no phy stats It's not very useful to copy back an empty ethtool_stats struct and return 0 if we didn't actually have any stats. This also allows for further simplification of this function in the future commits. Signed-off-by: Daniil Tatianin Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- net/ethtool/ioctl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index c2f1a542e6fa..932fa8225b2f 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -2099,7 +2099,8 @@ static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr) return n_stats; if (n_stats > S32_MAX / sizeof(u64)) return -ENOMEM; - WARN_ON_ONCE(!n_stats); + if (WARN_ON_ONCE(!n_stats)) + return -EOPNOTSUPP; if (copy_from_user(&stats, useraddr, sizeof(stats))) return -EFAULT; From fd4778581d61d8848b532f8cdc9b325138748437 Mon Sep 17 00:00:00 2001 From: Daniil Tatianin Date: Mon, 26 Dec 2022 14:48:24 +0300 Subject: [PATCH 048/112] net/ethtool/ioctl: remove if n_stats checks from ethtool_get_phy_stats Now that we always early return if we don't have any stats we can remove these checks as they're no longer necessary. Signed-off-by: Daniil Tatianin Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- net/ethtool/ioctl.c | 24 ++++++++++-------------- 1 file changed, 10 insertions(+), 14 deletions(-) diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index 932fa8225b2f..85f0cffdcec8 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -2107,28 +2107,24 @@ static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr) stats.n_stats = n_stats; - if (n_stats) { - data = vzalloc(array_size(n_stats, sizeof(u64))); - if (!data) - return -ENOMEM; + data = vzalloc(array_size(n_stats, sizeof(u64))); + if (!data) + return -ENOMEM; - if (phydev && !ops->get_ethtool_phy_stats && - phy_ops && phy_ops->get_stats) { - ret = phy_ops->get_stats(phydev, &stats, data); - if (ret < 0) - goto out; - } else { - ops->get_ethtool_phy_stats(dev, &stats, data); - } + if (phydev && !ops->get_ethtool_phy_stats && + phy_ops && phy_ops->get_stats) { + ret = phy_ops->get_stats(phydev, &stats, data); + if (ret < 0) + goto out; } else { - data = NULL; + ops->get_ethtool_phy_stats(dev, &stats, data); } ret = -EFAULT; if (copy_to_user(useraddr, &stats, sizeof(stats))) goto out; useraddr += sizeof(stats); - if (n_stats && copy_to_user(useraddr, data, array_size(n_stats, sizeof(u64)))) + if (copy_to_user(useraddr, data, array_size(n_stats, sizeof(u64)))) goto out; ret = 0; From 201ed315f9676809cd5b20a39206e964106d4f27 Mon Sep 17 00:00:00 2001 From: Daniil Tatianin Date: Mon, 26 Dec 2022 14:48:25 +0300 Subject: [PATCH 049/112] net/ethtool/ioctl: split ethtool_get_phy_stats into multiple helpers So that it's easier to follow and make sense of the branching and various conditions. Stats retrieval has been split into two separate functions ethtool_get_phy_stats_phydev & ethtool_get_phy_stats_ethtool. The former attempts to retrieve the stats using phydev & phy_ops, while the latter uses ethtool_ops. Actual n_stats validation & array allocation has been moved into a new ethtool_vzalloc_stats_array helper. This also fixes a potential NULL dereference of ops->get_ethtool_phy_stats where it was getting called in an else branch unconditionally without making sure it was actually present. Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Signed-off-by: Daniil Tatianin Reviewed-by: Andrew Lunn Signed-off-by: David S. Miller --- net/ethtool/ioctl.c | 102 ++++++++++++++++++++++++++++++-------------- 1 file changed, 69 insertions(+), 33 deletions(-) diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c index 85f0cffdcec8..646b3e490c71 100644 --- a/net/ethtool/ioctl.c +++ b/net/ethtool/ioctl.c @@ -2078,23 +2078,8 @@ static int ethtool_get_stats(struct net_device *dev, void __user *useraddr) return ret; } -static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr) +static int ethtool_vzalloc_stats_array(int n_stats, u64 **data) { - const struct ethtool_phy_ops *phy_ops = ethtool_phy_ops; - const struct ethtool_ops *ops = dev->ethtool_ops; - struct phy_device *phydev = dev->phydev; - struct ethtool_stats stats; - u64 *data; - int ret, n_stats; - - if (!phydev && (!ops->get_ethtool_phy_stats || !ops->get_sset_count)) - return -EOPNOTSUPP; - - if (phydev && !ops->get_ethtool_phy_stats && - phy_ops && phy_ops->get_sset_count) - n_stats = phy_ops->get_sset_count(phydev); - else - n_stats = ops->get_sset_count(dev, ETH_SS_PHY_STATS); if (n_stats < 0) return n_stats; if (n_stats > S32_MAX / sizeof(u64)) @@ -2102,31 +2087,82 @@ static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr) if (WARN_ON_ONCE(!n_stats)) return -EOPNOTSUPP; + *data = vzalloc(array_size(n_stats, sizeof(u64))); + if (!*data) + return -ENOMEM; + + return 0; +} + +static int ethtool_get_phy_stats_phydev(struct phy_device *phydev, + struct ethtool_stats *stats, + u64 **data) + { + const struct ethtool_phy_ops *phy_ops = ethtool_phy_ops; + int n_stats, ret; + + if (!phy_ops || !phy_ops->get_sset_count || !phy_ops->get_stats) + return -EOPNOTSUPP; + + n_stats = phy_ops->get_sset_count(phydev); + + ret = ethtool_vzalloc_stats_array(n_stats, data); + if (ret) + return ret; + + stats->n_stats = n_stats; + return phy_ops->get_stats(phydev, stats, *data); +} + +static int ethtool_get_phy_stats_ethtool(struct net_device *dev, + struct ethtool_stats *stats, + u64 **data) +{ + const struct ethtool_ops *ops = dev->ethtool_ops; + int n_stats, ret; + + if (!ops || !ops->get_sset_count || ops->get_ethtool_phy_stats) + return -EOPNOTSUPP; + + n_stats = ops->get_sset_count(dev, ETH_SS_PHY_STATS); + + ret = ethtool_vzalloc_stats_array(n_stats, data); + if (ret) + return ret; + + stats->n_stats = n_stats; + ops->get_ethtool_phy_stats(dev, stats, *data); + + return 0; +} + +static int ethtool_get_phy_stats(struct net_device *dev, void __user *useraddr) +{ + struct phy_device *phydev = dev->phydev; + struct ethtool_stats stats; + u64 *data = NULL; + int ret = -EOPNOTSUPP; + if (copy_from_user(&stats, useraddr, sizeof(stats))) return -EFAULT; - stats.n_stats = n_stats; + if (phydev) + ret = ethtool_get_phy_stats_phydev(phydev, &stats, &data); - data = vzalloc(array_size(n_stats, sizeof(u64))); - if (!data) - return -ENOMEM; + if (ret == -EOPNOTSUPP) + ret = ethtool_get_phy_stats_ethtool(dev, &stats, &data); - if (phydev && !ops->get_ethtool_phy_stats && - phy_ops && phy_ops->get_stats) { - ret = phy_ops->get_stats(phydev, &stats, data); - if (ret < 0) - goto out; - } else { - ops->get_ethtool_phy_stats(dev, &stats, data); + if (ret) + goto out; + + if (copy_to_user(useraddr, &stats, sizeof(stats))) { + ret = -EFAULT; + goto out; } - ret = -EFAULT; - if (copy_to_user(useraddr, &stats, sizeof(stats))) - goto out; useraddr += sizeof(stats); - if (copy_to_user(useraddr, data, array_size(n_stats, sizeof(u64)))) - goto out; - ret = 0; + if (copy_to_user(useraddr, data, array_size(stats.n_stats, sizeof(u64)))) + ret = -EFAULT; out: vfree(data); From ad425666a1f05d9b215a84cf010c3789b2ea8206 Mon Sep 17 00:00:00 2001 From: Chunhao Lin Date: Mon, 26 Dec 2022 20:31:52 +0800 Subject: [PATCH 050/112] r8169: move rtl_wol_enable_rx() and rtl_prepare_power_down() There is no functional change. Moving these two functions for following patch "r8169: fix dmar pte write access is not set error". Signed-off-by: Chunhao Lin Reviewed-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169_main.c | 44 +++++++++++------------ 1 file changed, 22 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index a9dcc98b6af1..acc2500342ca 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -2210,28 +2210,6 @@ static int rtl_set_mac_address(struct net_device *dev, void *p) return 0; } -static void rtl_wol_enable_rx(struct rtl8169_private *tp) -{ - if (tp->mac_version >= RTL_GIGA_MAC_VER_25) - RTL_W32(tp, RxConfig, RTL_R32(tp, RxConfig) | - AcceptBroadcast | AcceptMulticast | AcceptMyPhys); -} - -static void rtl_prepare_power_down(struct rtl8169_private *tp) -{ - if (tp->dash_type != RTL_DASH_NONE) - return; - - if (tp->mac_version == RTL_GIGA_MAC_VER_32 || - tp->mac_version == RTL_GIGA_MAC_VER_33) - rtl_ephy_write(tp, 0x19, 0xff64); - - if (device_may_wakeup(tp_to_dev(tp))) { - phy_speed_down(tp->phydev, false); - rtl_wol_enable_rx(tp); - } -} - static void rtl_init_rxcfg(struct rtl8169_private *tp) { switch (tp->mac_version) { @@ -2455,6 +2433,28 @@ static void rtl_enable_rxdvgate(struct rtl8169_private *tp) rtl_wait_txrx_fifo_empty(tp); } +static void rtl_wol_enable_rx(struct rtl8169_private *tp) +{ + if (tp->mac_version >= RTL_GIGA_MAC_VER_25) + RTL_W32(tp, RxConfig, RTL_R32(tp, RxConfig) | + AcceptBroadcast | AcceptMulticast | AcceptMyPhys); +} + +static void rtl_prepare_power_down(struct rtl8169_private *tp) +{ + if (tp->dash_type != RTL_DASH_NONE) + return; + + if (tp->mac_version == RTL_GIGA_MAC_VER_32 || + tp->mac_version == RTL_GIGA_MAC_VER_33) + rtl_ephy_write(tp, 0x19, 0xff64); + + if (device_may_wakeup(tp_to_dev(tp))) { + phy_speed_down(tp->phydev, false); + rtl_wol_enable_rx(tp); + } +} + static void rtl_set_tx_config_registers(struct rtl8169_private *tp) { u32 val = TX_DMA_BURST << TxDMAShift | From bb41c13c05c23d9bc46b4e37d8914078c6a40e3a Mon Sep 17 00:00:00 2001 From: Chunhao Lin Date: Mon, 26 Dec 2022 20:31:53 +0800 Subject: [PATCH 051/112] r8169: fix dmar pte write access is not set error When close device, if wol is enabled, rx will be enabled. When open device it will cause rx packet to be dma to the wrong memory address after pci_set_master() and system log will show blow messages. DMAR: DRHD: handling fault status reg 3 DMAR: [DMA Write] Request device [02:00.0] PASID ffffffff fault addr ffdd4000 [fault reason 05] PTE Write access is not set In this patch, driver disable tx/rx when close device. If wol is enabled, only enable rx filter and disable rxdv_gate(if support) to let hardware only receive packet to fifo but not to dma it. Signed-off-by: Chunhao Lin Reviewed-by: Heiner Kallweit Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169_main.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index acc2500342ca..24592d972523 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -2438,6 +2438,9 @@ static void rtl_wol_enable_rx(struct rtl8169_private *tp) if (tp->mac_version >= RTL_GIGA_MAC_VER_25) RTL_W32(tp, RxConfig, RTL_R32(tp, RxConfig) | AcceptBroadcast | AcceptMulticast | AcceptMyPhys); + + if (tp->mac_version >= RTL_GIGA_MAC_VER_40) + rtl_disable_rxdvgate(tp); } static void rtl_prepare_power_down(struct rtl8169_private *tp) @@ -3872,7 +3875,7 @@ static void rtl8169_tx_clear(struct rtl8169_private *tp) netdev_reset_queue(tp->dev); } -static void rtl8169_cleanup(struct rtl8169_private *tp, bool going_down) +static void rtl8169_cleanup(struct rtl8169_private *tp) { napi_disable(&tp->napi); @@ -3884,9 +3887,6 @@ static void rtl8169_cleanup(struct rtl8169_private *tp, bool going_down) rtl_rx_close(tp); - if (going_down && tp->dev->wol_enabled) - goto no_reset; - switch (tp->mac_version) { case RTL_GIGA_MAC_VER_28: case RTL_GIGA_MAC_VER_31: @@ -3907,7 +3907,7 @@ static void rtl8169_cleanup(struct rtl8169_private *tp, bool going_down) } rtl_hw_reset(tp); -no_reset: + rtl8169_tx_clear(tp); rtl8169_init_ring_indexes(tp); } @@ -3918,7 +3918,7 @@ static void rtl_reset_work(struct rtl8169_private *tp) netif_stop_queue(tp->dev); - rtl8169_cleanup(tp, false); + rtl8169_cleanup(tp); for (i = 0; i < NUM_RX_DESC; i++) rtl8169_mark_to_asic(tp->RxDescArray + i); @@ -4605,7 +4605,7 @@ static void rtl8169_down(struct rtl8169_private *tp) pci_clear_master(tp->pci_dev); rtl_pci_commit(tp); - rtl8169_cleanup(tp, true); + rtl8169_cleanup(tp); rtl_disable_exit_l1(tp); rtl_prepare_power_down(tp); } From c2052189f19bd98c80b5d46dc6e42330d2b3b35d Mon Sep 17 00:00:00 2001 From: Xuezhi Zhang Date: Tue, 27 Dec 2022 19:03:52 +0800 Subject: [PATCH 052/112] s390/qeth: convert sysfs snprintf to sysfs_emit Follow the advice of the Documentation/filesystems/sysfs.rst and show() should only use sysfs_emit() or sysfs_emit_at() when formatting the value to be returned to user space. Signed-off-by: Xuezhi Zhang Signed-off-by: David S. Miller --- drivers/s390/net/qeth_core_sys.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/s390/net/qeth_core_sys.c b/drivers/s390/net/qeth_core_sys.c index 406be169173c..d1adc4b83193 100644 --- a/drivers/s390/net/qeth_core_sys.c +++ b/drivers/s390/net/qeth_core_sys.c @@ -410,13 +410,13 @@ static ssize_t qeth_dev_isolation_show(struct device *dev, switch (card->options.isolation) { case ISOLATION_MODE_NONE: - return snprintf(buf, 6, "%s\n", ATTR_QETH_ISOLATION_NONE); + return sysfs_emit(buf, "%s\n", ATTR_QETH_ISOLATION_NONE); case ISOLATION_MODE_FWD: - return snprintf(buf, 9, "%s\n", ATTR_QETH_ISOLATION_FWD); + return sysfs_emit(buf, "%s\n", ATTR_QETH_ISOLATION_FWD); case ISOLATION_MODE_DROP: - return snprintf(buf, 6, "%s\n", ATTR_QETH_ISOLATION_DROP); + return sysfs_emit(buf, "%s\n", ATTR_QETH_ISOLATION_DROP); default: - return snprintf(buf, 5, "%s\n", "N/A"); + return sysfs_emit(buf, "%s\n", "N/A"); } } @@ -500,9 +500,9 @@ static ssize_t qeth_hw_trap_show(struct device *dev, struct qeth_card *card = dev_get_drvdata(dev); if (card->info.hwtrap) - return snprintf(buf, 5, "arm\n"); + return sysfs_emit(buf, "arm\n"); else - return snprintf(buf, 8, "disarm\n"); + return sysfs_emit(buf, "disarm\n"); } static ssize_t qeth_hw_trap_store(struct device *dev, From 40cab44b9089a41f71bbd0eff753eb91d5dafd68 Mon Sep 17 00:00:00 2001 From: Pedro Tammela Date: Tue, 27 Dec 2022 11:04:59 -0300 Subject: [PATCH 053/112] net/sched: fix retpoline wrapper compilation on configs without tc filters Rudi reports a compilation failure on x86_64 when CONFIG_NET_CLS or CONFIG_NET_CLS_ACT is not set but CONFIG_RETPOLINE is set. A misplaced '#endif' was causing the issue. Fixes: 7f0e810220e2 ("net/sched: add retpoline wrapper for tc") Tested-by: Rudi Heitbaum Signed-off-by: Pedro Tammela Signed-off-by: David S. Miller --- include/net/tc_wrapper.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/net/tc_wrapper.h b/include/net/tc_wrapper.h index ceed2fc089ff..d323fffb839a 100644 --- a/include/net/tc_wrapper.h +++ b/include/net/tc_wrapper.h @@ -216,6 +216,8 @@ skip: return tp->classify(skb, tp, res); } +#endif /* CONFIG_NET_CLS */ + static inline void tc_wrapper_init(void) { #ifdef CONFIG_X86 @@ -224,8 +226,6 @@ static inline void tc_wrapper_init(void) #endif } -#endif /* CONFIG_NET_CLS */ - #else #define TC_INDIRECT_SCOPE static From 1f0ae22ab470946143485a02cc1cd7e05c0f9120 Mon Sep 17 00:00:00 2001 From: Moshe Shemesh Date: Mon, 12 Dec 2022 10:42:15 +0200 Subject: [PATCH 054/112] net/mlx5: E-Switch, properly handle ingress tagged packets on VST Fix SRIOV VST mode behavior to insert cvlan when a guest tag is already present in the frame. Previous VST mode behavior was to drop packets or override existing tag, depending on the device version. In this patch we fix this behavior by correctly building the HW steering rule with a push vlan action, or for older devices we ask the FW to stack the vlan when a vlan is already present. Fixes: 07bab9502641 ("net/mlx5: E-Switch, Refactor eswitch ingress acl codes") Fixes: dfcb1ed3c331 ("net/mlx5: E-Switch, Vport ingress/egress ACLs rules for VST mode") Signed-off-by: Moshe Shemesh Reviewed-by: Mark Bloch Signed-off-by: Saeed Mahameed --- .../mellanox/mlx5/core/esw/acl/egress_lgcy.c | 7 +++- .../mellanox/mlx5/core/esw/acl/ingress_lgcy.c | 33 ++++++++++++++++--- .../net/ethernet/mellanox/mlx5/core/eswitch.c | 30 ++++++++++++----- .../net/ethernet/mellanox/mlx5/core/eswitch.h | 6 ++++ include/linux/mlx5/device.h | 5 +++ include/linux/mlx5/mlx5_ifc.h | 3 +- 6 files changed, 68 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c index 60a73990017c..6b4c9ffad95b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c @@ -67,6 +67,7 @@ static void esw_acl_egress_lgcy_groups_destroy(struct mlx5_vport *vport) int esw_acl_egress_lgcy_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { + bool vst_mode_steering = esw_vst_mode_is_steering(esw); struct mlx5_flow_destination drop_ctr_dst = {}; struct mlx5_flow_destination *dst = NULL; struct mlx5_fc *drop_counter = NULL; @@ -77,6 +78,7 @@ int esw_acl_egress_lgcy_setup(struct mlx5_eswitch *esw, */ int table_size = 2; int dest_num = 0; + int actions_flag; int err = 0; if (vport->egress.legacy.drop_counter) { @@ -119,8 +121,11 @@ int esw_acl_egress_lgcy_setup(struct mlx5_eswitch *esw, vport->vport, vport->info.vlan, vport->info.qos); /* Allowed vlan rule */ + actions_flag = MLX5_FLOW_CONTEXT_ACTION_ALLOW; + if (vst_mode_steering) + actions_flag |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP; err = esw_egress_acl_vlan_create(esw, vport, NULL, vport->info.vlan, - MLX5_FLOW_CONTEXT_ACTION_ALLOW); + actions_flag); if (err) goto out; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c index b1a5199260f6..093ed86a0acd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c @@ -139,11 +139,14 @@ static void esw_acl_ingress_lgcy_groups_destroy(struct mlx5_vport *vport) int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { + bool vst_mode_steering = esw_vst_mode_is_steering(esw); struct mlx5_flow_destination drop_ctr_dst = {}; struct mlx5_flow_destination *dst = NULL; struct mlx5_flow_act flow_act = {}; struct mlx5_flow_spec *spec = NULL; struct mlx5_fc *counter = NULL; + bool vst_check_cvlan = false; + bool vst_push_cvlan = false; /* The ingress acl table contains 4 groups * (2 active rules at the same time - * 1 allow rule from one of the first 3 groups. @@ -203,7 +206,26 @@ int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, goto out; } - if (vport->info.vlan || vport->info.qos) + if ((vport->info.vlan || vport->info.qos)) { + if (vst_mode_steering) + vst_push_cvlan = true; + else if (!MLX5_CAP_ESW(esw->dev, vport_cvlan_insert_always)) + vst_check_cvlan = true; + } + + if (vst_check_cvlan || vport->info.spoofchk) + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + + /* Create ingress allow rule */ + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW; + if (vst_push_cvlan) { + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH; + flow_act.vlan[0].prio = vport->info.qos; + flow_act.vlan[0].vid = vport->info.vlan; + flow_act.vlan[0].ethtype = ETH_P_8021Q; + } + + if (vst_check_cvlan) MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag); @@ -218,9 +240,6 @@ int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, ether_addr_copy(smac_v, vport->info.mac); } - /* Create ingress allow rule */ - spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW; vport->ingress.allow_rule = mlx5_add_flow_rules(vport->ingress.acl, spec, &flow_act, NULL, 0); if (IS_ERR(vport->ingress.allow_rule)) { @@ -232,6 +251,9 @@ int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, goto out; } + if (!vst_check_cvlan && !vport->info.spoofchk) + goto out; + memset(&flow_act, 0, sizeof(flow_act)); flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; /* Attach drop flow counter */ @@ -257,7 +279,8 @@ int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, return 0; out: - esw_acl_ingress_lgcy_cleanup(esw, vport); + if (err) + esw_acl_ingress_lgcy_cleanup(esw, vport); kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 527e4bffda8d..0dfd5742c6fe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -161,10 +161,17 @@ static int modify_esw_vport_cvlan(struct mlx5_core_dev *dev, u16 vport, esw_vport_context.vport_cvlan_strip, 1); if (set_flags & SET_VLAN_INSERT) { - /* insert only if no vlan in packet */ - MLX5_SET(modify_esw_vport_context_in, in, - esw_vport_context.vport_cvlan_insert, 1); - + if (MLX5_CAP_ESW(dev, vport_cvlan_insert_always)) { + /* insert either if vlan exist in packet or not */ + MLX5_SET(modify_esw_vport_context_in, in, + esw_vport_context.vport_cvlan_insert, + MLX5_VPORT_CVLAN_INSERT_ALWAYS); + } else { + /* insert only if no vlan in packet */ + MLX5_SET(modify_esw_vport_context_in, in, + esw_vport_context.vport_cvlan_insert, + MLX5_VPORT_CVLAN_INSERT_WHEN_NO_CVLAN); + } MLX5_SET(modify_esw_vport_context_in, in, esw_vport_context.cvlan_pcp, qos); MLX5_SET(modify_esw_vport_context_in, in, @@ -809,6 +816,7 @@ out_free: static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { + bool vst_mode_steering = esw_vst_mode_is_steering(esw); u16 vport_num = vport->vport; int flags; int err; @@ -839,8 +847,9 @@ static int esw_vport_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) flags = (vport->info.vlan || vport->info.qos) ? SET_VLAN_STRIP | SET_VLAN_INSERT : 0; - modify_esw_vport_cvlan(esw->dev, vport_num, vport->info.vlan, - vport->info.qos, flags); + if (esw->mode == MLX5_ESWITCH_OFFLOADS || !vst_mode_steering) + modify_esw_vport_cvlan(esw->dev, vport_num, vport->info.vlan, + vport->info.qos, flags); return 0; @@ -1848,6 +1857,7 @@ int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, u16 vport, u16 vlan, u8 qos, u8 set_flags) { struct mlx5_vport *evport = mlx5_eswitch_get_vport(esw, vport); + bool vst_mode_steering = esw_vst_mode_is_steering(esw); int err = 0; if (IS_ERR(evport)) @@ -1855,9 +1865,11 @@ int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, if (vlan > 4095 || qos > 7) return -EINVAL; - err = modify_esw_vport_cvlan(esw->dev, vport, vlan, qos, set_flags); - if (err) - return err; + if (esw->mode == MLX5_ESWITCH_OFFLOADS || !vst_mode_steering) { + err = modify_esw_vport_cvlan(esw->dev, vport, vlan, qos, set_flags); + if (err) + return err; + } evport->info.vlan = vlan; evport->info.qos = qos; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 5a85a5d32be7..92644fbb5081 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -527,6 +527,12 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, u16 vport, u16 vlan, u8 qos, u8 set_flags); +static inline bool esw_vst_mode_is_steering(struct mlx5_eswitch *esw) +{ + return (MLX5_CAP_ESW_EGRESS_ACL(esw->dev, pop_vlan) && + MLX5_CAP_ESW_INGRESS_ACL(esw->dev, push_vlan)); +} + static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev, u8 vlan_depth) { diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 5fe5d198b57a..29d4b201c7b2 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1090,6 +1090,11 @@ enum { MLX5_VPORT_ADMIN_STATE_AUTO = 0x2, }; +enum { + MLX5_VPORT_CVLAN_INSERT_WHEN_NO_CVLAN = 0x1, + MLX5_VPORT_CVLAN_INSERT_ALWAYS = 0x3, +}; + enum { MLX5_L3_PROT_TYPE_IPV4 = 0, MLX5_L3_PROT_TYPE_IPV6 = 1, diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index f3d1c62c98dd..a9ee7bc59c90 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -913,7 +913,8 @@ struct mlx5_ifc_e_switch_cap_bits { u8 vport_svlan_insert[0x1]; u8 vport_cvlan_insert_if_not_exist[0x1]; u8 vport_cvlan_insert_overwrite[0x1]; - u8 reserved_at_5[0x2]; + u8 reserved_at_5[0x1]; + u8 vport_cvlan_insert_always[0x1]; u8 esw_shared_ingress_acl[0x1]; u8 esw_uplink_ingress_acl[0x1]; u8 root_ft_on_other_esw[0x1]; From 2a35b2c2e6a252eda2134aae6a756861d9299531 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 18 Oct 2022 12:51:52 +0200 Subject: [PATCH 055/112] net/mlx5: Add forgotten cleanup calls into mlx5_init_once() error path There are two cleanup calls missing in mlx5_init_once() error path. Add them making the error path flow to be the same as mlx5_cleanup_once(). Fixes: 52ec462eca9b ("net/mlx5: Add reserved-gids support") Fixes: 7c39afb394c7 ("net/mlx5: PTP code migration to driver core section") Signed-off-by: Jiri Pirko Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 7f5db13e3550..ec5652f31dda 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1050,6 +1050,8 @@ err_rl_cleanup: err_tables_cleanup: mlx5_geneve_destroy(dev->geneve); mlx5_vxlan_destroy(dev->vxlan); + mlx5_cleanup_clock(dev); + mlx5_cleanup_reserved_gids(dev); mlx5_cq_debugfs_cleanup(dev); mlx5_fw_reset_cleanup(dev); err_events_cleanup: From 44aee8ea15ac205490a41b00cbafcccbf9f7f82b Mon Sep 17 00:00:00 2001 From: Shay Drory Date: Sun, 18 Dec 2022 12:42:14 +0200 Subject: [PATCH 056/112] net/mlx5: Fix io_eq_size and event_eq_size params validation io_eq_size and event_eq_size params are of param type DEVLINK_PARAM_TYPE_U32. But, the validation callback is addressing them as DEVLINK_PARAM_TYPE_U16. This cause mismatch in validation in big-endian systems, in which values in range were rejected while 268500991 was accepted. Fix it by checking the U32 value in the validation callback. Fixes: 0844fa5f7b89 ("net/mlx5: Let user configure io_eq_size param") Signed-off-by: Shay Drory Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/devlink.