From 2a5970d5aaff8f3e33ce3bfaa403ae88c40de40d Mon Sep 17 00:00:00 2001 From: Sagi Maimon Date: Tue, 15 Apr 2025 08:31:31 +0300 Subject: [PATCH 01/16] ptp: ocp: fix start time alignment in ptp_ocp_signal_set In ptp_ocp_signal_set, the start time for periodic signals is not aligned to the next period boundary. The current code rounds up the start time and divides by the period but fails to multiply back by the period, causing misaligned signal starts. Fix this by multiplying the rounded-up value by the period to ensure the start time is the closest next period. Fixes: 4bd46bb037f8e ("ptp: ocp: Use DIV64_U64_ROUND_UP for rounding.") Signed-off-by: Sagi Maimon Reviewed-by: Vadim Fedorenko Link: https://patch.msgid.link/20250415053131.129413-1-maimon.sagi@gmail.com Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_ocp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/ptp/ptp_ocp.c b/drivers/ptp/ptp_ocp.c index 7945c6be1f7c..faf6e027f89a 100644 --- a/drivers/ptp/ptp_ocp.c +++ b/drivers/ptp/ptp_ocp.c @@ -2067,6 +2067,7 @@ ptp_ocp_signal_set(struct ptp_ocp *bp, int gen, struct ptp_ocp_signal *s) if (!s->start) { /* roundup() does not work on 32-bit systems */ s->start = DIV64_U64_ROUND_UP(start_ns, s->period); + s->start *= s->period; s->start = ktime_add(s->start, s->phase); } -- 2.51.0 From 4798cfa2097f0833d54d8f5ce20ef14631917839 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 15 Apr 2025 08:15:52 -0700 Subject: [PATCH 02/16] net: don't try to ops lock uninitialized devs We need to be careful when operating on dev while in rtnl_create_link(). Some devices (vxlan) initialize netdev_ops in ->newlink, so later on. Avoid using netdev_lock_ops(), the device isn't registered so we cannot legally call its ops or generate any notifications for it. netdev_ops_assert_locked_or_invisible() is safe to use, it checks registration status first. Reported-by: syzbot+de1c7d68a10e3f123bdd@syzkaller.appspotmail.com Fixes: 04efcee6ef8d ("net: hold instance lock during NETDEV_CHANGE") Acked-by: Stanislav Fomichev Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20250415151552.768373-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/core/dev.c | 2 ++ net/core/rtnetlink.c | 5 +---- 2 files changed, 3 insertions(+), 4 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 5fcbc66d865e..1be7cb73a602 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -1520,6 +1520,8 @@ EXPORT_SYMBOL(netdev_features_change); void netif_state_change(struct net_device *dev) { + netdev_ops_assert_locked_or_invisible(dev); + if (dev->flags & IFF_UP) { struct netdev_notifier_change_info change_info = { .info.dev = dev, diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 39a5b72e861f..c5a7f41982a5 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -3676,11 +3676,8 @@ struct net_device *rtnl_create_link(struct net *net, const char *ifname, nla_len(tb[IFLA_BROADCAST])); if (tb[IFLA_TXQLEN]) dev->tx_queue_len = nla_get_u32(tb[IFLA_TXQLEN]); - if (tb[IFLA_OPERSTATE]) { - netdev_lock_ops(dev); + if (tb[IFLA_OPERSTATE]) set_operstate(dev, nla_get_u8(tb[IFLA_OPERSTATE])); - netdev_unlock_ops(dev); - } if (tb[IFLA_LINKMODE]) dev->link_mode = nla_get_u8(tb[IFLA_LINKMODE]); if (tb[IFLA_GROUP]) -- 2.51.0 From c7b67ddc3c999aa2f8d77be7ef1913298fe78f0e Mon Sep 17 00:00:00 2001 From: Hans Holmberg Date: Wed, 9 Apr 2025 12:39:56 +0000 Subject: [PATCH 03/16] xfs: document zoned rt specifics in admin-guide Document the lifetime, nolifetime and max_open_zones mount options added for zoned rt file systems. Also add documentation describing the max_open_zones sysfs attribute exposed in /sys/fs/xfs//zoned/ Fixes: 4e4d52075577 ("xfs: add the zoned space allocator") Signed-off-by: Hans Holmberg Reviewed-by: Christoph Hellwig Reviewed-by: Darrick J. Wong Signed-off-by: Carlos Maiolino --- Documentation/admin-guide/xfs.rst | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/Documentation/admin-guide/xfs.rst b/Documentation/admin-guide/xfs.rst index 7b0811d650f9..3e76276bd488 100644 --- a/Documentation/admin-guide/xfs.rst +++ b/Documentation/admin-guide/xfs.rst @@ -124,6 +124,14 @@ When mounting an XFS filesystem, the following options