From 5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Feb 2023 18:24:12 -0800 Subject: Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ... --- drivers/misc/habanalabs/common/state_dump.c | 718 ++++++++++++++++++++++++++++ 1 file changed, 718 insertions(+) create mode 100644 drivers/misc/habanalabs/common/state_dump.c (limited to 'drivers/misc/habanalabs/common/state_dump.c') diff --git a/drivers/misc/habanalabs/common/state_dump.c b/drivers/misc/habanalabs/common/state_dump.c new file mode 100644 index 000000000..74726907c --- /dev/null +++ b/drivers/misc/habanalabs/common/state_dump.c @@ -0,0 +1,718 @@ +// SPDX-License-Identifier: GPL-2.0 + +/* + * Copyright 2021 HabanaLabs, Ltd. + * All Rights Reserved. + */ + +#include +#include +#include "habanalabs.h" + +/** + * hl_format_as_binary - helper function, format an integer as binary + * using supplied scratch buffer + * @buf: the buffer to use + * @buf_len: buffer capacity + * @n: number to format + * + * Returns pointer to buffer + */ +char *hl_format_as_binary(char *buf, size_t buf_len, u32 n) +{ + int i; + u32 bit; + bool leading0 = true; + char *wrptr = buf; + + if (buf_len > 0 && buf_len < 3) { + *wrptr = '\0'; + return buf; + } + + wrptr[0] = '0'; + wrptr[1] = 'b'; + wrptr += 2; + /* Remove 3 characters from length for '0b' and '\0' termination */ + buf_len -= 3; + + for (i = 0; i < sizeof(n) * BITS_PER_BYTE && buf_len; ++i, n <<= 1) { + /* Writing bit calculation in one line would cause a false + * positive static code analysis error, so splitting. + */ + bit = n & (1 << (sizeof(n) * BITS_PER_BYTE - 1)); + bit = !!bit; + leading0 &= !bit; + if (!leading0) { + *wrptr = '0' + bit; + ++wrptr; + } + } + + *wrptr = '\0'; + + return buf; +} + +/** + * resize_to_fit - helper function, resize buffer to fit given amount of data + * @buf: destination buffer double pointer + * @size: pointer to the size container + * @desired_size: size the buffer must contain + * + * Returns 0 on success or error code on failure. + * On success, the size of buffer is at least desired_size. Buffer is allocated + * via vmalloc and must be freed with vfree. + */ +static int resize_to_fit(char **buf, size_t *size, size_t desired_size) +{ + char *resized_buf; + size_t new_size; + + if (*size >= desired_size) + return 0; + + /* Not enough space to print all, have to resize */ + new_size = max_t(size_t, PAGE_SIZE, round_up(desired_size, PAGE_SIZE)); + resized_buf = vmalloc(new_size); + if (!resized_buf) + return -ENOMEM; + memcpy(resized_buf, *buf, *size); + vfree(*buf); + *buf = resized_buf; + *size = new_size; + + return 1; +} + +/** + * hl_snprintf_resize() - print formatted data to buffer, resize as needed + * @buf: buffer double pointer, to be written to and resized, must be either + * NULL or allocated with vmalloc. + * @size: current size of the buffer + * @offset: current offset to write to + * @format: format of the data + * + * This function will write formatted data into the buffer. If buffer is not + * large enough, it will be resized using vmalloc. Size may be modified if the + * buffer was resized, offset will be advanced by the number of bytes written + * not including the terminating character + * + * Returns 0 on success or error code on failure + * + * Note that the buffer has to be manually released using vfree. + */ +int hl_snprintf_resize(char **buf, size_t *size, size_t *offset, + const char *format, ...) +{ + va_list args; + size_t length; + int rc; + + if (*buf == NULL && (*size != 0 || *offset != 0)) + return -EINVAL; + + va_start(args, format); + length = vsnprintf(*buf + *offset, *size - *offset, format, args); + va_end(args); + + rc = resize_to_fit(buf, size, *offset + length + 1); + if (rc < 0) + return rc; + else if (rc > 0) { + /* Resize was needed, write again */ + va_start(args, format); + length = vsnprintf(*buf + *offset, *size - *offset, format, + args); + va_end(args); + } + + *offset += length; + + return 0; +} + +/** + * hl_sync_engine_to_string - convert engine type enum to string literal + * @engine_type: engine type (TPC/MME/DMA) + * + * Return the resolved string literal + */ +const char *hl_sync_engine_to_string(enum hl_sync_engine_type engine_type) +{ + switch (engine_type) { + case ENGINE_DMA: + return "DMA"; + case ENGINE_MME: + return "MME"; + case ENGINE_TPC: + return "TPC"; + } + return "Invalid Engine Type"; +} + +/** + * hl_print_resize_sync_engine - helper function, format engine name and ID + * using hl_snprintf_resize + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + * @engine_type: engine type (TPC/MME/DMA) + * @engine_id: engine numerical id + * + * Returns 0 on success or error code on failure + */ +static int hl_print_resize_sync_engine(char **buf, size_t *size, size_t *offset, + enum hl_sync_engine_type engine_type, + u32 engine_id) +{ + return hl_snprintf_resize(buf, size, offset, "%s%u", + hl_sync_engine_to_string(engine_type), engine_id); +} + +/** + * hl_state_dump_get_sync_name - transform sync object id to name if available + * @hdev: pointer to the device + * @sync_id: sync object id + * + * Returns a name literal or NULL if not resolved. + * Note: returning NULL shall not be considered as a failure, as not all + * sync objects are named. + */ +const char *hl_state_dump_get_sync_name(struct hl_device *hdev, u32 sync_id) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + struct hl_hw_obj_name_entry *entry; + + hash_for_each_possible(sds->so_id_to_str_tb, entry, + node, sync_id) + if (sync_id == entry->id) + return entry->name; + + return NULL; +} + +/** + * hl_state_dump_get_monitor_name - transform monitor object dump to monitor + * name if available + * @hdev: pointer to the device + * @mon: monitor state dump + * + * Returns a name literal or NULL if not resolved. + * Note: returning NULL shall not be considered as a failure, as not all + * monitors are named. + */ +const char *hl_state_dump_get_monitor_name(struct hl_device *hdev, + struct hl_mon_state_dump *mon) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + struct hl_hw_obj_name_entry *entry; + + hash_for_each_possible(sds->monitor_id_to_str_tb, + entry, node, mon->id) + if (mon->id == entry->id) + return entry->name; + + return NULL; +} + +/** + * hl_state_dump_free_sync_to_engine_map - free sync object to engine map + * @map: sync object to engine map + * + * Note: generic free implementation, the allocation is implemented per ASIC. + */ +void hl_state_dump_free_sync_to_engine_map(struct hl_sync_to_engine_map *map) +{ + struct hl_sync_to_engine_map_entry *entry; + struct hlist_node *tmp_node; + int i; + + hash_for_each_safe(map->tb, i, tmp_node, entry, node) { + hash_del(&entry->node); + kfree(entry); + } +} + +/** + * hl_state_dump_get_sync_to_engine - transform sync_id to + * hl_sync_to_engine_map_entry if available for current id + * @map: sync object to engine map + * @sync_id: sync object id + * + * Returns the translation entry if found or NULL if not. + * Note, returned NULL shall not be considered as a failure as the map + * does not cover all possible, it is a best effort sync ids. + */ +static struct hl_sync_to_engine_map_entry * +hl_state_dump_get_sync_to_engine(struct hl_sync_to_engine_map *map, u32 sync_id) +{ + struct hl_sync_to_engine_map_entry *entry; + + hash_for_each_possible(map->tb, entry, node, sync_id) + if (entry->sync_id == sync_id) + return entry; + return NULL; +} + +/** + * hl_state_dump_read_sync_objects - read sync objects array + * @hdev: pointer to the device + * @index: sync manager block index starting with E_N + * + * Returns array of size SP_SYNC_OBJ_AMOUNT on success or NULL on failure + */ +static u32 *hl_state_dump_read_sync_objects(struct hl_device *hdev, u32 index) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + u32 *sync_objects; + s64 base_addr; /* Base addr can be negative */ + int i; + + base_addr = sds->props[SP_SYNC_OBJ_BASE_ADDR] + + sds->props[SP_NEXT_SYNC_OBJ_ADDR] * index; + + sync_objects = vmalloc(sds->props[SP_SYNC_OBJ_AMOUNT] * sizeof(u32)); + if (!sync_objects) + return NULL; + + for (i = 0; i < sds->props[SP_SYNC_OBJ_AMOUNT]; ++i) + sync_objects[i] = RREG32(base_addr + i * sizeof(u32)); + + return sync_objects; +} + +/** + * hl_state_dump_free_sync_objects - free sync objects array allocated by + * hl_state_dump_read_sync_objects + * @sync_objects: sync objects array + */ +static void hl_state_dump_free_sync_objects(u32 *sync_objects) +{ + vfree(sync_objects); +} + + +/** + * hl_state_dump_print_syncs_single_block - print active sync objects on a + * single block + * @hdev: pointer to the device + * @index: sync manager block index starting with E_N + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + * @map: sync engines names map + * + * Returns 0 on success or error code on failure + */ +static int +hl_state_dump_print_syncs_single_block(struct hl_device *hdev, u32 index, + char **buf, size_t *size, size_t *offset, + struct hl_sync_to_engine_map *map) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + const char *sync_name; + u32 *sync_objects = NULL; + int rc = 0, i; + + if (sds->sync_namager_names) { + rc = hl_snprintf_resize( + buf, size, offset, "%s\n", + sds->sync_namager_names[index]); + if (rc) + goto out; + } + + sync_objects = hl_state_dump_read_sync_objects(hdev, index); + if (!sync_objects) { + rc = -ENOMEM; + goto out; + } + + for (i = 0; i < sds->props[SP_SYNC_OBJ_AMOUNT]; ++i) { + struct hl_sync_to_engine_map_entry *entry; + u64 sync_object_addr; + + if (!sync_objects[i]) + continue; + + sync_object_addr = sds->props[SP_SYNC_OBJ_BASE_ADDR] + + sds->props[SP_NEXT_SYNC_OBJ_ADDR] * index + + i * sizeof(u32); + + rc = hl_snprintf_resize(buf, size, offset, "sync id: %u", i); + if (rc) + goto free_sync_objects; + sync_name = hl_state_dump_get_sync_name(hdev, i); + if (sync_name) { + rc = hl_snprintf_resize(buf, size, offset, " %s", + sync_name); + if (rc) + goto free_sync_objects; + } + rc = hl_snprintf_resize(buf, size, offset, ", value: %u", + sync_objects[i]); + if (rc) + goto free_sync_objects; + + /* Append engine string */ + entry = hl_state_dump_get_sync_to_engine(map, + (u32)sync_object_addr); + if (entry) { + rc = hl_snprintf_resize(buf, size, offset, + ", Engine: "); + if (rc) + goto free_sync_objects; + rc = hl_print_resize_sync_engine(buf, size, offset, + entry->engine_type, + entry->engine_id); + if (rc) + goto free_sync_objects; + } + + rc = hl_snprintf_resize(buf, size, offset, "\n"); + if (rc) + goto free_sync_objects; + } + +free_sync_objects: + hl_state_dump_free_sync_objects(sync_objects); +out: + return rc; +} + +/** + * hl_state_dump_print_syncs - print active sync objects + * @hdev: pointer to the device + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + * + * Returns 0 on success or error code on failure + */ +static int hl_state_dump_print_syncs(struct hl_device *hdev, + char **buf, size_t *size, + size_t *offset) + +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + struct hl_sync_to_engine_map *map; + u32 index; + int rc = 0; + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) + return -ENOMEM; + + rc = sds->funcs.gen_sync_to_engine_map(hdev, map); + if (rc) + goto free_map_mem; + + rc = hl_snprintf_resize(buf, size, offset, "Non zero sync objects:\n"); + if (rc) + goto out; + + if (sds->sync_namager_names) { + for (index = 0; sds->sync_namager_names[index]; ++index) { + rc = hl_state_dump_print_syncs_single_block( + hdev, index, buf, size, offset, map); + if (rc) + goto out; + } + } else { + for (index = 0; index < sds->props[SP_NUM_CORES]; ++index) { + rc = hl_state_dump_print_syncs_single_block( + hdev, index, buf, size, offset, map); + if (rc) + goto out; + } + } + +out: + hl_state_dump_free_sync_to_engine_map(map); +free_map_mem: + kfree(map); + + return rc; +} + +/** + * hl_state_dump_alloc_read_sm_block_monitors - read monitors for a specific + * block + * @hdev: pointer to the device + * @index: sync manager block index starting with E_N + * + * Returns an array of monitor data of size SP_MONITORS_AMOUNT or NULL + * on error + */ +static struct hl_mon_state_dump * +hl_state_dump_alloc_read_sm_block_monitors(struct hl_device *hdev, u32 index) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + struct hl_mon_state_dump *monitors; + s64 base_addr; /* Base addr can be negative */ + int i; + + monitors = vmalloc(sds->props[SP_MONITORS_AMOUNT] * + sizeof(struct hl_mon_state_dump)); + if (!monitors) + return NULL; + + base_addr = sds->props[SP_NEXT_SYNC_OBJ_ADDR] * index; + + for (i = 0; i < sds->props[SP_MONITORS_AMOUNT]; ++i) { + monitors[i].id = i; + monitors[i].wr_addr_low = + RREG32(base_addr + sds->props[SP_MON_OBJ_WR_ADDR_LOW] + + i * sizeof(u32)); + + monitors[i].wr_addr_high = + RREG32(base_addr + sds->props[SP_MON_OBJ_WR_ADDR_HIGH] + + i * sizeof(u32)); + + monitors[i].wr_data = + RREG32(base_addr + sds->props[SP_MON_OBJ_WR_DATA] + + i * sizeof(u32)); + + monitors[i].arm_data = + RREG32(base_addr + sds->props[SP_MON_OBJ_ARM_DATA] + + i * sizeof(u32)); + + monitors[i].status = + RREG32(base_addr + sds->props[SP_MON_OBJ_STATUS] + + i * sizeof(u32)); + } + + return monitors; +} + +/** + * hl_state_dump_free_monitors - free the monitors structure + * @monitors: monitors array created with + * hl_state_dump_alloc_read_sm_block_monitors + */ +static void hl_state_dump_free_monitors(struct hl_mon_state_dump *monitors) +{ + vfree(monitors); +} + +/** + * hl_state_dump_print_monitors_single_block - print active monitors on a + * single block + * @hdev: pointer to the device + * @index: sync manager block index starting with E_N + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + * + * Returns 0 on success or error code on failure + */ +static int hl_state_dump_print_monitors_single_block(struct hl_device *hdev, + u32 index, + char **buf, size_t *size, + size_t *offset) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + struct hl_mon_state_dump *monitors = NULL; + int rc = 0, i; + + if (sds->sync_namager_names) { + rc = hl_snprintf_resize( + buf, size, offset, "%s\n", + sds->sync_namager_names[index]); + if (rc) + goto out; + } + + monitors = hl_state_dump_alloc_read_sm_block_monitors(hdev, index); + if (!monitors) { + rc = -ENOMEM; + goto out; + } + + for (i = 0; i < sds->props[SP_MONITORS_AMOUNT]; ++i) { + if (!(sds->funcs.monitor_valid(&monitors[i]))) + continue; + + /* Monitor is valid, dump it */ + rc = sds->funcs.print_single_monitor(buf, size, offset, hdev, + &monitors[i]); + if (rc) + goto free_monitors; + + hl_snprintf_resize(buf, size, offset, "\n"); + } + +free_monitors: + hl_state_dump_free_monitors(monitors); +out: + return rc; +} + +/** + * hl_state_dump_print_monitors - print active monitors + * @hdev: pointer to the device + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + * + * Returns 0 on success or error code on failure + */ +static int