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index ddb197970c22..be59bb35d795 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -563,7 +563,7 @@ static int mlx5_devlink_eq_depth_validate(struct devlink *devlink, u32 id, union devlink_param_value val, struct netlink_ext_ack *extack) { - return (val.vu16 >= 64 && val.vu16 <= 4096) ? 0 : -EINVAL; + return (val.vu32 >= 64 && val.vu32 <= 4096) ? 0 : -EINVAL; } static const struct devlink_param mlx5_devlink_params[] = { From 9078e843efec530f279a155f262793c58b0746bd Mon Sep 17 00:00:00 2001 From: Shay Drory Date: Thu, 24 Nov 2022 13:34:12 +0200 Subject: [PATCH 057/112] net/mlx5: Avoid recovery in probe flows Currently, recovery is done without considering whether the device is still in probe flow. This may lead to recovery before device have finished probed successfully. e.g.: while mlx5_init_one() is running. Recovery flow is using functionality that is loaded only by mlx5_init_one(), and there is no point in running recovery without mlx5_init_one() finished successfully. Fix it by waiting for probe flow to finish and checking whether the device is probed before trying to perform recovery. Fixes: 51d138c2610a ("net/mlx5: Fix health error state handling") Signed-off-by: Shay Drory Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/health.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c index 86ed87d704f7..96417c5feed7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/health.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c @@ -674,6 +674,12 @@ static void mlx5_fw_fatal_reporter_err_work(struct work_struct *work) dev = container_of(priv, struct mlx5_core_dev, priv); devlink = priv_to_devlink(dev); + mutex_lock(&dev->intf_state_mutex); + if (test_bit(MLX5_DROP_NEW_HEALTH_WORK, &health->flags)) { + mlx5_core_err(dev, "health works are not permitted at this stage\n"); + return; + } + mutex_unlock(&dev->intf_state_mutex); enter_error_state(dev, false); if (IS_ERR_OR_NULL(health->fw_fatal_reporter)) { devl_lock(devlink); From c4ad5f2bdad56265b23d3635494ecdb205431807 Mon Sep 17 00:00:00 2001 From: Shay Drory Date: Wed, 9 Nov 2022 14:42:59 +0200 Subject: [PATCH 058/112] net/mlx5: Fix RoCE setting at HCA level mlx5 PF can disable RoCE for its VFs and SFs. In such case RoCE is marked as unsupported on those VFs/SFs. The cited patch added an option for disable (and enable) RoCE at HCA level. However, that commit didn't check whether RoCE is supported on the HCA and enabled user to try and set RoCE to on. Fix it by checking whether the HCA supports RoCE. Fixes: fbfa97b4d79f ("net/mlx5: Disable roce at HCA level") Signed-off-by: Shay Drory Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/devlink.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c index be59bb35d795..5bd83c0275f8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/devlink.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/devlink.c @@ -468,7 +468,7 @@ static int mlx5_devlink_enable_roce_validate(struct devlink *devlink, u32 id, bool new_state = val.vbool; if (new_state && !MLX5_CAP_GEN(dev, roce) && - !MLX5_CAP_GEN(dev, roce_rw_supported)) { + !(MLX5_CAP_GEN(dev, roce_rw_supported) && MLX5_CAP_GEN_MAX(dev, roce))) { NL_SET_ERR_MSG_MOD(extack, "Device doesn't support RoCE"); return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index ec5652f31dda..df134f6d32dc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -613,7 +613,7 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx) MLX5_SET(cmd_hca_cap, set_hca_cap, num_total_dynamic_vf_msix, MLX5_CAP_GEN_MAX(dev, num_total_dynamic_vf_msix)); - if (MLX5_CAP_GEN(dev, roce_rw_supported)) + if (MLX5_CAP_GEN(dev, roce_rw_supported) && MLX5_CAP_GEN_MAX(dev, roce)) MLX5_SET(cmd_hca_cap, set_hca_cap, roce, mlx5_is_roce_on(dev)); From b12d581e83e3ae1080c32ab83f123005bd89a840 Mon Sep 17 00:00:00 2001 From: Dragos Tatulea Date: Mon, 28 Nov 2022 15:24:21 +0200 Subject: [PATCH 059/112] net/mlx5e: IPoIB, Don't allow CQE compression to be turned on by default mlx5e_build_nic_params will turn CQE compression on if the hardware capability is enabled and the slow_pci_heuristic condition is detected. As IPoIB doesn't support CQE compression, make sure to disable the feature in the IPoIB profile init. Please note that the feature is not exposed to the user for IPoIB interfaces, so it can't be subsequently turned on. Fixes: b797a684b0dd ("net/mlx5e: Enable CQE compression when PCI is slower than link") Signed-off-by: Dragos Tatulea Reviewed-by: Gal Pressman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 7c5c500fd215..2c73c8445e63 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -71,6 +71,10 @@ static void mlx5i_build_nic_params(struct mlx5_core_dev *mdev, params->packet_merge.type = MLX5E_PACKET_MERGE_NONE; params->hard_mtu = MLX5_IB_GRH_BYTES + MLX5_IPOIB_HARD_LEN; params->tunneled_offload_en = false; + + /* CQE compression is not supported for IPoIB */ + params->rx_cqe_compress_def = false; + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS, params->rx_cqe_compress_def); } /* Called directly after IPoIB netdevice was created to initialize SW structs */ From f8c18a5749cf917096f75dd59885b7a0fe9298ba Mon Sep 17 00:00:00 2001 From: Tariq Toukan Date: Sun, 27 Nov 2022 09:21:28 +0200 Subject: [PATCH 060/112] net/mlx5e: Fix RX reporter for XSK RQs RX reporter mistakenly reads from the regular (inactive) RQ when XSK RQ is active. Fix it here. Fixes: 3db4c85cde7a ("net/mlx5e: xsk: Use queue indices starting from 0 for XSK queues") Signed-off-by: Tariq Toukan Reviewed-by: Gal Pressman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c index 5f6f95ad6888..1ae15b8536a8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/reporter_rx.c @@ -459,7 +459,11 @@ static int mlx5e_rx_reporter_diagnose(struct devlink_health_reporter *reporter, goto unlock; for (i = 0; i < priv->channels.num; i++) { - struct mlx5e_rq *rq = &priv->channels.c[i]->rq; + struct mlx5e_channel *c = priv->channels.c[i]; + struct mlx5e_rq *rq; + + rq = test_bit(MLX5E_CHANNEL_STATE_XSK, c->state) ? + &c->xskrq : &c->rq; err = mlx5e_rx_reporter_build_diagnose_output(rq, fmsg); if (err) From 849190e3e4ccf452fbe2240eace30a9ca83fb8d2 Mon Sep 17 00:00:00 2001 From: Chris Mi Date: Mon, 28 Nov 2022 13:54:29 +0800 Subject: [PATCH 061/112] net/mlx5e: CT: Fix ct debugfs folder name Need to use sprintf to build a string instead of sscanf. Otherwise dirname is null and both "ct_nic" and "ct_fdb" won't be created. But its redundant anyway as driver could be in switchdev mode but still add nic rules. So use "ct" as folder name. Fixes: 77422a8f6f61 ("net/mlx5e: CT: Add ct driver counters") Signed-off-by: Chris Mi Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c index a69849e0deed..313df8232db7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c @@ -2103,14 +2103,9 @@ out_err: static void mlx5_ct_tc_create_dbgfs(struct mlx5_tc_ct_priv *ct_priv) { - bool is_fdb = ct_priv->ns_type == MLX5_FLOW_NAMESPACE_FDB; struct mlx5_tc_ct_debugfs *ct_dbgfs = &ct_priv->debugfs; - char dirname[16] = {}; - if (sscanf(dirname, "ct_%s", is_fdb ? "fdb" : "nic") < 0) - return; - - ct_dbgfs->root = debugfs_create_dir(dirname, mlx5_debugfs_get_dev_root(ct_priv->dev)); + ct_dbgfs->root = debugfs_create_dir("ct", mlx5_debugfs_get_dev_root(ct_priv->dev)); debugfs_create_atomic_t("offloaded", 0400, ct_dbgfs->root, &ct_dbgfs->stats.offloaded); debugfs_create_atomic_t("rx_dropped", 0400, ct_dbgfs->root, From 2951b2e142ecf6e0115df785ba91e91b6da74602 Mon Sep 17 00:00:00 2001 From: Chris Mi Date: Mon, 5 Dec 2022 09:22:50 +0800 Subject: [PATCH 062/112] net/mlx5e: Always clear dest encap in neigh-update-del The cited commit introduced a bug for multiple encapsulations flow. If one dest encap becomes invalid, the flow is set slow path flag. But when other dests encap become invalid, they are not cleared due to slow path flag of the flow. When neigh-update-add is running, it will use invalid encap. Fix it by checking slow path flag after clearing dest encap. Fixes: 9a5f9cc794e1 ("net/mlx5e: Fix possible use-after-free deleting fdb rule") Signed-off-by: Chris Mi Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c index ff73d25bc6eb..2aaf8ab857b8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c @@ -222,7 +222,7 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, int err; list_for_each_entry(flow, flow_list, tmp_list) { - if (!mlx5e_is_offloaded_flow(flow) || flow_flag_test(flow, SLOW)) + if (!mlx5e_is_offloaded_flow(flow)) continue; attr = mlx5e_tc_get_encap_attr(flow); @@ -231,6 +231,13 @@ void mlx5e_tc_encap_flows_del(struct mlx5e_priv *priv, esw_attr->dests[flow->tmp_entry_index].flags &= ~MLX5_ESW_DEST_ENCAP_VALID; esw_attr->dests[flow->tmp_entry_index].pkt_reformat = NULL; + /* Clear pkt_reformat before checking slow path flag. Because + * in next iteration, the same flow is already set slow path + * flag, but still need to clear the pkt_reformat. + */ + if (flow_flag_test(flow, SLOW)) + continue; + /* update from encap rule to slow path rule */ spec = &flow->attr->parse_attr->spec; rule = mlx5e_tc_offload_to_slow_path(esw, flow, spec); From 1e267ab88dc44c48f556218f7b7f14c76f7aa066 Mon Sep 17 00:00:00 2001 From: Adham Faris Date: Wed, 14 Dec 2022 16:02:57 +0200 Subject: [PATCH 063/112] net/mlx5e: Fix hw mtu initializing at XDP SQ allocation Current xdp xmit functions logic (mlx5e_xmit_xdp_frame_mpwqe or mlx5e_xmit_xdp_frame), validates xdp packet length by comparing it to hw mtu (configured at xdp sq allocation) before xmiting it. This check does not account for ethernet fcs length (calculated and filled by the nic). Hence, when we try sending packets with length > (hw-mtu - ethernet-fcs-size), the device port drops it and tx_errors_phy is incremented. Desired behavior is to catch these packets and drop them by the driver. Fix this behavior in XDP SQ allocation function (mlx5e_alloc_xdpsq) by subtracting ethernet FCS header size (4 Bytes) from current hw mtu value, since ethernet FCS is calculated and written to ethernet frames by the nic. Fixes: d8bec2b29a82 ("net/mlx5e: Support bpf_xdp_adjust_head()") Signed-off-by: Adham Faris Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 8d36e2de53a9..cff5f2e29e1e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1305,7 +1305,7 @@ static int mlx5e_alloc_xdpsq(struct mlx5e_channel *c, sq->channel = c; sq->uar_map = mdev->mlx5e_res.hw_objs.bfreg.map; sq->min_inline_mode = params->tx_min_inline_mode; - sq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); + sq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu) - ETH_FCS_LEN; sq->xsk_pool = xsk_pool; sq->stats = sq->xsk_pool ? From e54638a8380bd9c146a883035fffd0a821813682 Mon Sep 17 00:00:00 2001 From: Maor Dickman Date: Sun, 1 Aug 2021 14:45:17 +0300 Subject: [PATCH 064/112] net/mlx5e: Set geneve_tlv_option_0_exist when matching on geneve option The cited patch added support of matching on geneve option by setting geneve_tlv_option_0_data mask and key but didn't set geneve_tlv_option_0_exist bit which is required on some HWs when matching geneve_tlv_option_0_data parameter, this may cause in some cases for packets to wrongly match on rules with different geneve option. Example of such case is packet with geneve_tlv_object class=789 and data=456 will wrongly match on rule with match geneve_tlv_object class=123 and data=456. Fix it by setting geneve_tlv_option_0_exist bit when supported by the HW when matching on geneve_tlv_option_0_data parameter. Fixes: 9272e3df3023 ("net/mlx5e: Geneve, Add support for encap/decap flows offload") Signed-off-by: Maor Dickman Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c index f5b26f5a7de4..054d80c4e65c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_geneve.c @@ -273,6 +273,11 @@ static int mlx5e_tc_tun_parse_geneve_options(struct mlx5e_priv *priv, geneve_tlv_option_0_data, be32_to_cpu(opt_data_key)); MLX5_SET(fte_match_set_misc3, misc_3_c, geneve_tlv_option_0_data, be32_to_cpu(opt_data_mask)); + if (MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev, + ft_field_support.geneve_tlv_option_0_exist)) { + MLX5_SET_TO_ONES(fte_match_set_misc, misc_c, geneve_tlv_option_0_exist); + MLX5_SET_TO_ONES(fte_match_set_misc, misc_v, geneve_tlv_option_0_exist); + } spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_3; From 4d1c1379d71777ddeda3e54f8fc26e9ecbfd1009 Mon Sep 17 00:00:00 2001 From: Eli Cohen Date: Thu, 15 Dec 2022 14:28:34 +0200 Subject: [PATCH 065/112] net/mlx5: Lag, fix failure to cancel delayed bond work Commit 0d4e8ed139d8 ("net/mlx5: Lag, avoid lockdep warnings") accidentally removed a call to cancel delayed bond work thus it may cause queued delay to expire and fall on an already destroyed work queue. Fix by restoring the call cancel_delayed_work_sync() before destroying the workqueue. This prevents call trace such as this: [ 329.230417] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ 329.231444] #PF: supervisor write access in kernel mode [ 329.232233] #PF: error_code(0x0002) - not-present page [ 329.233007] PGD 0 P4D 0 [ 329.233476] Oops: 0002 [#1] SMP [ 329.234012] CPU: 5 PID: 145 Comm: kworker/u20:4 Tainted: G OE 6.0.0-rc5_mlnx #1 [ 329.235282] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 [ 329.236868] Workqueue: mlx5_cmd_0000:08:00.1 cmd_work_handler [mlx5_core] [ 329.237886] RIP: 0010:_raw_spin_lock+0xc/0x20 [ 329.238585] Code: f0 0f b1 17 75 02 f3 c3 89 c6 e9 6f 3c 5f ff 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 0f 1f 44 00 00 31 c0 ba 01 00 00 00 0f b1 17 75 02 f3 c3 89 c6 e9 45 3c 5f ff 0f 1f 44 00 00 0f 1f [ 329.241156] RSP: 0018:ffffc900001b0e98 EFLAGS: 00010046 [ 329.241940] RAX: 0000000000000000 RBX: ffffffff82374ae0 RCX: 0000000000000000 [ 329.242954] RDX: 0000000000000001 RSI: 0000000000000014 RDI: 0000000000000000 [ 329.243974] RBP: ffff888106ccf000 R08: ffff8881004000c8 R09: ffff888100400000 [ 329.244990] R10: 0000000000000000 R11: ffffffff826669f8 R12: 0000000000002000 [ 329.246009] R13: 0000000000000005 R14: ffff888100aa7ce0 R15: ffff88852ca80000 [ 329.247030] FS: 0000000000000000(0000) GS:ffff88852ca80000(0000) knlGS:0000000000000000 [ 329.248260] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 329.249111] CR2: 0000000000000000 CR3: 000000016d675001 CR4: 0000000000770ee0 [ 329.250133] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 329.251152] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 329.252176] PKRU: 55555554 Fixes: 0d4e8ed139d8 ("net/mlx5: Lag, avoid lockdep warnings") Signed-off-by: Eli Cohen Reviewed-by: Maor Dickman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c index 32c3e0a649a7..ad32b80e8501 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c @@ -228,6 +228,7 @@ static void mlx5_ldev_free(struct kref *ref) if (ldev->nb.notifier_call) unregister_netdevice_notifier_net(&init_net, &ldev->nb); mlx5_lag_mp_cleanup(ldev); + cancel_delayed_work_sync(&ldev->bond_work); destroy_workqueue(ldev->wq); mlx5_lag_mpesw_cleanup(ldev); mutex_destroy(&ldev->lock); From 9ed1d9aeef5842ecacb660fce933613b58af1e00 Mon Sep 17 00:00:00 2001 From: Chuang Wang Date: Sat, 24 Dec 2022 21:31:46 +0800 Subject: [PATCH 066/112] bpf: Fix panic due to wrong pageattr of im->image In the scenario where livepatch and kretfunc coexist, the pageattr of im->image is rox after arch_prepare_bpf_trampoline in bpf_trampoline_update, and then modify_fentry or register_fentry returns -EAGAIN from bpf_tramp_ftrace_ops_func, the BPF_TRAMP_F_ORIG_STACK flag will be configured, and arch_prepare_bpf_trampoline will be re-executed. At this time, because the pageattr of im->image is rox, arch_prepare_bpf_trampoline will read and write im->image, which causes a fault. as follows: insmod livepatch-sample.ko # samples/livepatch/livepatch-sample.c bpftrace -e 'kretfunc:cmdline_proc_show {}' BUG: unable to handle page fault for address: ffffffffa0206000 PGD 322d067 P4D 322d067 PUD 322e063 PMD 1297e067 PTE d428061 Oops: 0003 [#1] PREEMPT SMP PTI CPU: 2 PID: 270 Comm: bpftrace Tainted: G E K 6.1.0 #5 RIP: 0010:arch_prepare_bpf_trampoline+0xed/0x8c0 RSP: 0018:ffffc90001083ad8 EFLAGS: 00010202 RAX: ffffffffa0206000 RBX: 0000000000000020 RCX: 0000000000000000 RDX: ffffffffa0206001 RSI: ffffffffa0206000 RDI: 0000000000000030 RBP: ffffc90001083b70 R08: 0000000000000066 R09: ffff88800f51b400 R10: 000000002e72c6e5 R11: 00000000d0a15080 R12: ffff8880110a68c8 R13: 0000000000000000 R14: ffff88800f51b400 R15: ffffffff814fec10 FS: 00007f87bc0dc780(0000) GS:ffff88803e600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffffffffa0206000 CR3: 0000000010b70000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: bpf_trampoline_update+0x25a/0x6b0 __bpf_trampoline_link_prog+0x101/0x240 bpf_trampoline_link_prog+0x2d/0x50 bpf_tracing_prog_attach+0x24c/0x530 bpf_raw_tp_link_attach+0x73/0x1d0 __sys_bpf+0x100e/0x2570 __x64_sys_bpf+0x1c/0x30 do_syscall_64+0x5b/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd With this patch, when modify_fentry or register_fentry returns -EAGAIN from bpf_tramp_ftrace_ops_func, the pageattr of im->image will be reset to nx+rw. Cc: stable@vger.kernel.org Fixes: 00963a2e75a8 ("bpf: Support bpf_trampoline on functions with IPMODIFY (e.g. livepatch)") Signed-off-by: Chuang Wang Acked-by: Jiri Olsa Acked-by: Song Liu Link: https://lore.kernel.org/r/20221224133146.780578-1-nashuiliang@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/trampoline.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c index 11f5ec0b8016..d0ed7d6f5eec 100644 --- a/kernel/bpf/trampoline.c +++ b/kernel/bpf/trampoline.c @@ -488,6 +488,10 @@ again: /* reset fops->func and fops->trampoline for re-register */ tr->fops->func = NULL; tr->fops->trampoline = 0; + + /* reset im->image memory attr for arch_prepare_bpf_trampoline */ + set_memory_nx((long)im->image, 1); + set_memory_rw((long)im->image, 1); goto again; } #endif From 8f161ca1105a6af6614333f13aa7be4aab8b633a Mon Sep 17 00:00:00 2001 From: Alexei Starovoitov Date: Wed, 28 Dec 2022 13:55:57 -0800 Subject: [PATCH 067/112] selftests/bpf: Temporarily disable part of btf_dump:var_data test. Commit 7443b296e699 ("x86/percpu: Move cpu_number next to current_task") moved global per_cpu variable 'cpu_number' into pcpu_hot structure. Therefore this part of var_data test is no longer valid. Disable it until better solution is found. Signed-off-by: Alexei Starovoitov --- tools/testing/selftests/bpf/prog_tests/btf_dump.