From afa4ceb0fb648655c9f04921ccc801feb034109c Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 29 Oct 2024 02:00:29 -0700 Subject: [PATCH 01/16] net: netconsole: selftests: Add userdata validation Extend netcons_basic selftest to verify the userdata functionality by: 1. Creating a test key in the userdata configfs directory 2. Writing a known value to the key 3. Validating the key-value pair appears in the captured network output This ensures the userdata feature is properly tested during selftests. Signed-off-by: Breno Leitao Link: https://patch.msgid.link/20241029090030.1793551-3-leitao@debian.org Signed-off-by: Jakub Kicinski --- .../selftests/drivers/net/netcons_basic.sh | 29 +++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/tools/testing/selftests/drivers/net/netcons_basic.sh b/tools/testing/selftests/drivers/net/netcons_basic.sh index 8d28e5189e91..182eb1a97e59 100755 --- a/tools/testing/selftests/drivers/net/netcons_basic.sh +++ b/tools/testing/selftests/drivers/net/netcons_basic.sh @@ -26,10 +26,13 @@ DSTIP=192.0.2.2 PORT="6666" MSG="netconsole selftest" +USERDATA_KEY="key" +USERDATA_VALUE="value" TARGET=$(mktemp -u netcons_XXXXX) DEFAULT_PRINTK_VALUES=$(cat /proc/sys/kernel/printk) NETCONS_CONFIGFS="/sys/kernel/config/netconsole" NETCONS_PATH="${NETCONS_CONFIGFS}"/"${TARGET}" +KEY_PATH="${NETCONS_PATH}/userdata/${USERDATA_KEY}" # NAMESPACE will be populated by setup_ns with a random value NAMESPACE="" @@ -122,6 +125,8 @@ function cleanup() { # delete netconsole dynamic reconfiguration echo 0 > "${NETCONS_PATH}"/enabled + # Remove key + rmdir "${KEY_PATH}" # Remove the configfs entry rmdir "${NETCONS_PATH}" @@ -136,6 +141,18 @@ function cleanup() { echo "${DEFAULT_PRINTK_VALUES}" > /proc/sys/kernel/printk } +function set_user_data() { + if [[ ! -d "${NETCONS_PATH}""/userdata" ]] + then + echo "Userdata path not available in ${NETCONS_PATH}/userdata" + exit "${ksft_skip}" + fi + + mkdir -p "${KEY_PATH}" + VALUE_PATH="${KEY_PATH}""/value" + echo "${USERDATA_VALUE}" > "${VALUE_PATH}" +} + function listen_port_and_save_to() { local OUTPUT=${1} # Just wait for 2 seconds @@ -146,6 +163,10 @@ function listen_port_and_save_to() { function validate_result() { local TMPFILENAME="$1" + # TMPFILENAME will contain something like: + # 6.11.1-0_fbk0_rc13_509_g30d75cea12f7,13,1822,115075213798,-;netconsole selftest: netcons_gtJHM + # key=value + # Check if the file exists if [ ! -f "$TMPFILENAME" ]; then echo "FAIL: File was not generated." >&2 @@ -158,6 +179,12 @@ function validate_result() { exit "${ksft_fail}" fi + if ! grep -q "${USERDATA_KEY}=${USERDATA_VALUE}" "${TMPFILENAME}"; then + echo "FAIL: ${USERDATA_KEY}=${USERDATA_VALUE} not found in ${TMPFILENAME}" >&2 + cat "${TMPFILENAME}" >&2 + exit "${ksft_fail}" + fi + # Delete the file once it is validated, otherwise keep it # for debugging purposes rm "${TMPFILENAME}" @@ -220,6 +247,8 @@ trap cleanup EXIT set_network # Create a dynamic target for netconsole create_dynamic_target +# Set userdata "key" with the "value" value +set_user_data # Listed for netconsole port inside the namespace and destination interface listen_port_and_save_to "${OUTPUT_FILE}" & # Wait for socat to start and listen to the port. -- 2.51.0 From 6b2d11e2d8fc130df4708be0b6b53fd3e6b54cf6 Mon Sep 17 00:00:00 2001 From: Dmitry Safonov <0x7f454c46@gmail.com> Date: Wed, 30 Oct 2024 04:22:33 +0000 Subject: [PATCH 02/16] net/tcp: Add missing lockdep annotations for TCP-AO hlist traversals Under CONFIG_PROVE_RCU_LIST + CONFIG_RCU_EXPERT hlist_for_each_entry_rcu() provides very helpful splats, which help to find possible issues. I missed CONFIG_RCU_EXPERT=y in my testing config the same as described in a3e4bf7f9675 ("configs/debug: make sure PROVE_RCU_LIST=y takes effect"). The fix itself is trivial: add the very same lockdep annotations as were used to dereference ao_info from the socket. Reported-by: Jakub Kicinski Closes: https://lore.kernel.org/netdev/20241028152645.35a8be66@kernel.org/ Signed-off-by: Dmitry Safonov <0x7f454c46@gmail.com> Link: https://patch.msgid.link/20241030-tcp-ao-hlist-lockdep-annotate-v1-1-bf641a64d7c6@gmail.com Signed-off-by: Jakub Kicinski --- include/net/tcp_ao.h | 3 ++- net/ipv4/tcp_ao.c | 42 +++++++++++++++++++++++------------------- net/ipv4/tcp_ipv4.c | 3 ++- net/ipv6/tcp_ipv6.c | 4 ++-- 4 files changed, 29 insertions(+), 23 deletions(-) diff --git a/include/net/tcp_ao.h b/include/net/tcp_ao.h index 1d46460d0fef..df655ce6987d 100644 --- a/include/net/tcp_ao.h +++ b/include/net/tcp_ao.h @@ -183,7 +183,8 @@ int tcp_ao_hash_skb(unsigned short int family, const u8 *tkey, int hash_offset, u32 sne); int tcp_parse_ao(struct sock *sk, int cmd, unsigned short int family, sockptr_t optval, int optlen); -struct tcp_ao_key *tcp_ao_established_key(struct tcp_ao_info *ao, +struct tcp_ao_key *tcp_ao_established_key(const struct sock *sk, + struct tcp_ao_info *ao, int sndid, int rcvid); int tcp_ao_copy_all_matching(const struct sock *sk, struct sock *newsk, struct request_sock *req, struct sk_buff *skb, diff --git a/net/ipv4/tcp_ao.c b/net/ipv4/tcp_ao.c index db6516092daf..bbb8d5f0eae7 100644 --- a/net/ipv4/tcp_ao.c +++ b/net/ipv4/tcp_ao.c @@ -109,12 +109,13 @@ bool tcp_ao_ignore_icmp(const struct sock *sk, int family, int type, int code) * it's known that the keys in ao_info are matching peer's * family/address/VRF/etc. */ -struct tcp_ao_key *tcp_ao_established_key(struct tcp_ao_info *ao, +struct tcp_ao_key *tcp_ao_established_key(const struct sock *sk, + struct tcp_ao_info *ao, int sndid, int rcvid) { struct tcp_ao_key *key; - hlist_for_each_entry_rcu(key, &ao->head, node) { + hlist_for_each_entry_rcu(key, &ao->head, node, lockdep_sock_is_held(sk)) { if ((sndid >= 0 && key->sndid != sndid) || (rcvid >= 0 && key->rcvid != rcvid)) continue; @@ -205,7 +206,7 @@ static struct tcp_ao_key *__tcp_ao_do_lookup(const struct sock *sk, int l3index, if (!ao) return NULL; - hlist_for_each_entry_rcu(key, &ao->head, node) { + hlist_for_each_entry_rcu(key, &ao->head, node, lockdep_sock_is_held(sk)) { u8 prefixlen = min(prefix, key->prefixlen); if (!tcp_ao_key_cmp(key, l3index, addr, prefixlen, @@ -793,7 +794,7 @@ int tcp_ao_prepare_reset(const struct sock *sk, struct sk_buff *skb, if (!ao_info) return -ENOENT; - *key = tcp_ao_established_key(ao_info, aoh->rnext_keyid, -1); + *key = tcp_ao_established_key(sk, ao_info, aoh->rnext_keyid, -1); if (!