/* * Copyright 2023 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 <linux/delay.h> #include <linux/firmware.h> #include <linux/module.h> #include <linux/pci.h> #include "amdgpu.h" #include "amdgpu_ucode.h" #include "amdgpu_trace.h" #include "gc/gc_12_0_0_offset.h" #include "gc/gc_12_0_0_sh_mask.h" #include "hdp/hdp_6_0_0_offset.h" #include "ivsrcid/gfx/irqsrcs_gfx_11_0_0.h" #include "soc15_common.h" #include "soc15.h" #include "sdma_v6_0_0_pkt_open.h" #include "nbio_v4_3.h" #include "sdma_common.h" #include "sdma_v7_0.h" #include "v12_structs.h" MODULE_FIRMWARE(…) …; MODULE_FIRMWARE(…) …; #define SDMA1_REG_OFFSET … #define SDMA0_HYP_DEC_REG_START … #define SDMA0_HYP_DEC_REG_END … #define SDMA1_HYP_DEC_REG_OFFSET … static void sdma_v7_0_set_ring_funcs(struct amdgpu_device *adev); static void sdma_v7_0_set_buffer_funcs(struct amdgpu_device *adev); static void sdma_v7_0_set_vm_pte_funcs(struct amdgpu_device *adev); static void sdma_v7_0_set_irq_funcs(struct amdgpu_device *adev); static int sdma_v7_0_start(struct amdgpu_device *adev); static u32 sdma_v7_0_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 internal_offset) { … } static unsigned sdma_v7_0_ring_init_cond_exec(struct amdgpu_ring *ring, uint64_t addr) { … } /** * sdma_v7_0_ring_get_rptr - get the current read pointer * * @ring: amdgpu ring pointer * * Get the current rptr from the hardware. */ static uint64_t sdma_v7_0_ring_get_rptr(struct amdgpu_ring *ring) { … } /** * sdma_v7_0_ring_get_wptr - get the current write pointer * * @ring: amdgpu ring pointer * * Get the current wptr from the hardware. */ static uint64_t sdma_v7_0_ring_get_wptr(struct amdgpu_ring *ring) { … } /** * sdma_v7_0_ring_set_wptr - commit the write pointer * * @ring: amdgpu ring pointer * * Write the wptr back to the hardware. */ static void sdma_v7_0_ring_set_wptr(struct amdgpu_ring *ring) { … } static void sdma_v7_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) { … } /** * sdma_v7_0_ring_emit_ib - Schedule an IB on the DMA engine * * @ring: amdgpu ring pointer * @job: job to retrieve vmid from * @ib: IB object to schedule * @flags: unused * * Schedule an IB in the DMA ring. */ static void sdma_v7_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job, struct amdgpu_ib *ib, uint32_t flags) { … } /** * sdma_v7_0_ring_emit_mem_sync - flush the IB by graphics cache rinse * * @ring: amdgpu ring pointer * * flush the IB by graphics cache rinse. */ static void sdma_v7_0_ring_emit_mem_sync(struct amdgpu_ring *ring) { … } /** * sdma_v7_0_ring_emit_hdp_flush - emit an hdp flush on the DMA ring * * @ring: amdgpu ring pointer * * Emit an hdp flush packet on the requested DMA ring. */ static void sdma_v7_0_ring_emit_hdp_flush(struct amdgpu_ring *ring) { … } /** * sdma_v7_0_ring_emit_fence - emit a fence on the DMA ring * * @ring: amdgpu ring pointer * @addr: address * @seq: fence seq number * @flags: fence flags * * Add a DMA fence packet to the ring to write * the fence seq number and DMA trap packet to generate * an interrupt if needed. */ static void sdma_v7_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq, unsigned flags) { … } /** * sdma_v7_0_gfx_stop - stop the gfx async dma engines * * @adev: amdgpu_device pointer * * Stop the gfx async dma ring buffers. */ static void sdma_v7_0_gfx_stop(struct amdgpu_device *adev) { … } /** * sdma_v7_0_rlc_stop - stop the compute async dma engines * * @adev: amdgpu_device pointer * * Stop the compute async dma queues. */ static void sdma_v7_0_rlc_stop(struct amdgpu_device *adev) { … } /** * sdma_v7_0_ctx_switch_enable - stop the async dma engines context switch * * @adev: amdgpu_device pointer * @enable: enable/disable the DMA MEs context switch. * * Halt or unhalt the async dma engines context switch. */ static void