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/sgi-gru/grutlbpurge.c | 318 +++++++++++++++++++++++++++++++++++++ 1 file changed, 318 insertions(+) create mode 100644 drivers/misc/sgi-gru/grutlbpurge.c (limited to 'drivers/misc/sgi-gru/grutlbpurge.c') diff --git a/drivers/misc/sgi-gru/grutlbpurge.c b/drivers/misc/sgi-gru/grutlbpurge.c new file mode 100644 index 000000000..10921cd26 --- /dev/null +++ b/drivers/misc/sgi-gru/grutlbpurge.c @@ -0,0 +1,318 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* + * SN Platform GRU Driver + * + * MMUOPS callbacks + TLB flushing + * + * This file handles emu notifier callbacks from the core kernel. The callbacks + * are used to update the TLB in the GRU as a result of changes in the + * state of a process address space. This file also handles TLB invalidates + * from the GRU driver. + * + * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include "gru.h" +#include "grutables.h" +#include + +#define gru_random() get_cycles() + +/* ---------------------------------- TLB Invalidation functions -------- + * get_tgh_handle + * + * Find a TGH to use for issuing a TLB invalidate. For GRUs that are on the + * local blade, use a fixed TGH that is a function of the blade-local cpu + * number. Normally, this TGH is private to the cpu & no contention occurs for + * the TGH. For offblade GRUs, select a random TGH in the range above the + * private TGHs. A spinlock is required to access this TGH & the lock must be + * released when the invalidate is completes. This sucks, but it is the best we + * can do. + * + * Note that the spinlock is IN the TGH handle so locking does not involve + * additional cache lines. + * + */ +static inline int get_off_blade_tgh(struct gru_state *gru) +{ + int n; + + n = GRU_NUM_TGH - gru->gs_tgh_first_remote; + n = gru_random() % n; + n += gru->gs_tgh_first_remote; + return n; +} + +static inline int get_on_blade_tgh(struct gru_state *gru) +{ + return uv_blade_processor_id() >> gru->gs_tgh_local_shift; +} + +static struct gru_tlb_global_handle *get_lock_tgh_handle(struct gru_state + *gru) +{ + struct gru_tlb_global_handle *tgh; + int n; + + preempt_disable(); + if (uv_numa_blade_id() == gru->gs_blade_id) + n = get_on_blade_tgh(gru); + else + n = get_off_blade_tgh(gru); + tgh = get_tgh_by_index(gru, n); + lock_tgh_handle(tgh); + + return tgh; +} + +static void get_unlock_tgh_handle(struct gru_tlb_global_handle *tgh) +{ + unlock_tgh_handle(tgh); + preempt_enable(); +} + +/* + * gru_flush_tlb_range + * + * General purpose TLB invalidation function. This function scans every GRU in + * the ENTIRE system (partition) looking for GRUs where the specified MM has + * been accessed by the GRU. For each GRU found, the TLB must be invalidated OR + * the ASID invalidated. Invalidating an ASID causes a new ASID to be assigned + * on the next fault. This effectively flushes the ENTIRE TLB for the MM at the + * cost of (possibly) a large number of future TLBmisses. + * + * The current algorithm is optimized based on the following (somewhat true) + * assumptions: + * - GRU contexts are not loaded into a GRU unless a reference is made to + * the data segment or control block (this is true, not an assumption). + * If a DS/CB is referenced, the user will also issue instructions that + * cause TLBmisses. It is not necessary to optimize for the case where + * contexts are loaded but no instructions cause TLB misses. (I know + * this will happen but I'm not optimizing for