*key) return -ENOENT; *traffic_key = snd_other_key(*key); @@ -979,7 +980,7 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb, */ key = READ_ONCE(info->rnext_key); if (key->rcvid != aoh->keyid) { - key = tcp_ao_established_key(info, -1, aoh->keyid); + key = tcp_ao_established_key(sk, info, -1, aoh->keyid); if (!key) goto key_not_found; } @@ -1003,7 +1004,7 @@ tcp_inbound_ao_hash(struct sock *sk, const struct sk_buff *skb, aoh->rnext_keyid, tcp_ao_hdr_maclen(aoh)); /* If the key is not found we do nothing. */ - key = tcp_ao_established_key(info, aoh->rnext_keyid, -1); + key = tcp_ao_established_key(sk, info, aoh->rnext_keyid, -1); if (key) /* pairs with tcp_ao_del_cmd */ WRITE_ONCE(info->current_key, key); @@ -1163,7 +1164,7 @@ void tcp_ao_established(struct sock *sk) if (!ao) return; - hlist_for_each_entry_rcu(key, &ao->head, node) + hlist_for_each_entry_rcu(key, &ao->head, node, lockdep_sock_is_held(sk)) tcp_ao_cache_traffic_keys(sk, ao, key); } @@ -1180,7 +1181,7 @@ void tcp_ao_finish_connect(struct sock *sk, struct sk_buff *skb) WRITE_ONCE(ao->risn, tcp_hdr(skb)->seq); ao->rcv_sne = 0; - hlist_for_each_entry_rcu(key, &ao->head, node) + hlist_for_each_entry_rcu(key, &ao->head, node, lockdep_sock_is_held(sk)) tcp_ao_cache_traffic_keys(sk, ao, key); } @@ -1256,14 +1257,14 @@ int tcp_ao_copy_all_matching(const struct sock *sk, struct sock *newsk, key_head = rcu_dereference(hlist_first_rcu(&new_ao->head)); first_key = hlist_entry_safe(key_head, struct tcp_ao_key, node); - key = tcp_ao_established_key(new_ao, tcp_rsk(req)->ao_keyid, -1); + key = tcp_ao_established_key(req_to_sk(req), new_ao, tcp_rsk(req)->ao_keyid, -1); if (key) new_ao->current_key = key; else new_ao->current_key = first_key; /* set rnext_key */ - key = tcp_ao_established_key(new_ao, -1, tcp_rsk(req)->ao_rcv_next); + key = tcp_ao_established_key(req_to_sk(req), new_ao, -1, tcp_rsk(req)->ao_rcv_next); if (key) new_ao->rnext_key = key; else @@ -1857,12 +1858,12 @@ static int tcp_ao_del_cmd(struct sock *sk, unsigned short int family, * if there's any. */ if (cmd.set_current) { - new_current = tcp_ao_established_key(ao_info, cmd.current_key, -1); + new_current = tcp_ao_established_key(sk, ao_info, cmd.current_key, -1); if (!new_current) return -ENOENT; } if (cmd.set_rnext) { - new_rnext = tcp_ao_established_key(ao_info, -1, cmd.rnext); + new_rnext = tcp_ao_established_key(sk, ao_info, -1, cmd.rnext); if (!new_rnext) return -ENOENT; } @@ -1902,7 +1903,8 @@ static int tcp_ao_del_cmd(struct sock *sk, unsigned short int family, * "It is presumed that an MKT affecting a particular * connection cannot be destroyed during an active connection" */ - hlist_for_each_entry_rcu(key, &ao_info->head, node) { + hlist_for_each_entry_rcu(key, &ao_info->head, node, + lockdep_sock_is_held(sk)) { if (cmd.sndid != key->sndid || cmd.rcvid != key->rcvid) continue; @@ -2000,14 +2002,14 @@ static int tcp_ao_info_cmd(struct sock *sk, unsigned short int family, * if there's any. */ if (cmd.set_current) { - new_current = tcp_ao_established_key(ao_info, cmd.current_key, -1); + new_current = tcp_ao_established_key(sk, ao_info, cmd.current_key, -1); if (!new_current) { err = -ENOENT; goto out; } } if (cmd.set_rnext) { - new_rnext = tcp_ao_established_key(ao_info, -1, cmd.rnext); + new_rnext = tcp_ao_established_key(sk, ao_info, -1, cmd.rnext); if (!new_rnext) { err = -ENOENT; goto out; @@ -2101,7 +2103,8 @@ int tcp_v4_parse_ao(struct sock *sk, int cmd, sockptr_t optval, int optlen) * The layout of the fields in the user and kernel structures is expected to * be the same (including in the 32bit vs 64bit case). */ -static int tcp_ao_copy_mkts_to_user(struct tcp_ao_info *ao_info, +static int tcp_ao_copy_mkts_to_user(const struct sock *sk, + struct tcp_ao_info *ao_info, sockptr_t optval, sockptr_t optlen) { struct tcp_ao_getsockopt opt_in, opt_out; @@ -2229,7 +2232,8 @@ static int tcp_ao_copy_mkts_to_user(struct tcp_ao_info *ao_info, /* May change in RX, while we're dumping, pre-fetch it */ current_key = READ_ONCE(ao_info->current_key); - hlist_for_each_entry_rcu(key, &ao_info->head, node) { + hlist_for_each_entry_rcu(key, &ao_info->head, node, + lockdep_sock_is_held(sk)) { if (opt_in.get_all) goto match; @@ -2309,7 +2313,7 @@ int tcp_ao_get_mkts(struct sock *sk, sockptr_t optval, sockptr_t optlen) if (!ao_info) return -ENOENT; - return tcp_ao_copy_mkts_to_user(ao_info, optval, optlen); + return tcp_ao_copy_mkts_to_user(sk, ao_info, optval, optlen); } int tcp_ao_get_sock_info(struct sock *sk, sockptr_t optval, sockptr_t optlen) @@ -2396,7 +2400,7 @@ int tcp_ao_set_repair(struct sock *sk, sockptr_t optval, unsigned int optlen) WRITE_ONCE(ao->snd_sne, cmd.snd_sne); WRITE_ONCE(ao->rcv_sne, cmd.rcv_sne); - hlist_for_each_entry_rcu(key, &ao->head, node) + hlist_for_each_entry_rcu(key, &ao->head, node, lockdep_sock_is_held(sk)) tcp_ao_cache_traffic_keys(sk, ao, key); return 0; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 9d3dd101ea71..a38c8b1f44db 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1053,7 +1053,8 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb) } if (aoh) - key.ao_key = tcp_ao_established_key(ao_info, aoh->rnext_keyid, -1); + key.ao_key = tcp_ao_established_key(sk, ao_info, + aoh->rnext_keyid, -1); } } if (key.ao_key) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 597920061a3a..c748eeae1453 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1172,8 +1172,8 @@ static void tcp_v6_timewait_ack(struct sock *sk, struct sk_buff *skb) if (tcp_parse_auth_options(tcp_hdr(skb), NULL, &aoh)) goto out; if (aoh) - key.ao_key = tcp_ao_established_key(ao_info, - aoh->rnext_keyid, -1); + key.ao_key = tcp_ao_established_key(sk, ao_info, + aoh->rnext_keyid, -1); } } if (key.ao_key) { -- 2.51.0 From a865276872ec4f129f8a582634be82dcc275dc2a Mon Sep 17 00:00:00 2001 From: Caleb