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(). ... --- arch/mips/mm/page.c | 682 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 682 insertions(+) create mode 100644 arch/mips/mm/page.c (limited to 'arch/mips/mm/page.c') diff --git a/arch/mips/mm/page.c b/arch/mips/mm/page.c new file mode 100644 index 000000000..d3b4459d0 --- /dev/null +++ b/arch/mips/mm/page.c @@ -0,0 +1,682 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2003, 04, 05 Ralf Baechle (ralf@linux-mips.org) + * Copyright (C) 2007 Maciej W. Rozycki + * Copyright (C) 2008 Thiemo Seufer + * Copyright (C) 2012 MIPS Technologies, Inc. + */ +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_SIBYTE_DMA_PAGEOPS +#include +#include +#include +#endif + +#include + +/* Registers used in the assembled routines. */ +#define ZERO 0 +#define AT 2 +#define A0 4 +#define A1 5 +#define A2 6 +#define T0 8 +#define T1 9 +#define T2 10 +#define T3 11 +#define T9 25 +#define RA 31 + +/* Handle labels (which must be positive integers). */ +enum label_id { + label_clear_nopref = 1, + label_clear_pref, + label_copy_nopref, + label_copy_pref_both, + label_copy_pref_store, +}; + +UASM_L_LA(_clear_nopref) +UASM_L_LA(_clear_pref) +UASM_L_LA(_copy_nopref) +UASM_L_LA(_copy_pref_both) +UASM_L_LA(_copy_pref_store) + +/* We need one branch and therefore one relocation per target label. */ +static struct uasm_label labels[5]; +static struct uasm_reloc relocs[5]; + +#define cpu_is_r4600_v1_x() ((read_c0_prid() & 0xfffffff0) == 0x00002010) +#define cpu_is_r4600_v2_x() ((read_c0_prid() & 0xfffffff0) == 0x00002020) + +/* + * R6 has a limited offset of the pref instruction. + * Skip it if the offset is more than 9 bits. + */ +#define _uasm_i_pref(a, b, c, d) \ +do { \ + if (cpu_has_mips_r6) { \ + if (c <= 0xff && c >= -0x100) \ + uasm_i_pref(a, b, c, d);\ + } else { \ + uasm_i_pref(a, b, c, d); \ + } \ +} while(0) + +static int pref_bias_clear_store; +static int pref_bias_copy_load; +static int pref_bias_copy_store; + +static u32 pref_src_mode; +static u32 pref_dst_mode; + +static int clear_word_size; +static int copy_word_size; + +static int half_clear_loop_size; +static int half_copy_loop_size; + +static int cache_line_size; +#define cache_line_mask() (cache_line_size - 1) + +static inline void +pg_addiu(u32 **buf, unsigned int reg1, unsigned int reg2, unsigned int off) +{ + if (cpu_has_64bit_gp_regs && + IS_ENABLED(CONFIG_CPU_DADDI_WORKAROUNDS) && + r4k_daddiu_bug()) { + if (off > 0x7fff) { + uasm_i_lui(buf, T9, uasm_rel_hi(off)); + uasm_i_addiu(buf, T9, T9, uasm_rel_lo(off)); + } else + uasm_i_addiu(buf, T9, ZERO, off); + uasm_i_daddu(buf, reg1, reg2, T9); + } else { + if (off > 0x7fff) { + uasm_i_lui(buf, T9, uasm_rel_hi(off)); + uasm_i_addiu(buf, T9, T9, uasm_rel_lo(off)); + UASM_i_ADDU(buf, reg1, reg2, T9); + } else + UASM_i_ADDIU(buf, reg1, reg2, off); + } +} + +static void set_prefetch_parameters(void) +{ + if (cpu_has_64bit_gp_regs || cpu_has_64bit_zero_reg) + clear_word_size = 8; + else + clear_word_size = 4; + + if (cpu_has_64bit_gp_regs) + copy_word_size = 8; + else + copy_word_size = 4; + + /* + * The pref's used here are using "streaming" hints, which cause the + * copied data to