are accepted. controls the size of each buffer and so is also relevant to this case. + lifetime (default) or nolifetime + Enable data placement based on write life time hints provided + by the user. This turns on co-allocation of data of similar + life times when statistically favorable to reduce garbage + collection cost. + + These options are only available for zoned rt file systems. + logbsize=value Set the size of each in-memory log buffer. The size may be specified in bytes, or in kilobytes with a "k" suffix. @@ -143,6 +151,14 @@ When mounting an XFS filesystem, the following options are accepted. optional, and the log section can be separate from the data section or contained within it. + max_open_zones=value + Specify the max number of zones to keep open for writing on a + zoned rt device. Many open zones aids file data separation + but may impact performance on HDDs. + + If ``max_open_zones`` is not specified, the value is determined + by the capabilities and the size of the zoned rt device. + noalign Data allocations will not be aligned at stripe unit boundaries. This is only relevant to filesystems created @@ -546,6 +562,19 @@ The interesting knobs for XFS workqueues are as follows: Zoned Filesystems ================= +For zoned file systems, the following attribute is exposed in: + + /sys/fs/xfs//zoned/ + + max_open_zones (Min: 1 Default: Varies Max: UINTMAX) + This read-only attribute exposes the maximum number of open zones + available for data placement. The value is determined at mount time and + is limited by the capabilities of the backing zoned device, file system + size and the max_open_zones mount option. + +Zoned Filesystems +================= + For zoned file systems, the following attributes are exposed in: /sys/fs/xfs//zoned/ -- 2.51.0 From d2d31ea8cd80b9830cdab624e94f9d41178fc99d Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 15 Apr 2025 15:53:48 +0200 Subject: [PATCH 04/16] netfilter: conntrack: fix erronous removal of offload bit The blamed commit exposes a possible issue with flow_offload_teardown(): We might remove the offload bit of a conntrack entry that has been offloaded again. 1. conntrack entry c1 is offloaded via flow f1 (f1->ct == c1). 2. f1 times out and is pushed back to slowpath, c1 offload bit is removed. Due to bug, f1 is not unlinked from rhashtable right away. 3. a new packet arrives for the flow and re-offload is triggered, i.e. f2->ct == c1. This is because lookup in flowtable skip entries with teardown bit set. 4. Next flowtable gc cycle finds f1 again 5. flow_offload_teardown() is called again for f1 and c1 offload bit is removed again, even though we have f2 referencing the same entry. This is harmless, but clearly not correct. Fix the bug that exposes this: set 'teardown = true' to have the gc callback unlink the flowtable entry from the table right away instead of the unintentional defer to the next round. Also prevent flow_offload_teardown() from fixing up the ct state more than once: We could also be called from the data path or a notifier, not only from the flowtable gc callback. NF_FLOW_TEARDOWN can never be unset, so we can use it as synchronization point: if we observe did not see a 0 -> 1 transition, then another CPU is already doing the ct state fixups for us. Fixes: 03428ca5cee9 ("netfilter: conntrack: rework offload nf_conn timeout extension logic") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_flow_table_core.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index 9d8361526f82..9441ac3d8c1a 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -383,8 +383,8 @@ static void flow_offload_del(struct nf_flowtable *flow_table, void flow_offload_teardown(struct flow_offload *flow) { clear_bit(IPS_OFFLOAD_BIT, &flow->ct->status); - set_bit(NF_FLOW_TEARDOWN, &flow->flags); - flow_offload_fixup_ct(flow); + if (!test_and_set_bit(NF_FLOW_TEARDOWN, &flow->flags)) + flow_offload_fixup_ct(flow); } EXPORT_SYMBOL_GPL(flow_offload_teardown); @@ -558,10 +558,12 @@ static void nf_flow_offload_gc_step(struct nf_flowtable *flow_table, if (nf_flow_has_expired(flow) || nf_ct_is_dying(flow->ct) || - nf_flow_custom_gc(flow_table, flow)) + nf_flow_custom_gc(flow_table, flow)) { flow_offload_teardown(flow); - else