hl_state_dump_print_monitors(struct hl_device *hdev, + char **buf, size_t *size, + size_t *offset) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + u32 index; + int rc = 0; + + rc = hl_snprintf_resize(buf, size, offset, + "Valid (armed) monitor objects:\n"); + if (rc) + goto out; + + if (sds->sync_namager_names) { + for (index = 0; sds->sync_namager_names[index]; ++index) { + rc = hl_state_dump_print_monitors_single_block( + hdev, index, buf, size, offset); + if (rc) + goto out; + } + } else { + for (index = 0; index < sds->props[SP_NUM_CORES]; ++index) { + rc = hl_state_dump_print_monitors_single_block( + hdev, index, buf, size, offset); + if (rc) + goto out; + } + } + +out: + return rc; +} + +/** + * hl_state_dump_print_engine_fences - print active fences for a specific + * engine + * @hdev: pointer to the device + * @engine_type: engine type to use + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + */ +static int +hl_state_dump_print_engine_fences(struct hl_device *hdev, + enum hl_sync_engine_type engine_type, + char **buf, size_t *size, size_t *offset) +{ + struct hl_state_dump_specs *sds = &hdev->state_dump_specs; + int rc = 0, i, n_fences; + u64 base_addr, next_fence; + + switch (engine_type) { + case ENGINE_TPC: + n_fences = sds->props[SP_NUM_OF_TPC_ENGINES]; + base_addr = sds->props[SP_TPC0_CMDQ]; + next_fence = sds->props[SP_NEXT_TPC]; + break; + case ENGINE_MME: + n_fences = sds->props[SP_NUM_OF_MME_ENGINES]; + base_addr = sds->props[SP_MME_CMDQ]; + next_fence = sds->props[SP_NEXT_MME]; + break; + case ENGINE_DMA: + n_fences = sds->props[SP_NUM_OF_DMA_ENGINES]; + base_addr = sds->props[SP_DMA_CMDQ]; + next_fence = sds->props[SP_DMA_QUEUES_OFFSET]; + break; + default: + return -EINVAL; + } + for (i = 0; i < n_fences; ++i) { + rc = sds->funcs.print_fences_single_engine( + hdev, + base_addr + next_fence * i + + sds->props[SP_FENCE0_CNT_OFFSET], + base_addr + next_fence * i + + sds->props[SP_CP_STS_OFFSET], + engine_type, i, buf, size, offset); + if (rc) + goto out; + } +out: + return rc; +} + +/** + * hl_state_dump_print_fences - print active fences + * @hdev: pointer to the device + * @buf: destination buffer double pointer to be used with hl_snprintf_resize + * @size: pointer to the size container + * @offset: pointer to the offset container + */ +static int hl_state_dump_print_fences(struct hl_device *hdev, char **buf, + size_t *size, size_t *offset) +{ + int rc = 0; + + rc = hl_snprintf_resize(buf, size, offset, "Valid (armed) fences:\n"); + if (rc) + goto out; + + rc = hl_state_dump_print_engine_fences(hdev, ENGINE_TPC, buf, size, offset); + if (rc) + goto out; + + rc = hl_state_dump_print_engine_fences(hdev, ENGINE_MME, buf, size, offset); + if (rc) + goto out; + + rc = hl_state_dump_print_engine_fences(hdev, ENGINE_DMA, buf, size, offset); + if (rc) + goto out; + +out: + return rc; +} + +/** + * hl_state_dump() - dump system state + * @hdev: pointer to device structure + */ +int hl_state_dump(struct hl_device *hdev) +{ + char *buf = NULL; + size_t offset = 0, size = 0; + int rc; + + rc = hl_snprintf_resize(&buf, &size, &offset, + "Timestamp taken on: %llu\n\n", + ktime_to_ns(ktime_get())); + if (rc) + goto err; + + rc = hl_state_dump_print_syncs(hdev, &buf, &size, &offset); + if (rc) + goto err; + + hl_snprintf_resize(&buf, &size, &offset, "\n"); + + rc = hl_state_dump_print_monitors(hdev, &buf, &size, &offset); + if (rc) + goto err; + + hl_snprintf_resize(&buf, &size, &offset, "\n"); + + rc = hl_state_dump_print_fences(hdev, &buf, &size, &offset); + if (rc) + goto err; + + hl_snprintf_resize(&buf, &size, &offset, "\n"); + + hl_debugfs_set_state_dump(hdev, buf, size); + + return 0; +err: + vfree(buf); + return rc; +} -- cgit v1.2.3