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/btf_dump.c b/tools/testing/selftests/bpf/prog_tests/btf_dump.c index 0ba2e8b9c6ac..e9ea38aa8248 100644 --- a/tools/testing/selftests/bpf/prog_tests/btf_dump.c +++ b/tools/testing/selftests/bpf/prog_tests/btf_dump.c @@ -801,7 +801,7 @@ static void test_btf_dump_struct_data(struct btf *btf, struct btf_dump *d, static void test_btf_dump_var_data(struct btf *btf, struct btf_dump *d, char *str) { -#if defined(__i386__) || defined(__x86_64__) || defined(__aarch64__) +#if 0 TEST_BTF_DUMP_VAR(btf, d, NULL, str, "cpu_number", int, BTF_F_COMPACT, "int cpu_number = (int)100", 100); #endif From 7ff94f276f8ea05df82eb115225e9b26f47a3347 Mon Sep 17 00:00:00 2001 From: Kui-Feng Lee Date: Fri, 16 Dec 2022 14:18:54 -0800 Subject: [PATCH 068/112] bpf: keep a reference to the mm, in case the task is dead. Fix the system crash that happens when a task iterator travel through vma of tasks. In task iterators, we used to access mm by following the pointer on the task_struct; however, the death of a task will clear the pointer, even though we still hold the task_struct. That can cause an unexpected crash for a null pointer when an iterator is visiting a task that dies during the visit. Keeping a reference of mm on the iterator ensures we always have a valid pointer to mm. Co-developed-by: Song Liu Signed-off-by: Song Liu Signed-off-by: Kui-Feng Lee Reported-by: Nathan Slingerland Acked-by: Yonghong Song Link: https://lore.kernel.org/r/20221216221855.4122288-2-kuifeng@meta.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/task_iter.c | 39 +++++++++++++++++++++++++++------------ 1 file changed, 27 insertions(+), 12 deletions(-) diff --git a/kernel/bpf/task_iter.c b/kernel/bpf/task_iter.c index c2a2182ce570..c4ab9d6cdbe9 100644 --- a/kernel/bpf/task_iter.c +++ b/kernel/bpf/task_iter.c @@ -438,6 +438,7 @@ struct bpf_iter_seq_task_vma_info { */ struct bpf_iter_seq_task_common common; struct task_struct *task; + struct mm_struct *mm; struct vm_area_struct *vma; u32 tid; unsigned long prev_vm_start; @@ -456,16 +457,19 @@ task_vma_seq_get_next(struct bpf_iter_seq_task_vma_info *info) enum bpf_task_vma_iter_find_op op; struct vm_area_struct *curr_vma; struct task_struct *curr_task; + struct mm_struct *curr_mm; u32 saved_tid = info->tid; /* If this function returns a non-NULL vma, it holds a reference to - * the task_struct, and holds read lock on vma->mm->mmap_lock. + * the task_struct, holds a refcount on mm->mm_users, and holds + * read lock on vma->mm->mmap_lock. * If this function returns NULL, it does not hold any reference or * lock. */ if (info->task) { curr_task = info->task; curr_vma = info->vma; + curr_mm = info->mm; /* In case of lock contention, drop mmap_lock to unblock * the writer. * @@ -504,13 +508,15 @@ task_vma_seq_get_next(struct bpf_iter_seq_task_vma_info *info) * 4.2) VMA2 and VMA2' covers different ranges, process * VMA2'. */ - if (mmap_lock_is_contended(curr_task->mm)) { + if (mmap_lock_is_contended(curr_mm)) { info->prev_vm_start = curr_vma->vm_start; info->prev_vm_end = curr_vma->vm_end; op = task_vma_iter_find_vma; - mmap_read_unlock(curr_task->mm); - if (mmap_read_lock_killable(curr_task->mm)) + mmap_read_unlock(curr_mm); + if (mmap_read_lock_killable(curr_mm)) { + mmput(curr_mm); goto finish; + } } else { op = task_vma_iter_next_vma; } @@ -535,42 +541,47 @@ again: op = task_vma_iter_find_vma; } - if (!curr_task->mm) + curr_mm = get_task_mm(curr_task); + if (!curr_mm) goto next_task; - if (mmap_read_lock_killable(curr_task->mm)) + if (mmap_read_lock_killable(curr_mm)) { + mmput(curr_mm); goto finish; + } } switch (op) { case task_vma_iter_first_vma: - curr_vma = find_vma(curr_task->mm, 0); + curr_vma = find_vma(curr_mm, 0); break; case task_vma_iter_next_vma: - curr_vma = find_vma(curr_task->mm, curr_vma->vm_end); + curr_vma = find_vma(curr_mm, curr_vma->vm_end); break; case task_vma_iter_find_vma: /* We dropped mmap_lock so it is necessary to use find_vma * to find the next vma. This is similar to the mechanism * in show_smaps_rollup(). */ - curr_vma = find_vma(curr_task->mm, info->prev_vm_end - 1); + curr_vma = find_vma(curr_mm, info->prev_vm_end - 1); /* case 1) and 4.2) above just use curr_vma */ /* check for case 2) or case 4.1) above */ if (curr_vma && curr_vma->vm_start == info->prev_vm_start && curr_vma->vm_end == info->prev_vm_end) - curr_vma = find_vma(curr_task->mm, curr_vma->vm_end); + curr_vma = find_vma(curr_mm, curr_vma->vm_end); break; } if (!curr_vma) { /* case 3) above, or case 2) 4.1) with vma->next == NULL */ - mmap_read_unlock(curr_task->mm); + mmap_read_unlock(curr_mm); + mmput(curr_mm); goto next_task; } info->task = curr_task; info->vma = curr_vma; + info->mm = curr_mm; return curr_vma; next_task: @@ -579,6 +590,7 @@ next_task: put_task_struct(curr_task); info->task = NULL; + info->mm = NULL; info->tid++; goto again; @@ -587,6 +599,7 @@ finish: put_task_struct(curr_task); info->task = NULL; info->vma = NULL; + info->mm = NULL; return NULL; } @@ -658,7 +671,9 @@ static void task_vma_seq_stop(struct seq_file *seq, void *v) */ info->prev_vm_start = ~0UL; info->prev_vm_end = info->vma->vm_end; - mmap_read_unlock(info->task->mm); + mmap_read_unlock(info->mm); + mmput(info->mm); + info->mm = NULL; put_task_struct(info->task); info->task = NULL; } From b7793c8db7d9beb903bb42f52872b5b46abdcb88 Mon Sep 17 00:00:00 2001 From: Kui-Feng Lee Date: Fri, 16 Dec 2022 14:18:55 -0800 Subject: [PATCH 069/112] selftests/bpf: add a test for iter/task_vma for short-lived processes When a task iterator traverses vma(s), it is possible task->mm might become invalid in the middle of traversal and this may cause kernel misbehave (e.g., crash) This test case creates iterators repeatedly and forks short-lived processes in the background to detect this bug. The test will last for 3 seconds to get the chance to trigger the issue. Signed-off-by: Kui-Feng Lee Acked-by: Yonghong Song Link: https://lore.kernel.org/r/20221216221855.4122288-3-kuifeng@meta.com Signed-off-by: Alexei Starovoitov --- .../selftests/bpf/prog_tests/bpf_iter.c | 73 +++++++++++++++++++ 1 file changed, 73 insertions(+) diff --git a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c index 6f8ed61fc4b4..3af6450763e9 100644 --- a/tools/testing/selftests/bpf/prog_tests/bpf_iter.c +++ b/tools/testing/selftests/bpf/prog_tests/bpf_iter.c @@ -1465,6 +1465,77 @@ out: bpf_iter_task_vma__destroy(skel); } +static void test_task_vma_dead_task(void) +{ + struct bpf_iter_task_vma *skel; + int wstatus, child_pid = -1; + time_t start_tm, cur_tm; + int err, iter_fd = -1; + int wait_sec = 3; + + skel = bpf_iter_task_vma__open(); + if (!ASSERT_OK_PTR(skel, "bpf_iter_task_vma__open")) + return; + + skel->bss->pid = getpid(); + + err = bpf_iter_task_vma__load(skel); + if (!ASSERT_OK(err, "bpf_iter_task_vma__load")) + goto out; + + skel->links.proc_maps = bpf_program__attach_iter( + skel->progs.proc_maps, NULL); + + if (!ASSERT_OK_PTR(skel->links.proc_maps, "bpf_program__attach_iter")) { + skel->links.proc_maps = NULL; + goto out; + } + + start_tm = time(NULL); + cur_tm = start_tm; + + child_pid = fork(); + if (child_pid == 0) { + /* Fork short-lived processes in the background. */ + while (cur_tm < start_tm + wait_sec) { + system("echo > /dev/null"); + cur_tm = time(NULL); + } + exit(0); + } + + if (!ASSERT_GE(child_pid, 0, "fork_child")) + goto out; + + while (cur_tm < start_tm + wait_sec) { + iter_fd = bpf_iter_create(bpf_link__fd(skel->links.proc_maps)); + if (!ASSERT_GE(iter_fd, 0, "create_iter")) + goto out; + + /* Drain all data from iter_fd. */ + while (cur_tm < start_tm + wait_sec) { + err = read_fd_into_buffer(iter_fd, task_vma_output, CMP_BUFFER_SIZE); + if (!ASSERT_GE(err, 0, "read_iter_fd")) + goto out; + + cur_tm = time(NULL); + + if (err == 0) + break; + } + + close(iter_fd); + iter_fd = -1; + } + + check_bpf_link_info(skel->progs.proc_maps); + +out: + waitpid(child_pid, &wstatus, 0); + close(iter_fd); + bpf_iter_task_vma__destroy(skel); +} + void test_bpf_sockmap_map_iter_fd(void) { struct bpf_iter_sockmap *skel; @@ -1586,6 +1657,8 @@ void test_bpf_iter(void) test_task_file(); if (test__start_subtest("task_vma")) test_task_vma(); + if (test__start_subtest("task_vma_dead_task")) + test_task_vma_dead_task(); if (test__start_subtest("task_btf")) test_task_btf(); if (test__start_subtest("tcp4")) From 45435d8da71f9f3e6860e6e6ea9667b6ec17ec64 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 23 Dec 2022 10:28:44 -0800 Subject: [PATCH 070/112] bpf: Always use maximal size for copy_array() Instead of counting on prior allocations to have sized allocations to the next kmalloc bucket size, always perform a krealloc that is at least ksize(dst) in size (which is a no-op), so the size can be correctly tracked by all the various allocation size trackers (KASAN, __alloc_size, etc). Reported-by: Hyunwoo Kim Link: https://lore.kernel.org/bpf/20221223094551.GA1439509@ubuntu Fixes: ceb35b666d42 ("bpf/verifier: Use kmalloc_size_roundup() to match ksize() usage") Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: John Fastabend Cc: Andrii Nakryiko Cc: Martin KaFai Lau Cc: Song Liu Cc: Yonghong Song Cc: KP Singh Cc: Stanislav Fomichev Cc: Hao Luo Cc: Jiri Olsa Cc: bpf@vger.kernel.org Signed-off-by: Kees Cook Link: https://lore.kernel.org/r/20221223182836.never.866-kees@kernel.org Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 243d06ce6842..85f96c1e9f62 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1054,6 +1054,8 @@ static void print_insn_state(struct bpf_verifier_env *env, */ static void *copy_array(void *dst, const void *src, size_t n, size_t size, gfp_t flags) { + size_t alloc_bytes; + void *orig = dst; size_t bytes; if (ZERO_OR_NULL_PTR(src)) @@ -1062,11 +1064,11 @@ static void *copy_array(void *dst, const void *src, size_t n, size_t size, gfp_t if (unlikely(check_mul_overflow(n, size, &bytes))) return NULL; - if (ksize(dst) < ksize(src)) { - kfree(dst); - dst = kmalloc_track_caller(kmalloc_size_roundup(bytes), flags); - if (!dst) - return NULL; + alloc_bytes = max(ksize(orig), kmalloc_size_roundup(bytes)); + dst = krealloc(orig, alloc_bytes, flags); + if (!dst) { + kfree(orig); + return NULL; } memcpy(dst, src, bytes); From 936a192f974018b4f6040f6f77b1cc1e75bd8666 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 26 Dec 2022 22:27:52 +0900 Subject: [PATCH 071/112] tcp: Add TIME_WAIT sockets in bhash2. Jiri Slaby reported regression of bind() with a simple repro. [0] The repro creates a TIME_WAIT socket and tries to bind() a new socket with the same local address and port. Before commit 28044fc1d495 ("net: Add a bhash2 table hashed by port and address"), the bind() failed with -EADDRINUSE, but now it succeeds. The cited commit should have put TIME_WAIT sockets into bhash2; otherwise, inet_bhash2_conflict() misses TIME_WAIT sockets when validating bind() requests if the address is not a wildcard one. The straight option is to move sk_bind2_node from struct sock to struct sock_common to add twsk to bhash2 as implemented as RFC. [1] However, the binary layout change in the struct sock could affect performances moving hot fields on different cachelines. To avoid that, we add another TIME_WAIT list in inet_bind2_bucket and check it while validating bind(). [0]: https://lore.kernel.org/netdev/6b971a4e-c7d8-411e-1f92-fda29b5b2fb9@kernel.org/ [1]: https://lore.kernel.org/netdev/20221221151258.25748-2-kuniyu@amazon.com/ Fixes: 28044fc1d495 ("net: Add a bhash2 table hashed by port and address") Reported-by: Jiri Slaby Suggested-by: Paolo Abeni Signed-off-by: Kuniyuki Iwashima Acked-by: Joanne Koong Signed-off-by: David S. Miller --- include/net/inet_hashtables.h | 4 ++++ include/net/inet_timewait_sock.h | 5 +++++ net/ipv4/inet_connection_sock.c | 28 +++++++++++++++++++++++----- net/ipv4/inet_hashtables.c | 8 +++++--- net/ipv4/inet_timewait_sock.c | 31 +++++++++++++++++++++++++++++-- 5 files changed, 66 insertions(+), 10 deletions(-) diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 69174093078f..99bd823e97f6 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -108,6 +108,10 @@ struct inet_bind2_bucket { struct hlist_node node; /* List of sockets hashed to this bucket */ struct hlist_head owners; + /* bhash has twsk in owners, but bhash2 has twsk in + * deathrow not to add a member in struct sock_common. + */ + struct hlist_head deathrow; }; static inline struct net *ib_net(const struct inet_bind_bucket *ib) diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index 5b47545f22d3..4a8e578405cb 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -73,9 +73,14 @@ struct inet_timewait_sock { u32 tw_priority; struct timer_list tw_timer; struct inet_bind_bucket *tw_tb; + struct inet_bind2_bucket *tw_tb2; + struct hlist_node tw_bind2_node; }; #define tw_tclass tw_tos +#define twsk_for_each_bound_bhash2(__tw, list) \ + hlist_for_each_entry(__tw, list, tw_bind2_node) + static inline struct inet_timewait_sock *inet_twsk(const struct sock *sk) { return (struct inet_timewait_sock *)sk; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index b366ab9148f2..848ffc3e0239 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -173,22 +173,40 @@ static bool inet_bind_conflict(const struct sock *sk, struct sock *sk2, return false; } +static bool __inet_bhash2_conflict(const struct sock *sk, struct sock *sk2, + kuid_t sk_uid, bool relax, + bool reuseport_cb_ok, bool reuseport_ok) +{ + if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) + return false; + + return inet_bind_conflict(sk, sk2, sk_uid, relax, + reuseport_cb_ok, reuseport_ok); +} + static bool inet_bhash2_conflict(const struct sock *sk, const struct inet_bind2_bucket *tb2, kuid_t sk_uid, bool relax, bool reuseport_cb_ok, bool reuseport_ok) { + struct inet_timewait_sock *tw2; struct sock *sk2; sk_for_each_bound_bhash2(sk2, &tb2->owners) { - if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) - continue; - - if (inet_bind_conflict(sk, sk2, sk_uid, relax, - reuseport_cb_ok, reuseport_ok)) + if (__inet_bhash2_conflict(sk, sk2, sk_uid, relax, + reuseport_cb_ok, reuseport_ok)) return true; } + + twsk_for_each_bound_bhash2(tw2, &tb2->deathrow) { + sk2 = (struct sock *)tw2; + + if (__inet_bhash2_conflict(sk, sk2, sk_uid, relax, + reuseport_cb_ok, reuseport_ok)) + return true; + } + return false; } diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index d039b4e732a3..24a38b56fab9 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -116,6 +116,7 @@ static void inet_bind2_bucket_init(struct inet_bind2_bucket *tb, #endif tb->rcv_saddr = sk->sk_rcv_saddr; INIT_HLIST_HEAD(&tb->owners); + INIT_HLIST_HEAD(&tb->deathrow); hlist_add_head(&tb->node, &head->chain); } @@ -137,7 +138,7 @@ struct inet_bind2_bucket *inet_bind2_bucket_create(struct kmem_cache *cachep, /* Caller must hold hashbucket lock for this tb with local BH disabled */ void inet_bind2_bucket_destroy(struct kmem_cache *cachep, struct inet_bind2_bucket *tb) { - if (hlist_empty(&tb->owners)) { + if (hlist_empty(&tb->owners) && hlist_empty(&tb->deathrow)) { __hlist_del(&tb->node); kmem_cache_free(cachep, tb); } @@ -1103,15 +1104,16 @@ ok: /* Head lock still held and bh's disabled */ inet_bind_hash(sk, tb, tb2, port); - spin_unlock(&head2->lock); - if (sk_unhashed(sk)) { inet_sk(sk)->inet_sport = htons(port); inet_ehash_nolisten(sk, (struct sock *)tw, NULL); } if (tw) inet_twsk_bind_unhash(tw, hinfo); + + spin_unlock(&head2->lock); spin_unlock(&head->lock); + if (tw) inet_twsk_deschedule_put(tw); local_bh_enable(); diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index 66fc940f9521..1d77d992e6e7 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -29,6 +29,7 @@ void inet_twsk_bind_unhash(struct inet_timewait_sock *tw, struct inet_hashinfo *hashinfo) { + struct inet_bind2_bucket *tb2 = tw->tw_tb2; struct inet_bind_bucket *tb = tw->tw_tb; if (!tb) @@ -37,6 +38,11 @@ void inet_twsk_bind_unhash(struct inet_timewait_sock *tw, __hlist_del(&tw->tw_bind_node); tw->tw_tb = NULL; inet_bind_bucket_destroy(hashinfo->bind_bucket_cachep, tb); + + __hlist_del(&tw->tw_bind2_node); + tw->tw_tb2 = NULL; + inet_bind2_bucket_destroy(hashinfo->bind2_bucket_cachep, tb2); + __sock_put((struct sock *)tw); } @@ -45,7 +51,7 @@ static void inet_twsk_kill(struct inet_timewait_sock *tw) { struct inet_hashinfo *hashinfo = tw->tw_dr->hashinfo; spinlock_t *lock = inet_ehash_lockp(hashinfo, tw->tw_hash); - struct inet_bind_hashbucket *bhead; + struct inet_bind_hashbucket *bhead, *bhead2; spin_lock(lock); sk_nulls_del_node_init_rcu((struct sock *)tw); @@ -54,9 +60,13 @@ static void inet_twsk_kill(struct inet_timewait_sock *tw) /* Disassociate with bind bucket. */ bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), tw->tw_num, hashinfo->bhash_size)]; + bhead2 = inet_bhashfn_portaddr(hashinfo, (struct sock *)tw, + twsk_net(tw), tw->tw_num); spin_lock(&bhead->lock); + spin_lock(&bhead2->lock); inet_twsk_bind_unhash(tw, hashinfo); + spin_unlock(&bhead2->lock); spin_unlock(&bhead->lock); refcount_dec(&tw->tw_dr->tw_refcount); @@ -93,6 +103,12 @@ static void inet_twsk_add_bind_node(struct inet_timewait_sock *tw, hlist_add_head(&tw->tw_bind_node, list); } +static void inet_twsk_add_bind2_node(struct inet_timewait_sock *tw, + struct hlist_head *list) +{ + hlist_add_head(&tw->tw_bind2_node, list); +} + /* * Enter the time wait state. This is called with locally disabled BH. * Essentially we whip up a timewait bucket, copy the relevant info into it @@ -105,17 +121,28 @@ void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, const struct inet_connection_sock *icsk = inet_csk(sk); struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, sk->sk_hash); spinlock_t *lock = inet_ehash_lockp(hashinfo, sk->sk_hash); - struct inet_bind_hashbucket *bhead; + struct inet_bind_hashbucket *bhead, *bhead2; + /* Step 1: Put TW into bind hash. Original socket stays there too. Note, that any socket with inet->num != 0 MUST be bound in binding cache, even if it is closed. */ bhead = &hashinfo->bhash[inet_bhashfn(twsk_net(tw), inet->inet_num, hashinfo->bhash_size)]; + bhead2 = inet_bhashfn_portaddr(hashinfo, sk, twsk_net(tw), inet->inet_num); + spin_lock(&bhead->lock); + spin_lock(&bhead2->lock); + tw->tw_tb = icsk->icsk_bind_hash; WARN_ON(!icsk->icsk_bind_hash); inet_twsk_add_bind_node(tw, &tw->tw_tb->owners); + + tw->tw_tb2 = icsk->icsk_bind2_hash; + WARN_ON(!icsk->icsk_bind2_hash); + inet_twsk_add_bind2_node(tw, &tw->tw_tb2->deathrow); + + spin_unlock(&bhead2->lock); spin_unlock(&bhead->lock); spin_lock(lock); From 2c042e8e54efb2b8e25ed0cb28224e79948dc8ce Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 26 Dec 2022 22:27:53 +0900 Subject: [PATCH 072/112] tcp: Add selftest for bind() and TIME_WAIT. bhash2 split the bind() validation logic into wildcard and non-wildcard cases. Let's add a test to catch future regression. Before the previous patch: # ./bind_timewait TAP version 13 1..2 # Starting 2 tests from 3 test cases. # RUN bind_timewait.localhost.1 ... # bind_timewait.c:87:1:Expected ret (0) == -1 (-1) # 1: Test terminated by assertion # FAIL bind_timewait.localhost.1 not ok 1 bind_timewait.localhost.1 # RUN bind_timewait.addrany.1 ... # OK bind_timewait.addrany.1 ok 2 bind_timewait.addrany.1 # FAILED: 1 / 2 tests passed. # Totals: pass:1 fail:1 xfail:0 xpass:0 skip:0 error:0 After: # ./bind_timewait TAP version 13 1..2 # Starting 2 tests from 3 test cases. # RUN bind_timewait.localhost.1 ... # OK bind_timewait.localhost.1 ok 1 bind_timewait.localhost.1 # RUN bind_timewait.addrany.1 ... # OK bind_timewait.addrany.1 ok 2 bind_timewait.addrany.1 # PASSED: 2 / 2 tests passed. # Totals: pass:2 fail:0 xfail:0 xpass:0 skip:0 error:0 Signed-off-by: Kuniyuki Iwashima Acked-by: Joanne Koong Signed-off-by: David S. Miller --- tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/bind_timewait.c | 92 +++++++++++++++++++++ 2 files changed, 93 insertions(+) create mode 100644 tools/testing/selftests/net/bind_timewait.c diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore index 9cc84114741d..a6911cae368c 100644 --- a/tools/testing/selftests/net/.gitignore +++ b/tools/testing/selftests/net/.gitignore @@ -1,5 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only bind_bhash +bind_timewait csum cmsg_sender diag_uid diff --git a/tools/testing/selftests/net/bind_timewait.c b/tools/testing/selftests/net/bind_timewait.c new file mode 100644 index 000000000000..cb9fdf51ea59 --- /dev/null +++ b/tools/testing/selftests/net/bind_timewait.c @@ -0,0 +1,92 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright Amazon.com Inc. or its affiliates. */ + +#include +#include + +#include "../kselftest_harness.h" + +FIXTURE(bind_timewait) +{ + struct sockaddr_in addr; + socklen_t addrlen; +}; + +FIXTURE_VARIANT(bind_timewait) +{ + __u32 addr_const; +}; + +FIXTURE_VARIANT_ADD(bind_timewait, localhost) +{ + .addr_const = INADDR_LOOPBACK +}; + +FIXTURE_VARIANT_ADD(bind_timewait, addrany) +{ + .addr_const = INADDR_ANY +}; + +FIXTURE_SETUP(bind_timewait) +{ + self->addr.sin_family = AF_INET; + self->addr.sin_port = 0; + self->addr.sin_addr.s_addr = htonl(variant->addr_const); + self->addrlen = sizeof(self->addr); +} + +FIXTURE_TEARDOWN(bind_timewait) +{ +} + +void create_timewait_socket(struct __test_metadata *_metadata, + FIXTURE_DATA(bind_timewait) *self) +{ + int server_fd, client_fd, child_fd, ret; + struct sockaddr_in addr; + socklen_t addrlen; + + server_fd = socket(AF_INET, SOCK_STREAM, 0); + ASSERT_GT(server_fd, 0); + + ret = bind(server_fd, (struct sockaddr *)&self->addr, self->addrlen); + ASSERT_EQ(ret, 0); + + ret = listen(server_fd, 1); + ASSERT_EQ(ret, 0); + + ret = getsockname(server_fd, (struct sockaddr *)&self->addr, &self->addrlen); + ASSERT_EQ(ret, 0); + + client_fd = socket(AF_INET, SOCK_STREAM, 0); + ASSERT_GT(client_fd, 0); + + ret = connect(client_fd, (struct sockaddr *)&self->addr, self->addrlen); + ASSERT_EQ(ret, 0); + + addrlen = sizeof(addr); + child_fd = accept(server_fd, (struct sockaddr *)&addr, &addrlen); + ASSERT_GT(child_fd, 0); + + close(child_fd); + close(client_fd); + close(server_fd); +} + +TEST_F(bind_timewait, 1) +{ + int fd, ret; + + create_timewait_socket(_metadata, self); + + fd = socket(AF_INET, SOCK_STREAM, 0); + ASSERT_GT(fd, 0); + + ret = bind(fd, (struct sockaddr *)&self->addr, self->addrlen); + ASSERT_EQ(ret, -1); + ASSERT_EQ(errno, EADDRINUSE); + + close(fd); +} + +TEST_HARNESS_MAIN From 6b57bffa5f675a01c7981ed271e8521e87441abd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Tue, 27 Dec 2022 22:45:07 +0100 Subject: [PATCH 073/112] net: ethernet: broadcom: bcm63xx_enet: Drop empty platform remove function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A remove callback just returning 0 is equivalent to no remove callback at all. So drop the useless function. Signed-off-by: Uwe Kleine-König Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bcm63xx_enet.