sdma_v7_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable) { … } /** * sdma_v7_0_enable - stop the async dma engines * * @adev: amdgpu_device pointer * @enable: enable/disable the DMA MEs. * * Halt or unhalt the async dma engines. */ static void sdma_v7_0_enable(struct amdgpu_device *adev, bool enable) { … } /** * sdma_v7_0_gfx_resume - setup and start the async dma engines * * @adev: amdgpu_device pointer * * Set up the gfx DMA ring buffers and enable them. * Returns 0 for success, error for failure. */ static int sdma_v7_0_gfx_resume(struct amdgpu_device *adev) { … } /** * sdma_v7_0_rlc_resume - setup and start the async dma engines * * @adev: amdgpu_device pointer * * Set up the compute DMA queues and enable them. * Returns 0 for success, error for failure. */ static int sdma_v7_0_rlc_resume(struct amdgpu_device *adev) { … } static void sdma_v12_0_free_ucode_buffer(struct amdgpu_device *adev) { … } /** * sdma_v7_0_load_microcode - load the sDMA ME ucode * * @adev: amdgpu_device pointer * * Loads the sDMA0/1 ucode. * Returns 0 for success, -EINVAL if the ucode is not available. */ static int sdma_v7_0_load_microcode(struct amdgpu_device *adev) { … } static int sdma_v7_0_soft_reset(void *handle) { … } static bool sdma_v7_0_check_soft_reset(void *handle) { … } /** * sdma_v7_0_start - setup and start the async dma engines * * @adev: amdgpu_device pointer * * Set up the DMA engines and enable them. * Returns 0 for success, error for failure. */ static int sdma_v7_0_start(struct amdgpu_device *adev) { … } static int sdma_v7_0_mqd_init(struct amdgpu_device *adev, void *mqd, struct amdgpu_mqd_prop *prop) { … } static void sdma_v7_0_set_mqd_funcs(struct amdgpu_device *adev) { … } /** * sdma_v7_0_ring_test_ring - simple async dma engine test * * @ring: amdgpu_ring structure holding ring information * * Test the DMA engine by writing using it to write an * value to memory. * Returns 0 for success, error for failure. */ static int sdma_v7_0_ring_test_ring(struct amdgpu_ring *ring) { … } /** * sdma_v7_0_ring_test_ib - test an IB on the DMA engine * * @ring: amdgpu_ring structure holding ring information * @timeout: timeout value in jiffies, or MAX_SCHEDULE_TIMEOUT * * Test a simple IB in the DMA ring. * Returns 0 on success, error on failure. */ static int sdma_v7_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) { … } /** * sdma_v7_0_vm_copy_pte - update PTEs by copying them from the GART * * @ib: indirect buffer to fill with commands * @pe: addr of the page entry * @src: src addr to copy from * @count: number of page entries to update * * Update PTEs by copying them from the GART using sDMA. */ static void sdma_v7_0_vm_copy_pte(struct amdgpu_ib *ib, uint64_t pe, uint64_t src, unsigned count) { … } /** * sdma_v7_0_vm_write_pte - update PTEs by writing them manually * * @ib: indirect buffer to fill with commands * @pe: addr of the page entry * @value: dst addr to write into pe * @count: number of page entries to update * @incr: increase next addr by incr bytes * * Update PTEs by writing them manually using sDMA. */ static void sdma_v7_0_vm_write_pte(struct amdgpu_ib *ib, uint64_t pe, uint64_t value, unsigned count, uint32_t incr) { … } /** * sdma_v7_0_vm_set_pte_pde - update the page tables using sDMA * * @ib: indirect buffer to fill with commands * @pe: addr of the page entry * @addr: dst addr to write into pe * @count: number of page entries to update * @incr: increase next addr by incr bytes * @flags: access flags * * Update the page tables using sDMA. */ static void sdma_v7_0_vm_set_pte_pde(struct amdgpu_ib *ib, uint64_t pe, uint64_t addr, unsigned count, uint32_t incr, uint64_t flags) { … } /** * sdma_v7_0_ring_pad_ib - pad the IB * * @ring: amdgpu ring pointer * @ib: indirect buffer to fill with padding * * Pad