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/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c | 580 ++++++++++++++++++++++ 1 file changed, 580 insertions(+) create mode 100644 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c new file mode 100644 index 000000000..e83cb1c09 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -0,0 +1,580 @@ +/* + * Copyright 2014 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +#include "amdgpu.h" +#include "amdgpu_amdkfd.h" +#include "cikd.h" +#include "cik_sdma.h" +#include "gfx_v7_0.h" +#include "gca/gfx_7_2_d.h" +#include "gca/gfx_7_2_enum.h" +#include "gca/gfx_7_2_sh_mask.h" +#include "oss/oss_2_0_d.h" +#include "oss/oss_2_0_sh_mask.h" +#include "gmc/gmc_7_1_d.h" +#include "gmc/gmc_7_1_sh_mask.h" +#include "cik_structs.h" + +enum hqd_dequeue_request_type { + NO_ACTION = 0, + DRAIN_PIPE, + RESET_WAVES +}; + +enum { + MAX_TRAPID = 8, /* 3 bits in the bitfield. */ + MAX_WATCH_ADDRESSES = 4 +}; + +static void lock_srbm(struct amdgpu_device *adev, uint32_t mec, uint32_t pipe, + uint32_t queue, uint32_t vmid) +{ + uint32_t value = PIPEID(pipe) | MEID(mec) | VMID(vmid) | QUEUEID(queue); + + mutex_lock(&adev->srbm_mutex); + WREG32(mmSRBM_GFX_CNTL, value); +} + +static void unlock_srbm(struct amdgpu_device *adev) +{ + WREG32(mmSRBM_GFX_CNTL, 0); + mutex_unlock(&adev->srbm_mutex); +} + +static void acquire_queue(struct amdgpu_device *adev, uint32_t pipe_id, + uint32_t queue_id) +{ + uint32_t mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1; + uint32_t pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec); + + lock_srbm(adev, mec, pipe, queue_id, 0); +} + +static void release_queue(struct amdgpu_device *adev) +{ + unlock_srbm(adev); +} + +static void kgd_program_sh_mem_settings(struct amdgpu_device *adev, uint32_t vmid, + uint32_t sh_mem_config, + uint32_t sh_mem_ape1_base, + uint32_t sh_mem_ape1_limit, + uint32_t sh_mem_bases) +{ + lock_srbm(adev, 0, 0, 0, vmid); + + WREG32(mmSH_MEM_CONFIG, sh_mem_config); + WREG32(mmSH_MEM_APE1_BASE, sh_mem_ape1_base); + WREG32(mmSH_MEM_APE1_LIMIT, sh_mem_ape1_limit); + WREG32(mmSH_MEM_BASES, sh_mem_bases); + + unlock_srbm(adev); +} + +static int kgd_set_pasid_vmid_mapping(struct amdgpu_device *adev, u32 pasid, + unsigned int vmid) +{ + /* + * We have to assume that there is no outstanding mapping. + * The ATC_VMID_PASID_MAPPING_UPDATE_STATUS bit could be 0 because + * a mapping is in progress or because a mapping finished and the + * SW cleared it. So the protocol is to always wait & clear. + */ + uint32_t pasid_mapping = (pasid == 0) ? 0 : (uint32_t)pasid | + ATC_VMID0_PASID_MAPPING__VALID_MASK; + + WREG32(mmATC_VMID0_PASID_MAPPING + vmid, pasid_mapping); + + while (!(RREG32(mmATC_VMID_PASID_MAPPING_UPDATE_STATUS) & (1U << vmid))) + cpu_relax(); + WREG32(mmATC_VMID_PASID_MAPPING_UPDATE_STATUS, 1U << vmid); + + /* Mapping vmid to pasid also for IH block */ + WREG32(mmIH_VMID_0_LUT + vmid, pasid_mapping); + + return 0; +} + +static int kgd_init_interrupts(struct amdgpu_device *adev, uint32_t pipe_id) +{ + uint32_t mec; + uint32_t pipe; + + mec = (pipe_id / adev->gfx.mec.num_pipe_per_mec) + 1; + pipe = (pipe_id % adev->gfx.mec.num_pipe_per_mec); + + lock_srbm(adev, mec, pipe, 