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c index d91fdb0c2649..2cf96892e565 100644 --- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c +++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c @@ -2784,17 +2784,11 @@ static int bcm_enet_shared_probe(struct platform_device *pdev) return 0; } -static int bcm_enet_shared_remove(struct platform_device *pdev) -{ - return 0; -} - /* this "shared" driver is needed because both macs share a single * address space */ struct platform_driver bcm63xx_enet_shared_driver = { .probe = bcm_enet_shared_probe, - .remove = bcm_enet_shared_remove, .driver = { .name = "bcm63xx_enet_shared", .owner = THIS_MODULE, From af691c94d022440476b76560d310d6fea790cc60 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Tue, 27 Dec 2022 22:45:08 +0100 Subject: [PATCH 074/112] net: ethernet: freescale: enetc: Drop empty platform remove function MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A remove callback just returning 0 is equivalent to no remove callback at all. So drop the useless function. Signed-off-by: Uwe Kleine-König Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/enetc/enetc_ierb.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ierb.c b/drivers/net/ethernet/freescale/enetc/enetc_ierb.c index 91f02c505028..b307bef4dc29 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ierb.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ierb.c @@ -127,11 +127,6 @@ static int enetc_ierb_probe(struct platform_device *pdev) return 0; } -static int enetc_ierb_remove(struct platform_device *pdev) -{ - return 0; -} - static const struct of_device_id enetc_ierb_match[] = { { .compatible = "fsl,ls1028a-enetc-ierb", }, {}, @@ -144,7 +139,6 @@ static struct platform_driver enetc_ierb_driver = { .of_match_table = enetc_ierb_match, }, .probe = enetc_ierb_probe, - .remove = enetc_ierb_remove, }; module_platform_driver(enetc_ierb_driver); From fec7352117fa301bfbc31bacc14bb9a579376b36 Mon Sep 17 00:00:00 2001 From: Jian Shen Date: Wed, 28 Dec 2022 14:27:49 +0800 Subject: [PATCH 075/112] net: hns3: refine the handling for VF heartbeat Currently, the PF check the VF alive by the KEEP_ALVE mailbox from VF. VF keep sending the mailbox per 2 seconds. Once PF lost the mailbox for more than 8 seconds, it will regards the VF is abnormal, and stop notifying the state change to VF, include link state, vf mac, reset, even though it receives the KEEP_ALIVE mailbox again. It's inreasonable. This patch fixes it. PF will record the state change which need to notify VF when lost the VF's KEEP_ALIVE mailbox. And notify VF when receive the mailbox again. Introduce a new flag HCLGE_VPORT_STATE_INITED, used to distinguish the case whether VF driver loaded or not. For VF will query these states when initializing, so it's unnecessary to notify it in this case. Fixes: aa5c4f175be6 ("net: hns3: add reset handling for VF when doing PF reset") Signed-off-by: Jian Shen Signed-off-by: Hao Lan Reported-by: kernel test robot Signed-off-by: David S. Miller --- .../hisilicon/hns3/hns3pf/hclge_main.c | 57 +++++++++++---- .../hisilicon/hns3/hns3pf/hclge_main.h | 7 ++ .../hisilicon/hns3/hns3pf/hclge_mbx.c | 71 ++++++++++++++++--- 3 files changed, 112 insertions(+), 23 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 6c2742f59c77..07ad5f35219e 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -3910,9 +3910,17 @@ static int hclge_set_all_vf_rst(struct hclge_dev *hdev, bool reset) return ret; } - if (!reset || !test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) + if (!reset || + !test_bit(HCLGE_VPORT_STATE_INITED, &vport->state)) continue; + if (!test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state) && + hdev->reset_type == HNAE3_FUNC_RESET) { + set_bit(HCLGE_VPORT_NEED_NOTIFY_RESET, + &vport->need_notify); + continue; + } + /* Inform VF to process the reset. * hclge_inform_reset_assert_to_vf may fail if VF * driver is not loaded. @@ -4609,18 +4617,25 @@ static void hclge_reset_service_task(struct hclge_dev *hdev) static void hclge_update_vport_alive(struct hclge_dev *hdev) { +#define HCLGE_ALIVE_SECONDS_NORMAL 8 + + unsigned long alive_time = HCLGE_ALIVE_SECONDS_NORMAL * HZ; int i; /* start from vport 1 for PF is always alive */ for (i = 1; i < hdev->num_alloc_vport; i++) { struct hclge_vport *vport = &hdev->vport[i]; - if (time_after(jiffies, vport->last_active_jiffies + 8 * HZ)) + if (!test_bit(HCLGE_VPORT_STATE_INITED, &vport->state) || + !test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) + continue; + if (time_after(jiffies, vport->last_active_jiffies + + alive_time)) { clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); - - /* If vf is not alive, set to default value */ - if (!test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) - vport->mps = HCLGE_MAC_DEFAULT_FRAME; + dev_warn(&hdev->pdev->dev, + "VF %u heartbeat timeout\n", + i - HCLGE_VF_VPORT_START_NUM); + } } } @@ -8064,9 +8079,11 @@ int hclge_vport_start(struct hclge_vport *vport) { struct hclge_dev *hdev = vport->back; + set_bit(HCLGE_VPORT_STATE_INITED, &vport->state); set_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); set_bit(HCLGE_VPORT_STATE_PROMISC_CHANGE, &vport->state); vport->last_active_jiffies = jiffies; + vport->need_notify = 0; if (test_bit(vport->vport_id, hdev->vport_config_block)) { if (vport->vport_id) { @@ -8084,7 +8101,9 @@ int hclge_vport_start(struct hclge_vport *vport) void hclge_vport_stop(struct hclge_vport *vport) { + clear_bit(HCLGE_VPORT_STATE_INITED, &vport->state); clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); + vport->need_notify = 0; } static int hclge_client_start(struct hnae3_handle *handle) @@ -9208,7 +9227,8 @@ static int hclge_set_vf_mac(struct hnae3_handle *handle, int vf, return 0; } - dev_info(&hdev->pdev->dev, "MAC of VF %d has been set to %s\n", + dev_info(&hdev->pdev->dev, + "MAC of VF %d has been set to %s, will be active after VF reset\n", vf, format_mac_addr); return 0; } @@ -10465,12 +10485,16 @@ static int hclge_set_vf_vlan_filter(struct hnae3_handle *handle, int vfid, * for DEVICE_VERSION_V3, vf doesn't need to know about the port based * VLAN state. */ - if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3 && - test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) - (void)hclge_push_vf_port_base_vlan_info(&hdev->vport[0], - vport->vport_id, - state, &vlan_info); - + if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3) { + if (test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) + (void)hclge_push_vf_port_base_vlan_info(&hdev->vport[0], + vport->vport_id, + state, + &vlan_info); + else + set_bit(HCLGE_VPORT_NEED_NOTIFY_VF_VLAN, + &vport->need_notify); + } return 0; } @@ -11941,7 +11965,7 @@ static void hclge_reset_vport_state(struct hclge_dev *hdev) int i; for (i = 0; i < hdev->num_alloc_vport; i++) { - hclge_vport_stop(vport); + clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); vport++; } } @@ -12955,6 +12979,11 @@ static void hclge_clear_vport_vf_info(struct hclge_vport *vport, int vfid) struct hclge_vlan_info vlan_info; int ret; + clear_bit(HCLGE_VPORT_STATE_INITED, &vport->state); + clear_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); + vport->need_notify = 0; + vport->mps = 0; + /* after disable sriov, clean VF rate configured by PF */ ret = hclge_tm_qs_shaper_cfg(vport, 0); if (ret) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 495b639b0dc2..13f23d606e77 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -995,9 +995,15 @@ enum HCLGE_VPORT_STATE { HCLGE_VPORT_STATE_MAC_TBL_CHANGE, HCLGE_VPORT_STATE_PROMISC_CHANGE, HCLGE_VPORT_STATE_VLAN_FLTR_CHANGE, + HCLGE_VPORT_STATE_INITED, HCLGE_VPORT_STATE_MAX }; +enum HCLGE_VPORT_NEED_NOTIFY { + HCLGE_VPORT_NEED_NOTIFY_RESET, + HCLGE_VPORT_NEED_NOTIFY_VF_VLAN, +}; + struct hclge_vlan_info { u16 vlan_proto; /* so far support 802.1Q only */ u16 qos; @@ -1044,6 +1050,7 @@ struct hclge_vport { struct hnae3_handle roce; unsigned long state; + unsigned long need_notify; unsigned long last_active_jiffies; u32 mps; /* Max packet size */ struct hclge_vf_info vf_info; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index a7b06c63143c..04ff9bf12185 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -124,17 +124,26 @@ static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len, return status; } +static int hclge_inform_vf_reset(struct hclge_vport *vport, u16 reset_type) +{ + __le16 msg_data; + u8 dest_vfid; + + dest_vfid = (u8)vport->vport_id; + msg_data = cpu_to_le16(reset_type); + + /* send this requested info to VF */ + return hclge_send_mbx_msg(vport, (u8 *)&msg_data, sizeof(msg_data), + HCLGE_MBX_ASSERTING_RESET, dest_vfid); +} + int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport) { struct hclge_dev *hdev = vport->back; - __le16 msg_data; u16 reset_type; - u8 dest_vfid; BUILD_BUG_ON(HNAE3_MAX_RESET > U16_MAX); - dest_vfid = (u8)vport->vport_id; - if (hdev->reset_type == HNAE3_FUNC_RESET) reset_type = HNAE3_VF_PF_FUNC_RESET; else if (hdev->reset_type == HNAE3_FLR_RESET) @@ -142,11 +151,7 @@ int hclge_inform_reset_assert_to_vf(struct hclge_vport *vport) else reset_type = HNAE3_VF_FUNC_RESET; - msg_data = cpu_to_le16(reset_type); - - /* send this requested info to VF */ - return hclge_send_mbx_msg(vport, (u8 *)&msg_data, sizeof(msg_data), - HCLGE_MBX_ASSERTING_RESET, dest_vfid); + return hclge_inform_vf_reset(vport, reset_type); } static void hclge_free_vector_ring_chain(struct hnae3_ring_chain_node *head) @@ -652,9 +657,56 @@ static int hclge_reset_vf(struct hclge_vport *vport) return hclge_func_reset_cmd(hdev, vport->vport_id); } +static void hclge_notify_vf_config(struct hclge_vport *vport) +{ + struct hclge_dev *hdev = vport->back; + struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); + struct hclge_port_base_vlan_config *vlan_cfg; + int ret; + + hclge_push_vf_link_status(vport); + if (test_bit(HCLGE_VPORT_NEED_NOTIFY_RESET, &vport->need_notify)) { + ret = hclge_inform_vf_reset(vport, HNAE3_VF_PF_FUNC_RESET); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to inform VF %u reset!", + vport->vport_id - HCLGE_VF_VPORT_START_NUM); + return; + } + vport->need_notify = 0; + return; + } + + if (ae_dev->dev_version < HNAE3_DEVICE_VERSION_V3 && + test_bit(HCLGE_VPORT_NEED_NOTIFY_VF_VLAN, &vport->need_notify)) { + vlan_cfg = &vport->port_base_vlan_cfg; + ret = hclge_push_vf_port_base_vlan_info(&hdev->vport[0], + vport->vport_id, + vlan_cfg->state, + &vlan_cfg->vlan_info); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to inform VF %u port base vlan!", + vport->vport_id - HCLGE_VF_VPORT_START_NUM); + return; + } + clear_bit(HCLGE_VPORT_NEED_NOTIFY_VF_VLAN, &vport->need_notify); + } +} + static void hclge_vf_keep_alive(struct hclge_vport *vport) { + struct hclge_dev *hdev = vport->back; + vport->last_active_jiffies = jiffies; + + if (test_bit(HCLGE_VPORT_STATE_INITED, &vport->state) && + !test_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state)) { + set_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); + dev_info(&hdev->pdev->dev, "VF %u is alive!", + vport->vport_id - HCLGE_VF_VPORT_START_NUM); + hclge_notify_vf_config(vport); + } } static int hclge_set_vf_mtu(struct hclge_vport *vport, @@ -954,6 +1006,7 @@ static int hclge_mbx_vf_uninit_handler(struct hclge_mbx_ops_param *param) hclge_rm_vport_all_mac_table(param->vport, true, HCLGE_MAC_ADDR_MC); hclge_rm_vport_all_vlan_table(param->vport, true); + param->vport->mps = 0; return 0; } From d530ece70f16f912e1d1bfeea694246ab78b0a4b Mon Sep 17 00:00:00 2001 From: Jiguang Xiao Date: Wed, 28 Dec 2022 16:14:47 +0800 Subject: [PATCH 076/112] net: amd-xgbe: add missed tasklet_kill The driver does not call tasklet_kill in several places. Add the calls to fix it. Fixes: 85b85c853401 ("amd-xgbe: Re-issue interrupt if interrupt status not cleared") Signed-off-by: Jiguang Xiao Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 3 +++ drivers/net/ethernet/amd/xgbe/xgbe-i2c.c | 4 +++- drivers/net/ethernet/amd/xgbe/xgbe-mdio.c | 4 +++- 3 files changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c index 7b666106feee..614c0278419b 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c @@ -1064,6 +1064,9 @@ static void xgbe_free_irqs(struct xgbe_prv_data *pdata) devm_free_irq(pdata->dev, pdata->dev_irq, pdata); + tasklet_kill(&pdata->tasklet_dev); + tasklet_kill(&pdata->tasklet_ecc); + if (pdata->vdata->ecc_support && (pdata->dev_irq != pdata->ecc_irq)) devm_free_irq(pdata->dev, pdata->ecc_irq, pdata); diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c index 22d4fc547a0a..a9ccc4258ee5 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c @@ -447,8 +447,10 @@ static void xgbe_i2c_stop(struct xgbe_prv_data *pdata) xgbe_i2c_disable(pdata); xgbe_i2c_clear_all_interrupts(pdata); - if (pdata->dev_irq != pdata->i2c_irq) + if (pdata->dev_irq != pdata->i2c_irq) { devm_free_irq(pdata->dev, pdata->i2c_irq, pdata); + tasklet_kill(&pdata->tasklet_i2c); + } } static int xgbe_i2c_start(struct xgbe_prv_data *pdata) diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c index 4e97b4869522..0c5c1b155683 100644 --- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c @@ -1390,8 +1390,10 @@ static void xgbe_phy_stop(struct xgbe_prv_data *pdata) /* Disable auto-negotiation */ xgbe_an_disable_all(pdata); - if (pdata->dev_irq != pdata->an_irq) + if (pdata->dev_irq != pdata->an_irq) { devm_free_irq(pdata->dev, pdata->an_irq, pdata); + tasklet_kill(&pdata->tasklet_an); + } pdata->phy_if.phy_impl.stop(pdata); From 1573c6882018f69991aead951d09423ce978adac Mon Sep 17 00:00:00 2001 From: Po-Hsu Lin Date: Thu, 29 Dec 2022 13:41:06 +0800 Subject: [PATCH 077/112] selftests: net: fix cmsg_so_mark.sh test hang This cmsg_so_mark.sh test will hang on non-amd64 systems because of the infinity loop for argument parsing in cmsg_sender. Variable "o" in cs_parse_args() for taking getopt() should be an int, otherwise it will be 255 when getopt() returns -1 on non-amd64 system and thus causing infinity loop. Link: https://lore.kernel.org/lkml/CA+G9fYsM2k7mrF7W4V_TrZ-qDauWM394=8yEJ=-t1oUg8_40YA@mail.gmail.com/t/ Signed-off-by: Po-Hsu Lin Signed-off-by: David S. Miller --- tools/testing/selftests/net/cmsg_sender.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/cmsg_sender.c b/tools/testing/selftests/net/cmsg_sender.c index 75dd83e39207..24b21b15ed3f 100644 --- a/tools/testing/selftests/net/cmsg_sender.c +++ b/tools/testing/selftests/net/cmsg_sender.c @@ -110,7 +110,7 @@ static void __attribute__((noreturn)) cs_usage(const char *bin) static void cs_parse_args(int argc, char *argv[]) { - char o; + int o; while ((o = getopt(argc, argv, "46sS:p:m:M:d:tf:F:c:C:l:L:H:")) != -1) { switch (o) { From 332b49ff637d6c1a75b971022a8b992cf3c57db1 Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:05 +0000 Subject: [PATCH 078/112] net: ena: Fix toeplitz initial hash value On driver initialization, RSS hash initial value is set to zero, instead of the default value. This happens because we pass NULL as the RSS key parameter, which caused us to never initialize the RSS hash value. This patch fixes it by making sure the initial value is set, no matter what the value of the RSS key is. Fixes: 91a65b7d3ed8 ("net: ena: fix potential crash when rxfh key is NULL") Signed-off-by: Nati Koler Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_com.c | 29 +++++++---------------- 1 file changed, 9 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c index 8c8b4c88c7de..451c3a1b6255 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_com.c @@ -2400,29 +2400,18 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, return -EOPNOTSUPP; } - switch (func) { - case ENA_ADMIN_TOEPLITZ: - if (key) { - if (key_len != sizeof(hash_key->key)) { - netdev_err(ena_dev->net_device, - "key len (%u) doesn't equal the supported size (%zu)\n", - key_len, sizeof(hash_key->key)); - return -EINVAL; - } - memcpy(hash_key->key, key, key_len); - rss->hash_init_val = init_val; - hash_key->key_parts = key_len / sizeof(hash_key->key[0]); + if ((func == ENA_ADMIN_TOEPLITZ) && key) { + if (key_len != sizeof(hash_key->key)) { + netdev_err(ena_dev->net_device, + "key len (%u) doesn't equal the supported size (%zu)\n", + key_len, sizeof(hash_key->key)); + return -EINVAL; } - break; - case ENA_ADMIN_CRC32: - rss->hash_init_val = init_val; - break; - default: - netdev_err(ena_dev->net_device, "Invalid hash function (%d)\n", - func); - return -EINVAL; + memcpy(hash_key->key, key, key_len); + hash_key->key_parts = key_len / sizeof(hash_key->key[0]); } + rss->hash_init_val = init_val; old_func = rss->hash_func; rss->hash_func = func; rc = ena_com_set_hash_function(ena_dev); From 9c9e539956fa67efb8a65e32b72a853740b33445 Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:06 +0000 Subject: [PATCH 079/112] net: ena: Don't register memory info on XDP exchange Since the queues aren't destroyed when we only exchange XDP programs, there's no need to re-register them again. Fixes: 548c4940b9f1 ("net: ena: Implement XDP_TX action") Signed-off-by: Shay Agroskin Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index a95529a69cbb..6ba9b06719a0 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -512,16 +512,18 @@ static void ena_xdp_exchange_program_rx_in_range(struct ena_adapter *adapter, struct bpf_prog *prog, int first, int count) { + struct bpf_prog *old_bpf_prog; struct ena_ring *rx_ring; int i = 0; for (i = first; i < count; i++) { rx_ring = &adapter->rx_ring[i]; - xchg(&rx_ring->xdp_bpf_prog, prog); - if (prog) { + old_bpf_prog = xchg(&rx_ring->xdp_bpf_prog, prog); + + if (!old_bpf_prog && prog) { ena_xdp_register_rxq_info(rx_ring); rx_ring->rx_headroom = XDP_PACKET_HEADROOM; - } else { + } else if (old_bpf_prog && !prog) { ena_xdp_unregister_rxq_info(rx_ring); rx_ring->rx_headroom = NET_SKB_PAD; } From c7f5e34d906320fdc996afa616676161c029cc02 Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:07 +0000 Subject: [PATCH 080/112] net: ena: Account for the number of processed bytes in XDP The size of packets that were forwarded or dropped by XDP wasn't added to the total processed bytes statistic. Fixes: 548c4940b9f1 ("net: ena: Implement XDP_TX action") Signed-off-by: Shay Agroskin Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 6ba9b06719a0..9ae86bd3d457 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1719,6 +1719,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, } if (xdp_verdict != XDP_PASS) { xdp_flags |= xdp_verdict; + total_len += ena_rx_ctx.ena_bufs[0].len; res_budget--; continue; } From 59811faa2c54dbcf44d575b5a8f6e7077da88dc2 Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:08 +0000 Subject: [PATCH 081/112] net: ena: Use bitmask to indicate packet redirection Redirecting packets with XDP Redirect is done in two phases: 1. A packet is passed by the driver to the kernel using xdp_do_redirect(). 