From patchwork Wed Dec 11 22:32:13 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tony Nguyen X-Patchwork-Id: 13904297 X-Patchwork-Delegate: kuba@kernel.org Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.14]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 81C8F204594 for ; Wed, 11 Dec 2024 22:32:43 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=192.198.163.14 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1733956365; cv=none; b=tY8BIIHpnguue6nE8tBe5n5RjqokK9EREzU5vXy0hXGAaaF4RJXW6y25eC4C3XQrqC2f/lnGsvem9aCnnEUDD9zLYEuPw518+dHAIK5Z75w81amQH1BwwRVY8Q9hpI6fhIMs8e1eGKvnvrd50/Ex6yAjyj3ygWkdz4rnTc6yVUk= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1733956365; c=relaxed/simple; bh=fsWPt0xSRolST8lDpW6tj5oYSZrfQZOKKYXI4FeBAls=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=nQxMf+nMkMs8si6JXn1RRUADaMFW/mhdpXKXeoKNssh1yk6eqbX5U6f+VcooAgLDXDvdDwhPa3hzF4VzKpew2K/EiI+/Lb6VgC+8VVzTiMo3VIuTmLphi1LFKaXwiaaIHjIy4wO7Jt3M11T/yiU1CpqjizAImQLS8nk5GUkc/zk= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com; spf=pass smtp.mailfrom=intel.com; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b=gcQzlakR; arc=none smtp.client-ip=192.198.163.14 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=intel.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=intel.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=intel.com header.i=@intel.com header.b="gcQzlakR" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1733956363; x=1765492363; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=fsWPt0xSRolST8lDpW6tj5oYSZrfQZOKKYXI4FeBAls=; b=gcQzlakRKy+WtsKEYYxwwCx/qh+Rl6Hmhg6lNX4JTPoD5nVIzHO6Ikjw 5W8YRJVTRev6V4mU2qJuC1VcfgCUmgXY0LymbwG6QfA15BYiQm8g3RwJx P7qz6RT+ati8RM/H0UQCFT0BAd6DP6p7/uJMHCRbtMof5sMUlj9jb4kSD fqjW7WGxmh7Cdia7DF39XsQQUXGeepWDNdO1OIehPti0fDwCdXJTInFUj Zc7ipD6Pxqg+gOUUAB/5fqQr8Kwtqnc39vqrnOEGe43wfq2Vpm+2aM/Rq q1XQPrFKpx8kGv/bQyg/OrMBiLQPwEvoD2mv97g4PztItheD8IJNlZdBZ w==; X-CSE-ConnectionGUID: R6JaGaO1Qj+zeoqxDaK2ug== X-CSE-MsgGUID: vwk1pNnfQJypEdKNpLPmyA== X-IronPort-AV: E=McAfee;i="6700,10204,11283"; a="34599641" X-IronPort-AV: E=Sophos;i="6.12,226,1728975600"; d="scan'208";a="34599641" Received: from fmviesa008.fm.intel.com ([10.60.135.148]) by fmvoesa108.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 11 Dec 2024 14:32:40 -0800 X-CSE-ConnectionGUID: GyblmMf+QLO2cysdZhJOrQ== X-CSE-MsgGUID: kNkMt0R0Ti660GsxSoj6qg== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="6.12,226,1728975600"; d="scan'208";a="96192940" Received: from anguy11-upstream.jf.intel.com ([10.166.9.133]) by fmviesa008.fm.intel.com with ESMTP; 11 Dec 2024 14:32:40 -0800 From: Tony Nguyen To: davem@davemloft.net, kuba@kernel.org, pabeni@redhat.com, edumazet@google.com, andrew+netdev@lunn.ch, netdev@vger.kernel.org Cc: Przemek Kitszel , anthony.l.nguyen@intel.com, wojciech.drewek@intel.com, mateusz.polchlopek@intel.com, joe@perches.com, horms@kernel.org, jiri@resnulli.us, apw@canonical.com, lukas.bulwahn@gmail.com, dwaipayanray1@gmail.com, Igor Bagnucki , Pucha Himasekhar Reddy Subject: [PATCH net-next 5/7] ice: add Tx hang devlink health reporter Date: Wed, 11 Dec 2024 14:32:13 -0800 Message-ID: <20241211223231.397203-6-anthony.l.nguyen@intel.com> X-Mailer: git-send-email 2.42.0 In-Reply-To: <20241211223231.397203-1-anthony.l.nguyen@intel.com> References: <20241211223231.397203-1-anthony.l.nguyen@intel.com> Precedence: bulk X-Mailing-List: netdev@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-Patchwork-Delegate: kuba@kernel.org From: Przemek Kitszel Add Tx hang devlink health reporter, see struct ice_tx_hang_event