Sander Mateos Date: Wed, 30 Oct 2024 18:23:25 -0600 Subject: [PATCH 03/16] dim: make dim_calc_stats() inputs const pointers Make the start and end arguments to dim_calc_stats() const pointers to clarify that the function does not modify their values. Signed-off-by: Caleb Sander Mateos Reviewed-by: Vladimir Oltean Reviewed-by: Florian Fainelli Reviewed-by: Arthur Kiyanovski Link: https://patch.msgid.link/20241031002326.3426181-1-csander@purestorage.com Signed-off-by: Jakub Kicinski --- include/linux/dim.h | 3 ++- lib/dim/dim.c | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/include/linux/dim.h b/include/linux/dim.h index 1b581ff25a15..84579a50ae7f 100644 --- a/include/linux/dim.h +++ b/include/linux/dim.h @@ -351,7 +351,8 @@ void dim_park_tired(struct dim *dim); * Takes into consideration counter wrap-around. * Returned boolean indicates whether curr_stats are reliable. */ -bool dim_calc_stats(struct dim_sample *start, struct dim_sample *end, +bool dim_calc_stats(const struct dim_sample *start, + const struct dim_sample *end, struct dim_stats *curr_stats); /** diff --git a/lib/dim/dim.c b/lib/dim/dim.c index 83b65ac74d73..97c3d084ebf0 100644 --- a/lib/dim/dim.c +++ b/lib/dim/dim.c @@ -54,7 +54,8 @@ void dim_park_tired(struct dim *dim) } EXPORT_SYMBOL(dim_park_tired); -bool dim_calc_stats(struct dim_sample *start, struct dim_sample *end, +bool dim_calc_stats(const struct dim_sample *start, + const struct dim_sample *end, struct dim_stats *curr_stats) { /* u32 holds up to 71 minutes, should be enough */ -- 2.51.0 From 61bf0009a7657d394d942c8ee961b9ea5f2168fe Mon Sep 17 00:00:00 2001 From: Caleb Sander Mateos Date: Wed, 30 Oct 2024 18:23:26 -0600 Subject: [PATCH 04/16] dim: pass dim_sample to net_dim() by reference net_dim() is currently passed a struct dim_sample argument by value. struct dim_sample is 24 bytes. Since this is greater 16 bytes, x86-64 passes it on the stack. All callers have already initialized dim_sample on the stack, so passing it by value requires pushing a duplicated copy to the stack. Either witing to the stack and immediately reading it, or perhaps dereferencing addresses relative to the stack pointer in a chain of push instructions, seems to perform quite poorly. In a heavy TCP workload, mlx5e_handle_rx_dim() consumes 3% of CPU time, 94% of which is attributed to the first push instruction to copy dim_sample on the stack for the call to net_dim(): // Call ktime_get() 0.26 |4ead2: call 4ead7 // Pass the address of struct dim in %rdi |4ead7: lea 0x3d0(%rbx),%rdi // Set dim_sample.pkt_ctr |4eade: mov %r13d,0x8(%rsp) // Set dim_sample.byte_ctr |4eae3: mov %r12d,0xc(%rsp) // Set dim_sample.event_ctr 0.15 |4eae8: mov %bp,0x10(%rsp) // Duplicate dim_sample on the stack 94.16 |4eaed: push 0x10(%rsp) 2.79 |4eaf1: push 0x10(%rsp) 0.07 |4eaf5: push %rax // Call net_dim() 0.21 |4eaf6: call 4eafb To allow the caller to reuse the struct dim_sample already on the stack, pass the struct dim_sample by reference to net_dim(). Signed-off-by: Caleb Sander Mateos Reviewed-by: Vladimir Oltean Reviewed-by: Shannon Nelson Reviewed-by: Florian Fainelli Reviewed-by: Arthur Kiyanovski Reviewed-by: Louis Peens Link: https://patch.msgid.link/20241031002326.3426181-2-csander@purestorage.com Signed-off-by: Jakub Kicinski --- Documentation/networking/net_dim.rst | 2 +- drivers/net/ethernet/amazon/ena/ena_netdev.c | 2 +- drivers/net/ethernet/broadcom/bcmsysport.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 ++-- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 2 +- drivers/net/ethernet/freescale/enetc/enetc.c | 2 +- drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 4 ++-- drivers/net/ethernet/intel/ice/ice_txrx.c | 4 ++-- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 4 ++-- drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 2 +- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c | 4 ++-- drivers/net/ethernet/netronome/nfp/nfd3/dp.c | 4 ++-- drivers/net/ethernet/netronome/nfp/nfdk/dp.c | 4 ++-- drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 2 +- drivers/net/virtio_net.c | 2 +- drivers/soc/fsl/dpio/dpio-service.c | 2 +- include/linux/dim.h | 2 +- lib/dim/net_dim.c | 10 +++++----- 19 files changed, 31 insertions(+), 31 deletions(-) diff --git a/Documentation/networking/net_dim.rst b/Documentation/networking/net_dim.rst index 8908fd7b0a8d..4377998e6826 100644 --- a/Documentation/networking/net_dim.rst +++ b/Documentation/networking/net_dim.rst @@ -156,7 +156,7 @@ usage is not complete but it should make the outline of the usage clear. my_entity->bytes, &dim_sample); /* Call net DIM */ - net_dim(&my_entity->dim, dim_sample); + net_dim(&my_entity->dim, &dim_sample); ... } diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 96df20854eb9..63c8a2328142 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -1383,7 +1383,7 @@ static void ena_adjust_adaptive_rx_intr_moderation(struct ena_napi *ena_napi) rx_ring->rx_stats.bytes, &dim_sample); - net_dim(&ena_napi->dim, dim_sample); + net_dim(&ena_napi->dim, &dim_sample); rx_ring->per_napi_packets = 0; } diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c index caff6e87a488..031e9e0cca53 100644 --- a/drivers/net/ethernet/broadcom/bcmsysport.c +++ b/drivers/net/ethernet/broadcom/bcmsysport.c @@ -1029,7 +1029,7 @@ static int bcm_sysport_poll(struct napi_struct *napi, int budget) if (priv->dim.use_dim) { dim_update_sample(priv->dim.event_ctr, priv->dim.packets, priv->dim.bytes, &dim_sample); - net_dim(&priv->dim.dim, dim_sample); + net_dim(&priv->dim.dim, &dim_sample); } return work_done; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 6dd6541d8619..ca42b81133d7 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -3102,7 +3102,7 @@ static int bnxt_poll(struct napi_struct *napi, int budget) cpr->rx_packets, cpr->rx_bytes, &dim_sample); - net_dim(&cpr->dim, dim_sample); + net_dim(&cpr->dim, &dim_sample); } return work_done; } @@ -3233,7 +3233,7 @@ poll_done: cpr_rx->rx_packets, cpr_rx->rx_bytes, &dim_sample); - net_dim(&cpr->dim, dim_sample); + net_dim(&cpr->dim, &dim_sample); } return