2. After finishing polling for new packets the driver lets the kernel know that it can now process the redirected packet using xdp_do_flush_map(). The packets' redirection is handled in the napi context of the queue that called xdp_do_redirect() To avoid calling xdp_do_flush_map() each time the driver first checks whether any packets were redirected, using xdp_flags |= xdp_verdict; and if (xdp_flags & XDP_REDIRECT) xdp_do_flush_map() essentially treating XDP instructions as a bitmask, which isn't the case: enum xdp_action { XDP_ABORTED = 0, XDP_DROP, XDP_PASS, XDP_TX, XDP_REDIRECT, }; Given the current possible values of xdp_action, the current design doesn't have a bug (since XDP_REDIRECT = 100b), but it is still flawed. This patch makes the driver use a bitmask instead, to avoid future issues. Fixes: a318c70ad152 ("net: ena: introduce XDP redirect implementation") Signed-off-by: Shay Agroskin Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 26 ++++++++++++-------- drivers/net/ethernet/amazon/ena/ena_netdev.h | 9 +++++++ 2 files changed, 25 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 9ae86bd3d457..a67f55e5f755 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -374,9 +374,9 @@ static int ena_xdp_xmit(struct net_device *dev, int n, static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp) { + u32 verdict = ENA_XDP_PASS; struct bpf_prog *xdp_prog; struct ena_ring *xdp_ring; - u32 verdict = XDP_PASS; struct xdp_frame *xdpf; u64 *xdp_stat; @@ -393,7 +393,7 @@ static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp) if (unlikely(!xdpf)) { trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict); xdp_stat = &rx_ring->rx_stats.xdp_aborted; - verdict = XDP_ABORTED; + verdict = ENA_XDP_DROP; break; } @@ -409,29 +409,35 @@ static int ena_xdp_execute(struct ena_ring *rx_ring, struct xdp_buff *xdp) spin_unlock(&xdp_ring->xdp_tx_lock); xdp_stat = &rx_ring->rx_stats.xdp_tx; + verdict = ENA_XDP_TX; break; case XDP_REDIRECT: if (likely(!xdp_do_redirect(rx_ring->netdev, xdp, xdp_prog))) { xdp_stat = &rx_ring->rx_stats.xdp_redirect; + verdict = ENA_XDP_REDIRECT; break; } trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict); xdp_stat = &rx_ring->rx_stats.xdp_aborted; - verdict = XDP_ABORTED; + verdict = ENA_XDP_DROP; break; case XDP_ABORTED: trace_xdp_exception(rx_ring->netdev, xdp_prog, verdict); xdp_stat = &rx_ring->rx_stats.xdp_aborted; + verdict = ENA_XDP_DROP; break; case XDP_DROP: xdp_stat = &rx_ring->rx_stats.xdp_drop; + verdict = ENA_XDP_DROP; break; case XDP_PASS: xdp_stat = &rx_ring->rx_stats.xdp_pass; + verdict = ENA_XDP_PASS; break; default: bpf_warn_invalid_xdp_action(rx_ring->netdev, xdp_prog, verdict); xdp_stat = &rx_ring->rx_stats.xdp_invalid; + verdict = ENA_XDP_DROP; } ena_increase_stat(xdp_stat, 1, &rx_ring->syncp); @@ -1621,12 +1627,12 @@ static int ena_xdp_handle_buff(struct ena_ring *rx_ring, struct xdp_buff *xdp) * we expect, then we simply drop it */ if (unlikely(rx_ring->ena_bufs[0].len > ENA_XDP_MAX_MTU)) - return XDP_DROP; + return ENA_XDP_DROP; ret = ena_xdp_execute(rx_ring, xdp); /* The xdp program might expand the headers */ - if (ret == XDP_PASS) { + if (ret == ENA_XDP_PASS) { rx_info->page_offset = xdp->data - xdp->data_hard_start; rx_ring->ena_bufs[0].len = xdp->data_end - xdp->data; } @@ -1665,7 +1671,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, xdp_init_buff(&xdp, ENA_PAGE_SIZE, &rx_ring->xdp_rxq); do { - xdp_verdict = XDP_PASS; + xdp_verdict = ENA_XDP_PASS; skb = NULL; ena_rx_ctx.ena_bufs = rx_ring->ena_bufs; ena_rx_ctx.max_bufs = rx_ring->sgl_size; @@ -1693,7 +1699,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, xdp_verdict = ena_xdp_handle_buff(rx_ring, &xdp); /* allocate skb and fill it */ - if (xdp_verdict == XDP_PASS) + if (xdp_verdict == ENA_XDP_PASS) skb = ena_rx_skb(rx_ring, rx_ring->ena_bufs, ena_rx_ctx.descs, @@ -1711,13 +1717,13 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, /* Packets was passed for transmission, unmap it * from RX side. */ - if (xdp_verdict == XDP_TX || xdp_verdict == XDP_REDIRECT) { + if (xdp_verdict & ENA_XDP_FORWARDED) { ena_unmap_rx_buff(rx_ring, &rx_ring->rx_buffer_info[req_id]); rx_ring->rx_buffer_info[req_id].page = NULL; } } - if (xdp_verdict != XDP_PASS) { + if (xdp_verdict != ENA_XDP_PASS) { xdp_flags |= xdp_verdict; total_len += ena_rx_ctx.ena_bufs[0].len; res_budget--; @@ -1763,7 +1769,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, ena_refill_rx_bufs(rx_ring, refill_required); } - if (xdp_flags & XDP_REDIRECT) + if (xdp_flags & ENA_XDP_REDIRECT) xdp_do_flush_map(); return work_done; diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index 1bdce99bf688..290ae9bf47ee 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -409,6 +409,15 @@ enum ena_xdp_errors_t { ENA_XDP_NO_ENOUGH_QUEUES, }; +enum ENA_XDP_ACTIONS { + ENA_XDP_PASS = 0, + ENA_XDP_TX = BIT(0), + ENA_XDP_REDIRECT = BIT(1), + ENA_XDP_DROP = BIT(2) +}; + +#define ENA_XDP_FORWARDED (ENA_XDP_TX | ENA_XDP_REDIRECT) + static inline bool ena_xdp_present(struct ena_adapter *adapter) { return !!adapter->xdp_bpf_prog; From c7062aaee099f2f43d6f07a71744b44b94b94b34 Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:09 +0000 Subject: [PATCH 082/112] net: ena: Fix rx_copybreak value update Make the upper bound on rx_copybreak tighter, by making sure it is smaller than the minimum of mtu and ENA_PAGE_SIZE. With the current upper bound of mtu, rx_copybreak can be larger than a page. Such large rx_copybreak will not bring any performance benefit to the user and therefore makes no sense. In addition, the value update was only reflected in the adapter structure, but not applied for each ring, causing it to not take effect. Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)") Signed-off-by: Osama Abboud Signed-off-by: Arthur Kiyanovski Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_ethtool.c | 6 +----- drivers/net/ethernet/amazon/ena/ena_netdev.c | 18 ++++++++++++++++++ drivers/net/ethernet/amazon/ena/ena_netdev.h | 2 ++ 3 files changed, 21 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c index 48ae6d810f8f..8da79eedc057 100644 --- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c +++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c @@ -887,11 +887,7 @@ static int ena_set_tunable(struct net_device *netdev, switch (tuna->id) { case ETHTOOL_RX_COPYBREAK: len = *(u32 *)data; - if (len > adapter->netdev->mtu) { - ret = -EINVAL; - break; - } - adapter->rx_copybreak = len; + ret = ena_set_rx_copybreak(adapter, len); break; default: ret = -EINVAL; diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index a67f55e5f755..80a726932e81 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -2814,6 +2814,24 @@ int ena_update_queue_sizes(struct ena_adapter *adapter, return dev_was_up ? ena_up(adapter) : 0; } +int ena_set_rx_copybreak(struct ena_adapter *adapter, u32 rx_copybreak) +{ + struct ena_ring *rx_ring; + int i; + + if (rx_copybreak > min_t(u16, adapter->netdev->mtu, ENA_PAGE_SIZE)) + return -EINVAL; + + adapter->rx_copybreak = rx_copybreak; + + for (i = 0; i < adapter->num_io_queues; i++) { + rx_ring = &adapter->rx_ring[i]; + rx_ring->rx_copybreak = rx_copybreak; + } + + return 0; +} + int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count) { struct ena_com_dev *ena_dev = adapter->ena_dev; diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index 290ae9bf47ee..f9d862b630fa 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -392,6 +392,8 @@ int ena_update_queue_sizes(struct ena_adapter *adapter, int ena_update_queue_count(struct ena_adapter *adapter, u32 new_channel_count); +int ena_set_rx_copybreak(struct ena_adapter *adapter, u32 rx_copybreak); + int ena_get_sset_count(struct net_device *netdev, int sset); static inline void ena_reset_device(struct ena_adapter *adapter, From e712f3e4920b3a1a5e6b536827d118e14862896c Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:10 +0000 Subject: [PATCH 083/112] net: ena: Set default value for RX interrupt moderation RX ring can be NULL in XDP use cases where only TX queues are configured. In this scenario, the RX interrupt moderation value sent to the device remains in its default value of 0. In this change, setting the default value of the RX interrupt moderation to be the same as of the TX. Fixes: 548c4940b9f1 ("net: ena: Implement XDP_TX action") Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 80a726932e81..99f80c2d560a 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1823,8 +1823,9 @@ static void ena_adjust_adaptive_rx_intr_moderation(struct ena_napi *ena_napi) static void ena_unmask_interrupt(struct ena_ring *tx_ring, struct ena_ring *rx_ring) { + u32 rx_interval = tx_ring->smoothed_interval; struct ena_eth_io_intr_reg intr_reg; - u32 rx_interval = 0; + /* Rx ring can be NULL when for XDP tx queues which don't have an * accompanying rx_ring pair. */ From a8ee104f986e720cea52133885cc822d459398c7 Mon Sep 17 00:00:00 2001 From: David Arinzon Date: Thu, 29 Dec 2022 07:30:11 +0000 Subject: [PATCH 084/112] net: ena: Update NUMA TPH hint register upon NUMA node update The device supports a PCIe optimization hint, which indicates on which NUMA the queue is currently processed. This hint is utilized by PCIe in order to reduce its access time by accessing the correct NUMA resources and maintaining cache coherence. The driver calls the register update for the hint (called TPH - TLP Processing Hint) during the NAPI loop. Though the update is expected upon a NUMA change (when a queue is moved from one NUMA to the other), the current logic performs a register update when the queue is moved to a different CPU, but the CPU is not necessarily in a different NUMA. The changes include: 1. Performing the TPH update only when the queue has switched a NUMA node. 2. Moving the TPH update call to be triggered only when NAPI was scheduled from interrupt context, as opposed to a busy-polling loop. This is due to the fact that during busy-polling, the frequency of CPU switches for a particular queue is significantly higher, thus, the likelihood to switch NUMA is much higher. Therefore, providing the frequent updates to the device upon a NUMA update are unlikely to be beneficial. Fixes: 1738cd3ed342 ("net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)") Signed-off-by: David Arinzon Signed-off-by: David S. Miller --- drivers/net/ethernet/amazon/ena/ena_netdev.c | 27 +++++++++++++------- drivers/net/ethernet/amazon/ena/ena_netdev.h | 6 +++-- 2 files changed, 22 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 99f80c2d560a..e8ad5ea31aff 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -680,6 +680,7 @@ static void ena_init_io_rings_common(struct ena_adapter *adapter, ring->ena_dev = adapter->ena_dev; ring->per_napi_packets = 0; ring->cpu = 0; + ring->numa_node = 0; ring->no_interrupt_event_cnt = 0; u64_stats_init(&ring->syncp); } @@ -783,6 +784,7 @@ static int ena_setup_tx_resources(struct ena_adapter *adapter, int qid) tx_ring->next_to_use = 0; tx_ring->next_to_clean = 0; tx_ring->cpu = ena_irq->cpu; + tx_ring->numa_node = node; return 0; err_push_buf_intermediate_buf: @@ -915,6 +917,7 @@ static int ena_setup_rx_resources(struct ena_adapter *adapter, rx_ring->next_to_clean = 0; rx_ring->next_to_use = 0; rx_ring->cpu = ena_irq->cpu; + rx_ring->numa_node = node; return 0; } @@ -1863,20 +1866,27 @@ static void ena_update_ring_numa_node(struct ena_ring *tx_ring, if (likely(tx_ring->cpu == cpu)) goto out; + tx_ring->cpu = cpu; + if (rx_ring) + rx_ring->cpu = cpu; + numa_node = cpu_to_node(cpu); + + if (likely(tx_ring->numa_node == numa_node)) + goto out; + put_cpu(); if (numa_node != NUMA_NO_NODE) { ena_com_update_numa_node(tx_ring->ena_com_io_cq, numa_node); - if (rx_ring) + tx_ring->numa_node = numa_node; + if (rx_ring) { + rx_ring->numa_node = numa_node; ena_com_update_numa_node(rx_ring->ena_com_io_cq, numa_node); + } } - tx_ring->cpu = cpu; - if (rx_ring) - rx_ring->cpu = cpu; - return; out: put_cpu(); @@ -1997,11 +2007,10 @@ static int ena_io_poll(struct napi_struct *napi, int budget) if (ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev)) ena_adjust_adaptive_rx_intr_moderation(ena_napi); + ena_update_ring_numa_node(tx_ring, rx_ring); ena_unmask_interrupt(tx_ring, rx_ring); } - ena_update_ring_numa_node(tx_ring, rx_ring); - ret = rx_work_done; } else { ret = budget; @@ -2386,7 +2395,7 @@ static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid) ctx.mem_queue_type = ena_dev->tx_mem_queue_type; ctx.msix_vector = msix_vector; ctx.queue_size = tx_ring->ring_size; - ctx.numa_node = cpu_to_node(tx_ring->cpu); + ctx.numa_node = tx_ring->numa_node; rc = ena_com_create_io_queue(ena_dev, &ctx); if (rc) { @@ -2454,7 +2463,7 @@ static int ena_create_io_rx_queue(struct ena_adapter *adapter, int qid) ctx.mem_queue_type = ENA_ADMIN_PLACEMENT_POLICY_HOST; ctx.msix_vector = msix_vector; ctx.queue_size = rx_ring->ring_size; - ctx.numa_node = cpu_to_node(rx_ring->cpu); + ctx.numa_node = rx_ring->numa_node; rc = ena_com_create_io_queue(ena_dev, &ctx); if (rc) { diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index f9d862b630fa..2cb141079474 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -262,9 +262,11 @@ struct ena_ring { bool disable_meta_caching; u16 no_interrupt_event_cnt; - /* cpu for TPH */ + /* cpu and NUMA for TPH */ int cpu; - /* number of tx/rx_buffer_info's entries */ + int numa_node; + + /* number of tx/rx_buffer_info's entries */ int ring_size; enum ena_admin_placement_policy_type tx_mem_queue_type; From d039535850ee47079d59527e96be18d8e0daa84b Mon Sep 17 00:00:00 2001 From: Miaoqian Lin Date: Thu, 29 Dec 2022 10:29:25 +0400 Subject: [PATCH 085/112] net: phy: xgmiitorgmii: Fix refcount leak in xgmiitorgmii_probe of_phy_find_device() return device node with refcount incremented. Call put_device() to relese it when not needed anymore. Fixes: ab4e6ee578e8 ("net: phy: xgmiitorgmii: Check phy_driver ready before accessing") Signed-off-by: Miaoqian Lin Signed-off-by: David S. Miller --- drivers/net/phy/xilinx_gmii2rgmii.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/xilinx_gmii2rgmii.c b/drivers/net/phy/xilinx_gmii2rgmii.c index 8dcb49ed1f3d..7fd9fe6a602b 100644 --- a/drivers/net/phy/xilinx_gmii2rgmii.c +++ b/drivers/net/phy/xilinx_gmii2rgmii.c @@ -105,6 +105,7 @@ static int xgmiitorgmii_probe(struct mdio_device *mdiodev) if (!priv->phy_dev->drv) { dev_info(dev, "Attached phy not ready\n"); + put_device(&priv->phy_dev->mdio.dev); return -EPROBE_DEFER; } From 9807ae69746196ee4bbffe7d22d22ab2b61c6ed0 Mon Sep 17 00:00:00 2001 From: Christian Marangi Date: Thu, 29 Dec 2022 17:33:32 +0100 Subject: [PATCH 086/112] net: dsa: qca8k: fix wrong length value for mgmt eth packet The assumption that Documentation was right about how this value work was wrong. It was discovered that the length value of the mgmt header is in step of word size. As an example to process 4 byte of data the correct length to set is 2. To process 8 byte 4, 12 byte 6, 16 byte 8... Odd values will always return the next size on the ack packet. (length of 3 (6 byte) will always return 8 bytes of data) This means that a value of 15 (0xf) actually means reading/writing 32 bytes of data instead of 16 bytes. This behaviour is totally absent and not documented in the switch Documentation. In fact from Documentation the max value that mgmt eth can process is 16 byte of data while in reality it can process 32 bytes at once. To handle this we always round up the length after deviding it for word size. We check if the result is odd and we round another time to align to what the switch will provide in the ack packet. The workaround for the length limit of 15 is still needed as the length reg max value is 0xf(15) Reported-by: Ronald Wahl Tested-by: Ronald Wahl Fixes: 90386223f44e ("net: dsa: qca8k: add support for larger read/write size with mgmt Ethernet") Signed-off-by: Christian Marangi Cc: stable@vger.kernel.org # v5.18+ Signed-off-by: David S. Miller --- drivers/net/dsa/qca/qca8k-8xxx.c | 45 +++++++++++++++++++++++++------- 1 file changed, 35 insertions(+), 10 deletions(-) diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c index c5c3b4e92f28..46151320b2a8 100644 --- a/drivers/net/dsa/qca/qca8k-8xxx.c +++ b/drivers/net/dsa/qca/qca8k-8xxx.c @@ -146,7 +146,16 @@ static void qca8k_rw_reg_ack_handler(struct dsa_switch *ds, struct sk_buff *skb) command = get_unaligned_le32(&mgmt_ethhdr->command); cmd = FIELD_GET(QCA_HDR_MGMT_CMD, command); + len = FIELD_GET(QCA_HDR_MGMT_LENGTH, command); + /* Special case for len of 15 as this is the max value for len and needs to + * be increased before converting it from word to dword. + */ + if (len == 15) + len++; + + /* We can ignore odd value, we always round up them in the alloc function. */ + len *= sizeof(u16); /* Make sure the seq match the requested packet */ if (get_unaligned_le32(&mgmt_ethhdr->seq) == mgmt_eth_data->seq) @@ -193,17 +202,33 @@ static struct sk_buff *qca8k_alloc_mdio_header(enum mdio_cmd cmd, u32 reg, u32 * if (!skb) return NULL; - /* Max value for len reg is 15 (0xf) but the switch actually return 16 byte - * Actually for some reason the steps are: - * 0: nothing - * 1-4: first 4 byte - * 5-6: first 12 byte - * 7-15: all 16 byte + /* Hdr mgmt length value is in step of word size. + * As an example to process 4 byte of data the correct length to set is 2. + * To process 8 byte 4, 12 byte 6, 16 byte 8... + * + * Odd values will always return the next size on the ack packet. + * (length of 3 (6 byte) will always return 8 bytes of data) + * + * This means that a value of 15 (0xf) actually means reading/writing 32 bytes + * of data. + * + * To correctly calculate the length we devide the requested len by word and + * round up. + * On the ack function we can skip the odd check as we already handle the + * case here. */ - if (len == 16) - real_len = 15; - else - real_len = len; + real_len = DIV_ROUND_UP(len, sizeof(u16)); + + /* We check if the result len is odd and we round up another time to + * the next size. (length of 3 will be increased to 4 as switch will always + * return 8 bytes) + */ + if (real_len % sizeof(u16) != 0) + real_len++; + + /* Max reg value is 0xf(15) but switch will always return the next size (32 byte) */ + if (real_len == 16) + real_len--; skb_reset_mac_header(skb); skb_set_network_header(skb, skb->len); From d9dba91be71f03cc75bcf39fc0d5d99ff33f1ae0 Mon Sep 17 00:00:00 2001 From: Christian Marangi Date: Thu, 29 Dec 2022 17:33:33 +0100 Subject: [PATCH 087/112] net: dsa: tag_qca: fix wrong MGMT_DATA2 size It was discovered that MGMT_DATA2 can contain up to 28 bytes of data instead of the 12 bytes written in the Documentation by accounting the limit of 16 bytes declared in Documentation subtracting the first 4 byte in the packet header. Update the define with the real world value. Tested-by: Ronald Wahl Fixes: c2ee8181fddb ("net: dsa: tag_qca: add define for handling mgmt Ethernet packet") Signed-off-by: Christian Marangi Cc: stable@vger.kernel.org # v5.18+ Signed-off-by: David S. Miller --- include/linux/dsa/tag_qca.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/dsa/tag_qca.h b/include/linux/dsa/tag_qca.h index b1b5720d89a5..ee657452f122 100644 --- a/include/linux/dsa/tag_qca.h +++ b/include/linux/dsa/tag_qca.h @@ -45,8 +45,8 @@ struct sk_buff; QCA_HDR_MGMT_COMMAND_LEN + \ QCA_HDR_MGMT_DATA1_LEN) -#define QCA_HDR_MGMT_DATA2_LEN 12 /* Other 12 byte for the mdio data */ -#define QCA_HDR_MGMT_PADDING_LEN 34 /* Padding to reach the min Ethernet packet */ +#define QCA_HDR_MGMT_DATA2_LEN 28 /* Other 28 byte for the mdio data */ +#define QCA_HDR_MGMT_PADDING_LEN 18 /* Padding to reach the min Ethernet packet */ #define QCA_HDR_MGMT_PKT_LEN (QCA_HDR_MGMT_HEADER_LEN + \ QCA_HDR_LEN + \ From 03cb9e6d0b32b768e3d9d473c5c4ca1100877664 Mon Sep 17 00:00:00 2001 From: Christian Marangi Date: Thu, 29 Dec 2022 17:33:34 +0100 Subject: [PATCH 088/112] Revert "net: dsa: qca8k: cache lo and hi for mdio write" This reverts commit 2481d206fae7884cd07014fd1318e63af35e99eb. The Documentation is very confusing about the topic. The cache logic for hi and lo is wrong and actually miss some regs to be actually written. What the Documentation actually intended was that it's possible to skip writing hi OR lo if half of the reg is not needed to be written or read. Revert the change in favor of a better and correct implementation. Reported-by: Ronald Wahl Signed-off-by: Christian Marangi Cc: stable@vger.kernel.org # v5.18+ Signed-off-by: David S. Miller --- drivers/net/dsa/qca/qca8k-8xxx.c | 61 +++++++------------------------- drivers/net/dsa/qca/qca8k.h | 5 --- 2 files changed, 12 insertions(+), 54 deletions(-) diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c index 46151320b2a8..fbcd5c2b13ae 100644 --- a/drivers/net/dsa/qca/qca8k-8xxx.c +++ b/drivers/net/dsa/qca/qca8k-8xxx.c @@ -36,44 +36,6 @@ qca8k_split_addr(u32 regaddr, u16 *r1, u16 *r2, u16 *page) *page = regaddr & 0x3ff; } -static int -qca8k_set_lo(struct qca8k_priv *priv, int phy_id, u32 regnum, u16 lo) -{ - u16 *cached_lo = &priv->mdio_cache.lo; - struct mii_bus *bus = priv->bus; - int ret; - - if (lo == *cached_lo) - return 0; - - ret = bus->write(bus, phy_id, regnum, lo); - if (ret < 0) - dev_err_ratelimited(&bus->dev, - "failed to write qca8k 32bit lo register\n"); - - *cached_lo = lo; - return 0; -} - -static int -qca8k_set_hi(struct qca8k_priv *priv, int phy_id, u32 regnum, u16 hi) -{ - u16 *cached_hi = &priv->mdio_cache.hi; - struct mii_bus *bus = priv->bus; - int ret; - - if (hi == *cached_hi) - return 0; - - ret = bus->write(bus, phy_id, regnum, hi); - if (ret < 0) - dev_err_ratelimited(&bus->dev, - "failed to write qca8k 32bit hi register\n"); - - *cached_hi = hi; - return 0; -} - static int qca8k_mii_read32(struct mii_bus *bus, int phy_id, u32 regnum, u32 *val) { @@ -97,7 +59,7 @@ qca8k_mii_read32(struct mii_bus *bus, int phy_id, u32 regnum, u32 *val) } static void -qca8k_mii_write32(struct qca8k_priv *priv, int phy_id, u32 regnum, u32 val) +qca8k_mii_write32(struct mii_bus *bus, int phy_id, u32 regnum, u32 val) { u16 lo, hi; int ret; @@ -105,9 +67,12 @@ qca8k_mii_write32(struct qca8k_priv *priv, int phy_id, u32 regnum, u32 val) lo = val & 0xffff; hi = (u16)(val >> 16); - ret = qca8k_set_lo(priv, phy_id, regnum, lo); + ret = bus->write(bus, phy_id, regnum, lo); if (ret >= 0) - ret = qca8k_set_hi(priv, phy_id, regnum + 1, hi); + ret = bus->write(bus, phy_id, regnum + 1, hi); + if (ret < 0) + dev_err_ratelimited(&bus->dev, + "failed to write qca8k 32bit register\n"); } static int @@ -442,7 +407,7 @@ qca8k_regmap_write(void *ctx, uint32_t reg, uint32_t val) if (ret < 0) goto exit; - qca8k_mii_write32(priv, 0x10 | r2, r1, val); + qca8k_mii_write32(bus, 0x10 | r2, r1, val); exit: mutex_unlock(&bus->mdio_lock); @@ -475,7 +440,7 @@ qca8k_regmap_update_bits(void *ctx, uint32_t reg, uint32_t mask, uint32_t write_ val &= ~mask; val |= write_val; - qca8k_mii_write32(priv, 0x10 | r2, r1, val); + qca8k_mii_write32(bus, 0x10 | r2, r1, val); exit: mutex_unlock(&bus->mdio_lock); @@ -750,14 +715,14 @@ qca8k_mdio_write(struct qca8k_priv *priv, int phy, int regnum, u16 data) if (ret) goto exit; - qca8k_mii_write32(priv, 0x10 | r2, r1, val); + qca8k_mii_write32(bus, 0x10 | r2, r1, val); ret = qca8k_mdio_busy_wait(bus, QCA8K_MDIO_MASTER_CTRL, QCA8K_MDIO_MASTER_BUSY); exit: /* even if the busy_wait timeouts try to clear the MASTER_EN */ - qca8k_mii_write32(priv, 0x10 | r2, r1, 0); + qca8k_mii_write32(bus, 0x10 | r2, r1, 0); mutex_unlock(&bus->mdio_lock); @@ -787,7 +752,7 @@ qca8k_mdio_read(struct qca8k_priv *priv, int phy, int regnum) if (ret) goto exit; - qca8k_mii_write32(priv, 0x10 | r2, r1, val); + qca8k_mii_write32(bus, 0x10 | r2, r1, val); ret = qca8k_mdio_busy_wait(bus, QCA8K_MDIO_MASTER_CTRL, QCA8K_MDIO_MASTER_BUSY); @@ -798,7 +763,7 @@ qca8k_mdio_read(struct qca8k_priv *priv, int phy, int regnum) exit: /* even if the busy_wait timeouts try to clear the MASTER_EN */ - qca8k_mii_write32(priv, 0x10 | r2, r1, 0); + qca8k_mii_write32(bus, 0x10 | r2, r1, 0); mutex_unlock(&bus->mdio_lock); @@ -1968,8 +1933,6 @@ qca8k_sw_probe(struct mdio_device *mdiodev) } priv->mdio_cache.page = 0xffff; - priv->mdio_cache.lo = 0xffff; - priv->mdio_cache.hi = 0xffff; /* Check the detected switch id */ ret = qca8k_read_switch_id(priv); diff --git a/drivers/net/dsa/qca/qca8k.h b/drivers/net/dsa/qca/qca8k.h