to see what is reported. Subsequent commits will extend it by more info, for now it dumps descriptors with little metadata. Reviewed-by: Igor Bagnucki Reviewed-by: Wojciech Drewek Co-developed-by: Mateusz Polchlopek Signed-off-by: Mateusz Polchlopek Tested-by: Pucha Himasekhar Reddy (A Contingent worker at Intel) Signed-off-by: Przemek Kitszel Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/Makefile | 1 + .../net/ethernet/intel/ice/devlink/health.c | 188 ++++++++++++++++++ .../net/ethernet/intel/ice/devlink/health.h | 48 +++++ drivers/net/ethernet/intel/ice/ice.h | 2 + drivers/net/ethernet/intel/ice/ice_main.c | 18 +- 5 files changed, 252 insertions(+), 5 deletions(-) create mode 100644 drivers/net/ethernet/intel/ice/devlink/health.c create mode 100644 drivers/net/ethernet/intel/ice/devlink/health.h diff --git a/drivers/net/ethernet/intel/ice/Makefile b/drivers/net/ethernet/intel/ice/Makefile index 56aa23aee472..9e0d9f710441 100644 --- a/drivers/net/ethernet/intel/ice/Makefile +++ b/drivers/net/ethernet/intel/ice/Makefile @@ -32,6 +32,7 @@ ice-y := ice_main.o \ ice_parser_rt.o \ ice_idc.o \ devlink/devlink.o \ + devlink/health.o \ devlink/port.o \ ice_sf_eth.o \ ice_sf_vsi_vlan_ops.o \ diff --git a/drivers/net/ethernet/intel/ice/devlink/health.c b/drivers/net/ethernet/intel/ice/devlink/health.c new file mode 100644 index 000000000000..b8c5a1c372dc --- /dev/null +++ b/drivers/net/ethernet/intel/ice/devlink/health.c @@ -0,0 +1,188 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2024, Intel Corporation. */ + +#include "health.h" +#include "ice.h" + +#define ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, obj, name) \ + devlink_fmsg_put(fmsg, #name, (obj)->name) + +/** + * ice_devlink_health_report - boilerplate to call given @reporter + * + * @reporter: devlink health reporter to call, do nothing on NULL + * @msg: message to pass up, "event name" is fine + * @priv_ctx: typically some event struct + */ +static void ice_devlink_health_report(struct devlink_health_reporter *reporter, + const char *msg, void *priv_ctx) +{ + if (!reporter) + return; + + /* We do not do auto recovering, so return value of the below function + * will always be 0, thus we do ignore it. + */ + devlink_health_report(reporter, msg, priv_ctx); +} + +/** + * ice_fmsg_put_ptr - put hex value of pointer into fmsg + * + * @fmsg: devlink fmsg under construction + * @name: name to pass + * @ptr: 64 bit value to print as hex and put into fmsg + */ +static void ice_fmsg_put_ptr(struct devlink_fmsg *fmsg, const char *name, + void *ptr) +{ + char buf[sizeof(ptr) * 3]; + + sprintf(buf, "%p", ptr); + devlink_fmsg_put(fmsg, name, buf); +} + +struct ice_tx_hang_event { + u32 head; + u32 intr; + u16 vsi_num; + u16 queue; + u16 next_to_clean; + u16 next_to_use; + struct ice_tx_ring *tx_ring; +}; + +static int ice_tx_hang_reporter_dump(struct devlink_health_reporter *reporter, + struct devlink_fmsg *fmsg, void *priv_ctx, + struct netlink_ext_ack *extack) +{ + struct ice_tx_hang_event *event = priv_ctx; + + if (!event) + return 0; + + devlink_fmsg_obj_nest_start(fmsg); + ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, event, head); + ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, event, intr); + ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, event, vsi_num); + ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, event, queue); + ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, event, next_to_clean); + ICE_DEVLINK_FMSG_PUT_FIELD(fmsg, event, next_to_use); + devlink_fmsg_put(fmsg, "irq-mapping", event->tx_ring->q_vector->name); + ice_fmsg_put_ptr(fmsg, "desc-ptr", event->tx_ring->desc); + ice_fmsg_put_ptr(fmsg, "dma-ptr", (void *)(long)event->tx_ring->dma); + devlink_fmsg_binary_pair_put(fmsg, "desc", event->tx_ring->desc, + event->tx_ring->count * sizeof(struct ice_tx_desc)); + devlink_fmsg_obj_nest_end(fmsg); + + return 0; +} + +void ice_prep_tx_hang_report(struct ice_pf *pf, struct ice_tx_ring *tx_ring, + u16 vsi_num, u32 head, u32 intr) +{ + struct ice_health_tx_hang_buf *buf = &pf->health_reporters.tx_hang_buf; + + buf->tx_ring = tx_ring; + buf->vsi_num = vsi_num; + buf->head = head; + buf->intr = intr; +} + +void ice_report_tx_hang(struct ice_pf *pf) +{ + struct ice_health_tx_hang_buf *buf = &pf->health_reporters.tx_hang_buf; + struct ice_tx_ring *tx_ring = buf->tx_ring; + + struct ice_tx_hang_event ev = { + .head = buf->head, + .intr = buf->intr, + .vsi_num = buf->vsi_num, + .queue = tx_ring->q_index, + .next_to_clean = tx_ring->next_to_clean, + .next_to_use = tx_ring->next_to_use, + .tx_ring = tx_ring, + }; + + ice_devlink_health_report(pf->health_reporters.tx_hang, "Tx hang", &ev); +} + +static struct devlink_health_reporter * +ice_init_devlink_rep(struct ice_pf *pf, + const struct devlink_health_reporter_ops *ops) +{ + struct devlink *devlink = priv_to_devlink(pf); + struct devlink_health_reporter *rep; + const u64 graceful_period = 0; + + rep = devl_health_reporter_create(devlink, ops, graceful_period, pf); + if (IS_ERR(rep)) { + struct device *dev = ice_pf_to_dev(pf); + + dev_err(dev, "failed to create devlink %s health report er", + ops->name); + return NULL; + } + return rep; +} + +#define ICE_DEFINE_HEALTH_REPORTER_OPS(_name) \ + static const struct devlink_health_reporter_ops ice_ ## _name ## _reporter_ops = { \ + .name = #_name, \ + .dump = ice_ ## _name ## _reporter_dump, \ +} + +ICE_DEFINE_HEALTH_REPORTER_OPS(tx_hang); + +/** + * ice_health_init - allocate and init all ice devlink health reporters and + * accompanied data + * + * @pf: PF struct + */ +void ice_health_init(struct ice_pf *pf) +{ + struct ice_health *reps = &pf->health_reporters; + + reps->tx_hang = ice_init_devlink_rep(pf, &ice_tx_hang_reporter_ops); +} + +/** + * ice_deinit_devl_reporter - destroy given devlink health reporter + * @reporter: reporter to destroy + */ +static void ice_deinit_devl_reporter(struct devlink_health_reporter *reporter) +{ + if (reporter) + devl_health_reporter_destroy(reporter); +} + +/** + * ice_health_deinit - deallocate all ice devlink health reporters and + * accompanied data + * + * @pf: PF struct + */ +void ice_health_deinit(struct ice_pf *pf) +{ + ice_deinit_devl_reporter(pf->health_reporters.tx_hang); +} + +static +void ice_health_assign_healthy_state(struct devlink_health_reporter *reporter) +{ + if (reporter) + devlink_health_reporter_state_update(reporter, + DEVLINK_HEALTH_REPORTER_STATE_HEALTHY); +} + +/** + * ice_health_clear - clear devlink health issues after a reset + * @pf: the PF device structure + * + * Mark the PF in healthy state again after a reset has completed. + */ +void ice_health_clear(struct ice_pf *pf) +{ + ice_health_assign_healthy_state(pf->health_reporters.tx_hang); +} diff --git a/drivers/net/ethernet/intel/ice/devlink/health.h b/drivers/net/ethernet/intel/ice/devlink/health.h new file mode 100644 index 000000000000..92e40d790e72 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/devlink/health.h @@ -0,0 +1,48 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2024, Intel Corporation. */ + +#ifndef _HEALTH_H_ +#define _HEALTH_H_ + +#include + +/** + * DOC: health.h + * + * This header file stores