if (!teardown) + teardown = true; + } else if (!teardown) { nf_flow_table_extend_ct_timeout(flow->ct); + } if (teardown) { if (test_bit(NF_FLOW_HW, &flow->flags)) { -- 2.51.0 From 75bc744466444ef417b5f709f72b91c83301bcd1 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 15 Apr 2025 14:35:41 +0530 Subject: [PATCH 05/16] net: ti: icssg-prueth: Fix kernel warning while bringing down network interface During network interface initialization, the NIC driver needs to register its Rx queue with the XDP, to ensure the incoming XDP buffer carries a pointer reference to this info and is stored inside xdp_rxq_info. While this struct isn't tied to XDP prog, if there are any changes in Rx queue, the NIC driver needs to stop the Rx queue by unregistering with XDP before purging and reallocating memory. Drop page_pool destroy during Rx channel reset as this is already handled by XDP during xdp_rxq_info_unreg (Rx queue unregister), failing to do will cause the following warning: warning logs: https://gist.github.com/MeghanaMalladiTI/eb627e5dc8de24e42d7d46572c13e576 Fixes: 46eeb90f03e0 ("net: ti: icssg-prueth: Use page_pool API for RX buffer allocation") Signed-off-by: Meghana Malladi Reviewed-by: Simon Horman Reviewed-by: Roger Quadros Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20250415090543.717991-2-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index 14002b026452..ec643fb69d30 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -1215,9 +1215,6 @@ void prueth_reset_rx_chan(struct prueth_rx_chn *chn, prueth_rx_cleanup); if (disable) k3_udma_glue_disable_rx_chn(chn->rx_chn); - - page_pool_destroy(chn->pg_pool); - chn->pg_pool = NULL; } EXPORT_SYMBOL_GPL(prueth_reset_rx_chan); -- 2.51.0 From 8ed2fa661350f0b49edb765d18173b5c766c3686 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 15 Apr 2025 14:35:42 +0530 Subject: [PATCH 06/16] net: ti: icssg-prueth: Fix possible NULL pointer dereference inside emac_xmit_xdp_frame() There is an error check inside emac_xmit_xdp_frame() function which is called when the driver wants to transmit XDP frame, to check if the allocated tx descriptor is NULL, if true to exit and return ICSSG_XDP_CONSUMED implying failure in transmission. In this case trying to free a descriptor which is NULL will result in kernel crash due to NULL pointer dereference. Fix this error handling and increase netdev tx_dropped stats in the caller of this function if the function returns ICSSG_XDP_CONSUMED. Fixes: 62aa3246f462 ("net: ti: icssg-prueth: Add XDP support") Reported-by: Dan Carpenter Closes: https://lore.kernel.org/all/70d8dd76-0c76-42fc-8611-9884937c82f5@stanley.mountain/ Signed-off-by: Meghana Malladi Reviewed-by: Simon Horman Reviewed-by: Roger Quadros Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20250415090543.717991-3-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index ec643fb69d30..b4be76e13a2f 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -583,7 +583,7 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac, first_desc = k3_cppi_desc_pool_alloc(tx_chn->desc_pool); if (!first_desc) { netdev_dbg(ndev, "xdp tx: failed to allocate descriptor\n"); - goto drop_free_descs; /* drop */ + return ICSSG_XDP_CONSUMED; /* drop */ } if (page) { /* already DMA mapped by page_pool */ @@ -671,8 +671,10 @@ static u32 emac_run_xdp(struct prueth_emac *emac, struct xdp_buff *xdp, q_idx = smp_processor_id() % emac->tx_ch_num; result = emac_xmit_xdp_frame(emac, xdpf, page, q_idx); - if (result == ICSSG_XDP_CONSUMED) + if (result == ICSSG_XDP_CONSUMED) { + ndev->stats.tx_dropped++; goto drop; + } dev_sw_netstats_rx_add(ndev, xdpf->len); return result; -- 2.51.0 From 7349c9e9979333abfce42da5f9025598083b59c9 Mon Sep 17 00:00:00 2001 From: Meghana Malladi Date: Tue, 15 Apr 2025 14:35:43 +0530 Subject: [PATCH 07/16] net: ti: icss-iep: Fix possible NULL pointer dereference for perout request The ICSS IEP driver tracks perout and pps enable state with flags. Currently when disabling pps and perout signals during icss_iep_exit(), results in NULL pointer dereference for perout. To fix the null pointer dereference issue, the icss_iep_perout_enable_hw function can be modified to directly clear the IEP CMP registers when disabling PPS or PEROUT, without referencing the ptp_perout_request structure, as its contents are irrelevant in this case. Fixes: 9b115361248d ("net: ti: icssg-prueth: Fix clearing of IEP_CMP_CFG registers during iep_init") Reported-by: Dan Carpenter Closes: https://lore.kernel.org/all/7b1c7c36-363a-4085-b26c-4f210bee1df6@stanley.mountain/ Signed-off-by: Meghana Malladi Reviewed-by: Jacob Keller Link: https://patch.msgid.link/20250415090543.717991-4-m-malladi@ti.