0, 0); + + WREG32(mmCPC_INT_CNTL, CP_INT_CNTL_RING0__TIME_STAMP_INT_ENABLE_MASK | + CP_INT_CNTL_RING0__OPCODE_ERROR_INT_ENABLE_MASK); + + unlock_srbm(adev); + + return 0; +} + +static inline uint32_t get_sdma_rlc_reg_offset(struct cik_sdma_rlc_registers *m) +{ + uint32_t retval; + + retval = m->sdma_engine_id * SDMA1_REGISTER_OFFSET + + m->sdma_queue_id * KFD_CIK_SDMA_QUEUE_OFFSET; + + pr_debug("RLC register offset for SDMA%d RLC%d: 0x%x\n", + m->sdma_engine_id, m->sdma_queue_id, retval); + + return retval; +} + +static inline struct cik_mqd *get_mqd(void *mqd) +{ + return (struct cik_mqd *)mqd; +} + +static inline struct cik_sdma_rlc_registers *get_sdma_mqd(void *mqd) +{ + return (struct cik_sdma_rlc_registers *)mqd; +} + +static int kgd_hqd_load(struct amdgpu_device *adev, void *mqd, + uint32_t pipe_id, uint32_t queue_id, + uint32_t __user *wptr, uint32_t wptr_shift, + uint32_t wptr_mask, struct mm_struct *mm) +{ + struct cik_mqd *m; + uint32_t *mqd_hqd; + uint32_t reg, wptr_val, data; + bool valid_wptr = false; + + m = get_mqd(mqd); + + acquire_queue(adev, pipe_id, queue_id); + + /* HQD registers extend from CP_MQD_BASE_ADDR to CP_MQD_CONTROL. */ + mqd_hqd = &m->cp_mqd_base_addr_lo; + + for (reg = mmCP_MQD_BASE_ADDR; reg <= mmCP_MQD_CONTROL; reg++) + WREG32(reg, mqd_hqd[reg - mmCP_MQD_BASE_ADDR]); + + /* Copy userspace write pointer value to register. + * Activate doorbell logic to monitor subsequent changes. + */ + data = REG_SET_FIELD(m->cp_hqd_pq_doorbell_control, + CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 1); + WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, data); + + /* read_user_ptr may take the mm->mmap_lock. + * release srbm_mutex to avoid circular dependency between + * srbm_mutex->mmap_lock->reservation_ww_class_mutex->srbm_mutex. + */ + release_queue(adev); + valid_wptr = read_user_wptr(mm, wptr, wptr_val); + acquire_queue(adev, pipe_id, queue_id); + if (valid_wptr) + WREG32(mmCP_HQD_PQ_WPTR, (wptr_val << wptr_shift) & wptr_mask); + + data = REG_SET_FIELD(m->cp_hqd_active, CP_HQD_ACTIVE, ACTIVE, 1); + WREG32(mmCP_HQD_ACTIVE, data); + + release_queue(adev); + + return 0; +} + +static int kgd_hqd_dump(struct amdgpu_device *adev, + uint32_t pipe_id, uint32_t queue_id, + uint32_t (**dump)[2], uint32_t *n_regs) +{ + uint32_t i = 0, reg; +#define HQD_N_REGS (35+4) +#define DUMP_REG(addr) do { \ + if (WARN_ON_ONCE(i >= HQD_N_REGS)) \ + break; \ + (*dump)[i][0] = (addr) << 2; \ + (*dump)[i++][1] = RREG32(addr); \ + } while (0) + + *dump = kmalloc_array(HQD_N_REGS * 2, sizeof(uint32_t), GFP_KERNEL); + if (*dump == NULL) + return -ENOMEM; + + acquire_queue(adev, pipe_id, queue_id); + + DUMP_REG(mmCOMPUTE_STATIC_THREAD_MGMT_SE0); + DUMP_REG(mmCOMPUTE_STATIC_THREAD_MGMT_SE1); + DUMP_REG(mmCOMPUTE_STATIC_THREAD_MGMT_SE2); + DUMP_REG(mmCOMPUTE_STATIC_THREAD_MGMT_SE3); + + for (reg = mmCP_MQD_BASE_ADDR; reg <= mmCP_MQD_CONTROL; reg++) + DUMP_REG(reg); + + release_queue(adev); + + WARN_ON_ONCE(i != HQD_N_REGS); + *n_regs = i; + + return 0; +} + +static int kgd_hqd_sdma_load(struct amdgpu_device *adev, void *mqd, + uint32_t __user *wptr, struct mm_struct *mm) +{ + struct cik_sdma_rlc_registers *m; + unsigned long end_jiffies; + uint32_t