everything that is needed for broadly understood + * devlink health mechanism for ice driver. + */ + +struct ice_pf; +struct ice_tx_ring; + +/** + * struct ice_health - stores ice devlink health reporters and accompanied data + * @tx_hang: devlink health reporter for tx_hang event + * @tx_hang_buf: pre-allocated place to put info for Tx hang reporter from + * non-sleeping context + * @tx_ring: ring that the hang occured on + * @head: descriptior head + * @intr: interrupt register value + * @vsi_num: VSI owning the queue that the hang occured on + */ +struct ice_health { + struct devlink_health_reporter *tx_hang; + struct_group_tagged(ice_health_tx_hang_buf, tx_hang_buf, + struct ice_tx_ring *tx_ring; + u32 head; + u32 intr; + u16 vsi_num; + ); +}; + + +void ice_health_init(struct ice_pf *pf); +void ice_health_deinit(struct ice_pf *pf); +void ice_health_clear(struct ice_pf *pf); + +void ice_prep_tx_hang_report(struct ice_pf *pf, struct ice_tx_ring *tx_ring, + u16 vsi_num, u32 head, u32 intr); +void ice_report_tx_hang(struct ice_pf *pf); + +#endif /* _HEALTH_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 2f5d6f974185..71e05d30f0fd 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -78,6 +78,7 @@ #include "ice_irq.h" #include "ice_dpll.h" #include "ice_adapter.h" +#include "devlink/health.h" #define ICE_BAR0 0 #define ICE_REQ_DESC_MULTIPLE 32 @@ -665,6 +666,7 @@ struct ice_pf { struct ice_agg_node vf_agg_node[ICE_MAX_VF_AGG_NODES]; struct ice_dplls dplls; struct device *hwmon_dev; + struct ice_health health_reporters; u8 num_quanta_prof_used; }; diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index d641dd8b8184..316f5109bd3f 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -2364,9 +2364,11 @@ static void ice_service_task(struct work_struct *work) struct ice_pf *pf = container_of(work, struct ice_pf, serv_task); unsigned long start_time = jiffies; - /* subtasks */ + if (pf->health_reporters.tx_hang_buf.tx_ring) { + ice_report_tx_hang(pf); + pf->health_reporters.tx_hang_buf.tx_ring = NULL; + } - /* process reset requests first */ ice_reset_subtask(pf); /* bail if a reset/recovery cycle is pending or rebuild failed */ @@ -5087,6 +5089,7 @@ static int ice_init_devlink(struct ice_pf *pf) return err; ice_devlink_init_regions(pf); + ice_health_init(pf); ice_devlink_register(pf); return 0; @@ -5095,6 +5098,7 @@ static int ice_init_devlink(struct ice_pf *pf) static void ice_deinit_devlink(struct ice_pf *pf) { ice_devlink_unregister(pf); + ice_health_deinit(pf); ice_devlink_destroy_regions(pf); ice_devlink_unregister_params(pf); } @@ -7793,6 +7797,8 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) /* if we get here, reset flow is successful */ clear_bit(ICE_RESET_FAILED, pf->state); + ice_health_clear(pf); + ice_plug_aux_dev(pf); if (ice_is_feature_supported(pf, ICE_F_SRIOV_LAG)) ice_lag_rebuild(pf); @@ -8283,16 +8289,18 @@ void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) if (tx_ring) { struct ice_hw *hw = &pf->hw; - u32 head, val = 0; + u32 head, intr = 0; head = FIELD_GET(QTX_COMM_HEAD_HEAD_M, rd32(hw, QTX_COMM_HEAD(vsi->txq_map[txqueue]))); /* Read interrupt register */ - val = rd32(hw, GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); + intr = rd32(hw, GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %u, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n", vsi->vsi_num, txqueue, tx_ring->next_to_clean, - head, tx_ring->next_to_use, val); + head, tx_ring->next_to_use, intr); + + ice_prep_tx_hang_report(pf, tx_ring, vsi->vsi_num, head, intr); } pf->tx_timeout_last_recovery = jiffies;