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icss_iep.c | 121 +++++++++++------------ 1 file changed, 58 insertions(+), 63 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icss_iep.c b/drivers/net/ethernet/ti/icssg/icss_iep.c index b4a34c57b7b4..2a1c43316f46 100644 --- a/drivers/net/ethernet/ti/icssg/icss_iep.c +++ b/drivers/net/ethernet/ti/icssg/icss_iep.c @@ -412,6 +412,22 @@ static int icss_iep_perout_enable_hw(struct icss_iep *iep, int ret; u64 cmp; + if (!on) { + /* Disable CMP 1 */ + regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, + IEP_CMP_CFG_CMP_EN(1), 0); + + /* clear CMP regs */ + regmap_write(iep->map, ICSS_IEP_CMP1_REG0, 0); + if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT) + regmap_write(iep->map, ICSS_IEP_CMP1_REG1, 0); + + /* Disable sync */ + regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0); + + return 0; + } + /* Calculate width of the signal for PPS/PEROUT handling */ ts.tv_sec = req->on.sec; ts.tv_nsec = req->on.nsec; @@ -430,64 +446,39 @@ static int icss_iep_perout_enable_hw(struct icss_iep *iep, if (ret) return ret; - if (on) { - /* Configure CMP */ - regmap_write(iep->map, ICSS_IEP_CMP1_REG0, lower_32_bits(cmp)); - if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT) - regmap_write(iep->map, ICSS_IEP_CMP1_REG1, upper_32_bits(cmp)); - /* Configure SYNC, based on req on width */ - regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG, - div_u64(ns_width, iep->def_inc)); - regmap_write(iep->map, ICSS_IEP_SYNC0_PERIOD_REG, 0); - regmap_write(iep->map, ICSS_IEP_SYNC_START_REG, - div_u64(ns_start, iep->def_inc)); - regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0); /* one-shot mode */ - /* Enable CMP 1 */ - regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, - IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1)); - } else { - /* Disable CMP 1 */ - regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, - IEP_CMP_CFG_CMP_EN(1), 0); - - /* clear regs */ - regmap_write(iep->map, ICSS_IEP_CMP1_REG0, 0); - if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT) - regmap_write(iep->map, ICSS_IEP_CMP1_REG1, 0); - } + /* Configure CMP */ + regmap_write(iep->map, ICSS_IEP_CMP1_REG0, lower_32_bits(cmp)); + if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT) + regmap_write(iep->map, ICSS_IEP_CMP1_REG1, upper_32_bits(cmp)); + /* Configure SYNC, based on req on width */ + regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG, + div_u64(ns_width, iep->def_inc)); + regmap_write(iep->map, ICSS_IEP_SYNC0_PERIOD_REG, 0); + regmap_write(iep->map, ICSS_IEP_SYNC_START_REG, + div_u64(ns_start, iep->def_inc)); + regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0); /* one-shot mode */ + /* Enable CMP 1 */ + regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, + IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1)); } else { - if (on) { - u64 start_ns; - - iep->period = ((u64)req->period.sec * NSEC_PER_SEC) + - req->period.nsec; - start_ns = ((u64)req->period.sec * NSEC_PER_SEC) - + req->period.nsec; - icss_iep_update_to_next_boundary(iep, start_ns); - - regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG, - div_u64(ns_width, iep->def_inc)); - regmap_write(iep->map, ICSS_IEP_SYNC_START_REG, - div_u64(ns_start, iep->def_inc)); - /* Enable Sync in single shot mode */ - regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, - IEP_SYNC_CTRL_SYNC_N_EN(0) | IEP_SYNC_CTRL_SYNC_EN); - /* Enable CMP 1 */ - regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, - IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1)); - } else { - /* Disable CMP 1 */ - regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, - IEP_CMP_CFG_CMP_EN(1), 0); - - /* clear CMP regs */ - regmap_write(iep->map, ICSS_IEP_CMP1_REG0, 0); - if (iep->plat_data->flags & ICSS_IEP_64BIT_COUNTER_SUPPORT) - regmap_write(iep->map, ICSS_IEP_CMP1_REG1, 0); - - /* Disable sync */ - regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, 0); - } + u64 start_ns; + + iep->period = ((u64)req->period.sec * NSEC_PER_SEC) + + req->period.nsec; + start_ns = ((u64)req->period.sec * NSEC_PER_SEC) + + req->period.nsec; + icss_iep_update_to_next_boundary(iep, start_ns); + + regmap_write(iep->map, ICSS_IEP_SYNC_PWIDTH_REG, + div_u64(ns_width, iep->def_inc)); + regmap_write(iep->map, ICSS_IEP_SYNC_START_REG, + div_u64(ns_start, iep->def_inc)); + /* Enable Sync in single shot mode */ + regmap_write(iep->map, ICSS_IEP_SYNC_CTRL_REG, + IEP_SYNC_CTRL_SYNC_N_EN(0) | IEP_SYNC_CTRL_SYNC_EN); + /* Enable CMP 1 */ + regmap_update_bits(iep->map, ICSS_IEP_CMP_CFG_REG, + IEP_CMP_CFG_CMP_EN(1), IEP_CMP_CFG_CMP_EN(1)); } return 0; @@ -498,11 +489,21 @@ static int icss_iep_perout_enable(struct icss_iep *iep, { int ret = 0; + if (!on) + goto disable; + /* Reject requests with unsupported flags */ if (req->flags & ~(PTP_PEROUT_DUTY_CYCLE | PTP_PEROUT_PHASE)) return -EOPNOTSUPP; + /* Set default "on" time (1ms) for the signal if not passed by the app */ + if (!(req->flags & PTP_PEROUT_DUTY_CYCLE)) { + req->on.sec = 0; + req->on.nsec = NSEC_PER_MSEC; + } + +disable: mutex_lock(&iep->ptp_clk_mutex); if (iep->pps_enabled) { @@ -513,12 +514,6 @@ static int icss_iep_perout_enable(struct icss_iep *iep, if (iep->perout_enabled == !!on) goto exit; - /* Set default "on" time (1ms) for the signal if not passed by the app */ - if (!(req->flags & PTP_PEROUT_DUTY_CYCLE)) { - req->on.sec = 0; - req->on.nsec = NSEC_PER_MSEC; - } - ret = icss_iep_perout_enable_hw(iep, req, on); if (!ret) iep->perout_enabled = !!on; -- 2.51.0 From 6bc2b6c6f16d8e60de518d26da1bc6bc436cf71d Mon Sep 17 00:00:00 2001 From: Bo-Cun Chen Date: Wed, 16 Apr 2025 01:50:46 +0100 Subject: [PATCH 08/16] net: ethernet: mtk_eth_soc: reapply mdc divider on reset In the current method, the MDC divider was reset to the default setting of 2.5MHz after the NETSYS SER. Therefore, we need to reapply the MDC divider configuration function in mtk_hw_init() after reset. Fixes: c0a440031d431 ("net: ethernet: mtk_eth_soc: set MDIO bus clock frequency") Signed-off-by: Bo-Cun Chen Signed-off-by: Daniel Golle Link: https://patch.msgid.link/8ab7381447e6cdcb317d5b5a6ddd90a1734efcb0.1744764277.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 39 +++++++++++++-------- drivers/net/ethernet/mediatek/mtk_eth_soc.h | 1 + 2 files changed, 25 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 43197b28b3e7..1a235283b0e9 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -871,9 +871,25 @@ static const struct phylink_mac_ops mtk_phylink_ops = { .mac_enable_tx_lpi = mtk_mac_enable_tx_lpi, }; +static void mtk_mdio_config(struct mtk_eth *eth) +{ + u32 val; + + /* Configure MDC Divider */ + val = FIELD_PREP(PPSC_MDC_CFG, eth->mdc_divider); + + /* Configure MDC Turbo Mode */ + if (mtk_is_netsys_v3_or_greater(eth)) + mtk_m32(eth, 0, MISC_MDC_TURBO, MTK_MAC_MISC_V3); + else + val |= PPSC_MDC_TURBO; + + mtk_m32(eth, PPSC_MDC_CFG, val, MTK_PPSC); +} + static int mtk_mdio_init(struct mtk_eth *eth) { - unsigned int max_clk = 2500000, divider; + unsigned int max_clk = 2500000; struct device_node *mii_np; int ret; u32 val; @@ -908,20 +924,9 @@ static int mtk_mdio_init(struct mtk_eth *eth) } max_clk = val; } - divider = min_t(unsigned int, DIV_ROUND_UP(MDC_MAX_FREQ, max_clk), 63); - - /* Configure MDC Turbo Mode */ - if (mtk_is_netsys_v3_or_greater(eth)) - mtk_m32(eth, 0, MISC_MDC_TURBO, MTK_MAC_MISC_V3); - - /* Configure MDC Divider */ - val = FIELD_PREP(PPSC_MDC_CFG, divider); - if (!mtk_is_netsys_v3_or_greater(eth)) - val |= PPSC_MDC_TURBO; - mtk_m32(eth, PPSC_MDC_CFG, val, MTK_PPSC); - - dev_dbg(eth->dev, "MDC is running on %d Hz\n", MDC_MAX_FREQ / divider); - + eth->mdc_divider = min_t(unsigned int, DIV_ROUND_UP(MDC_MAX_FREQ, max_clk), 