sdma_rlc_reg_offset; + uint32_t data; + + m = get_sdma_mqd(mqd); + sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m); + + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, + m->sdma_rlc_rb_cntl & (~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK)); + + end_jiffies = msecs_to_jiffies(2000) + jiffies; + while (true) { + data = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS); + if (data & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK) + break; + if (time_after(jiffies, end_jiffies)) { + pr_err("SDMA RLC not idle in %s\n", __func__); + return -ETIME; + } + usleep_range(500, 1000); + } + + data = REG_SET_FIELD(m->sdma_rlc_doorbell, SDMA0_RLC0_DOORBELL, + ENABLE, 1); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, data); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR, + m->sdma_rlc_rb_rptr); + + if (read_user_wptr(mm, wptr, data)) + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR, data); + else + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_WPTR, + m->sdma_rlc_rb_rptr); + + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_VIRTUAL_ADDR, + m->sdma_rlc_virtual_addr); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE, m->sdma_rlc_rb_base); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_BASE_HI, + m->sdma_rlc_rb_base_hi); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_LO, + m->sdma_rlc_rb_rptr_addr_lo); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR_ADDR_HI, + m->sdma_rlc_rb_rptr_addr_hi); + + data = REG_SET_FIELD(m->sdma_rlc_rb_cntl, SDMA0_RLC0_RB_CNTL, + RB_ENABLE, 1); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, data); + + return 0; +} + +static int kgd_hqd_sdma_dump(struct amdgpu_device *adev, + uint32_t engine_id, uint32_t queue_id, + uint32_t (**dump)[2], uint32_t *n_regs) +{ + uint32_t sdma_offset = engine_id * SDMA1_REGISTER_OFFSET + + queue_id * KFD_CIK_SDMA_QUEUE_OFFSET; + uint32_t i = 0, reg; +#undef HQD_N_REGS +#define HQD_N_REGS (19+4) + + *dump = kmalloc_array(HQD_N_REGS * 2, sizeof(uint32_t), GFP_KERNEL); + if (*dump == NULL) + return -ENOMEM; + + for (reg = mmSDMA0_RLC0_RB_CNTL; reg <= mmSDMA0_RLC0_DOORBELL; reg++) + DUMP_REG(sdma_offset + reg); + for (reg = mmSDMA0_RLC0_VIRTUAL_ADDR; reg <= mmSDMA0_RLC0_WATERMARK; + reg++) + DUMP_REG(sdma_offset + reg); + + WARN_ON_ONCE(i != HQD_N_REGS); + *n_regs = i; + + return 0; +} + +static bool kgd_hqd_is_occupied(struct amdgpu_device *adev, + uint64_t queue_address, uint32_t pipe_id, + uint32_t queue_id) +{ + uint32_t act; + bool retval = false; + uint32_t low, high; + + acquire_queue(adev, pipe_id, queue_id); + act = RREG32(mmCP_HQD_ACTIVE); + if (act) { + low = lower_32_bits(queue_address >> 8); + high = upper_32_bits(queue_address >> 8); + + if (low == RREG32(mmCP_HQD_PQ_BASE) && + high == RREG32(mmCP_HQD_PQ_BASE_HI)) + retval = true; + } + release_queue(adev); + return retval; +} + +static bool kgd_hqd_sdma_is_occupied(struct amdgpu_device *adev, void *mqd) +{ + struct cik_sdma_rlc_registers *m; + uint32_t sdma_rlc_reg_offset; + uint32_t sdma_rlc_rb_cntl; + + m = get_sdma_mqd(mqd); + sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m); + + sdma_rlc_rb_cntl = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL); + + if (sdma_rlc_rb_cntl & SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK) + return true; + + return false; +} + +static int kgd_hqd_destroy(struct amdgpu_device *adev, void *mqd, + enum kfd_preempt_type reset_type, + unsigned int utimeout, uint32_t pipe_id, + uint32_t queue_id) +{ + uint32_t temp; + enum hqd_dequeue_request_type type; + unsigned long flags, end_jiffies; + int retry; + + if (amdgpu_in_reset(adev)) + return -EIO; + + acquire_queue(adev, pipe_id, queue_id); + WREG32(mmCP_HQD_PQ_DOORBELL_CONTROL, 0); + + switch (reset_type) { + case KFD_PREEMPT_TYPE_WAVEFRONT_DRAIN: + type = DRAIN_PIPE; + break; + case KFD_PREEMPT_TYPE_WAVEFRONT_RESET: + type = RESET_WAVES; + break; + default: + type = DRAIN_PIPE; + break; + } + + /* Workaround: If IQ timer is active and the wait time is close to or + * equal to 0, dequeueing is not safe. Wait until either the wait time + * is larger or timer is cleared. Also, ensure that IQ_REQ_PEND is + * cleared before continuing. Also, ensure wait times are set to at + * least 0x3. + */ + local_irq_save(flags); + preempt_disable(); + retry = 5000; /* wait for 500 usecs at maximum */ + while (true) { + temp = RREG32(mmCP_HQD_IQ_TIMER); + if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, PROCESSING_IQ)) { + pr_debug("HW is processing IQ\n"); + goto loop; + } + if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, ACTIVE)) { + if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, RETRY_TYPE) + == 3) /* SEM-rearm is safe */ + break; + /* Wait time 3 is safe for CP, but our MMIO read/write + * time is close to 1 microsecond, so check for 10 to + * leave more buffer room + */ + if (REG_GET_FIELD(temp, CP_HQD_IQ_TIMER, WAIT_TIME) + >= 10) + break; + pr_debug("IQ timer is active\n"); + } else + break; +loop: + if (!retry) { + pr_err("CP HQD IQ timer status time out\n"); + break; + } + ndelay(100); + --retry; + } + retry = 1000; + while (true) { + temp = RREG32(mmCP_HQD_DEQUEUE_REQUEST); + if (!(temp & CP_HQD_DEQUEUE_REQUEST__IQ_REQ_PEND_MASK)) + break; + pr_debug("Dequeue request is pending\n"); + + if (!retry) { + pr_err("CP HQD dequeue request time out\n"); + break; + } + ndelay(100); + --retry; + } + local_irq_restore(flags); + preempt_enable(); + + WREG32(mmCP_HQD_DEQUEUE_REQUEST, type); + + end_jiffies = (utimeout * HZ / 1000) + jiffies; + while (true) { + temp = RREG32(mmCP_HQD_ACTIVE); + if (!(temp & CP_HQD_ACTIVE__ACTIVE_MASK)) + break; + if (time_after(jiffies, end_jiffies)) { + pr_err("cp queue preemption time out\n"); + release_queue(adev); + return -ETIME; + } + usleep_range(500, 1000); + } + + release_queue(adev); + return 0; +} + +static int kgd_hqd_sdma_destroy(struct amdgpu_device *adev, void *mqd, + unsigned int utimeout) +{ + struct cik_sdma_rlc_registers *m; + uint32_t sdma_rlc_reg_offset; + uint32_t temp; + unsigned long end_jiffies = (utimeout * HZ / 1000) + jiffies; + + m = get_sdma_mqd(mqd); + sdma_rlc_reg_offset = get_sdma_rlc_reg_offset(m); + + temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL); + temp = temp & ~SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK; + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, temp); + + while (true) { + temp = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_CONTEXT_STATUS); + if (temp & SDMA0_RLC0_CONTEXT_STATUS__IDLE_MASK) + break; + if (time_after(jiffies, end_jiffies)) { + pr_err("SDMA RLC not idle in %s\n", __func__); + return -ETIME; + } + usleep_range(500, 1000); + } + + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_DOORBELL, 0); + WREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL, + RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_CNTL) | + SDMA0_RLC0_RB_CNTL__RB_ENABLE_MASK); + + m->sdma_rlc_rb_rptr = RREG32(sdma_rlc_reg_offset + mmSDMA0_RLC0_RB_RPTR); + + return 0; +} + +static int kgd_wave_control_execute(struct amdgpu_device *adev, + uint32_t gfx_index_val, + uint32_t sq_cmd) +{ + uint32_t data; + + mutex_lock(&adev->grbm_idx_mutex); + + WREG32(mmGRBM_GFX_INDEX, gfx_index_val); + WREG32(mmSQ_CMD, sq_cmd); + + /* Restore the GRBM_GFX_INDEX register */ + + data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | + GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | + GRBM_GFX_INDEX__SE_BROADCAST_WRITES_MASK; + + WREG32(mmGRBM_GFX_INDEX, data); + + mutex_unlock(&adev->grbm_idx_mutex); + + return 0; +} + +static bool get_atc_vmid_pasid_mapping_info(struct amdgpu_device *adev, + uint8_t vmid, uint16_t *p_pasid) +{ + uint32_t value; + + value = RREG32(mmATC_VMID0_PASID_MAPPING + vmid); + *p_pasid = value & ATC_VMID0_PASID_MAPPING__PASID_MASK; + + return !!(value & ATC_VMID0_PASID_MAPPING__VALID_MASK); +} + +static void set_scratch_backing_va(struct amdgpu_device *adev, + uint64_t va, uint32_t vmid) +{ + lock_srbm(adev, 0, 0, 0, vmid); + WREG32(mmSH_HIDDEN_PRIVATE_BASE_VMID, va); + unlock_srbm(adev); +} + +static void set_vm_context_page_table_base(struct amdgpu_device *adev, + uint32_t vmid, uint64_t page_table_base) +{ + if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid)) { + pr_err("trying to set page table base for wrong VMID\n"); + return; + } + WREG32(mmVM_CONTEXT8_PAGE_TABLE_BASE_ADDR + vmid - 8, + lower_32_bits(page_table_base)); +} + + /** + * read_vmid_from_vmfault_reg - read vmid from register + * + * adev: amdgpu_device pointer + * @vmid: vmid pointer + * read vmid from register (CIK). + */ +static uint32_t read_vmid_from_vmfault_reg(struct amdgpu_device *adev) +{ + uint32_t status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS); + + return REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID); +} + +const struct kfd2kgd_calls gfx_v7_kfd2kgd = { + .program_sh_mem_settings = kgd_program_sh_mem_settings, + .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping, + .init_interrupts = kgd_init_interrupts, + .hqd_load = kgd_hqd_load, + .hqd_sdma_load = kgd_hqd_sdma_load, + .hqd_dump = kgd_hqd_dump, + .hqd_sdma_dump = kgd_hqd_sdma_dump, + .hqd_is_occupied = kgd_hqd_is_occupied, + .hqd_sdma_is_occupied = kgd_hqd_sdma_is_occupied, + .hqd_destroy = kgd_hqd_destroy, + .hqd_sdma_destroy = kgd_hqd_sdma_destroy, + .wave_control_execute = kgd_wave_control_execute, + .get_atc_vmid_pasid_mapping_info = get_atc_vmid_pasid_mapping_info, + .set_scratch_backing_va = set_scratch_backing_va, + .set_vm_context_page_table_base = set_vm_context_page_table_base, + .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg, +}; -- cgit v1.2.3