work_done; } diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 10966ab15373..53a949eb9180 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2405,7 +2405,7 @@ static int bcmgenet_rx_poll(struct napi_struct *napi, int budget) if (ring->dim.use_dim) { dim_update_sample(ring->dim.event_ctr, ring->dim.packets, ring->dim.bytes, &dim_sample); - net_dim(&ring->dim.dim, dim_sample); + net_dim(&ring->dim.dim, &dim_sample); } return work_done; diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index c09370eab319..05dedea6185a 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -718,7 +718,7 @@ static void enetc_rx_net_dim(struct enetc_int_vector *v) v->rx_ring.stats.packets, v->rx_ring.stats.bytes, &dim_sample); - net_dim(&v->rx_dim, dim_sample); + net_dim(&v->rx_dim, &dim_sample); } static int enetc_bd_ready_count(struct enetc_bdr *tx_ring, int ci) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index b09f0cca34dc..fbfd3ee5648f 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -4478,7 +4478,7 @@ static void hns3_update_rx_int_coalesce(struct hns3_enet_tqp_vector *tqp_vector) dim_update_sample(tqp_vector->event_cnt, rx_group->total_packets, rx_group->total_bytes, &sample); - net_dim(&rx_group->dim, sample); + net_dim(&rx_group->dim, &sample); } static void hns3_update_tx_int_coalesce(struct hns3_enet_tqp_vector *tqp_vector) @@ -4491,7 +4491,7 @@ static void hns3_update_tx_int_coalesce(struct hns3_enet_tqp_vector *tqp_vector) dim_update_sample(tqp_vector->event_cnt, tx_group->total_packets, tx_group->total_bytes, &sample); - net_dim(&tx_group->dim, sample); + net_dim(&tx_group->dim, &sample); } static int hns3_nic_common_poll(struct napi_struct *napi, int budget) diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index 8208055d6e7f..5d2d7736fd5f 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -1352,14 +1352,14 @@ static void ice_net_dim(struct ice_q_vector *q_vector) struct dim_sample dim_sample; __ice_update_sample(q_vector, tx, &dim_sample, true); - net_dim(&tx->dim, dim_sample); + net_dim(&tx->dim, &dim_sample); } if (ITR_IS_DYNAMIC(rx)) { struct dim_sample dim_sample; __ice_update_sample(q_vector, rx, &dim_sample, false); - net_dim(&rx->dim, dim_sample); + net_dim(&rx->dim, &dim_sample); } } diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index d4e6f0e10487..da2a5becf62f 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -3679,7 +3679,7 @@ static void idpf_net_dim(struct idpf_q_vector *q_vector) idpf_update_dim_sample(q_vector, &dim_sample, &q_vector->tx_dim, packets, bytes); - net_dim(&q_vector->tx_dim, dim_sample); + net_dim(&q_vector->tx_dim, &dim_sample); check_rx_itr: if (!IDPF_ITR_IS_DYNAMIC(q_vector->rx_intr_mode)) @@ -3698,7 +3698,7 @@ check_rx_itr: idpf_update_dim_sample(q_vector, &dim_sample, &q_vector->rx_dim, packets, bytes); - net_dim(&q_vector->rx_dim, dim_sample); + net_dim(&q_vector->rx_dim, &dim_sample); } /** diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c index 933e18ba2fb2..7aaf32e9aa95 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c @@ -527,7 +527,7 @@ static void otx2_adjust_adaptive_coalese(struct otx2_nic *pfvf, struct otx2_cq_p rx_frames + tx_frames, rx_bytes + tx_bytes, &dim_sample); - net_dim(&cq_poll->dim, dim_sample); + net_dim(&cq_poll->dim, &dim_sample); } int otx2_napi_handler(struct napi_struct *napi, int budget) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index f01ceee5f02d..53485142938c 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -2227,7 +2227,7 @@ rx_done: eth->rx_bytes += bytes; dim_update_sample(eth->rx_events, eth->rx_packets, eth->rx_bytes, &dim_sample); - net_dim(ð->rx_dim, dim_sample); + net_dim(ð->rx_dim, &dim_sample); if (xdp_flush) xdp_do_flush(); @@ -2377,7 +2377,7 @@ static int mtk_poll_tx(struct mtk_eth *eth, int budget) dim_update_sample(eth->tx_events, eth->tx_packets, eth->tx_bytes, &dim_sample); - net_dim(ð->tx_dim, dim_sample); + net_dim(ð->tx_dim, &dim_sample); if (mtk_queue_stopped(eth) && (atomic_read(&ring->free_count) > ring->thresh)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index 5873fde65c2e..417098f0b2bb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -55,7 +55,7 @@ static void mlx5e_handle_tx_dim(struct mlx5e_txqsq *sq) return; dim_update_sample(sq->cq.event_ctr, stats->packets, stats->bytes, &dim_sample); - net_dim(sq->dim, dim_sample); + net_dim(sq->dim, &dim_sample); } static void mlx5e_handle_rx_dim(struct mlx5e_rq *rq) @@ -67,7 +67,7 @@ static void mlx5e_handle_rx_dim(struct mlx5e_rq *rq) return; dim_update_sample(rq->cq.event_ctr, stats->packets, stats->bytes, &dim_sample); - net_dim(rq->dim, dim_sample); + net_dim(rq->dim, &dim_sample); } void mlx5e_trigger_irq(struct mlx5e_icosq *sq) diff --git a/drivers/net/ethernet/netronome/nfp/nfd3/dp.c b/drivers/net/ethernet/netronome/nfp/nfd3/dp.c index d215efc6cad0..f1c6c47564b1 100644 --- a/drivers/net/ethernet/netronome/nfp/nfd3/dp.c +++ b/drivers/net/ethernet/netronome/nfp/nfd3/dp.c @@ -1179,7 +1179,7 @@ int nfp_nfd3_poll(struct napi_struct *napi, int budget) } while (u64_stats_fetch_retry(&r_vec->rx_sync, start)); dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample); - net_dim(&r_vec->rx_dim, dim_sample); + net_dim(&r_vec->rx_dim, &dim_sample); } if (r_vec->nfp_net->tx_coalesce_adapt_on && r_vec->tx_ring) { @@ -1194,7 +1194,7 @@ int nfp_nfd3_poll(struct napi_struct *napi, int budget) } while (u64_stats_fetch_retry(&r_vec->tx_sync, start)); dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample); - net_dim(&r_vec->tx_dim, dim_sample); + net_dim(&r_vec->tx_dim, &dim_sample); } return pkts_polled; diff --git a/drivers/net/ethernet/netronome/nfp/nfdk/dp.c b/drivers/net/ethernet/netronome/nfp/nfdk/dp.c index dae5af7d1845..ebeb6ab4465c 100644 --- a/drivers/net/ethernet/netronome/nfp/nfdk/dp.c +++ b/drivers/net/ethernet/netronome/nfp/nfdk/dp.c @@ -1289,7 +1289,7 @@ int nfp_nfdk_poll(struct napi_struct *napi, int budget) } while (u64_stats_fetch_retry(&r_vec->rx_sync, start)); dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample); - net_dim(&r_vec->rx_dim, dim_sample); + net_dim(&r_vec->rx_dim, &dim_sample); } if (r_vec->nfp_net->tx_coalesce_adapt_on && r_vec->tx_ring) { @@ -1304,7 +1304,7 @@ int nfp_nfdk_poll(struct napi_struct *napi, int budget) } while (u64_stats_fetch_retry(&r_vec->tx_sync, start)); dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample); - net_dim(&r_vec->tx_dim, dim_sample); + net_dim(&r_vec->tx_dim, &dim_sample); } return pkts_polled; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index 0eeda7e502db..2ac59564ded1 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -928,7 +928,7 @@ static void ionic_dim_update(struct ionic_qcq *qcq, int napi_mode) dim_update_sample(qcq->cq.bound_intr->rearm_count, pkts, bytes, &dim_sample); - net_dim(&qcq->dim, dim_sample); + net_dim(&qcq->dim, &dim_sample); } int ionic_tx_napi(struct napi_struct *napi, int budget) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 792e9eadbfc3..869586c17ffd 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -2804,7 +2804,7 @@ static void virtnet_rx_dim_update(struct virtnet_info *vi, struct receive_queue u64_stats_read(&rq->stats.bytes), &cur_sample); - net_dim(&rq->dim, cur_sample); + net_dim(&rq->dim, &cur_sample); rq->packets_in_napi = 0; } diff --git a/drivers/soc/fsl/dpio/dpio-service.c b/drivers/soc/fsl/dpio/dpio-service.c index b811446e0fa5..0b60ed16297c 100644 --- a/drivers/soc/fsl/dpio/dpio-service.c +++ b/drivers/soc/fsl/dpio/dpio-service.c @@ -891,7 +891,7 @@ void dpaa2_io_update_net_dim(struct dpaa2_io *d, __u64 frames, __u64 bytes) d->frames += frames; dim_update_sample(d->event_ctr, d->frames, d->bytes, &dim_sample); - net_dim(&d->rx_dim, dim_sample); + net_dim(&d->rx_dim, &dim_sample); spin_unlock(&d->dim_lock); } diff --git a/include/linux/dim.h b/include/linux/dim.h index 84579a50ae7f..06543fd40fcc 100644 --- a/include/linux/dim.h +++ b/include/linux/dim.h @@ -425,7 +425,7 @@ struct dim_cq_moder net_dim_get_def_tx_moderation(u8 cq_period_mode); * This is the main logic of the algorithm, where data is processed in order * to decide on next required action. */ -void net_dim(struct dim *dim, struct dim_sample end_sample); +void net_dim(struct dim *dim, const struct dim_sample *end_sample); /* RDMA DIM */ diff --git a/lib/dim/net_dim.c b/lib/dim/net_dim.c index d7e7028e9b19..d6aa09a979b3 100644 --- a/lib/dim/net_dim.c +++ b/lib/dim/net_dim.c @@ -347,7 +347,7 @@ static bool net_dim_decision(struct dim_stats *curr_stats, struct dim *dim) return dim->profile_ix != prev_ix; } -void net_dim(struct dim *dim, struct dim_sample end_sample) +void net_dim(struct dim *dim, const struct dim_sample *end_sample) { struct dim_stats curr_stats; u16 nevents; @@ -355,11 +355,11 @@ void net_dim(struct dim *dim, struct dim_sample end_sample) switch (dim->state) { case DIM_MEASURE_IN_PROGRESS: nevents = BIT_GAP(BITS_PER_TYPE(u16), - end_sample.event_ctr, + end_sample->event_ctr, dim->start_sample.event_ctr); if (nevents < DIM_NEVENTS) break; - if (!dim_calc_stats(&dim->start_sample, &end_sample, &curr_stats)) + if (!dim_calc_stats(&dim->start_sample, end_sample, &curr_stats)) break; if (net_dim_decision(&curr_stats, dim)) { dim->state = DIM_APPLY_NEW_PROFILE; @@ -368,8 +368,8 @@ void net_dim(struct dim *dim, struct dim_sample end_sample) } fallthrough; case DIM_START_MEASURE: - dim_update_sample(end_sample.event_ctr, end_sample.pkt_ctr, - end_sample.byte_ctr, &dim->start_sample); + dim_update_sample(end_sample->event_ctr, end_sample->pkt_ctr, + end_sample->byte_ctr, &dim->start_sample); dim->state = DIM_MEASURE_IN_PROGRESS; break; case DIM_APPLY_NEW_PROFILE: -- 2.51.0 From 2e570cd187e3b5c8e56627523e0c12e2ffc4745f Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Thu, 31 Oct 2024 15:28:18 +0100 Subject: [PATCH 05/16] net: dsa: mt7530: Add TBF qdisc offload support MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Introduce port_setup_tc callback in mt7530 dsa driver in order to enable dsa ports rate shaping via hw Token Bucket Filter (TBF) for hw switched traffic. Tested-by: Arınç ÜNAL Signed-off-by: Lorenzo Bianconi Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20241031-mt7530-tc-offload-v2-1-cb242ad954a0@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/dsa/mt7530.c | 49 ++++++++++++++++++++++++++++++++++++++++ drivers/net/dsa/mt7530.h | 12 ++++++++++ 2 files changed, 61 insertions(+) diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c index d84ee1b419a6..086b8b3d5b40 100644 --- a/drivers/net/dsa/mt7530.c +++ b/drivers/net/dsa/mt7530.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "mt7530.h" @@ -3146,6 +3147,53 @@ mt753x_conduit_state_change(struct dsa_switch *ds, mt7530_rmw(priv, MT753X_MFC, MT7530_CPU_EN | MT7530_CPU_PORT_MASK, val); } +static int mt753x_tc_setup_qdisc_tbf(struct dsa_switch *ds, int port, + struct tc_tbf_qopt_offload *qopt) +{ + struct tc_tbf_qopt_offload_replace_params *p = &qopt->replace_params; + struct mt7530_priv *priv = ds->priv; + u32 rate = 0; + + switch (qopt->command) { + case TC_TBF_REPLACE: + rate = div_u64(p->rate.rate_bytes_ps, 1000) << 3; /* kbps */ + fallthrough; + case TC_TBF_DESTROY: { + u32 val, tick; + + mt7530_rmw(priv, MT753X_GERLCR, EGR_BC_MASK, + EGR_BC_CRC_IPG_PREAMBLE); + + /* if rate is greater than 10Mbps tick is 1/32 ms, + * 1ms otherwise + */ + tick = rate > 10000 ? 