63); + mtk_mdio_config(eth); + dev_dbg(eth->dev, "MDC is running on %d Hz\n", MDC_MAX_FREQ / eth->mdc_divider); ret = of_mdiobus_register(eth->mii_bus, mii_np); err_put_node: @@ -3974,6 +3979,10 @@ static int mtk_hw_init(struct mtk_eth *eth, bool reset) else mtk_hw_reset(eth); + /* No MT7628/88 support yet */ + if (reset && !MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628)) + mtk_mdio_config(eth); + if (mtk_is_netsys_v3_or_greater(eth)) { /* Set FE to PDMAv2 if necessary */ val = mtk_r32(eth, MTK_FE_GLO_MISC); diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h index 90a377ab4359..39709649ea8d 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h @@ -1271,6 +1271,7 @@ struct mtk_eth { struct clk *clks[MTK_CLK_MAX]; struct mii_bus *mii_bus; + unsigned int mdc_divider; struct work_struct pending_work; unsigned long state; -- 2.51.0 From 6b02eb372c6776c9abb8bc81cf63f96039c24664 Mon Sep 17 00:00:00 2001 From: Bo-Cun Chen Date: Wed, 16 Apr 2025 01:51:07 +0100 Subject: [PATCH 09/16] net: ethernet: mtk_eth_soc: correct the max weight of the queue limit for 100Mbps Without this patch, the maximum weight of the queue limit will be incorrect when linked at 100Mbps due to an apparent typo. Fixes: f63959c7eec31 ("net: ethernet: mtk_eth_soc: implement multi-queue support for per-port queues") Signed-off-by: Bo-Cun Chen Signed-off-by: Daniel Golle Link: https://patch.msgid.link/74111ba0bdb13743313999ed467ce564e8189006.1744764277.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 1a235283b0e9..5a3cfb8908a1 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -734,7 +734,7 @@ static void mtk_set_queue_speed(struct mtk_eth *eth, unsigned int idx, case SPEED_100: val |= MTK_QTX_SCH_MAX_RATE_EN | FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 103) | - FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 3); + FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 3) | FIELD_PREP(MTK_QTX_SCH_MAX_RATE_WEIGHT, 1); break; case SPEED_1000: @@ -757,7 +757,7 @@ static void mtk_set_queue_speed(struct mtk_eth *eth, unsigned int idx, case SPEED_100: val |= MTK_QTX_SCH_MAX_RATE_EN | FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 1) | - FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 5); + FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 5) | FIELD_PREP(MTK_QTX_SCH_MAX_RATE_WEIGHT, 1); break; case SPEED_1000: -- 2.51.0 From 1b66124135f5f8640bd540fadda4b20cdd23114b Mon Sep 17 00:00:00 2001 From: Bo-Cun Chen Date: Wed, 16 Apr 2025 01:51:25 +0100 Subject: [PATCH 10/16] net: ethernet: mtk_eth_soc: revise QDMA packet scheduler settings The QDMA packet scheduler suffers from a performance issue. Fix this by picking up changes from MediaTek's SDK which change to use Token Bucket instead of Leaky Bucket and fix the SPEED_1000 configuration. Fixes: 160d3a9b1929 ("net: ethernet: mtk_eth_soc: introduce MTK_NETSYS_V2 support") Signed-off-by: Bo-Cun Chen Signed-off-by: Daniel Golle Link: https://patch.msgid.link/18040f60f9e2f5855036b75b28c4332a2d2ebdd8.1744764277.git.daniel@makrotopia.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 5a3cfb8908a1..bdb98c9d8b1c 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -762,8 +762,8 @@ static void mtk_set_queue_speed(struct mtk_eth *eth, unsigned int idx, break; case SPEED_1000: val |= MTK_QTX_SCH_MAX_RATE_EN | - FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 10) | - FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 5) | + FIELD_PREP(MTK_QTX_SCH_MAX_RATE_MAN, 1) | + FIELD_PREP(MTK_QTX_SCH_MAX_RATE_EXP, 6) | FIELD_PREP(MTK_QTX_SCH_MAX_RATE_WEIGHT, 10); break; default: @@ -3320,7 +3320,7 @@ static int mtk_start_dma(struct mtk_eth *eth) if (mtk_is_netsys_v2_or_greater(eth)) val |= MTK_MUTLI_CNT | MTK_RESV_BUF | MTK_WCOMP_EN | MTK_DMAD_WR_WDONE | - MTK_CHK_DDONE_EN | MTK_LEAKY_BUCKET_EN; + MTK_CHK_DDONE_EN; else val |= MTK_RX_BT_32DWORDS; mtk_w32(eth, val, reg_map->qdma.glo_cfg); -- 2.51.0 From 750d0ac001e85b754404178ee8ce01cbc76a03be Mon Sep 17 00:00:00 2001 From: Maxime Chevallier Date: Wed, 16 Apr 2025 14:54:48 +0200 Subject: [PATCH 11/16] MAINTAINERS: Add entry for Socfpga DWMAC ethernet glue driver