it). + * - GRU instructions to invalidate TLB entries are SLOOOOWWW - normally + * a few usec but in unusual cases, it could be longer. Avoid if + * possible. + * - intrablade process migration between cpus is not frequent but is + * common. + * - a GRU context is not typically migrated to a different GRU on the + * blade because of intrablade migration + * - interblade migration is rare. Processes migrate their GRU context to + * the new blade. + * - if interblade migration occurs, migration back to the original blade + * is very very rare (ie., no optimization for this case) + * - most GRU instruction operate on a subset of the user REGIONS. Code + * & shared library regions are not likely targets of GRU instructions. + * + * To help improve the efficiency of TLB invalidation, the GMS data + * structure is maintained for EACH address space (MM struct). The GMS is + * also the structure that contains the pointer to the mmu callout + * functions. This structure is linked to the mm_struct for the address space + * using the mmu "register" function. The mmu interfaces are used to + * provide the callbacks for TLB invalidation. The GMS contains: + * + * - asid[maxgrus] array. ASIDs are assigned to a GRU when a context is + * loaded into the GRU. + * - asidmap[maxgrus]. bitmap to make it easier to find non-zero asids in + * the above array + * - ctxbitmap[maxgrus]. Indicates the contexts that are currently active + * in the GRU for the address space. This bitmap must be passed to the + * GRU to do an invalidate. + * + * The current algorithm for invalidating TLBs is: + * - scan the asidmap for GRUs where the context has been loaded, ie, + * asid is non-zero. + * - for each gru found: + * - if the ctxtmap is non-zero, there are active contexts in the + * GRU. TLB invalidate instructions must be issued to the GRU. + * - if the ctxtmap is zero, no context is active. Set the ASID to + * zero to force a full TLB invalidation. This is fast but will + * cause a lot of TLB misses if the context is reloaded onto the + * GRU + * + */ + +void gru_flush_tlb_range(struct gru_mm_struct *gms, unsigned long start, + unsigned long len) +{ + struct gru_state *gru; + struct gru_mm_tracker *asids; + struct gru_tlb_global_handle *tgh; + unsigned long num; + int grupagesize, pagesize, pageshift, gid, asid; + + /* ZZZ TODO - handle huge pages */ + pageshift = PAGE_SHIFT; + pagesize = (1UL << pageshift); + grupagesize = GRU_PAGESIZE(pageshift); + num = min(((len + pagesize - 1) >> pageshift), GRUMAXINVAL); + + STAT(flush_tlb); + gru_dbg(grudev, "gms %p, start 0x%lx, len 0x%lx, asidmap 0x%lx\n", gms, + start, len, gms->ms_asidmap[0]); + + spin_lock(&gms->ms_asid_lock); + for_each_gru_in_bitmap(gid, gms->ms_asidmap) { + STAT(flush_tlb_gru); + gru = GID_TO_GRU(gid); + asids = gms->ms_asids + gid; + asid = asids->mt_asid; + if (asids->mt_ctxbitmap && asid) { + STAT(flush_tlb_gru_tgh); + asid = GRUASID(asid, start); + gru_dbg(grudev, + " FLUSH gruid %d, asid 0x%x, vaddr 0x%lx, vamask 0x%x, num %ld, cbmap 0x%x\n", + gid, asid, start, grupagesize, num, asids->mt_ctxbitmap); + tgh = get_lock_tgh_handle(gru); + tgh_invalidate(tgh, start, ~0, asid, grupagesize, 0, + num - 1, asids->mt_ctxbitmap); + get_unlock_tgh_handle(tgh); + } else { + STAT(flush_tlb_gru_zero_asid); + asids->mt_asid = 0; + __clear_bit(gru->gs_gid, gms->ms_asidmap); + gru_dbg(grudev, + " CLEARASID gruid %d, asid 0x%x, cbtmap 0x%x, asidmap 