2 : 7; + val = FIELD_PREP(ERLCR_CIR_MASK, (rate >> 5)) | + FIELD_PREP(ERLCR_EN_MASK, !!rate) | + FIELD_PREP(ERLCR_EXP_MASK, tick) | + ERLCR_TBF_MODE_MASK | + FIELD_PREP(ERLCR_MANT_MASK, 0xf); + mt7530_write(priv, MT753X_ERLCR_P(port), val); + break; + } + default: + return -EOPNOTSUPP; + } + + return 0; +} + +static int mt753x_setup_tc(struct dsa_switch *ds, int port, + enum tc_setup_type type, void *type_data) +{ + switch (type) { + case TC_SETUP_QDISC_TBF: + return mt753x_tc_setup_qdisc_tbf(ds, port, type_data); + default: + return -EOPNOTSUPP; + } +} + static int mt7988_setup(struct dsa_switch *ds) { struct mt7530_priv *priv = ds->priv; @@ -3193,6 +3241,7 @@ const struct dsa_switch_ops mt7530_switch_ops = { .get_mac_eee = mt753x_get_mac_eee, .set_mac_eee = mt753x_set_mac_eee, .conduit_state_change = mt753x_conduit_state_change, + .port_setup_tc = mt753x_setup_tc, }; EXPORT_SYMBOL_GPL(mt7530_switch_ops); diff --git a/drivers/net/dsa/mt7530.h b/drivers/net/dsa/mt7530.h index 6ad33a9f6b1d..448200689f49 100644 --- a/drivers/net/dsa/mt7530.h +++ b/drivers/net/dsa/mt7530.h @@ -248,6 +248,18 @@ enum mt7530_vlan_egress_attr { #define AGE_UNIT_MAX 0xfff #define AGE_UNIT(x) (AGE_UNIT_MASK & (x)) +#define MT753X_ERLCR_P(x) (0x1040 + ((x) * 0x100)) +#define ERLCR_CIR_MASK GENMASK(31, 16) +#define ERLCR_EN_MASK BIT(15) +#define ERLCR_EXP_MASK GENMASK(11, 8) +#define ERLCR_TBF_MODE_MASK BIT(7) +#define ERLCR_MANT_MASK GENMASK(6, 0) + +#define MT753X_GERLCR 0x10e0 +#define EGR_BC_MASK GENMASK(7, 0) +#define EGR_BC_CRC 0x4 /* crc */ +#define EGR_BC_CRC_IPG_PREAMBLE 0x18 /* crc + ipg + preamble */ + /* Register for port STP state control */ #define MT7530_SSP_P(x) (0x2000 + ((x) * 0x100)) #define FID_PST(fid, state) (((state) & 0x3) << ((fid) * 2)) -- 2.51.0 From 5c87206cdb537f67c51f3f9a229258dce77d9a23 Mon Sep 17 00:00:00 2001 From: Colin Ian King Date: Thu, 31 Oct 2024 13:50:42 +0000 Subject: [PATCH 06/16] ptp: fc3: remove redundant check on variable ret The check on ret has already been performed a few statements earlier and ret has not been re-assigned and so the re-checking is redundant. Clean up the code by removing the redundant check. Signed-off-by: Colin Ian King Link: https://patch.msgid.link/20241031135042.3250614-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_fc3.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/ptp/ptp_fc3.c b/drivers/ptp/ptp_fc3.c index e14e149b746e..879b82f03535 100644 --- a/drivers/ptp/ptp_fc3.c +++ b/drivers/ptp/ptp_fc3.c @@ -986,11 +986,6 @@ static int idtfc3_probe(struct platform_device *pdev) mutex_unlock(idtfc3->lock); - if (err) { - ptp_clock_unregister(idtfc3->ptp_clock); - return err; - } - platform_set_drvdata(pdev, idtfc3); return 0; -- 2.51.0 From d847548c7ef44ac01c1c102ed19744c8e26ada9b Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Fri, 1 Nov 2024 16:13:31 -0500 Subject: [PATCH 07/16] dt-bindings: net: snps,dwmac: Fix "snps,kbbe" type The driver and description indicate "snps,kbbe" is a boolean, not an uint32. Signed-off-by: Rob Herring (Arm) Reviewed-by: Krzysztof Kozlowski Link: https://patch.msgid.link/20241101211331.24605-2-robh@kernel.org Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/snps,dwmac.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml index 4e2ba1bf788c..f48a0f44cf2d 100644 --- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml +++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml @@ -560,7 +560,7 @@ properties: max read outstanding req. limit snps,kbbe: - $ref: /schemas/types.yaml#/definitions/uint32 + $ref: /schemas/types.yaml#/definitions/flag description: do not cross 1KiB boundary. -- 2.51.0 From 8a6631f1cece09047fa44608d21d520ca65ce7d8 Mon Sep 17 00:00:00 2001 From: Oleksij Rempel Date: Wed, 30 Oct 2024 09:52:24 +0100 Subject: [PATCH 08/16] net: macb: avoid redundant lookup for "mdio" child node in MDIO setup Pass the "mdio" child node directly to `macb_mdiobus_register` to avoid performing the node lookup twice. Signed-off-by: Oleksij Rempel Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20241030085224.2632426-1-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 23 +++++++++-------------- 1 file changed, 9 insertions(+), 14 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index ebe886b98891..daa416fb1724 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -915,20 +915,15 @@ static int macb_mii_probe(struct net_device *dev) return 0; } -static int macb_mdiobus_register(struct macb *bp) +static int macb_mdiobus_register(struct macb *bp, struct device_node *mdio_np) { struct device_node *child, *np = bp->pdev->dev.of_node; /* If we have a child named mdio, probe it instead of looking for PHYs * directly under the MAC node */ - child = of_get_child_by_name(np, "mdio"); - if (child) { - int ret = of_mdiobus_register(bp->mii_bus, child); - - of_node_put(child); - return ret; - } + if (mdio_np) + return of_mdiobus_register(bp->mii_bus, mdio_np); /* Only create the PHY from the device tree if at least one PHY is * described. Otherwise scan the entire MDIO bus. We do this to support @@ -950,17 +945,15 @@ static int macb_mdiobus_register(struct macb *bp) static int macb_mii_init(struct macb *bp) { - struct device_node *child, *np = bp->pdev->dev.of_node; + struct device_node *mdio_np, *np = bp->pdev->dev.of_node; int err = -ENXIO; /* With fixed-link, we don't need to register the MDIO bus, * except if we have a child named "mdio" in the device tree. * In that case, some devices may be attached to the MACB's MDIO bus. */ - child = of_get_child_by_name(np, "mdio"); - if (child) - of_node_put(child); - else if (of_phy_is_fixed_link(np)) + mdio_np = of_get_child_by_name(np, "mdio"); + if (!mdio_np && of_phy_is_fixed_link(np)) return macb_mii_probe(bp->dev); /* Enable management port */ @@ -984,7 +977,7 @@ static int macb_mii_init(struct macb *bp) dev_set_drvdata(&bp->dev->dev, bp->mii_bus); - err = macb_mdiobus_register(bp); + err = macb_mdiobus_register(bp, mdio_np); if (err) goto err_out_free_mdiobus; @@ -999,6 +992,8 @@ err_out_unregister_bus: err_out_free_mdiobus: mdiobus_free(bp->mii_bus); err_out: + of_node_put(mdio_np); + return err; } -- 2.51.0 From 0c30d6eedd1ec0c1382bcab9576d26413cd278a3 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Wed, 30 Oct 2024 13:43:11 +0100 Subject: [PATCH 09/16] ipvlan: Prepare ipvlan_process_v4_outbound() to future .flowi4_tos conversion. Use ip4h_dscp() to get the DSCP from the IPv4 header, then convert the dscp_t value to __u8 with inet_dscp_to_dsfield(). Then, when we'll convert .flowi4_tos to dscp_t, we'll just have to drop the inet_dscp_to_dsfield() call. Signed-off-by: Guillaume Nault Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/f48335504a05b3587e0081a9b4511e0761571ca5.1730292157.git.gnault@redhat.com Signed-off-by: Jakub Kicinski --- drivers/net/ipvlan/ipvlan_core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ipvlan/ipvlan_core.c b/drivers/net/ipvlan/ipvlan_core.c index b1afcb8740de..fd591ddb3884 100644 --- a/drivers/net/ipvlan/ipvlan_core.c +++ b/drivers/net/ipvlan/ipvlan_core.c @@ -3,6 +3,7 @@ */ #include +#include #include "ipvlan.h" @@ -422,7 +423,7 @@ static noinline_for_stack int ipvlan_process_v4_outbound(struct sk_buff *skb) int err, ret = NET_XMIT_DROP; struct flowi4 fl4 = { .flowi4_oif = dev->ifindex, - .flowi4_tos = ip4h->tos & INET_DSCP_MASK, + .flowi4_tos = inet_dscp_to_dsfield(ip4h_dscp(ip4h)), .flowi4_flags = FLOWI_FLAG_ANYSRC, .flowi4_mark = skb->mark, .daddr = ip4h->daddr, -- 2.51.0 From 937677f481259b5291001ef7c68242d366e23b64 Mon Sep 17 00:00:00 2001 From: Guillaume Nault Date: Wed, 30 Oct 2024 14:27:19 +0100 Subject: [PATCH 10/16] vrf: Prepare vrf_process_v4_outbound() to future .flowi4_tos conversion. Use ip4h_dscp() to get the DSCP from the IPv4 header, then convert the dscp_t value to __u8 with inet_dscp_to_dsfield(). Then, when we'll convert .flowi4_tos to dscp_t, we'll just have to drop the inet_dscp_to_dsfield() call. Signed-off-by: Guillaume Nault Reviewed-by: David Ahern Link: https://patch.msgid.link/6be084229008dcfa7a4e2758befccfd2217a331e.1730294788.git.gnault@redhat.com Signed-off-by: Jakub Kicinski --- drivers/net/vrf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 4087f72f0d2b..67d25f4f94ef 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -521,7 +521,7 @@ static netdev_tx_t vrf_process_v4_outbound(struct sk_buff *skb, /* needed to match OIF rule */ fl4.flowi4_l3mdev = vrf_dev->ifindex; fl4.flowi4_iif = LOOPBACK_IFINDEX; - fl4.flowi4_tos = ip4h->tos & INET_DSCP_MASK; + fl4.flowi4_tos = inet_dscp_to_dsfield(ip4h_dscp(ip4h)); fl4.flowi4_flags = FLOWI_FLAG_ANYSRC; fl4.flowi4_proto = ip4h->protocol; fl4.daddr = ip4h->daddr; -- 2.51.0 From 96111f1ec6bf14b6367395bf4d36797155ae354e Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Wed, 30 Oct 2024 13:37:16 -0700 Subject: [PATCH 11/16] net: ibm: emac: tah: use devm for kzalloc Simplifies the probe function by removing gotos. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20241030203727.6039-2-rosenp@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/emac/tah.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/ibm/emac/tah.c b/drivers/net/ethernet/ibm/emac/tah.c index c605c8ff933e..267c23ec15d7 100644 --- a/drivers/net/ethernet/ibm/emac/tah.c +++ b/drivers/net/ethernet/ibm/emac/tah.c @@ -90,28 +90,25 @@ static int tah_probe(struct platform_device *ofdev) struct device_node *np = ofdev->dev.of_node; struct tah_instance *dev; struct resource regs; - int rc; - rc = -ENOMEM; - dev = kzalloc(sizeof(struct tah_instance), GFP_KERNEL); - if (dev == NULL) - goto err_gone; + dev = devm_kzalloc(&ofdev->dev, sizeof(struct tah_instance), + GFP_KERNEL); + if (!dev) + return -ENOMEM; mutex_init(&dev->lock); dev->ofdev = ofdev; - rc = -ENXIO; if (of_address_to_resource(np, 0, ®s)) { printk(KERN_ERR "%pOF: Can't get registers address\n", np); - goto err_free; + return -ENXIO; } - rc = -ENOMEM; dev->base = (struct tah_regs __iomem *)ioremap(regs.start, sizeof(struct tah_regs)); if (dev->base == NULL) { printk(KERN_ERR "%pOF: Can't map device registers!\n", np); - goto err_free; + return -ENOMEM; } platform_set_drvdata(ofdev, dev); @@ -123,11 +120,6 @@ static int tah_probe(struct platform_device *ofdev) wmb(); return 0; - - err_free: - kfree(dev); - err_gone: - return rc; } static void tah_remove(struct platform_device *ofdev) @@ -137,7 +129,6 @@ static void tah_remove(struct platform_device *ofdev) WARN_ON(dev->users != 0); iounmap(dev->base); - kfree(dev); } static const struct of_device_id tah_match[] = -- 2.51.0 From 18082a84a7f06e9a4029daaf5a927e1e3f34f7ad Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Wed, 30 Oct 2024 13:37:17 -0700 Subject: [PATCH 12/16] net: ibm: emac: tah: use devm for mutex_init It seems that since inception, this driver never called mutex_destroy in _remove. Use devm to handle this automatically. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20241030203727.6039-3-rosenp@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/emac/tah.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ibm/emac/tah.c b/drivers/net/ethernet/ibm/emac/tah.c index 267c23ec15d7..4b325505053b 100644 --- a/drivers/net/ethernet/ibm/emac/tah.c +++ b/drivers/net/ethernet/ibm/emac/tah.c @@ -90,13 +90,17 @@ static int tah_probe(struct platform_device *ofdev) struct device_node *np = ofdev->dev.of_node; struct tah_instance *dev; struct resource regs; + int err; dev = devm_kzalloc(&ofdev->dev, sizeof(struct tah_instance), GFP_KERNEL); if (!dev) return -ENOMEM; - mutex_init(&dev->lock); + err = devm_mutex_init(&ofdev->dev, &dev->lock); + if (err) + return err; + dev->ofdev = ofdev; if (of_address_to_resource(np, 0, ®s)) { -- 2.51.0 From 9f3ea8d70d6c4bde8cabf0a57efafdc139d88217 Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Wed, 30 Oct 2024 13:37:18 -0700 Subject: [PATCH 13/16] net: ibm: emac: tah: devm_platform_get_resources Simplifies the probe function by a bit and allows removing the _remove function such that devm now handles all cleanup. printk gets converted to dev_err as np is now gone. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20241030203727.6039-4-rosenp@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/emac/tah.c | 26 ++++---------------------- 1 file changed, 4 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/ibm/emac/tah.c b/drivers/net/ethernet/ibm/emac/tah.c index 4b325505053b..09f6373ed2f9 100644 --- a/drivers/net/ethernet/ibm/emac/tah.c +++ b/drivers/net/ethernet/ibm/emac/tah.c @@ -87,9 +87,7 @@ void *tah_dump_regs(struct platform_device *ofdev, void *buf) static int tah_probe(struct platform_device *ofdev) { - struct device_node *np = ofdev->dev.of_node; struct tah_instance *dev; - struct resource regs; int err; dev = devm_kzalloc(&ofdev->dev, sizeof(struct tah_instance), @@ -103,16 +101,10 @@ static int tah_probe(struct platform_device *ofdev) dev->ofdev = ofdev; - if (of_address_to_resource(np, 0, ®s)) { - printk(KERN_ERR "%pOF: Can't get registers address\n", np); - return -ENXIO; - } - - dev->base = (struct tah_regs __iomem *)ioremap(regs.start, - sizeof(struct tah_regs)); - if (dev->base == NULL) { - printk(KERN_ERR "%pOF: Can't map device registers!