Socfpga's DWMAC glue comes in a variety of flavours with multiple options when it comes to physical interfaces, making it not so easy to test. Having access to a Cyclone5 with RGMII as well as Lynx PCS variants, add myself as a maintainer to help with reviews and testing. Signed-off-by: Maxime Chevallier Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20250416125453.306029-1-maxime.chevallier@bootlin.com Signed-off-by: Jakub Kicinski --- MAINTAINERS | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 8c7d796131a8..067f72e8af10 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3191,6 +3191,12 @@ M: Dinh Nguyen S: Maintained F: drivers/clk/socfpga/ +ARM/SOCFPGA DWMAC GLUE LAYER +M: Maxime Chevallier +S: Maintained +F: Documentation/devicetree/bindings/net/socfpga-dwmac.txt +F: drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c + ARM/SOCFPGA EDAC BINDINGS M: Matthew Gerlach S: Maintained -- 2.51.0 From 91037037ee3d611ce17f39d75f79c7de394b122a Mon Sep 17 00:00:00 2001 From: Henry Martin Date: Fri, 18 Apr 2025 10:38:13 +0800 Subject: [PATCH 12/16] net/mlx5: Fix null-ptr-deref in mlx5_create_{inner_,}ttc_table() Add NULL check for mlx5_get_flow_namespace() returns in mlx5_create_inner_ttc_table() and mlx5_create_ttc_table() to prevent NULL pointer dereference. Fixes: 137f3d50ad2a ("net/mlx5: Support matching on l4_type for ttc_table") Signed-off-by: Henry Martin Reviewed-by: Mark Bloch Reviewed-by: Michal Swiatkowski Link: https://patch.msgid.link/20250418023814.71789-2-bsdhenrymartin@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c index eb3bd9c7f66e..228d0f6570d4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c @@ -655,6 +655,11 @@ struct mlx5_ttc_table *mlx5_create_inner_ttc_table(struct mlx5_core_dev *dev, } ns = mlx5_get_flow_namespace(dev, params->ns_type); + if (!ns) { + kvfree(ttc); + return ERR_PTR(-EOPNOTSUPP); + } + groups = use_l4_type ? &inner_ttc_groups[TTC_GROUPS_USE_L4_TYPE] : &inner_ttc_groups[TTC_GROUPS_DEFAULT]; @@ -728,6 +733,11 @@ struct mlx5_ttc_table *mlx5_create_ttc_table(struct mlx5_core_dev *dev, } ns = mlx5_get_flow_namespace(dev, params->ns_type); + if (!ns) { + kvfree(ttc); + return ERR_PTR(-EOPNOTSUPP); + } + groups = use_l4_type ? &ttc_groups[TTC_GROUPS_USE_L4_TYPE] : &ttc_groups[TTC_GROUPS_DEFAULT]; -- 2.51.0 From fa8fd315127ca48c65e7e6692a84ffcf3d07168e Mon Sep 17 00:00:00 2001 From: Henry Martin Date: Fri, 18 Apr 2025 10:38:14 +0800 Subject: [PATCH 13/16] net/mlx5: Move ttc allocation after switch case to prevent leaks Relocate the memory allocation for ttc table after the switch statement that validates params->ns_type in both mlx5_create_inner_ttc_table() and mlx5_create_ttc_table(). This ensures memory is only allocated after confirming valid input, eliminating potential memory leaks when invalid ns_type cases occur. Fixes: 137f3d50ad2a ("net/mlx5: Support matching on l4_type for ttc_table") Signed-off-by: Henry Martin Reviewed-by: Michal Swiatkowski Reviewed-by: Mark Bloch Link: https://patch.msgid.link/20250418023814.71789-3-bsdhenrymartin@gmail.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c index 228d0f6570d4..ca9ecec358b2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/fs_ttc.c @@ -637,10 +637,6 @@ struct mlx5_ttc_table *mlx5_create_inner_ttc_table(struct mlx5_core_dev *dev, bool use_l4_type; int err; - ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL); - if (!ttc) - return ERR_PTR(-ENOMEM); - switch (params->ns_type) { case MLX5_FLOW_NAMESPACE_PORT_SEL: use_l4_type = MLX5_CAP_GEN_2(dev, pcc_ifa2) && @@ -654,6 +650,10 @@ struct mlx5_ttc_table *mlx5_create_inner_ttc_table(struct mlx5_core_dev *dev, return ERR_PTR(-EINVAL); } + ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL); + if (!ttc) + return ERR_PTR(-ENOMEM); + ns = mlx5_get_flow_namespace(dev, params->ns_type); if (!ns) { kvfree(ttc); @@ -715,10 +715,6 @@ struct mlx5_ttc_table *mlx5_create_ttc_table(struct mlx5_core_dev *dev, bool use_l4_type; int err; - ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL); - if (!ttc) - return ERR_PTR(-ENOMEM); - switch (params->ns_type) { case MLX5_FLOW_NAMESPACE_PORT_SEL: use_l4_type = MLX5_CAP_GEN_2(dev, pcc_ifa2) && @@ -732,6 +728,10 @@ struct mlx5_ttc_table *mlx5_create_ttc_table(struct mlx5_core_dev *dev, return ERR_PTR(-EINVAL); } + ttc = kvzalloc(sizeof(*ttc), GFP_KERNEL); + if (!ttc) + return ERR_PTR(-ENOMEM); + ns = mlx5_get_flow_namespace(dev, params->ns_type); if (!ns) { kvfree(ttc); -- 2.51.0 From d3153c3b42707d26c81083b426f2ef0951bce545 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Thu, 17 Apr 2025 18:53:17 -0700 Subject: [PATCH 14/16] net: fix the missing unlock for detached devices The combined condition was left as is when we converted from __dev_get_by_index() to netdev_get_by_index_lock(). There was no need to undo anything with the former, for the latter we need an unlock. Fixes: 1d22d3060b9b ("net: drop rtnl_lock for queue_mgmt operations") Reviewed-by: Mina Almasry Link: https://patch.msgid.link/20250418015317.1954107-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/core/netdev-genl.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c index 5d7af50fe702..230743bdbb14 100644 --- a/net/core/netdev-genl.c +++ b/net/core/netdev-genl.c @@ -861,14 +861,17 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct genl_info *info) mutex_lock(&priv->lock); + err = 0; netdev = netdev_get_by_index_lock(genl_info_net(info), ifindex); - if (!netdev || !netif_device_present(netdev)) { + if (!netdev) { err = -ENODEV; goto err_unlock_sock; } - - if (!netdev_need_ops_lock(netdev)) { + if (!netif_device_present(netdev)) + err = -ENODEV; + else if (!netdev_need_ops_lock(netdev)) err = -EOPNOTSUPP; + if (err) { NL_SET_BAD_ATTR(info->extack, info->attrs[NETDEV_A_DEV_IFINDEX]); goto err_unlock; -- 2.51.0 From c083da15f06c808c44444bfcef3c939a07ad394b Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 17 Apr 2025 11:15:01 +0100 Subject: [PATCH 15/16] MAINTAINERS: Add ism.h to S390 NETWORKING DRIVERS ism.h appears to be part of s390 networking drivers so add it to the corresponding section in MAINTAINERS. This aids developers, and tooling such as get_maintainer.pl alike to CC patches to the appropriate people and mailing lists. And is in keeping with an ongoing effort for NETWORKING entries in MAINTAINERS to more accurately reflect the way code is maintained. Signed-off-by: Simon Horman Link: https://patch.msgid.link/20250417-ism-maint-v1-1-b001be8545ce@kernel.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index 067f72e8af10..aef5f9a8acf5 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -21270,6 +21270,7 @@ L: linux-s390@vger.kernel.org L: netdev@vger.kernel.org S: Supported F: drivers/s390/net/ +F: include/linux/ism.h S390 PCI SUBSYSTEM M: Niklas Schnelle -- 2.51.0 From e00c1517f2bc73186a18ac2cb1d6c5fee7e95239 Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 17 Apr 2025 11:15:02 +0100 Subject: [PATCH 16/16] MAINTAINERS: Add s390 networking drivers to NETWORKING DRIVERS These files are already correctly covered by the S390 NETWORKING DRIVERS section. In practice commits for these drivers feed into the Networking subsystem. So it seems appropriate to also list them under NETWORKING DRIVERS. This aids developers, and tooling such as get_maintainer.pl alike to CC patches to all the appropriate people and mailing lists. And is in keeping with an ongoing effort for NETWORKING entries in MAINTAINERS to more accurately reflect the way code is maintained. Signed-off-by: Simon Horman Link: https://patch.msgid.link/20250417-ism-maint-v1-2-b001be8545ce@kernel.org Signed-off-by: Jakub Kicinski --- MAINTAINERS | 2 ++ 1 file changed, 2 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index aef5f9a8acf5..d980da72ec60 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16770,6 +16770,7 @@ F: Documentation/networking/net_cachelines/net_device.rst F: drivers/connector/ F: drivers/net/ F: drivers/ptp/ +F: drivers/s390/net/ F: include/dt-bindings/net/ F: include/linux/cn_proc.h F: include/linux/etherdevice.h @@ -16779,6 +16780,7 @@ F: include/linux/fddidevice.h F: include/linux/hippidevice.h F: include/linux/if_* F: include/linux/inetdevice.h +F: include/linux/ism.h F: include/linux/netdev* F: include/linux/platform_data/wiznet.h F: include/uapi/linux/cn_proc.h -- 2.51.0