index 0b7a5cb12321..03514f7a20be 100644 --- a/drivers/net/dsa/qca/qca8k.h +++ b/drivers/net/dsa/qca/qca8k.h @@ -375,11 +375,6 @@ struct qca8k_mdio_cache { * mdio writes */ u16 page; -/* lo and hi can also be cached and from Documentation we can skip one - * extra mdio write if lo or hi is didn't change. - */ - u16 lo; - u16 hi; }; struct qca8k_pcs { From cfbd6de588ef659c198083205dc954a6d3ed2aec Mon Sep 17 00:00:00 2001 From: Christian Marangi Date: Thu, 29 Dec 2022 17:33:35 +0100 Subject: [PATCH 089/112] net: dsa: qca8k: introduce single mii read/write lo/hi It may be useful to read/write just the lo or hi half of a reg. This is especially useful for phy poll with the use of mdio master. The mdio master reg is composed by the first 16 bit related to setup and the other half with the returned data or data to write. Refactor the mii function to permit single mii read/write of lo or hi half of the reg. Tested-by: Ronald Wahl Signed-off-by: Christian Marangi Signed-off-by: David S. Miller --- drivers/net/dsa/qca/qca8k-8xxx.c | 110 ++++++++++++++++++++++++------- 1 file changed, 86 insertions(+), 24 deletions(-) diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c index fbcd5c2b13ae..92c4bfef7c97 100644 --- a/drivers/net/dsa/qca/qca8k-8xxx.c +++ b/drivers/net/dsa/qca/qca8k-8xxx.c @@ -37,42 +37,104 @@ qca8k_split_addr(u32 regaddr, u16 *r1, u16 *r2, u16 *page) } static int -qca8k_mii_read32(struct mii_bus *bus, int phy_id, u32 regnum, u32 *val) +qca8k_mii_write_lo(struct mii_bus *bus, int phy_id, u32 regnum, u32 val) +{ + int ret; + u16 lo; + + lo = val & 0xffff; + ret = bus->write(bus, phy_id, regnum, lo); + if (ret < 0) + dev_err_ratelimited(&bus->dev, + "failed to write qca8k 32bit lo register\n"); + + return ret; +} + +static int +qca8k_mii_write_hi(struct mii_bus *bus, int phy_id, u32 regnum, u32 val) +{ + int ret; + u16 hi; + + hi = (u16)(val >> 16); + ret = bus->write(bus, phy_id, regnum, hi); + if (ret < 0) + dev_err_ratelimited(&bus->dev, + "failed to write qca8k 32bit hi register\n"); + + return ret; +} + +static int +qca8k_mii_read_lo(struct mii_bus *bus, int phy_id, u32 regnum, u32 *val) { int ret; ret = bus->read(bus, phy_id, regnum); - if (ret >= 0) { - *val = ret; - ret = bus->read(bus, phy_id, regnum + 1); - *val |= ret << 16; - } - - if (ret < 0) { - dev_err_ratelimited(&bus->dev, - "failed to read qca8k 32bit register\n"); - *val = 0; - return ret; - } + if (ret < 0) + goto err; + *val = ret & 0xffff; return 0; + +err: + dev_err_ratelimited(&bus->dev, + "failed to read qca8k 32bit lo register\n"); + *val = 0; + + return ret; +} + +static int +qca8k_mii_read_hi(struct mii_bus *bus, int phy_id, u32 regnum, u32 *val) +{ + int ret; + + ret = bus->read(bus, phy_id, regnum); + if (ret < 0) + goto err; + + *val = ret << 16; + return 0; + +err: + dev_err_ratelimited(&bus->dev, + "failed to read qca8k 32bit hi register\n"); + *val = 0; + + return ret; +} + +static int +qca8k_mii_read32(struct mii_bus *bus, int phy_id, u32 regnum, u32 *val) +{ + u32 hi, lo; + int ret; + + *val = 0; + + ret = qca8k_mii_read_lo(bus, phy_id, regnum, &lo); + if (ret < 0) + goto err; + + ret = qca8k_mii_read_hi(bus, phy_id, regnum + 1, &hi); + if (ret < 0) + goto err; + + *val = lo | hi; + +err: + return ret; } static void qca8k_mii_write32(struct mii_bus *bus, int phy_id, u32 regnum, u32 val) { - u16 lo, hi; - int ret; + if (qca8k_mii_write_lo(bus, phy_id, regnum, val) < 0) + return; - lo = val & 0xffff; - hi = (u16)(val >> 16); - - ret = bus->write(bus, phy_id, regnum, lo); - if (ret >= 0) - ret = bus->write(bus, phy_id, regnum + 1, hi); - if (ret < 0) - dev_err_ratelimited(&bus->dev, - "failed to write qca8k 32bit register\n"); + qca8k_mii_write_hi(bus, phy_id, regnum + 1, val); } static int From a4165830ca237f2b3318faf62562bce8ce12a389 Mon Sep 17 00:00:00 2001 From: Christian Marangi Date: Thu, 29 Dec 2022 17:33:36 +0100 Subject: [PATCH 090/112] net: dsa: qca8k: improve mdio master read/write by using single lo/hi Improve mdio master read/write by using singe mii read/write lo/hi. In a read and write we need to poll the mdio master regs in a busy loop to check for a specific bit present in the upper half of the reg. We can ignore the other half since it won't contain useful data. This will save an additional useless read for each read and write operation. In a read operation the returned data is present in the mdio master reg lower half. We can ignore the other half since it won't contain useful data. This will save an additional useless read for each read operation. In a read operation it's needed to just set the hi half of the mdio master reg as the lo half will be replaced by the result. This will save an additional useless write for each read operation. Tested-by: Ronald Wahl Signed-off-by: Christian Marangi Signed-off-by: David S. Miller --- drivers/net/dsa/qca/qca8k-8xxx.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/dsa/qca/qca8k-8xxx.c b/drivers/net/dsa/qca/qca8k-8xxx.c index 92c4bfef7c97..2f224b166bbb 100644 --- a/drivers/net/dsa/qca/qca8k-8xxx.c +++ b/drivers/net/dsa/qca/qca8k-8xxx.c @@ -740,9 +740,9 @@ qca8k_mdio_busy_wait(struct mii_bus *bus, u32 reg, u32 mask) qca8k_split_addr(reg, &r1, &r2, &page); - ret = read_poll_timeout(qca8k_mii_read32, ret1, !(val & mask), 0, + ret = read_poll_timeout(qca8k_mii_read_hi, ret1, !(val & mask), 0, QCA8K_BUSY_WAIT_TIMEOUT * USEC_PER_MSEC, false, - bus, 0x10 | r2, r1, &val); + bus, 0x10 | r2, r1 + 1, &val); /* Check if qca8k_read has failed for a different reason * before returnting -ETIMEDOUT @@ -784,7 +784,7 @@ qca8k_mdio_write(struct qca8k_priv *priv, int phy, int regnum, u16 data) exit: /* even if the busy_wait timeouts try to clear the MASTER_EN */ - qca8k_mii_write32(bus, 0x10 | r2, r1, 0); + qca8k_mii_write_hi(bus, 0x10 | r2, r1 + 1, 0); mutex_unlock(&bus->mdio_lock); @@ -814,18 +814,18 @@ qca8k_mdio_read(struct qca8k_priv *priv, int phy, int regnum) if (ret) goto exit; - qca8k_mii_write32(bus, 0x10 | r2, r1, val); + qca8k_mii_write_hi(bus, 0x10 | r2, r1 + 1, val); ret = qca8k_mdio_busy_wait(bus, QCA8K_MDIO_MASTER_CTRL, QCA8K_MDIO_MASTER_BUSY); if (ret) goto exit; - ret = qca8k_mii_read32(bus, 0x10 | r2, r1, &val); + ret = qca8k_mii_read_lo(bus, 0x10 | r2, r1, &val); exit: /* even if the busy_wait timeouts try to clear the MASTER_EN */ - qca8k_mii_write32(bus, 0x10 | r2, r1, 0); + qca8k_mii_write_hi(bus, 0x10 | r2, r1 + 1, 0); mutex_unlock(&bus->mdio_lock); From 6d4cfcf97986cc67635630a2bc1f8d5c92ecdbba Mon Sep 17 00:00:00 2001 From: Sean Anderson Date: Thu, 29 Dec 2022 15:21:20 -0500 Subject: [PATCH 091/112] net: phy: Update documentation for get_rate_matching Now that phylink no longer calls phy_get_rate_matching with PHY_INTERFACE_MODE_NA, phys no longer need to support it. Remove the documentation mandating support. Fixes: 7642cc28fd37 ("net: phylink: fix PHY validation with rate adaption") Signed-off-by: Sean Anderson Signed-off-by: David S. Miller --- include/linux/phy.h | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/include/linux/phy.h b/include/linux/phy.h index 71eeb4e3b1fd..6378c997ded5 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -826,10 +826,7 @@ struct phy_driver { * whether to advertise lower-speed modes for that interface. It is * assumed that if a rate matching mode is supported on an interface, * then that interface's rate can be adapted to all slower link speeds - * supported by the phy. If iface is %PHY_INTERFACE_MODE_NA, and the phy - * supports any kind of rate matching for any interface, then it must - * return that rate matching mode (preferring %RATE_MATCH_PAUSE to - * %RATE_MATCH_CRS). If the interface is not supported, this should + * supported by the phy. If the interface is not supported, this should * return %RATE_MATCH_NONE. */ int (*get_rate_matching)(struct phy_device *phydev, From 9c4d7f45d60745a1cea0e841fa5e3444c398d2f1 Mon Sep 17 00:00:00 2001 From: Po-Hsu Lin Date: Fri, 30 Dec 2022 17:18:28 +0800 Subject: [PATCH 092/112] selftests: net: fix cleanup_v6() for arp_ndisc_evict_nocarrier The cleanup_v6() will cause the arp_ndisc_evict_nocarrier script exit with 255 (No such file or directory), even the tests are good: # selftests: net: arp_ndisc_evict_nocarrier.sh # run arp_evict_nocarrier=1 test # RTNETLINK answers: File exists # ok # run arp_evict_nocarrier=0 test # RTNETLINK answers: File exists # ok # run all.arp_evict_nocarrier=0 test # RTNETLINK answers: File exists # ok # run ndisc_evict_nocarrier=1 test # ok # run ndisc_evict_nocarrier=0 test # ok # run all.ndisc_evict_nocarrier=0 test # ok not ok 1 selftests: net: arp_ndisc_evict_nocarrier.sh # exit=255 This is because it's trying to modify the parameter for ipv4 instead. Also, tests for ipv6 (run_ndisc_evict_nocarrier_enabled() and run_ndisc_evict_nocarrier_disabled() are working on veth1, reflect this fact in cleanup_v6(). Fixes: f86ca07eb531 ("selftests: net: add arp_ndisc_evict_nocarrier") Signed-off-by: Po-Hsu Lin Signed-off-by: David S. Miller --- tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh b/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh index b5af08af8559..b4ec1eeee6c9 100755 --- a/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh +++ b/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh @@ -24,8 +24,8 @@ cleanup_v6() ip netns del me ip netns del peer - sysctl -w net.ipv4.conf.veth0.ndisc_evict_nocarrier=1 >/dev/null 2>&1 - sysctl -w net.ipv4.conf.all.ndisc_evict_nocarrier=1 >/dev/null 2>&1 + sysctl -w net.ipv6.conf.veth1.ndisc_evict_nocarrier=1 >/dev/null 2>&1 + sysctl -w net.ipv6.conf.all.ndisc_evict_nocarrier=1 >/dev/null 2>&1 } create_ns() From 1856628baa17032531916984808d1bdfd62700d4 Mon Sep 17 00:00:00 2001 From: Po-Hsu Lin Date: Fri, 30 Dec 2022 17:18:29 +0800 Subject: [PATCH 093/112] selftests: net: return non-zero for failures reported in arp_ndisc_evict_nocarrier Return non-zero return value if there is any failure reported in this script during the test. Otherwise it can only reflect the status of the last command. Fixes: f86ca07eb531 ("selftests: net: add arp_ndisc_evict_nocarrier") Signed-off-by: Po-Hsu Lin Signed-off-by: David S. Miller --- .../selftests/net/arp_ndisc_evict_nocarrier.sh | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh b/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh index b4ec1eeee6c9..4a110bb01e53 100755 --- a/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh +++ b/tools/testing/selftests/net/arp_ndisc_evict_nocarrier.sh @@ -18,6 +18,7 @@ readonly V4_ADDR1=10.0.10.2 readonly V6_ADDR0=2001:db8:91::1 readonly V6_ADDR1=2001:db8:91::2 nsid=100 +ret=0 cleanup_v6() { @@ -61,7 +62,7 @@ setup_v6() { if [ $? -ne 0 ]; then cleanup_v6 echo "failed" - exit + exit 1 fi # Set veth2 down, which will put veth1 in NOCARRIER state @@ -88,7 +89,7 @@ setup_v4() { if [ $? -ne 0 ]; then cleanup_v4 echo "failed" - exit + exit 1 fi # Set veth1 down, which will put veth0 in NOCARRIER state @@ -115,6 +116,7 @@ run_arp_evict_nocarrier_enabled() { if [ $? -eq 0 ];then echo "failed" + ret=1 else echo "ok" fi @@ -134,6 +136,7 @@ run_arp_evict_nocarrier_disabled() { echo "ok" else echo "failed" + ret=1 fi cleanup_v4 @@ -164,6 +167,7 @@ run_ndisc_evict_nocarrier_enabled() { if [ $? -eq 0 ];then echo "failed" + ret=1 else echo "ok" fi @@ -182,6 +186,7 @@ run_ndisc_evict_nocarrier_disabled() { echo "ok" else echo "failed" + ret=1 fi cleanup_v6 @@ -198,6 +203,7 @@ run_ndisc_evict_nocarrier_disabled_all() { echo "ok" else echo "failed" + ret=1 fi cleanup_v6 @@ -218,3 +224,4 @@ if [ "$(id -u)" -ne 0 ];then fi run_all_tests +exit $ret From d9d71a89f28d27ac772c429b37d52668d011df7a Mon Sep 17 00:00:00 2001 From: Alex Elder Date: Fri, 30 Dec 2022 16:33:04 -0600 Subject: [PATCH 094/112] net: ipa: use proper endpoint mask for suspend It is now possible for a system to have more than 32 endpoints. As a result, registers related to endpoint suspend are parameterized, with 32 endpoints represented in one more registers. In ipa_interrupt_suspend_control(), the IPA_SUSPEND_EN register offset is determined properly, but the bit mask used still assumes the number of enpoints won't exceed 32. This is a bug. Fix it. Fixes: f298ba785e2d ("net: ipa: add a parameter to suspend registers") Signed-off-by: Alex Elder Signed-off-by: David S. Miller --- drivers/net/ipa/ipa_interrupt.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ipa/ipa_interrupt.c b/drivers/net/ipa/ipa_interrupt.c index a49f66efacb8..d458a35839cc 100644 --- a/drivers/net/ipa/ipa_interrupt.c +++ b/drivers/net/ipa/ipa_interrupt.c @@ -132,10 +132,10 @@ static void ipa_interrupt_suspend_control(struct ipa_interrupt *interrupt, u32 endpoint_id, bool enable) { struct ipa *ipa = interrupt->ipa; + u32 mask = BIT(endpoint_id % 32); u32 unit = endpoint_id / 32; const struct ipa_reg *reg; u32 offset; - u32 mask; u32 val; WARN_ON(!test_bit(endpoint_id, ipa->available)); @@ -148,7 +148,6 @@ static void ipa_interrupt_suspend_control(struct ipa_interrupt *interrupt, offset = ipa_reg_n_offset(reg, unit); val = ioread32(ipa->reg_virt + offset); - mask = BIT(endpoint_id); if (enable) val |= mask; else From a3542b0ccd58f9fd42f34afa9daea435279a7c1c Mon Sep 17 00:00:00 2001 From: Samuel Holland Date: Sat, 31 Dec 2022 16:05:46 -0600 Subject: [PATCH 095/112] dt-bindings: net: sun8i-emac: Add phy-supply property This property has always been supported by the Linux driver; see commit 9f93ac8d4085 ("net-next: stmmac: Add dwmac-sun8i"). In fact, the original driver submission includes the phy-supply code but no mention of it in the binding, so the omission appears to be accidental. In addition, the property is documented in the binding for the previous hardware generation, allwinner,sun7i-a20-gmac. Document phy-supply in the binding to fix devicetree validation for the 25+ boards that already use this property. Fixes: 0441bde003be ("dt-bindings: net-next: Add DT bindings documentation for Allwinner dwmac-sun8i") Acked-by: Rob Herring Reviewed-by: Andre Przywara Signed-off-by: Samuel Holland Signed-off-by: David S. Miller --- .../devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml | 3 +++ 1 file changed, 3 insertions(+) diff --git a/Documentation/devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml b/Documentation/devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml index 1432fda3b603..47bc2057e629 100644 --- a/Documentation/devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml +++ b/Documentation/devicetree/bindings/net/allwinner,sun8i-a83t-emac.yaml @@ -40,6 +40,9 @@ properties: clock-names: const: stmmaceth + phy-supply: + description: PHY regulator + syscon: $ref: /schemas/types.yaml#/definitions/phandle description: From 91e2286160edd29d3fea8efff2dcda7df321878d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micha=C5=82=20Grzelak?= Date: Thu, 29 Dec 2022 15:22:19 +0100 Subject: [PATCH 096/112] dt-bindings: net: marvell,orion-mdio: Fix examples MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As stated in marvell-orion-mdio.txt deleted in commit 0781434af811f ("dt-bindings: net: orion-mdio: Convert to JSON schema") if 'interrupts' property is present, width of 'reg' should be 0x84. Otherwise, width of 'reg' should be 0x4. Fix 'examples:' and add constraints checking whether 'interrupts' property is present and validate it against fixed values in reg. Signed-off-by: Michał Grzelak Reviewed-by: Krzysztof Kozlowski Signed-off-by: David S. Miller --- .../bindings/net/marvell,orion-mdio.yaml | 30 ++++++++++++++++--- 1 file changed, 26 insertions(+), 4 deletions(-) diff --git a/Documentation/devicetree/bindings/net/marvell,orion-mdio.yaml b/Documentation/devicetree/bindings/net/marvell,orion-mdio.yaml index d2906b4a0f59..e35da8b01dc2 100644 --- a/Documentation/devicetree/bindings/net/marvell,orion-mdio.yaml +++ b/Documentation/devicetree/bindings/net/marvell,orion-mdio.yaml @@ -16,9 +16,6 @@ description: | 8k has a second unit which provides an interface with the xMDIO bus. This driver handles these interfaces. -allOf: - - $ref: "mdio.yaml#" - properties: compatible: enum: @@ -39,13 +36,38 @@ required: - compatible - reg +allOf: + - $ref: mdio.yaml# + + - if: + required: + - interrupts + + then: + properties: + reg: + items: + - items: + - $ref: /schemas/types.yaml#/definitions/cell + - const: 0x84 + + else: + properties: + reg: + items: + - items: + - $ref: /schemas/types.yaml#/definitions/cell + - enum: + - 0x4 + - 0x10 + unevaluatedProperties: false examples: - | mdio@d0072004 { compatible = "marvell,orion-mdio"; - reg = <0xd0072004 0x4>; + reg = <0xd0072004 0x84>; #address-cells = <1>; #size-cells = <0>; interrupts = <30>; From a2965c7be0522eaa18808684b7b82b248515511b Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 1 Jan 2023 16:57:43 -0500 Subject: [PATCH 097/112] net: sched: atm: dont intepret cls results when asked to drop If asked to drop a packet via TC_ACT_SHOT it is unsafe to assume res.class contains a valid pointer Fixes: b0188d4dbe5f ("[NET_SCHED]: sch_atm: Lindent") Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/sched/sch_atm.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index f52255fea652..4a981ca90b0b 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -393,10 +393,13 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch, result = tcf_classify(skb, NULL, fl, &res, true); if (result < 0) continue; + if (result == TC_ACT_SHOT) + goto done; + flow = (struct atm_flow_data *)res.class; if (!flow) flow = lookup_flow(sch, res.classid); - goto done; + goto drop; } } flow = NULL; From caa4b35b4317d5147b3ab0fbdc9c075c7d2e9c12 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 1 Jan 2023 16:57:44 -0500 Subject: [PATCH 098/112] net: sched: cbq: dont intepret cls results when asked to drop If asked to drop a packet via TC_ACT_SHOT it is unsafe to assume that res.class contains a valid pointer Sample splat reported by Kyle Zeng [ 5.405624] 0: reclassify loop, rule prio 0, protocol 800 [ 5.406326] ================================================================== [ 5.407240] BUG: KASAN: slab-out-of-bounds in cbq_enqueue+0x54b/0xea0 [ 5.407987] Read of size 1 at addr ffff88800e3122aa by task poc/299 [ 5.408731] [ 5.408897] CPU: 0 PID: 299 Comm: poc Not tainted 5.10.155+ #15 [ 5.409516] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 [ 5.410439] Call Trace: [ 5.410764] dump_stack+0x87/0xcd [ 5.411153] print_address_description+0x7a/0x6b0 [ 5.411687] ? vprintk_func+0xb9/0xc0 [ 5.411905] ? printk+0x76/0x96 [ 5.412110] ? cbq_enqueue+0x54b/0xea0 [ 5.412323] kasan_report+0x17d/0x220 [ 5.412591] ? cbq_enqueue+0x54b/0xea0 [ 5.412803] __asan_report_load1_noabort+0x10/0x20 [ 5.413119] cbq_enqueue+0x54b/0xea0 [ 5.413400] ? __kasan_check_write+0x10/0x20 [ 5.413679] __dev_queue_xmit+0x9c0/0x1db0 [ 5.413922] dev_queue_xmit+0xc/0x10 [ 5.414136] ip_finish_output2+0x8bc/0xcd0 [ 5.414436] __ip_finish_output+0x472/0x7a0 [ 5.414692] ip_finish_output+0x5c/0x190 [ 5.414940] ip_output+0x2d8/0x3c0 [ 5.415150] ? ip_mc_finish_output+0x320/0x320 [ 5.415429] __ip_queue_xmit+0x753/0x1760 [ 5.415664] ip_queue_xmit+0x47/0x60 [ 5.415874] __tcp_transmit_skb+0x1ef9/0x34c0 [ 5.416129] tcp_connect+0x1f5e/0x4cb0 [ 5.416347] tcp_v4_connect+0xc8d/0x18c0 [ 5.416577] __inet_stream_connect+0x1ae/0xb40 [ 5.416836] ? local_bh_enable+0x11/0x20 [ 5.417066] ? lock_sock_nested+0x175/0x1d0 [ 5.417309] inet_stream_connect+0x5d/0x90 [ 5.417548] ? __inet_stream_connect+0xb40/0xb40 [ 5.417817] __sys_connect+0x260/0x2b0 [ 5.418037] __x64_sys_connect+0x76/0x80 [ 5.418267] do_syscall_64+0x31/0x50 [ 5.418477] entry_SYSCALL_64_after_hwframe+0x61/0xc6 [ 5.418770] RIP: 0033:0x473bb7 [ 5.418952] Code: 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2a 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 18 89 54 24 0c 48 89 34 24 89 [ 5.420046] RSP: 002b:00007fffd20eb0f8 EFLAGS: 00000246 ORIG_RAX: 000000000000002a [ 5.420472] RAX: ffffffffffffffda RBX: 00007fffd20eb578 RCX: 0000000000473bb7 [ 5.420872] RDX: 0000000000000010 RSI: 00007fffd20eb110 RDI: 0000000000000007 [ 5.421271] RBP: 00007fffd20eb150 R08: 0000000000000001 R09: 0000000000000004 [ 5.421671] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 [ 5.422071] R13: 00007fffd20eb568 R14: 00000000004fc740 R15: 0000000000000002 [ 5.422471] [ 5.422562] Allocated by task 299: [ 5.422782] __kasan_kmalloc+0x12d/0x160 [ 5.423007] kasan_kmalloc+0x5/0x10 [ 5.423208] kmem_cache_alloc_trace+0x201/0x2e0 [ 5.423492] tcf_proto_create+0x65/0x290 [ 5.423721] tc_new_tfilter+0x137e/0x1830 [ 5.423957] rtnetlink_rcv_msg+0x730/0x9f0 [ 5.424197] netlink_rcv_skb+0x166/0x300 [ 5.424428] rtnetlink_rcv+0x11/0x20 [ 5.424639] netlink_unicast+0x673/0x860 [ 5.424870] netlink_sendmsg+0x6af/0x9f0 [ 5.425100] __sys_sendto+0x58d/0x5a0 [ 5.425315] __x64_sys_sendto+0xda/0xf0 [ 5.425539] do_syscall_64+0x31/0x50 [ 5.425764] entry_SYSCALL_64_after_hwframe+0x61/0xc6 [ 5.426065] [ 5.426157] The buggy address belongs to the object at ffff88800e312200 [ 5.426157] which belongs to the cache kmalloc-128 of size 128 [ 5.426955] The buggy address is located 42 bytes to the right of [ 5.426955] 128-byte region [ffff88800e312200, ffff88800e312280) [ 5.427688] The buggy address belongs to the page: [ 5.427992] page:000000009875fabc refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0xe312 [ 5.428562] flags: 0x100000000000200(slab) [ 5.428812] raw: 0100000000000200 dead000000000100 dead000000000122 ffff888007843680 [ 5.429325] raw: 0000000000000000 0000000000100010 00000001ffffffff ffff88800e312401 [ 5.429875] page dumped because: kasan: bad access detected [ 5.430214] page->mem_cgroup:ffff88800e312401 [ 5.430471] [ 5.430564] Memory state around the buggy address: [ 5.430846] ffff88800e312180: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 5.431267] ffff88800e312200: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fc [ 5.431705] >ffff88800e312280: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 5.432123] ^ [ 5.432391] ffff88800e312300: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fc [ 5.432810] ffff88800e312380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 5.433229] ================================================================== [ 5.433648] Disabling lock debugging due to kernel taint Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Kyle Zeng Signed-off-by: Jamal Hadi Salim Signed-off-by: David S. Miller --- net/sched/sch_cbq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c index 6568e17c4c63..36db5f6782f2 100644 --- a/net/sched/sch_cbq.c +++ b/net/sched/sch_cbq.c @@ -230,6 +230,8 @@ cbq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) result = tcf_classify(skb, NULL, fl, &res, true); if (!fl || result < 0) goto fallback; + if (result == TC_ACT_SHOT) + return NULL; cl = (void *)res.class; if (!cl) { @@ -250,8 +252,6 @@ cbq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) case TC_ACT_TRAP: *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; fallthrough; - case TC_ACT_SHOT: - return NULL; case TC_ACT_RECLASSIFY: return cbq_reclassify(skb, cl); } From 43d253781f6321c6a07a5fe4ee72103a679a5f6b Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 1 Jan 2023 23:17:37 -0800 Subject: [PATCH 099/112] net: sched: htb: fix htb_classify() kernel-doc Fix W=1 kernel-doc warning: net/sched/sch_htb.c:214: warning: expecting prototype for htb_classify(). Prototype was for HTB_DIRECT() instead by moving the HTB_DIRECT() macro above the function. Add kernel-doc notation for function parameters as well. Signed-off-by: Randy Dunlap Cc: Jamal Hadi Salim Cc: Cong Wang Cc: Jiri Pirko Cc: "David S. Miller" Cc: Eric Dumazet Cc: Jakub Kicinski Cc: Paolo Abeni Signed-off-by: David S. Miller --- net/sched/sch_htb.