\n", np); - return -ENOMEM; + dev->base = devm_platform_ioremap_resource(ofdev, 0); + if (IS_ERR(dev->base)) { + dev_err(&ofdev->dev, "can't map device registers"); + return PTR_ERR(dev->base); } platform_set_drvdata(ofdev, dev); @@ -126,15 +118,6 @@ static int tah_probe(struct platform_device *ofdev) return 0; } -static void tah_remove(struct platform_device *ofdev) -{ - struct tah_instance *dev = platform_get_drvdata(ofdev); - - WARN_ON(dev->users != 0); - - iounmap(dev->base); -} - static const struct of_device_id tah_match[] = { { @@ -153,7 +136,6 @@ static struct platform_driver tah_driver = { .of_match_table = tah_match, }, .probe = tah_probe, - .remove = tah_remove, }; int __init tah_init(void) -- 2.51.0 From 070239c07ac1623c114a3f43fe37381a5ae3a9ce Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Wed, 30 Oct 2024 13:37:19 -0700 Subject: [PATCH 14/16] net: ibm: emac: rgmii: use devm for kzalloc Simplifies the probe function by removing gotos. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20241030203727.6039-5-rosenp@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/emac/rgmii.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/ibm/emac/rgmii.c b/drivers/net/ethernet/ibm/emac/rgmii.c index 317c22d09172..7bafe2edfc50 100644 --- a/drivers/net/ethernet/ibm/emac/rgmii.c +++ b/drivers/net/ethernet/ibm/emac/rgmii.c @@ -219,28 +219,25 @@ static int rgmii_probe(struct platform_device *ofdev) struct device_node *np = ofdev->dev.of_node; struct rgmii_instance *dev; struct resource regs; - int rc; - rc = -ENOMEM; - dev = kzalloc(sizeof(struct rgmii_instance), GFP_KERNEL); - if (dev == NULL) - goto err_gone; + dev = devm_kzalloc(&ofdev->dev, sizeof(struct rgmii_instance), + GFP_KERNEL); + if (!dev) + return -ENOMEM; mutex_init(&dev->lock); dev->ofdev = ofdev; - rc = -ENXIO; if (of_address_to_resource(np, 0, ®s)) { printk(KERN_ERR "%pOF: Can't get registers address\n", np); - goto err_free; + return -ENXIO; } - rc = -ENOMEM; dev->base = (struct rgmii_regs __iomem *)ioremap(regs.start, sizeof(struct rgmii_regs)); if (dev->base == NULL) { printk(KERN_ERR "%pOF: Can't map device registers!\n", np); - goto err_free; + return -ENOMEM; } /* Check for RGMII flags */ @@ -266,11 +263,6 @@ static int rgmii_probe(struct platform_device *ofdev) platform_set_drvdata(ofdev, dev); return 0; - - err_free: - kfree(dev); - err_gone: - return rc; } static void rgmii_remove(struct platform_device *ofdev) @@ -280,7 +272,6 @@ static void rgmii_remove(struct platform_device *ofdev) WARN_ON(dev->users != 0); iounmap(dev->base); - kfree(dev); } static const struct of_device_id rgmii_match[] = -- 2.51.0 From 01902fe2bdd75d028dd004fa2dfc95bc65f055bc Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Wed, 30 Oct 2024 13:37:20 -0700 Subject: [PATCH 15/16] net: ibm: emac: rgmii: use devm for mutex_init It seems that since inception, this driver never called mutex_destroy in _remove. Use devm to handle this automatically. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20241030203727.6039-6-rosenp@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/emac/rgmii.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ibm/emac/rgmii.c b/drivers/net/ethernet/ibm/emac/rgmii.c index 7bafe2edfc50..9063f0a17e25 100644 --- a/drivers/net/ethernet/ibm/emac/rgmii.c +++ b/drivers/net/ethernet/ibm/emac/rgmii.c @@ -219,13 +219,17 @@ static int rgmii_probe(struct platform_device *ofdev) struct device_node *np = ofdev->dev.of_node; struct rgmii_instance *dev; struct resource regs; + int err; dev = devm_kzalloc(&ofdev->dev, sizeof(struct rgmii_instance), GFP_KERNEL); if (!dev) return -ENOMEM; - mutex_init(&dev->lock); + err = devm_mutex_init(&ofdev->dev, &dev->lock); + if (err) + return err; + dev->ofdev = ofdev; if (of_address_to_resource(np, 0, ®s)) { -- 2.51.0 From 9fb40aeeb52171ace8bdceb5f8000ec56d0582a0 Mon Sep 17 00:00:00 2001 From: Rosen Penev Date: Wed, 30 Oct 2024 13:37:21 -0700 Subject: [PATCH 16/16] net: ibm: emac: rgmii: devm_platform_get_resource Simplifies the probe function by a bit and allows removing the _remove function such that devm now handles all cleanup. printk gets converted to dev_err as np is now gone. Signed-off-by: Rosen Penev Link: https://patch.msgid.link/20241030203727.6039-7-rosenp@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/emac/rgmii.c | 26 ++++---------------------- 1 file changed, 4 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/ibm/emac/rgmii.c b/drivers/net/ethernet/ibm/emac/rgmii.c index 9063f0a17e25..b544dd8633b7 100644 --- a/drivers/net/ethernet/ibm/emac/rgmii.c +++ b/drivers/net/ethernet/ibm/emac/rgmii.c @@ -216,9 +216,7 @@ void *rgmii_dump_regs(struct platform_device *ofdev, void *buf) static int rgmii_probe(struct platform_device *ofdev) { - struct device_node *np = ofdev->dev.of_node; struct rgmii_instance *dev; - struct resource regs; int err; dev = devm_kzalloc(&ofdev->dev, sizeof(struct rgmii_instance), @@ -232,16 +230,10 @@ static int rgmii_probe(struct platform_device *ofdev) dev->ofdev = ofdev; - if (of_address_to_resource(np, 0, ®s)) { - printk(KERN_ERR "%pOF: Can't get registers address\n", np); - return -ENXIO; - } - - dev->base = (struct rgmii_regs __iomem *)ioremap(regs.start, - sizeof(struct rgmii_regs)); - if (dev->base == NULL) { - printk(KERN_ERR "%pOF: Can't map device registers!\n", np); - return -ENOMEM; + dev->base = devm_platform_ioremap_resource(ofdev, 0); + if (IS_ERR(dev->base)) { + dev_err(&ofdev->dev, "can't map device registers"); + return PTR_ERR(dev->base); } /* Check for RGMII flags */ @@ -269,15 +261,6 @@ static int rgmii_probe(struct platform_device *ofdev) return 0; } -static void rgmii_remove(struct platform_device *ofdev) -{ - struct rgmii_instance *dev = platform_get_drvdata(ofdev); - - WARN_ON(dev->users != 0); - - iounmap(dev->base); -} - static const struct of_device_id rgmii_match[] = { { @@ -295,7 +278,6 @@ static struct platform_driver rgmii_driver = { .of_match_table = rgmii_match, }, .probe = rgmii_probe, - .remove = rgmii_remove, }; int __init rgmii_init(void) -- 2.51.0