0x%lx\n", + gid, asid, asids->mt_ctxbitmap, + gms->ms_asidmap[0]); + } + } + spin_unlock(&gms->ms_asid_lock); +} + +/* + * Flush the entire TLB on a chiplet. + */ +void gru_flush_all_tlb(struct gru_state *gru) +{ + struct gru_tlb_global_handle *tgh; + + gru_dbg(grudev, "gid %d\n", gru->gs_gid); + tgh = get_lock_tgh_handle(gru); + tgh_invalidate(tgh, 0, ~0, 0, 1, 1, GRUMAXINVAL - 1, 0xffff); + get_unlock_tgh_handle(tgh); +} + +/* + * MMUOPS notifier callout functions + */ +static int gru_invalidate_range_start(struct mmu_notifier *mn, + const struct mmu_notifier_range *range) +{ + struct gru_mm_struct *gms = container_of(mn, struct gru_mm_struct, + ms_notifier); + + STAT(mmu_invalidate_range); + atomic_inc(&gms->ms_range_active); + gru_dbg(grudev, "gms %p, start 0x%lx, end 0x%lx, act %d\n", gms, + range->start, range->end, atomic_read(&gms->ms_range_active)); + gru_flush_tlb_range(gms, range->start, range->end - range->start); + + return 0; +} + +static void gru_invalidate_range_end(struct mmu_notifier *mn, + const struct mmu_notifier_range *range) +{ + struct gru_mm_struct *gms = container_of(mn, struct gru_mm_struct, + ms_notifier); + + /* ..._and_test() provides needed barrier */ + (void)atomic_dec_and_test(&gms->ms_range_active); + + wake_up_all(&gms->ms_wait_queue); + gru_dbg(grudev, "gms %p, start 0x%lx, end 0x%lx\n", + gms, range->start, range->end); +} + +static struct mmu_notifier *gru_alloc_notifier(struct mm_struct *mm) +{ + struct gru_mm_struct *gms; + + gms = kzalloc(sizeof(*gms), GFP_KERNEL); + if (!gms) + return ERR_PTR(-ENOMEM); + STAT(gms_alloc); + spin_lock_init(&gms->ms_asid_lock); + init_waitqueue_head(&gms->ms_wait_queue); + + return &gms->ms_notifier; +} + +static void gru_free_notifier(struct mmu_notifier *mn) +{ + kfree(container_of(mn, struct gru_mm_struct, ms_notifier)); + STAT(gms_free); +} + +static const struct mmu_notifier_ops gru_mmuops = { + .invalidate_range_start = gru_invalidate_range_start, + .invalidate_range_end = gru_invalidate_range_end, + .alloc_notifier = gru_alloc_notifier, + .free_notifier = gru_free_notifier, +}; + +struct gru_mm_struct *gru_register_mmu_notifier(void) +{ + struct mmu_notifier *mn; + + mn = mmu_notifier_get_locked(&gru_mmuops, current->mm); + if (IS_ERR(mn)) + return ERR_CAST(mn); + + return container_of(mn, struct gru_mm_struct, ms_notifier); +} + +void gru_drop_mmu_notifier(struct gru_mm_struct *gms) +{ + mmu_notifier_put(&gms->ms_notifier); +} + +/* + * Setup TGH parameters. There are: + * - 24 TGH handles per GRU chiplet + * - a portion (MAX_LOCAL_TGH) of the handles are reserved for + * use by blade-local cpus + * - the rest are used by off-blade cpus. This usage is + * less frequent than blade-local usage. + * + * For now, use 16 handles for local flushes, 8 for remote flushes. If the blade + * has less tan or equal to 16 cpus, each cpu has a unique handle that it can + * use. + */ +#define MAX_LOCAL_TGH 16 + +void gru_tgh_flush_init(struct gru_state *gru) +{ + int cpus, shift = 0, n; + + cpus = uv_blade_nr_possible_cpus(gru->gs_blade_id); + + /* n = cpus rounded up to next power of 2 */ + if (cpus) { + n = 1 << fls(cpus - 1); + + /* + * shift count for converting local cpu# to TGH index + * 0 if cpus <= MAX_LOCAL_TGH, + * 1 if cpus <= 2*MAX_LOCAL_TGH, + * etc + */ + shift = max(0, fls(n - 1) - fls(MAX_LOCAL_TGH - 1)); + } + gru->gs_tgh_local_shift = shift; + + /* first starting TGH index to use for remote purges */ + gru->gs_tgh_first_remote = (cpus + (1 << shift) - 1) >> shift; + +} -- cgit v1.2.3