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c index e5b4bbf3ce3d..2238edece1a4 100644 --- a/net/sched/sch_htb.c +++ b/net/sched/sch_htb.c @@ -199,8 +199,14 @@ static unsigned long htb_search(struct Qdisc *sch, u32 handle) { return (unsigned long)htb_find(handle, sch); } + +#define HTB_DIRECT ((struct htb_class *)-1L) + /** * htb_classify - classify a packet into class + * @skb: the socket buffer + * @sch: the active queue discipline + * @qerr: pointer for returned status code * * It returns NULL if the packet should be dropped or -1 if the packet * should be passed directly thru. In all other cases leaf class is returned. @@ -211,8 +217,6 @@ static unsigned long htb_search(struct Qdisc *sch, u32 handle) * have no valid leaf we try to use MAJOR:default leaf. It still unsuccessful * then finish and return direct queue. */ -#define HTB_DIRECT ((struct htb_class *)-1L) - static struct htb_class *htb_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) { From 06bf62944144a92d83dd14fd1378d2a288259561 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Mon, 2 Jan 2023 08:55:56 +0200 Subject: [PATCH 100/112] vxlan: Fix memory leaks in error path The memory allocated by vxlan_vnigroup_init() is not freed in the error path, leading to memory leaks [1]. Fix by calling vxlan_vnigroup_uninit() in the error path. The leaks can be reproduced by annotating gro_cells_init() with ALLOW_ERROR_INJECTION() and then running: # echo "100" > /sys/kernel/debug/fail_function/probability # echo "1" > /sys/kernel/debug/fail_function/times # echo "gro_cells_init" > /sys/kernel/debug/fail_function/inject # printf %#x -12 > /sys/kernel/debug/fail_function/gro_cells_init/retval # ip link add name vxlan0 type vxlan dstport 4789 external vnifilter RTNETLINK answers: Cannot allocate memory [1] unreferenced object 0xffff88810db84a00 (size 512): comm "ip", pid 330, jiffies 4295010045 (age 66.016s) hex dump (first 32 bytes): f8 d5 76 0e 81 88 ff ff 01 00 00 00 00 00 00 02 ..v............. 03 00 04 00 48 00 00 00 00 00 00 01 04 00 01 00 ....H........... backtrace: [] kmalloc_trace+0x2a/0x60 [] vxlan_vnigroup_init+0x4c/0x160 [] vxlan_init+0x1ae/0x280 [] register_netdevice+0x57a/0x16d0 [] __vxlan_dev_create+0x7c7/0xa50 [] vxlan_newlink+0xd6/0x130 [] __rtnl_newlink+0x112b/0x18a0 [] rtnl_newlink+0x6c/0xa0 [] rtnetlink_rcv_msg+0x43f/0xd40 [] netlink_rcv_skb+0x170/0x440 [] netlink_unicast+0x53f/0x810 [] netlink_sendmsg+0x958/0xe70 [] ____sys_sendmsg+0x78f/0xa90 [] ___sys_sendmsg+0x13a/0x1e0 [] __sys_sendmsg+0x11c/0x1f0 [] do_syscall_64+0x38/0x80 unreferenced object 0xffff88810e76d5f8 (size 192): comm "ip", pid 330, jiffies 4295010045 (age 66.016s) hex dump (first 32 bytes): 04 00 00 00 00 00 00 00 db e1 4f e7 00 00 00 00 ..........O..... 08 d6 76 0e 81 88 ff ff 08 d6 76 0e 81 88 ff ff ..v.......v..... backtrace: [] __kmalloc_node+0x4e/0x90 [] kvmalloc_node+0xa6/0x1f0 [] bucket_table_alloc.isra.0+0x83/0x460 [] rhashtable_init+0x43b/0x7c0 [] vxlan_vnigroup_init+0x6c/0x160 [] vxlan_init+0x1ae/0x280 [] register_netdevice+0x57a/0x16d0 [] __vxlan_dev_create+0x7c7/0xa50 [] vxlan_newlink+0xd6/0x130 [] __rtnl_newlink+0x112b/0x18a0 [] rtnl_newlink+0x6c/0xa0 [] rtnetlink_rcv_msg+0x43f/0xd40 [] netlink_rcv_skb+0x170/0x440 [] netlink_unicast+0x53f/0x810 [] netlink_sendmsg+0x958/0xe70 [] ____sys_sendmsg+0x78f/0xa90 Fixes: f9c4bb0b245c ("vxlan: vni filtering support on collect metadata device") Signed-off-by: Ido Schimmel Reviewed-by: Nikolay Aleksandrov Signed-off-by: David S. Miller --- drivers/net/vxlan/vxlan_core.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index 92224b36787a..b1b179effe2a 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -2917,16 +2917,23 @@ static int vxlan_init(struct net_device *dev) vxlan_vnigroup_init(vxlan); dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; - - err = gro_cells_init(&vxlan->gro_cells, dev); - if (err) { - free_percpu(dev->tstats); - return err; + if (!dev->tstats) { + err = -ENOMEM; + goto err_vnigroup_uninit; } + err = gro_cells_init(&vxlan->gro_cells, dev); + if (err) + goto err_free_percpu; + return 0; + +err_free_percpu: + free_percpu(dev->tstats); +err_vnigroup_uninit: + if (vxlan->cfg.flags & VXLAN_F_VNIFILTER) + vxlan_vnigroup_uninit(vxlan); + return err; } static void vxlan_fdb_delete_default(struct vxlan_dev *vxlan, __be32 vni) From 588ab2dc25f60efeb516b4abedb6c551949cc185 Mon Sep 17 00:00:00 2001 From: Horatiu Vultur Date: Mon, 2 Jan 2023 13:12:15 +0100 Subject: [PATCH 101/112] net: sparx5: Fix reading of the MAC address There is an issue with the checking of the return value of 'of_get_mac_address', which returns 0 on success and negative value on failure. The driver interpretated the result the opposite way. Therefore if there was a MAC address defined in the DT, then the driver was generating a random MAC address otherwise it would use address 0. Fix this by checking correctly the return value of 'of_get_mac_address' Fixes: b74ef9f9cb91 ("net: sparx5: Do not use mac_addr uninitialized in mchp_sparx5_probe()") Signed-off-by: Horatiu Vultur Signed-off-by: David S. Miller --- drivers/net/ethernet/microchip/sparx5/sparx5_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_main.c b/drivers/net/ethernet/microchip/sparx5/sparx5_main.c index d25f4f09faa0..3c5d4fe99373 100644 --- a/drivers/net/ethernet/microchip/sparx5/sparx5_main.c +++ b/drivers/net/ethernet/microchip/sparx5/sparx5_main.c @@ -834,7 +834,7 @@ static int mchp_sparx5_probe(struct platform_device *pdev) if (err) goto cleanup_config; - if (!of_get_mac_address(np, sparx5->base_mac)) { + if (of_get_mac_address(np, sparx5->base_mac)) { dev_info(sparx5->dev, "MAC addr was not set, use random MAC\n"); eth_random_addr(sparx5->base_mac); sparx5->base_mac[5] = 0; From a31d47be64b9b74f8cfedffe03e0a8a1f9e51f23 Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Fri, 30 Dec 2022 13:24:37 +0100 Subject: [PATCH 102/112] netfilter: ipset: fix hash:net,port,net hang with /0 subnet MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The hash:net,port,net set type supports /0 subnets. However, the patch commit 5f7b51bf09baca8e titled "netfilter: ipset: Limit the maximal range of consecutive elements to add/delete" did not take into account it and resulted in an endless loop. The bug is actually older but the patch 5f7b51bf09baca8e brings it out earlier. Handle /0 subnets properly in hash:net,port,net set types. Fixes: 5f7b51bf09ba ("netfilter: ipset: Limit the maximal range of consecutive elements to add/delete") Reported-by: Марк Коренберг Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- net/netfilter/ipset/ip_set_hash_netportnet.c | 40 ++++++++++---------- 1 file changed, 21 insertions(+), 19 deletions(-) diff --git a/net/netfilter/ipset/ip_set_hash_netportnet.c b/net/netfilter/ipset/ip_set_hash_netportnet.c index 19bcdb3141f6..005a7ce87217 100644 --- a/net/netfilter/ipset/ip_set_hash_netportnet.c +++ b/net/netfilter/ipset/ip_set_hash_netportnet.c @@ -173,17 +173,26 @@ hash_netportnet4_kadt(struct ip_set *set, const struct sk_buff *skb, return adtfn(set, &e, &ext, &opt->ext, opt->cmdflags); } +static u32 +hash_netportnet4_range_to_cidr(u32 from, u32 to, u8 *cidr) +{ + if (from == 0 && to == UINT_MAX) { + *cidr = 0; + return to; + } + return ip_set_range_to_cidr(from, to, cidr); +} + static int hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_netportnet4 *h = set->data; + struct hash_netportnet4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netportnet4_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 ip = 0, ip_to = 0, p = 0, port, port_to; - u32 ip2_from = 0, ip2_to = 0, ip2, ipn; - u64 n = 0, m = 0; + u32 ip2_from = 0, ip2_to = 0, ip2, i = 0; bool with_ports = false; int ret; @@ -285,19 +294,6 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], } else { ip_set_mask_from_to(ip2_from, ip2_to, e.cidr[1]); } - ipn = ip; - do { - ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr[0]); - n++; - } while (ipn++ < ip_to); - ipn = ip2_from; - do { - ipn = ip_set_range_to_cidr(ipn, ip2_to, &e.cidr[1]); - m++; - } while (ipn++ < ip2_to); - - if (n*m*(port_to - port + 1) > IPSET_MAX_RANGE) - return -ERANGE; if (retried) { ip = ntohl(h->next.ip[0]); @@ -310,13 +306,19 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[], do { e.ip[0] = htonl(ip); - ip = ip_set_range_to_cidr(ip, ip_to, &e.cidr[0]); + ip = hash_netportnet4_range_to_cidr(ip, ip_to, &e.cidr[0]); for (; p <= port_to; p++) { e.port = htons(p); do { + i++; e.ip[1] = htonl(ip2); - ip2 = ip_set_range_to_cidr(ip2, ip2_to, - &e.cidr[1]); + if (i > IPSET_MAX_RANGE) { + hash_netportnet4_data_next(&h->next, + &e); + return -ERANGE; + } + ip2 = hash_netportnet4_range_to_cidr(ip2, + ip2_to, &e.cidr[1]); ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; From 5e29dc36bd5e2166b834ceb19990d9e68a734d7d Mon Sep 17 00:00:00 2001 From: Jozsef Kadlecsik Date: Fri, 30 Dec 2022 13:24:38 +0100 Subject: [PATCH 103/112] netfilter: ipset: Rework long task execution when adding/deleting entries When adding/deleting large number of elements in one step in ipset, it can take a reasonable amount of time and can result in soft lockup errors. The patch 5f7b51bf09ba ("netfilter: ipset: Limit the maximal range of consecutive elements to add/delete") tried to fix it by limiting the max elements to process at all. However it was not enough, it is still possible that we get hung tasks. Lowering the limit is not reasonable, so the approach in this patch is as follows: rely on the method used at resizing sets and save the state when we reach a smaller internal batch limit, unlock/lock and proceed from the saved state. Thus we can avoid long continuous tasks and at the same time removed the limit to add/delete large number of elements in one step. The nfnl mutex is held during the whole operation which prevents one to issue other ipset commands in parallel. Fixes: 5f7b51bf09ba ("netfilter: ipset: Limit the maximal range of consecutive elements to add/delete") Reported-by: syzbot+9204e7399656300bf271@syzkaller.appspotmail.com Signed-off-by: Jozsef Kadlecsik Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/ipset/ip_set.h | 2 +- net/netfilter/ipset/ip_set_core.c | 7 ++++--- net/netfilter/ipset/ip_set_hash_ip.c | 14 ++++++------- net/netfilter/ipset/ip_set_hash_ipmark.c | 13 ++++++------ net/netfilter/ipset/ip_set_hash_ipport.c | 13 ++++++------ net/netfilter/ipset/ip_set_hash_ipportip.c | 13 ++++++------ net/netfilter/ipset/ip_set_hash_ipportnet.c | 13 +++++++----- net/netfilter/ipset/ip_set_hash_net.c | 17 +++++++-------- net/netfilter/ipset/ip_set_hash_netiface.c | 15 ++++++-------- net/netfilter/ipset/ip_set_hash_netnet.c | 23 +++++++-------------- net/netfilter/ipset/ip_set_hash_netport.c | 19 +++++++---------- 11 files changed, 68 insertions(+), 81 deletions(-) diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h index ab934ad951a8..e8c350a3ade1 100644 --- a/include/linux/netfilter/ipset/ip_set.h +++ b/include/linux/netfilter/ipset/ip_set.h @@ -197,7 +197,7 @@ struct ip_set_region { }; /* Max range where every element is added/deleted in one step */ -#define IPSET_MAX_RANGE (1<<20) +#define IPSET_MAX_RANGE (1<<14) /* The max revision number supported by any set type + 1 */ #define IPSET_REVISION_MAX 9 diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c index e7ba5b6dd2b7..46ebee9400da 100644 --- a/net/netfilter/ipset/ip_set_core.c +++ b/net/netfilter/ipset/ip_set_core.c @@ -1698,9 +1698,10 @@ call_ad(struct net *net, struct sock *ctnl, struct sk_buff *skb, ret = set->variant->uadt(set, tb, adt, &lineno, flags, retried); ip_set_unlock(set); retried = true; - } while (ret == -EAGAIN && - set->variant->resize && - (ret = set->variant->resize(set, retried)) == 0); + } while (ret == -ERANGE || + (ret == -EAGAIN && + set->variant->resize && + (ret = set->variant->resize(set, retried)) == 0)); if (!ret || (ret == -IPSET_ERR_EXIST && eexist)) return 0; diff --git a/net/netfilter/ipset/ip_set_hash_ip.c b/net/netfilter/ipset/ip_set_hash_ip.c index e30513cefd90..c9f4e3859663 100644 --- a/net/netfilter/ipset/ip_set_hash_ip.c +++ b/net/netfilter/ipset/ip_set_hash_ip.c @@ -100,11 +100,11 @@ static int hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_ip4 *h = set->data; + struct hash_ip4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ip4_elem e = { 0 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 ip = 0, ip_to = 0, hosts; + u32 ip = 0, ip_to = 0, hosts, i = 0; int ret = 0; if (tb[IPSET_ATTR_LINENO]) @@ -149,14 +149,14 @@ hash_ip4_uadt(struct ip_set *set, struct nlattr *tb[], hosts = h->netmask == 32 ? 1 : 2 << (32 - h->netmask - 1); - /* 64bit division is not allowed on 32bit */ - if (((u64)ip_to - ip + 1) >> (32 - h->netmask) > IPSET_MAX_RANGE) - return -ERANGE; - if (retried) ip = ntohl(h->next.ip); - for (; ip <= ip_to;) { + for (; ip <= ip_to; i++) { e.ip = htonl(ip); + if (i > IPSET_MAX_RANGE) { + hash_ip4_data_next(&h->next, &e); + return -ERANGE; + } ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; diff --git a/net/netfilter/ipset/ip_set_hash_ipmark.c b/net/netfilter/ipset/ip_set_hash_ipmark.c index 153de3457423..a22ec1a6f6ec 100644 --- a/net/netfilter/ipset/ip_set_hash_ipmark.c +++ b/net/netfilter/ipset/ip_set_hash_ipmark.c @@ -97,11 +97,11 @@ static int hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_ipmark4 *h = set->data; + struct hash_ipmark4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ipmark4_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 ip, ip_to = 0; + u32 ip, ip_to = 0, i = 0; int ret; if (tb[IPSET_ATTR_LINENO]) @@ -148,13 +148,14 @@ hash_ipmark4_uadt(struct ip_set *set, struct nlattr *tb[], ip_set_mask_from_to(ip, ip_to, cidr); } - if (((u64)ip_to - ip + 1) > IPSET_MAX_RANGE) - return -ERANGE; - if (retried) ip = ntohl(h->next.ip); - for (; ip <= ip_to; ip++) { + for (; ip <= ip_to; ip++, i++) { e.ip = htonl(ip); + if (i > IPSET_MAX_RANGE) { + hash_ipmark4_data_next(&h->next, &e); + return -ERANGE; + } ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) diff --git a/net/netfilter/ipset/ip_set_hash_ipport.c b/net/netfilter/ipset/ip_set_hash_ipport.c index 2ffbd0b78a8c..e977b5a9c48d 100644 --- a/net/netfilter/ipset/ip_set_hash_ipport.c +++ b/net/netfilter/ipset/ip_set_hash_ipport.c @@ -112,11 +112,11 @@ static int hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_ipport4 *h = set->data; + struct hash_ipport4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ipport4_elem e = { .ip = 0 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 ip, ip_to = 0, p = 0, port, port_to; + u32 ip, ip_to = 0, p = 0, port, port_to, i = 0; bool with_ports = false; int ret; @@ -184,17 +184,18 @@ hash_ipport4_uadt(struct ip_set *set, struct nlattr *tb[], swap(port, port_to); } - if (((u64)ip_to - ip + 1)*(port_to - port + 1) > IPSET_MAX_RANGE) - return -ERANGE; - if (retried) ip = ntohl(h->next.ip); for (; ip <= ip_to; ip++) { p = retried && ip == ntohl(h->next.ip) ? ntohs(h->next.port) : port; - for (; p <= port_to; p++) { + for (; p <= port_to; p++, i++) { e.ip = htonl(ip); e.port = htons(p); + if (i > IPSET_MAX_RANGE) { + hash_ipport4_data_next(&h->next, &e); + return -ERANGE; + } ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) diff --git a/net/netfilter/ipset/ip_set_hash_ipportip.c b/net/netfilter/ipset/ip_set_hash_ipportip.c index 334fb1ad0e86..39a01934b153 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportip.c +++ b/net/netfilter/ipset/ip_set_hash_ipportip.c @@ -108,11 +108,11 @@ static int hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_ipportip4 *h = set->data; + struct hash_ipportip4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ipportip4_elem e = { .ip = 0 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 ip, ip_to = 0, p = 0, port, port_to; + u32 ip, ip_to = 0, p = 0, port, port_to, i = 0; bool with_ports = false; int ret; @@ -180,17 +180,18 @@ hash_ipportip4_uadt(struct ip_set *set, struct nlattr *tb[], swap(port, port_to); } - if (((u64)ip_to - ip + 1)*(port_to - port + 1) > IPSET_MAX_RANGE) - return -ERANGE; - if (retried) ip = ntohl(h->next.ip); for (; ip <= ip_to; ip++) { p = retried && ip == ntohl(h->next.ip) ? ntohs(h->next.port) : port; - for (; p <= port_to; p++) { + for (; p <= port_to; p++, i++) { e.ip = htonl(ip); e.port = htons(p); + if (i > IPSET_MAX_RANGE) { + hash_ipportip4_data_next(&h->next, &e); + return -ERANGE; + } ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) diff --git a/net/netfilter/ipset/ip_set_hash_ipportnet.c b/net/netfilter/ipset/ip_set_hash_ipportnet.c index 7df94f437f60..5c6de605a9fb 100644 --- a/net/netfilter/ipset/ip_set_hash_ipportnet.c +++ b/net/netfilter/ipset/ip_set_hash_ipportnet.c @@ -160,12 +160,12 @@ static int hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_ipportnet4 *h = set->data; + struct hash_ipportnet4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_ipportnet4_elem e = { .cidr = HOST_MASK - 1 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 ip = 0, ip_to = 0, p = 0, port, port_to; - u32 ip2_from = 0, ip2_to = 0, ip2; + u32 ip2_from = 0, ip2_to = 0, ip2, i = 0; bool with_ports = false; u8 cidr; int ret; @@ -253,9 +253,6 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], swap(port, port_to); } - if (((u64)ip_to - ip + 1)*(port_to - port + 1) > IPSET_MAX_RANGE) - return -ERANGE; - ip2_to = ip2_from; if (tb[IPSET_ATTR_IP2_TO]) { ret = ip_set_get_hostipaddr4(tb[IPSET_ATTR_IP2_TO], &ip2_to); @@ -282,9 +279,15 @@ hash_ipportnet4_uadt(struct ip_set *set, struct nlattr *tb[], for (; p <= port_to; p++) { e.port = htons(p); do { + i++; e.ip2 = htonl(ip2); ip2 = ip_set_range_to_cidr(ip2, ip2_to, &cidr); e.cidr = cidr - 1; + if (i > IPSET_MAX_RANGE) { + hash_ipportnet4_data_next(&h->next, + &e); + return -ERANGE; + } ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) diff --git a/net/netfilter/ipset/ip_set_hash_net.c b/net/netfilter/ipset/ip_set_hash_net.c index 1422739d9aa2..ce0a9ce5a91f 100644 --- a/net/netfilter/ipset/ip_set_hash_net.c +++ b/net/netfilter/ipset/ip_set_hash_net.c @@ -136,11 +136,11 @@ static int hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_net4 *h = set->data; + struct hash_net4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_net4_elem e = { .cidr = HOST_MASK }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 ip = 0, ip_to = 0, ipn, n = 0; + u32 ip = 0, ip_to = 0, i = 0; int ret; if (tb[IPSET_ATTR_LINENO]) @@ -188,19 +188,16 @@ hash_net4_uadt(struct ip_set *set, struct nlattr *tb[], if (ip + UINT_MAX == ip_to) return -IPSET_ERR_HASH_RANGE; } - ipn = ip; - do { - ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr); - n++; - } while (ipn++ < ip_to); - - if (n > IPSET_MAX_RANGE) - return -ERANGE; if (retried) ip = ntohl(h->next.ip); do { + i++; e.ip = htonl(ip); + if (i > IPSET_MAX_RANGE) { + hash_net4_data_next(&h->next, &e); + return -ERANGE; + } ip = ip_set_range_to_cidr(ip, ip_to, &e.cidr); ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) diff --git a/net/netfilter/ipset/ip_set_hash_netiface.c b/net/netfilter/ipset/ip_set_hash_netiface.c index 9810f5bf63f5..031073286236 100644 --- a/net/netfilter/ipset/ip_set_hash_netiface.c +++ b/net/netfilter/ipset/ip_set_hash_netiface.c @@ -202,7 +202,7 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[], ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netiface4_elem e = { .cidr = HOST_MASK, .elem = 1 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 ip = 0, ip_to = 0, ipn, n = 0; + u32 ip = 0, ip_to = 0, i = 0; int ret; if (tb[IPSET_ATTR_LINENO]) @@ -256,19 +256,16 @@ hash_netiface4_uadt(struct ip_set *set, struct nlattr *tb[], } else { ip_set_mask_from_to(ip, ip_to, e.cidr); } - ipn = ip; - do { - ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr); - n++; - } while (ipn++ < ip_to); - - if (n > IPSET_MAX_RANGE) - return -ERANGE; if (retried) ip = ntohl(h->next.ip); do { + i++; e.ip = htonl(ip); + if (i > IPSET_MAX_RANGE) { + hash_netiface4_data_next(&h->next, &e); + return -ERANGE; + } ip = ip_set_range_to_cidr(ip, ip_to, &e.cidr); ret = adtfn(set, &e, &ext, &ext, flags); diff --git a/net/netfilter/ipset/ip_set_hash_netnet.c b/net/netfilter/ipset/ip_set_hash_netnet.c index cdfb78c6e0d3..8fbe649c9dd3 100644 --- a/net/netfilter/ipset/ip_set_hash_netnet.c +++ b/net/netfilter/ipset/ip_set_hash_netnet.c @@ -166,13 +166,12 @@ static int hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_netnet4 *h = set->data; + struct hash_netnet4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netnet4_elem e = { }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); u32 ip = 0, ip_to = 0; - u32 ip2 = 0, ip2_from = 0, ip2_to = 0, ipn; - u64 n = 0, m = 0; + u32 ip2 = 0, ip2_from = 0, ip2_to = 0, i = 0; int ret; if (tb[IPSET_ATTR_LINENO]) @@ -248,19 +247,6 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], } else { ip_set_mask_from_to(ip2_from, ip2_to, e.cidr[1]); } - ipn = ip; - do { - ipn = ip_set_range_to_cidr(ipn, ip_to, &e.cidr[0]); - n++; - } while (ipn++ < ip_to); - ipn = ip2_from; - do { - ipn = ip_set_range_to_cidr(ipn, ip2_to, &e.cidr[1]); - m++; - } while (ipn++ < ip2_to); - - if (n*m > IPSET_MAX_RANGE) - return -ERANGE; if (retried) { ip = ntohl(h->next.ip[0]); @@ -273,7 +259,12 @@ hash_netnet4_uadt(struct ip_set *set, struct nlattr *tb[], e.ip[0] = htonl(ip); ip = ip_set_range_to_cidr(ip, ip_to, &e.cidr[0]); do { + i++; e.ip[1] = htonl(ip2); + if (i > IPSET_MAX_RANGE) { + hash_netnet4_data_next(&h->next, &e); + return -ERANGE; + } ip2 = ip_set_range_to_cidr(ip2, ip2_to, &e.cidr[1]); ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) diff --git a/net/netfilter/ipset/ip_set_hash_netport.c b/net/netfilter/ipset/ip_set_hash_netport.c index 09cf72eb37f8..d1a0628df4ef 100644 --- a/net/netfilter/ipset/ip_set_hash_netport.c +++ b/net/netfilter/ipset/ip_set_hash_netport.c @@ -154,12 +154,11 @@ static int hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], enum ipset_adt adt, u32 *lineno, u32 flags, bool retried) { - const struct hash_netport4 *h = set->data; + struct hash_netport4 *h = set->data; ipset_adtfn adtfn = set->variant->adt[adt]; struct hash_netport4_elem e = { .cidr = HOST_MASK - 1 }; struct ip_set_ext ext = IP_SET_INIT_UEXT(set); - u32 port, port_to, p = 0, ip = 0, ip_to = 0, ipn; - u64 n = 0; + u32 port, port_to, p = 0, ip = 0, ip_to = 0, i = 0; bool with_ports = false; u8 cidr; int ret; @@ -236,14 +235,6 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], } else { ip_set_mask_from_to(ip, ip_to, e.cidr + 1); } - ipn = ip; - do { - ipn = ip_set_range_to_cidr(ipn, ip_to, &cidr); - n++; - } while (ipn++ < ip_to); - - if (n*(port_to - port + 1) > IPSET_MAX_RANGE) - return -ERANGE; if (retried) { ip = ntohl(h->next.ip); @@ -255,8 +246,12 @@ hash_netport4_uadt(struct ip_set *set, struct nlattr *tb[], e.ip = htonl(ip); ip = ip_set_range_to_cidr(ip, ip_to, &cidr); e.cidr = cidr - 1; - for (; p <= port_to; p++) { + for (; p <= port_to; p++, i++) { e.port = htons(p); + if (i > IPSET_MAX_RANGE) { + hash_netport4_data_next(&h->next, &e); + return -ERANGE; + } ret = adtfn(set, &e, &ext, &ext, flags); if (ret && !ip_set_eexist(ret, flags)) return ret; From 9c807965483f42df1d053b7436eedd6cf28ece6f Mon Sep 17 00:00:00 2001 From: Daniil Tatianin Date: Mon, 2 Jan 2023 12:53:35 +0300 Subject: [PATCH 104/112] drivers/net/bonding/bond_3ad: return when there's no aggregator Otherwise we would dereference a NULL aggregator pointer when calling __set_agg_ports_ready on the line below. Found by Linux Verification Center (linuxtesting.org) with the SVACE static analysis tool. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Daniil Tatianin Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/bonding/bond_3ad.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index 455b555275f1..c99ffe6c683a 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -1549,6 +1549,7 @@ static void ad_port_selection_logic(struct port *port, bool *update_slave_arr) slave_err(bond->dev, port->slave->dev, "Port %d did not find a suitable aggregator\n", port->actor_port_number); + return; } } /* if all aggregator's ports are READY_N == TRUE, set ready=TRUE From 4af1b64f80fbe1275fb02c5f1c0cef099a4a231f Mon Sep 17 00:00:00 2001 From: Geetha sowjanya Date: Tue, 3 Jan 2023 09:20:12 +0530 Subject: [PATCH 105/112] octeontx2-pf: Fix lmtst ID used in aura free Current code uses per_cpu pointer to get the lmtst_id mapped to the core on which aura_free() is executed. Using per_cpu pointer without preemption disable causing mismatch between lmtst_id and core on which pointer gets freed. This patch fixes the issue by disabling preemption around aura_free. Fixes: ef6c8da71eaf ("octeontx2-pf: cn10K: Reserve LMTST lines per core") Signed-off-by: Sunil Goutham Signed-off-by: Geetha sowjanya Signed-off-by: David S. Miller --- .../marvell/octeontx2/nic/otx2_common.c | 30 +++++++++++++------ 1 file changed, 21 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index 9e10e7471b88..88f8772a61cd 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -1012,6 +1012,7 @@ static void otx2_pool_refill_task(struct work_struct *work) rbpool = cq->rbpool; free_ptrs = cq->pool_ptrs; + get_cpu(); while (cq->pool_ptrs) { if (otx2_alloc_rbuf(pfvf, rbpool, &bufptr)) { /* Schedule a WQ if we fails to free atleast half of the @@ -1031,6 +1032,7 @@ static void otx2_pool_refill_task(struct work_struct *work) pfvf->hw_ops->aura_freeptr(pfvf, qidx, bufptr + OTX2_HEAD_ROOM); cq->pool_ptrs--; } + put_cpu(); cq->refill_task_sched = false; } @@ -1368,6 +1370,7 @@ int otx2_sq_aura_pool_init(struct otx2_nic *pfvf) if (err) goto fail; + get_cpu(); /* Allocate pointers and free them to aura/pool */ for (qidx = 0; qidx < hw->tot_tx_queues; qidx++) { pool_id = otx2_get_pool_idx(pfvf, AURA_NIX_SQ, qidx); @@ -1376,18 +1379,24 @@ int otx2_sq_aura_pool_init(struct otx2_nic *pfvf) sq = &qset->sq[qidx]; sq->sqb_count = 0; sq->sqb_ptrs = kcalloc(num_sqbs, sizeof(*sq->sqb_ptrs), GFP_KERNEL); - if (!sq->sqb_ptrs) - return -ENOMEM; + if (!sq->sqb_ptrs) { + err = -ENOMEM; + goto err_mem; + } for (ptr = 0; ptr < num_sqbs; ptr++) { - if (otx2_alloc_rbuf(pfvf, pool, &bufptr)) - return -ENOMEM; + err = otx2_alloc_rbuf(pfvf, pool, &bufptr); + if (err) + goto err_mem; pfvf->hw_ops->aura_freeptr(pfvf, pool_id, bufptr); sq->sqb_ptrs[sq->sqb_count++] = (u64)bufptr; } } - return 0; +err_mem: + put_cpu(); + return err ? -ENOMEM : 0; + fail: otx2_mbox_reset(&pfvf->mbox.mbox, 0); otx2_aura_pool_free(pfvf); @@ -1426,18 +1435,21 @@ int otx2_rq_aura_pool_init(struct otx2_nic *pfvf) if (err) goto fail; + get_cpu(); /* Allocate pointers and free them to aura/pool */ for (pool_id = 0; pool_id < hw->rqpool_cnt; pool_id++) { pool = &pfvf->qset.pool[pool_id]; for (ptr = 0; ptr < num_ptrs; ptr++) { - if (otx2_alloc_rbuf(pfvf, pool, &bufptr)) - return -ENOMEM; + err = otx2_alloc_rbuf(pfvf, pool, &bufptr); + if (err) + goto err_mem; pfvf->hw_ops->aura_freeptr(pfvf, pool_id, bufptr + OTX2_HEAD_ROOM); } } - - return 0; +err_mem: + put_cpu(); + return err ? -ENOMEM : 0; fail: otx2_mbox_reset(&pfvf->mbox.mbox, 0); otx2_aura_pool_free(pfvf); From 7dc61838541928895abae6d2355258e02a251bba Mon Sep 17 00:00:00 2001 From: Sean Anderson Date: Tue, 3 Jan 2023 01:50:38 -0500 Subject: [PATCH 106/112] net: dpaa: Fix dtsec check for PCS availability We want to fail if the PCS is not available, not if it is available. Fix this condition. Fixes: 5d93cfcf7360 ("net: dpaa: Convert to phylink") Reported-by: Christian Zigotzky Signed-off-by: Sean Anderson Reviewed-by: Jiri Pirko Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fman/fman_dtsec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/freescale/fman/fman_dtsec.c b/drivers/net/ethernet/freescale/fman/fman_dtsec.c index d00bae15a901..d528ca681b6f 100644 --- a/drivers/net/ethernet/freescale/fman/fman_dtsec.c +++ b/drivers/net/ethernet/freescale/fman/fman_dtsec.c @@ -1430,7 +1430,7 @@ int dtsec_initialization(struct mac_device *mac_dev, dtsec->dtsec_drv_param->tx_pad_crc = true; phy_node = of_parse_phandle(mac_node, "tbi-handle", 0); - if (!phy_node || of_device_is_available(phy_node)) { + if (!phy_node || !of_device_is_available(phy_node)) { of_node_put(phy_node); err = -EINVAL; dev_err_probe(mac_dev->dev, err, From c7dd13805f8b8fc1ce3b6d40f6aff47e66b72ad2 Mon Sep 17 00:00:00 2001 From: Szymon Heidrich Date: Tue, 3 Jan 2023 10:17:09 +0100 Subject: [PATCH 107/112] usb: rndis_host: Secure rndis_query check against int overflow Variables off and len typed as uint32 in rndis_query function are controlled by incoming RNDIS response message thus their value may be manipulated. Setting off to a unexpectetly large value will cause the sum with len and 8 to overflow and pass the implemented validation step. Consequently the response pointer will be referring to a location past the expected buffer boundaries allowing information leakage e.g. via RNDIS_OID_802_3_PERMANENT_ADDRESS OID. Fixes: ddda08624013 ("USB: rndis_host, various cleanups") Signed-off-by: Szymon Heidrich Signed-off-by: David S. Miller --- drivers/net/usb/rndis_host.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c index f79333fe1783..7b3739b29c8f 100644 --- a/drivers/net/usb/rndis_host.c +++ b/drivers/net/usb/rndis_host.c @@ -255,7 +255,8 @@ static int rndis_query(struct usbnet *dev, struct usb_interface *intf, off = le32_to_cpu(u.get_c->offset); len = le32_to_cpu(u.get_c->len); - if (unlikely((8 + off + len) > CONTROL_BUFFER_SIZE)) + if (unlikely((off > CONTROL_BUFFER_SIZE - 8) || + (len > CONTROL_BUFFER_SIZE - 8 - off))) goto response_error; if (*reply_len != -1 && len != *reply_len) From 558016722e9d5bc0ac79c246ccd14a8a4eb028d4 Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat (VMware)" Date: Tue, 3 Jan 2023 14:09:41 -0800 Subject: [PATCH 108/112] MAINTAINERS: Update maintainers for ptp_vmw driver Vivek has decided to transfer the maintainership of the VMware virtual PTP clock driver (ptp_vmw) to Srivatsa and Deep. Update the MAINTAINERS file to reflect this change, and also add Alexey as a reviewer for the driver. Signed-off-by: Srivatsa S. Bhat (VMware) Acked-by: Vivek Thampi Acked-by: Deep Shah Acked-by: Alexey Makhalov Signed-off-by: David S. Miller --- MAINTAINERS | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 7f0b7181e60a..758878c0eddf 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -22243,7 +22243,9 @@ F: drivers/scsi/vmw_pvscsi.c F: drivers/scsi/vmw_pvscsi.h VMWARE VIRTUAL PTP CLOCK DRIVER -M: Vivek Thampi +M: Srivatsa S. Bhat (VMware) +M: Deep Shah +R: Alexey Makhalov R: VMware PV-Drivers Reviewers L: netdev@vger.kernel.org S: Supported From 5401c3e0992860b11fb4b25796e4c4f1921740df Mon Sep 17 00:00:00 2001 From: Caleb Sander Date: Tue, 3 Jan 2023 16:30:21 -0700 Subject: [PATCH 109/112] qed: allow sleep in qed_mcp_trace_dump() By default, qed_mcp_cmd_and_union() delays 10us at a time in a loop that can run 500K times, so calls to qed_mcp_nvm_rd_cmd() may block the current thread for over 5s. We observed thread scheduling delays over 700ms in production, with stacktraces pointing to this code as the culprit. qed_mcp_trace_dump() is called from ethtool, so sleeping is permitted. It already can sleep in qed_mcp_halt(), which calls qed_mcp_cmd(). Add a "can sleep" parameter to qed_find_nvram_image() and qed_nvram_read() so they can sleep during qed_mcp_trace_dump(). qed_mcp_trace_get_meta_info() and qed_mcp_trace_read_meta(), called only by qed_mcp_trace_dump(), allow these functions to sleep. I can't tell if the other caller (qed_grc_dump_mcp_hw_dump()) can sleep, so keep b_can_sleep set to false when it calls these functions. An example stacktrace from a custom warning we added to the kernel showing a thread that has not scheduled despite long needing resched: [ 2745.362925,17] ------------[ cut here ]------------ [ 2745.362941,17] WARNING: CPU: 23 PID: 5640 at arch/x86/kernel/irq.c:233 do_IRQ+0x15e/0x1a0() [ 2745.362946,17] Thread not rescheduled for 744 ms after irq 99 [ 2745.362956,17] Modules linked in: ... [ 2745.363339,17] CPU: 23 PID: 5640 Comm: lldpd Tainted: P O 4.4.182+ #202104120910+6d1da174272d.61x [ 2745.363343,17] Hardware name: FOXCONN MercuryB/Quicksilver Controller, BIOS H11P1N09 07/08/2020 [ 2745.363346,17] 0000000000000000 ffff885ec07c3ed8 ffffffff8131eb2f ffff885ec07c3f20 [ 2745.363358,17] ffffffff81d14f64 ffff885ec07c3f10 ffffffff81072ac2 ffff88be98ed0000 [ 2745.363369,17] 0000000000000063 0000000000000174 0000000000000074 0000000000000000 [ 2745.363379,17] Call Trace: [ 2745.363382,17] [] dump_stack+0x8e/0xcf [ 2745.363393,17] [] warn_slowpath_common+0x82/0xc0 [ 2745.363398,17] [] warn_slowpath_fmt+0x4c/0x50 [ 2745.363404,17] [] ? rcu_irq_exit+0xae/0xc0 [ 2745.363408,17] [] do_IRQ+0x15e/0x1a0 [ 2745.363413,17] [] common_interrupt+0x89/0x89 [ 2745.363416,17] [] ? delay_tsc+0x24/0x50 [ 2745.363425,17] [] __udelay+0x34/0x40 [ 2745.363457,17] [] qed_mcp_cmd_and_union+0x36f/0x7d0 [qed] [ 2745.363473,17] [] qed_mcp_nvm_rd_cmd+0x4d/0x90 [qed] [ 2745.363490,17] [] qed_mcp_trace_dump+0x4a7/0x630 [qed] [ 2745.363504,17] [] ? qed_fw_asserts_dump+0x1d6/0x1f0 [qed] [ 2745.363520,17] [] qed_dbg_mcp_trace_get_dump_buf_size+0x37/0x80 [qed] [ 2745.363536,17] [] qed_dbg_feature_size+0x61/0xa0 [qed] [ 2745.363551,17] [] qed_dbg_all_data_size+0x247/0x260 [qed] [ 2745.363560,17] [] qede_get_regs_len+0x30/0x40 [qede] [ 2745.363566,17] [] ethtool_get_drvinfo+0xe3/0x190 [ 2745.363570,17] [] dev_ethtool+0x1362/0x2140 [ 2745.363575,17] [] ? finish_task_switch+0x76/0x260 [ 2745.363580,17] [] ? __schedule+0x3c6/0x9d0 [ 2745.363585,17] [] ? hrtimer_start_range_ns+0x1d0/0x370 [ 2745.363589,17] [] ? dev_get_by_name_rcu+0x6b/0x90 [ 2745.363594,17] [] dev_ioctl+0xe8/0x710 [ 2745.363599,17] [] sock_do_ioctl+0x48/0x60 [ 2745.363603,17] [] sock_ioctl+0x1c7/0x280 [ 2745.363608,17] [] ? seccomp_phase1+0x83/0x220 [ 2745.363612,17] [] do_vfs_ioctl+0x2b3/0x4e0 [ 2745.363616,17] [] SyS_ioctl+0x41/0x70 [ 2745.363619,17] [] entry_SYSCALL_64_fastpath+0x1e/0x79 [ 2745.363622,17] ---[ end trace f6954aa440266421 ]--- Fixes: c965db4446291 ("qed: Add support for debug data collection") Signed-off-by: Caleb Sander Acked-by: Alok Prasad Link: https://lore.kernel.org/r/20230103233021.1457646-1-csander@purestorage.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/qlogic/qed/qed_debug.c | 28 +++++++++++++++------ 1 file changed, 20 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/qlogic/qed/qed_debug.c b/drivers/net/ethernet/qlogic/qed/qed_debug.c index 86ecb080b153..cdcead614e9f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_debug.c +++ b/drivers/net/ethernet/qlogic/qed/qed_debug.c @@ -1832,7 +1832,8 @@ static enum dbg_status qed_find_nvram_image(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 image_type, u32 *nvram_offset_bytes, - u32 *nvram_size_bytes) + u32 *nvram_size_bytes, + bool b_can_sleep) { u32 ret_mcp_resp, ret_mcp_param, ret_txn_size; struct mcp_file_att file_att; @@ -1846,7 +1847,8 @@ static enum dbg_status qed_find_nvram_image(struct qed_hwfn *p_hwfn, &ret_mcp_resp, &ret_mcp_param, &ret_txn_size, - (u32 *)&file_att, false); + (u32 *)&file_att, + b_can_sleep); /* Check response */ if (nvm_result || (ret_mcp_resp & FW_MSG_CODE_MASK) != @@ -1873,7 +1875,9 @@ static enum dbg_status qed_find_nvram_image(struct qed_hwfn *p_hwfn, static enum dbg_status qed_nvram_read(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 nvram_offset_bytes, - u32 nvram_size_bytes, u32 *ret_buf) + u32 nvram_size_bytes, + u32 *ret_buf, + bool b_can_sleep) { u32 ret_mcp_resp, ret_mcp_param, ret_read_size, bytes_to_copy; s32 bytes_left = nvram_size_bytes; @@ -1899,7 +1903,7 @@ static enum dbg_status qed_nvram_read(struct qed_hwfn *p_hwfn, &ret_mcp_resp, &ret_mcp_param, &ret_read_size, (u32 *)((u8 *)ret_buf + read_offset), - false)) + b_can_sleep)) return DBG_STATUS_NVRAM_READ_FAILED; /* Check response */ @@ -3380,7 +3384,8 @@ static u32 qed_grc_dump_mcp_hw_dump(struct qed_hwfn *p_hwfn, p_ptt, NVM_TYPE_HW_DUMP_OUT, &hw_dump_offset_bytes, - &hw_dump_size_bytes); + &hw_dump_size_bytes, + false); if (status != DBG_STATUS_OK) return 0; @@ -3397,7 +3402,9 @@ static u32 qed_grc_dump_mcp_hw_dump(struct qed_hwfn *p_hwfn, status = qed_nvram_read(p_hwfn, p_ptt, hw_dump_offset_bytes, - hw_dump_size_bytes, dump_buf + offset); + hw_dump_size_bytes, + dump_buf + offset, + false); if (status != DBG_STATUS_OK) { DP_NOTICE(p_hwfn, "Failed to read MCP HW Dump image from NVRAM\n"); @@ -4123,7 +4130,9 @@ static enum dbg_status qed_mcp_trace_get_meta_info(struct qed_hwfn *p_hwfn, return qed_find_nvram_image(p_hwfn, p_ptt, nvram_image_type, - trace_meta_offset, trace_meta_size); + trace_meta_offset, + trace_meta_size, + true); } /* Reads the MCP Trace meta data from NVRAM into the specified buffer */ @@ -4139,7 +4148,10 @@ static enum dbg_status qed_mcp_trace_read_meta(struct qed_hwfn *p_hwfn, /* Read meta data from NVRAM */ status = qed_nvram_read(p_hwfn, p_ptt, - nvram_offset_in_bytes, size_in_bytes, buf); + nvram_offset_in_bytes, + size_in_bytes, + buf, + true); if (status != DBG_STATUS_OK) return status; From 2c02d41d71f90a5168391b6a5f2954112ba2307c Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 3 Jan 2023 12:19:17 +0100 Subject: [PATCH 110/112] net/ulp: prevent ULP without clone op from entering the LISTEN status When an ULP-enabled socket enters the LISTEN status, the listener ULP data pointer is copied inside the child/accepted sockets by sk_clone_lock(). The relevant ULP can take care of de-duplicating the context pointer via the clone() operation, but only MPTCP and SMC implement such op. Other ULPs may end-up with a double-free at socket disposal time. We can't simply clear the ULP data at clone time, as TLS replaces the socket ops with custom ones assuming a valid TLS ULP context is available. Instead completely prevent clone-less ULP sockets from entering the LISTEN status. Fixes: 734942cc4ea6 ("tcp: ULP infrastructure") Reported-by: slipper Signed-off-by: Paolo Abeni Link: https://lore.kernel.org/r/4b80c3d1dbe3d0ab072f80450c202d9bc88b4b03.1672740602.git.pabeni@redhat.com Signed-off-by: Jakub Kicinski --- net/ipv4/inet_connection_sock.c | 14 ++++++++++++++ net/ipv4/tcp_ulp.c | 4 ++++ 2 files changed, 18 insertions(+) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 848ffc3e0239..d1f837579398 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1200,12 +1200,26 @@ void inet_csk_prepare_forced_close(struct sock *sk) } EXPORT_SYMBOL(inet_csk_prepare_forced_close); +static int inet_ulp_can_listen(const struct sock *sk) +{ + const struct inet_connection_sock *icsk = inet_csk(sk); + + if (icsk->icsk_ulp_ops && !icsk->icsk_ulp_ops->clone) + return -EINVAL; + + return 0; +} + int inet_csk_listen_start(struct sock *sk) { struct inet_connection_sock *icsk = inet_csk(sk); struct inet_sock *inet = inet_sk(sk); int err; + err = inet_ulp_can_listen(sk); + if (unlikely(err)) + return err; + reqsk_queue_alloc(&icsk->icsk_accept_queue); sk->sk_ack_backlog = 0; diff --git a/net/ipv4/tcp_ulp.c b/net/ipv4/tcp_ulp.c index 9ae50b1bd844..05b6077b9f2c 100644 --- a/net/ipv4/tcp_ulp.c +++ b/net/ipv4/tcp_ulp.c @@ -139,6 +139,10 @@ static int __tcp_set_ulp(struct sock *sk, const struct tcp_ulp_ops *ulp_ops) if (sk->sk_socket) clear_bit(SOCK_SUPPORT_ZC, &sk->sk_socket->flags); + err = -EINVAL; + if (!ulp_ops->clone && sk->sk_state == TCP_LISTEN) + goto out_err; + err = ulp_ops->init(sk); if (err) goto out_err; From 1ac88557447088ccd15eb2f2520ce46d463c8e0b Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 3 Jan 2023 19:27:36 +0000 Subject: [PATCH 111/112] inet: control sockets should not use current thread task_frag Because ICMP handlers run from softirq contexts, they must not use current thread task_frag. Previously, all sockets allocated by inet_ctl_sock_create() would use the per-socket page fragment, with no chance of recursion. Fixes: 98123866fcf3 ("Treewide: Stop corrupting socket's task_frag") Reported-by: syzbot+bebc6f1acdf4cbb79b03@syzkaller.appspotmail.com Signed-off-by: Eric Dumazet Cc: Benjamin Coddington Acked-by: Guillaume Nault Link: https://lore.kernel.org/r/20230103192736.454149-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/af_inet.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index ab4a06be489b..6c0ec2789943 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1665,6 +1665,7 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, if (rc == 0) { *sk = sock->sk; (*sk)->sk_allocation = GFP_ATOMIC; + (*sk)->sk_use_task_frag = false; /* * Unhash it so that IP input processing does not even see it, * we do not wish this socket to see incoming packets. From fe69230f05897b3de758427b574fc98025dfc907 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 4 Jan 2023 14:51:46 +0800 Subject: [PATCH 112/112] caif: fix memory leak in cfctrl_linkup_request() When linktype is unknown or kzalloc failed in cfctrl_linkup_request(), pkt is not released. Add release process to error path. Fixes: b482cd2053e3 ("net-caif: add CAIF core protocol stack") Fixes: 8d545c8f958f ("caif: Disconnect without waiting for response") Signed-off-by: Zhengchao Shao Reviewed-by: Jiri Pirko Link: https://lore.kernel.org/r/20230104065146.1153009-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- net/caif/cfctrl.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/caif/cfctrl.c b/net/caif/cfctrl.c index cc405d8c7c30..8480684f2762 100644 --- a/net/caif/cfctrl.c +++ b/net/caif/cfctrl.c @@ -269,11 +269,15 @@ int cfctrl_linkup_request(struct cflayer *layer, default: pr_warn("Request setup of bad link type = %d\n", param->linktype); + cfpkt_destroy(pkt); return -EINVAL; } req = kzalloc(sizeof(*req), GFP_KERNEL); - if (!req) + if (!req) { + cfpkt_destroy(pkt); return -ENOMEM; + } + req->client_layer = user_layer; req->cmd = CFCTRL_CMD_LINK_SETUP; req->param = *param;