the IB with NOPs to a boundary multiple of 8. */ static void sdma_v7_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib) { … } /** * sdma_v7_0_ring_emit_pipeline_sync - sync the pipeline * * @ring: amdgpu_ring pointer * * Make sure all previous operations are completed (CIK). */ static void sdma_v7_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring) { … } /** * sdma_v7_0_ring_emit_vm_flush - vm flush using sDMA * * @ring: amdgpu_ring pointer * @vmid: vmid number to use * @pd_addr: address * * Update the page table base and flush the VM TLB * using sDMA. */ static void sdma_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr) { … } static void sdma_v7_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val) { … } static void sdma_v7_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, uint32_t val, uint32_t mask) { … } static void sdma_v7_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, uint32_t reg0, uint32_t reg1, uint32_t ref, uint32_t mask) { … } static int sdma_v7_0_early_init(void *handle) { … } static int sdma_v7_0_sw_init(void *handle) { … } static int sdma_v7_0_sw_fini(void *handle) { … } static int sdma_v7_0_hw_init(void *handle) { … } static int sdma_v7_0_hw_fini(void *handle) { … } static int sdma_v7_0_suspend(void *handle) { … } static int sdma_v7_0_resume(void *handle) { … } static bool sdma_v7_0_is_idle(void *handle) { … } static int sdma_v7_0_wait_for_idle(void *handle) { … } static int sdma_v7_0_ring_preempt_ib(struct amdgpu_ring *ring) { … } static int sdma_v7_0_set_trap_irq_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, unsigned type, enum amdgpu_interrupt_state state) { … } static int sdma_v7_0_process_trap_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { … } static int sdma_v7_0_process_illegal_inst_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { … } static int sdma_v7_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { … } static int sdma_v7_0_set_powergating_state(void *handle, enum amd_powergating_state state) { … } static void sdma_v7_0_get_clockgating_state(void *handle, u64 *flags) { … } const struct amd_ip_funcs sdma_v7_0_ip_funcs = …; static const struct amdgpu_ring_funcs sdma_v7_0_ring_funcs = …; static void sdma_v7_0_set_ring_funcs(struct amdgpu_device *adev) { … } static const struct amdgpu_irq_src_funcs sdma_v7_0_trap_irq_funcs = …; static const struct amdgpu_irq_src_funcs sdma_v7_0_illegal_inst_irq_funcs = …; static void sdma_v7_0_set_irq_funcs(struct amdgpu_device *adev) { … } /** * sdma_v7_0_emit_copy_buffer - copy buffer using the sDMA engine * * @ib: indirect buffer to fill with commands * @src_offset: src GPU address * @dst_offset: dst GPU address * @byte_count: number of bytes to xfer * @copy_flags: copy flags for the buffers * * Copy GPU buffers using the DMA engine. * Used by the amdgpu ttm implementation to move pages if * registered as the asic copy callback. */ static void sdma_v7_0_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, uint32_t byte_count, uint32_t copy_flags) { … } /** * sdma_v7_0_emit_fill_buffer - fill buffer using the sDMA engine * * @ib: indirect buffer to fill * @src_data: value to write to buffer * @dst_offset: dst GPU address * @byte_count: number of bytes to xfer * * Fill GPU buffers using the DMA engine. */ static void sdma_v7_0_emit_fill_buffer(struct amdgpu_ib *ib, uint32_t src_data, uint64_t dst_offset, uint32_t byte_count) { … } static const struct amdgpu_buffer_funcs sdma_v7_0_buffer_funcs = …; static void sdma_v7_0_set_buffer_funcs(struct amdgpu_device *adev) { … } static const struct amdgpu_vm_pte_funcs sdma_v7_0_vm_pte_funcs = …; static void sdma_v7_0_set_vm_pte_funcs(struct amdgpu_device *adev) { … } const struct amdgpu_ip_block_version sdma_v7_0_ip_block = …;
Codebase Browser
linux