be kicked out of the cache sooner. A page copy often + * ends up copying a lot more data than is commonly used, so this seems + * to make sense in terms of reducing cache pollution, but I've no real + * performance data to back this up. + */ + if (cpu_has_prefetch) { + /* + * XXX: Most prefetch bias values in here are based on + * guesswork. + */ + cache_line_size = cpu_dcache_line_size(); + switch (current_cpu_type()) { + case CPU_R5500: + case CPU_TX49XX: + /* These processors only support the Pref_Load. */ + pref_bias_copy_load = 256; + break; + + case CPU_R10000: + case CPU_R12000: + case CPU_R14000: + case CPU_R16000: + /* + * Those values have been experimentally tuned for an + * Origin 200. + */ + pref_bias_clear_store = 512; + pref_bias_copy_load = 256; + pref_bias_copy_store = 256; + pref_src_mode = Pref_LoadStreamed; + pref_dst_mode = Pref_StoreStreamed; + break; + + case CPU_SB1: + case CPU_SB1A: + pref_bias_clear_store = 128; + pref_bias_copy_load = 128; + pref_bias_copy_store = 128; + /* + * SB1 pass1 Pref_LoadStreamed/Pref_StoreStreamed + * hints are broken. + */ + if (current_cpu_type() == CPU_SB1 && + (current_cpu_data.processor_id & 0xff) < 0x02) { + pref_src_mode = Pref_Load; + pref_dst_mode = Pref_Store; + } else { + pref_src_mode = Pref_LoadStreamed; + pref_dst_mode = Pref_StoreStreamed; + } + break; + + case CPU_LOONGSON64: + /* Loongson-3 only support the Pref_Load/Pref_Store. */ + pref_bias_clear_store = 128; + pref_bias_copy_load = 128; + pref_bias_copy_store = 128; + pref_src_mode = Pref_Load; + pref_dst_mode = Pref_Store; + break; + + default: + pref_bias_clear_store = 128; + pref_bias_copy_load = 256; + pref_bias_copy_store = 128; + pref_src_mode = Pref_LoadStreamed; + if (cpu_has_mips_r6) + /* + * Bit 30 (Pref_PrepareForStore) has been + * removed from MIPS R6. Use bit 5 + * (Pref_StoreStreamed). + */ + pref_dst_mode = Pref_StoreStreamed; + else + pref_dst_mode = Pref_PrepareForStore; + break; + } + } else { + if (cpu_has_cache_cdex_s) + cache_line_size = cpu_scache_line_size(); + else if (cpu_has_cache_cdex_p) + cache_line_size = cpu_dcache_line_size(); + } + /* + * Too much unrolling will overflow the available space in + * clear_space_array / copy_page_array. + */ + half_clear_loop_size = min(16 * clear_word_size, + max(cache_line_size >> 1, + 4 * clear_word_size)); + half_copy_loop_size = min(16 * copy_word_size, + max(cache_line_size >> 1, + 4 * copy_word_size)); +} + +static void build_clear_store(u32 **buf, int off) +{ + if (cpu_has_64bit_gp_regs || cpu_has_64bit_zero_reg) { + uasm_i_sd(buf, ZERO, off, A0); + } else { + uasm_i_sw(buf, ZERO, off, A0); + } +} + +static inline void build_clear_pref(u32 **buf, int off) +{ + if (off & cache_line_mask()) + return; + + if (pref_bias_clear_store) { + _uasm_i_pref(buf, pref_dst_mode, pref_bias_clear_store + off, + A0); + } else if (cache_line_size == (half_clear_loop_size << 1)) { + if (cpu_has_cache_cdex_s) { + uasm_i_cache(buf, Create_Dirty_Excl_SD, off, A0); + } else if (cpu_has_cache_cdex_p) { + if (IS_ENABLED(CONFIG_WAR_R4600_V1_HIT_CACHEOP) && + cpu_is_r4600_v1_x()) { + uasm_i_nop(buf); + uasm_i_nop(buf); + uasm_i_nop(buf); + uasm_i_nop(buf); + } + + if (IS_ENABLED(CONFIG_WAR_R4600_V2_HIT_CACHEOP) && + cpu_is_r4600_v2_x()) + uasm_i_lw(buf, ZERO, ZERO, AT); + + uasm_i_cache(buf, Create_Dirty_Excl_D, off, A0); + } + } +} + +extern u32 __clear_page_start; +extern u32 __clear_page_end; +extern u32 __copy_page_start; +extern u32 __copy_page_end; + +void build_clear_page(void) +{ + int off; + u32 *buf = &__clear_page_start; + struct uasm_label *l = labels; + struct uasm_reloc *r = relocs; + int i; + static atomic_t run_once = ATOMIC_INIT(0); + + if (atomic_xchg(&run_once, 1)) { + return; + } + + memset(labels, 0, sizeof(labels)); + memset(relocs, 0, sizeof(relocs)); + + set_prefetch_parameters(); + + /* + * This algorithm makes the following assumptions: + * - The prefetch bias is a multiple of 2 words. + * - The prefetch bias is less than one page. + */ + BUG_ON(pref_bias_clear_store % (2 * clear_word_size)); + BUG_ON(PAGE_SIZE < pref_bias_clear_store); + + off = PAGE_SIZE - pref_bias_clear_store; + if (off > 0xffff || !pref_bias_clear_store) + pg_addiu(&buf, A2, A0, off); + else + uasm_i_ori(&buf, A2, A0, off); + + if (IS_ENABLED(CONFIG_WAR_R4600_V2_HIT_CACHEOP) && cpu_is_r4600_v2_x()) + uasm_i_lui(&buf, AT, uasm_rel_hi(0xa0000000)); + + off = cache_line_size ? min(8, pref_bias_clear_store / cache_line_size) + * cache_line_size : 0; + while (off) { + build_clear_pref(&buf, -off); + off -= cache_line_size; + } + uasm_l_clear_pref(&l, buf); + do { + build_clear_pref(&buf, off); + build_clear_store(&buf, off); + off += clear_word_size; + } while (off < half_clear_loop_size); + pg_addiu(&buf, A0, A0, 2 * off); + off = -off; + do { + build_clear_pref(&buf, off); + if (off == -clear_word_size) + uasm_il_bne(&buf, &r, A0, A2, label_clear_pref); + build_clear_store(&buf, off); + off += clear_word_size; + } while (off < 0); + + if (pref_bias_clear_store) { + pg_addiu(&buf, A2, A0, pref_bias_clear_store); + uasm_l_clear_nopref(&l, buf); + off = 0; + do { + build_clear_store(&buf, off); + off += clear_word_size; + } while (off < half_clear_loop_size); + pg_addiu(&buf, A0, A0, 2 * off); + off = -off; + do { + if (off == -clear_word_size) + uasm_il_bne(&buf, &r, A0, A2, + label_clear_nopref); + build_clear_store(&buf, off); + off += clear_word_size; + } while (off < 0); + } + + uasm_i_jr(&buf, RA); + uasm_i_nop(&buf); + + BUG_ON(buf > &__clear_page_end); + + uasm_resolve_relocs(relocs, labels); + + pr_debug("Synthesized clear page handler (%u instructions).\n", + (u32)(buf - &__clear_page_start)); + + pr_debug("\t.set push\n"); + pr_debug("\t.set noreorder\n"); + for (i = 0; i < (buf - &__clear_page_start); i++) + pr_debug("\t.word 0x%08x\n", (&__clear_page_start)[i]); + pr_debug("\t.set pop\n"); +} + +static void build_copy_load(u32 **buf, int reg, int off) +{ + if (cpu_has_64bit_gp_regs) { + uasm_i_ld(buf, reg, off, A1); + } else { + uasm_i_lw(buf, reg, off, A1); + } +} + +static void build_copy_store(u32 **buf, int reg, int off) +{ + if (cpu_has_64bit_gp_regs) { + uasm_i_sd(buf, reg, off, A0); + } else { + uasm_i_sw(buf, reg, off, A0); + } +} + +static inline void build_copy_load_pref(u32 **buf, int off) +{ + if (off & cache_line_mask()) + return; + + if (pref_bias_copy_load) + _uasm_i_pref(buf, pref_src_mode, pref_bias_copy_load + off, A1); +} + +static inline void build_copy_store_pref(u32 **buf, int off) +{ + if (off & cache_line_mask()) + return; + + if (pref_bias_copy_store) { + _uasm_i_pref(buf, pref_dst_mode, pref_bias_copy_store + off, + A0); + } else if (cache_line_size == (half_copy_loop_size << 1)) { + if (cpu_has_cache_cdex_s) { + uasm_i_cache(buf, Create_Dirty_Excl_SD, off, A0); + } else if (cpu_has_cache_cdex_p) { + if (IS_ENABLED(CONFIG_WAR_R4600_V1_HIT_CACHEOP) && + cpu_is_r4600_v1_x()) { + uasm_i_nop(buf); + uasm_i_nop(buf); + uasm_i_nop(buf); + uasm_i_nop(buf); + } + + if (IS_ENABLED(CONFIG_WAR_R4600_V2_HIT_CACHEOP) && + cpu_is_r4600_v2_x()) + uasm_i_lw(buf, ZERO, ZERO, AT); + + uasm_i_cache(buf, Create_Dirty_Excl_D, off, A0); + } + } +} + +void build_copy_page(void) +{ + int off; + u32 *buf = &__copy_page_start; + struct uasm_label *l = labels; + struct uasm_reloc *r = relocs; + int i; + static atomic_t run_once = ATOMIC_INIT(0); + + if (atomic_xchg(&run_once, 1)) { + return; + } + + memset(labels, 0, sizeof(labels)); + memset(relocs, 0, sizeof(relocs)); + + set_prefetch_parameters(); + + /* + * This algorithm makes the following assumptions: + * - All prefetch biases are multiples of 8 words. + * - The prefetch biases are less than one page. + * - The store prefetch bias isn't greater than the load + * prefetch bias. + */ + BUG_ON(pref_bias_copy_load % (8 * copy_word_size)); + BUG_ON(pref_bias_copy_store % (8 * copy_word_size)); + BUG_ON(PAGE_SIZE < pref_bias_copy_load); + BUG_ON(pref_bias_copy_store > pref_bias_copy_load); + + off = PAGE_SIZE - pref_bias_copy_load; + if (off > 0xffff || !pref_bias_copy_load) + pg_addiu(&buf, A2, A0, off); + else + uasm_i_ori(&buf, A2, A0, off); + + if (IS_ENABLED(CONFIG_WAR_R4600_V2_HIT_CACHEOP) && cpu_is_r4600_v2_x()) + uasm_i_lui(&buf, AT, uasm_rel_hi(0xa0000000)); + + off = cache_line_size ? min(8, pref_bias_copy_load / cache_line_size) * + cache_line_size : 0; + while (off) { + build_copy_load_pref(&buf, -off); + off -= cache_line_size; + } + off = cache_line_size ? min(8, pref_bias_copy_store / cache_line_size) * + cache_line_size : 0; + while (off) { + build_copy_store_pref(&buf, -off); + off -= cache_line_size; + } + uasm_l_copy_pref_both(&l, buf); + do { + build_copy_load_pref(&buf, off); + build_copy_load(&buf, T0, off); + build_copy_load_pref(&buf, off + copy_word_size); + build_copy_load(&buf, T1, off + copy_word_size); + build_copy_load_pref(&buf, off + 2 * copy_word_size); + build_copy_load(&buf, T2, off + 2 * copy_word_size); + build_copy_load_pref(&buf, off + 3 * copy_word_size); + build_copy_load(&buf, T3, off + 3 * copy_word_size); + build_copy_store_pref(&buf, off); + build_copy_store(&buf, T0, off); + build_copy_store_pref(&buf, off + copy_word_size); + build_copy_store(&buf, T1, off + copy_word_size); + build_copy_store_pref(&buf, off + 2 * copy_word_size); + build_copy_store(&buf, T2, off + 2 * copy_word_size); + build_copy_store_pref(&buf, off + 3 * copy_word_size); + build_copy_store(&buf, T3, off + 3 * copy_word_size); + off += 4 * copy_word_size; + } while (off < half_copy_loop_size); + pg_addiu(&buf, A1, A1, 2 * off); + pg_addiu(&buf, A0, A0, 2 * off); + off = -off; + do { + build_copy_load_pref(&buf, off); + build_copy_load(&buf, T0, off); + build_copy_load_pref(&buf, off + copy_word_size); + build_copy_load(&buf, T1, off + copy_word_size); + build_copy_load_pref(&buf, off + 2 * copy_word_size); + build_copy_load(&buf, T2, off + 2 * copy_word_size); + build_copy_load_pref(&buf, off + 3 * copy_word_size); + build_copy_load(&buf, T3, off + 3 * copy_word_size); + build_copy_store_pref(&buf, off); + build_copy_store(&buf, T0, off); + build_copy_store_pref(&buf, off + copy_word_size); + build_copy_store(&buf, T1, off + copy_word_size); + build_copy_store_pref(&buf, off + 2 * copy_word_size); + build_copy_store(&buf, T2, off + 2 * copy_word_size); + build_copy_store_pref(&buf, off + 3 * copy_word_size); + if (off == -(4 * copy_word_size)) + uasm_il_bne(&buf, &r, A2, A0, label_copy_pref_both); + build_copy_store(&buf, T3, off + 3 * copy_word_size); + off += 4 * copy_word_size; + } while (off < 0); + + if (pref_bias_copy_load - pref_bias_copy_store) { + pg_addiu(&buf, A2, A0, + pref_bias_copy_load - pref_bias_copy_store); + uasm_l_copy_pref_store(&l, buf); + off = 0; + do { + build_copy_load(&buf, T0, off); + build_copy_load(&buf, T1, off + copy_word_size); + build_copy_load(&buf, T2, off + 2 * copy_word_size); + build_copy_load(&buf, T3, off + 3 * copy_word_size); + build_copy_store_pref(&buf, off); + build_copy_store(&buf, T0, off); + build_copy_store_pref(&buf, off + copy_word_size); + build_copy_store(&buf, T1, off + copy_word_size); + build_copy_store_pref(&buf, off + 2 * copy_word_size); + build_copy_store(&buf, T2, off + 2 * copy_word_size); + build_copy_store_pref(&buf, off + 3 * copy_word_size); + build_copy_store(&buf, T3, off + 3 * copy_word_size); + off += 4 * copy_word_size; + } while (off < half_copy_loop_size); + pg_addiu(&buf, A1, A1, 2 * off); + pg_addiu(&buf, A0, A0, 2 * off); + off = -off; + do { + build_copy_load(&buf, T0, off); + build_copy_load(&buf, T1, off + copy_word_size); + build_copy_load(&buf, T2, off + 2 * copy_word_size); + build_copy_load(&buf, T3, off + 3 * copy_word_size); + build_copy_store_pref(&buf, off); + build_copy_store(&buf, T0, off); + build_copy_store_pref(&buf, off + copy_word_size); + build_copy_store(&buf, T1, off + copy_word_size); + build_copy_store_pref(&buf, off + 2 * copy_word_size); + build_copy_store(&buf, T2, off + 2 * copy_word_size); + build_copy_store_pref(&buf, off + 3 * copy_word_size); + if (off == -(4 * copy_word_size)) + uasm_il_bne(&buf, &r, A2, A0, + label_copy_pref_store); + build_copy_store(&buf, T3, off + 3 * copy_word_size); + off += 4 * copy_word_size; + } while (off < 0); + } + + if (pref_bias_copy_store) { + pg_addiu(&buf, A2, A0, pref_bias_copy_store); + uasm_l_copy_nopref(&l, buf); + off = 0; + do { + build_copy_load(&buf, T0, off); + build_copy_load(&buf, T1, off + copy_word_size); + build_copy_load(&buf, T2, off + 2 * copy_word_size); + build_copy_load(&buf, T3, off + 3 * copy_word_size); + build_copy_store(&buf, T0, off); + build_copy_store(&buf, T1, off + copy_word_size); + build_copy_store(&buf, T2, off + 2 * copy_word_size); + build_copy_store(&buf, T3, off + 3 * copy_word_size); + off += 4 * copy_word_size; + } while (off < half_copy_loop_size); + pg_addiu(&buf, A1, A1, 2 * off); + pg_addiu(&buf, A0, A0, 2 * off); + off = -off; + do { + build_copy_load(&buf, T0, off); + build_copy_load(&buf, T1, off + copy_word_size); + build_copy_load(&buf, T2, off + 2 * copy_word_size); + build_copy_load(&buf, T3, off + 3 * copy_word_size); + build_copy_store(&buf, T0, off); + build_copy_store(&buf, T1, off + copy_word_size); + build_copy_store(&buf, T2, off + 2 * copy_word_size); + if (off == -(4 * copy_word_size)) + uasm_il_bne(&buf, &r, A2, A0, + label_copy_nopref); + build_copy_store(&buf, T3, off + 3 * copy_word_size); + off += 4 * copy_word_size; + } while (off < 0); + } + + uasm_i_jr(&buf, RA); + uasm_i_nop(&buf); + + BUG_ON(buf > &__copy_page_end); + + uasm_resolve_relocs(relocs, labels); + + pr_debug("Synthesized copy page handler (%u instructions).\n", + (u32)(buf - &__copy_page_start)); + + pr_debug("\t.set push\n"); + pr_debug("\t.set noreorder\n"); + for (i = 0; i < (buf - &__copy_page_start); i++) + pr_debug("\t.word 0x%08x\n", (&__copy_page_start)[i]); + pr_debug("\t.set pop\n"); +} + +#ifdef CONFIG_SIBYTE_DMA_PAGEOPS +extern void clear_page_cpu(void *page); +extern void copy_page_cpu(void *to, void *from); + +/* + * Pad descriptors to cacheline, since each is exclusively owned by a + * particular CPU. + */ +struct dmadscr { + u64 dscr_a; + u64 dscr_b; + u64 pad_a; + u64 pad_b; +} ____cacheline_aligned_in_smp page_descr[DM_NUM_CHANNELS]; + +void clear_page(void *page) +{ + u64 to_phys = CPHYSADDR((unsigned long)page); + unsigned int cpu = smp_processor_id(); + + /* if the page is not in KSEG0, use old way */ + if ((long)KSEGX((unsigned long)page) != (long)CKSEG0) + return clear_page_cpu(page); + + page_descr[cpu].dscr_a = to_phys | M_DM_DSCRA_ZERO_MEM | + M_DM_DSCRA_L2C_DEST | M_DM_DSCRA_INTERRUPT; + page_descr[cpu].dscr_b = V_DM_DSCRB_SRC_LENGTH(PAGE_SIZE); + __raw_writeq(1, IOADDR(A_DM_REGISTER(cpu, R_DM_DSCR_COUNT))); + + /* + * Don't really want to do it this way, but there's no + * reliable way to delay completion detection. + */ + while (!(__raw_readq(IOADDR(A_DM_REGISTER(cpu, R_DM_DSCR_BASE_DEBUG))) + & M_DM_DSCR_BASE_INTERRUPT)) + ; + __raw_readq(IOADDR(A_DM_REGISTER(cpu, R_DM_DSCR_BASE))); +} +EXPORT_SYMBOL(clear_page); + +void copy_page(void *to, void *from) +{ + u64 from_phys = CPHYSADDR((unsigned long)from); + u64 to_phys = CPHYSADDR((unsigned long)to); + unsigned int cpu = smp_processor_id(); + + /* if any page is not in KSEG0, use old way */ + if ((long)KSEGX((unsigned long)to) != (long)CKSEG0 + || (long)KSEGX((unsigned long)from) != (long)CKSEG0) + return copy_page_cpu(to, from); + + page_descr[cpu].dscr_a = to_phys | M_DM_DSCRA_L2C_DEST | + M_DM_DSCRA_INTERRUPT; + page_descr[cpu].dscr_b = from_phys | V_DM_DSCRB_SRC_LENGTH(PAGE_SIZE); + __raw_writeq(1, IOADDR(A_DM_REGISTER(cpu, R_DM_DSCR_COUNT))); + + /* + * Don't really want to do it this way, but there's no + * reliable way to delay completion detection. + */ + while (!(__raw_readq(IOADDR(A_DM_REGISTER(cpu, R_DM_DSCR_BASE_DEBUG))) + & M_DM_DSCR_BASE_INTERRUPT)) + ; + __raw_readq(IOADDR(A_DM_REGISTER(cpu, R_DM_DSCR_BASE))); +} +EXPORT_SYMBOL(copy_page); + +#endif /* CONFIG_SIBYTE_DMA_PAGEOPS */ -- cgit v1.2.3