diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c | 136 |
1 files changed, 55 insertions, 81 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index 4509bd4cce2d..92ce580647cd 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -57,22 +57,19 @@ static void sdma_v2_4_set_irq_funcs(struct amdgpu_device *adev); MODULE_FIRMWARE("amdgpu/topaz_sdma.bin"); MODULE_FIRMWARE("amdgpu/topaz_sdma1.bin"); -static const u32 sdma_offsets[SDMA_MAX_INSTANCE] = -{ +static const u32 sdma_offsets[SDMA_MAX_INSTANCE] = { SDMA0_REGISTER_OFFSET, SDMA1_REGISTER_OFFSET }; -static const u32 golden_settings_iceland_a11[] = -{ +static const u32 golden_settings_iceland_a11[] = { mmSDMA0_CHICKEN_BITS, 0xfc910007, 0x00810007, mmSDMA0_CLK_CTRL, 0xff000fff, 0x00000000, mmSDMA1_CHICKEN_BITS, 0xfc910007, 0x00810007, mmSDMA1_CLK_CTRL, 0xff000fff, 0x00000000, }; -static const u32 iceland_mgcg_cgcg_init[] = -{ +static const u32 iceland_mgcg_cgcg_init[] = { mmSDMA0_CLK_CTRL, 0xff000ff0, 0x00000100, mmSDMA1_CLK_CTRL, 0xff000ff0, 0x00000100 }; @@ -113,10 +110,9 @@ static void sdma_v2_4_init_golden_registers(struct amdgpu_device *adev) static void sdma_v2_4_free_microcode(struct amdgpu_device *adev) { int i; - for (i = 0; i < adev->sdma.num_instances; i++) { - release_firmware(adev->sdma.instance[i].fw); - adev->sdma.instance[i].fw = NULL; - } + + for (i = 0; i < adev->sdma.num_instances; i++) + amdgpu_ucode_release(&adev->sdma.instance[i].fw); } /** @@ -131,7 +127,6 @@ static void sdma_v2_4_free_microcode(struct amdgpu_device *adev) static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) { const char *chip_name; - char fw_name[30]; int err = 0, i; struct amdgpu_firmware_info *info = NULL; const struct common_firmware_header *header = NULL; @@ -143,18 +138,19 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) case CHIP_TOPAZ: chip_name = "topaz"; break; - default: BUG(); + default: + BUG(); } for (i = 0; i < adev->sdma.num_instances; i++) { if (i == 0) - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma.bin", chip_name); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_sdma.bin", chip_name); else - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_sdma1.bin", chip_name); - err = request_firmware(&adev->sdma.instance[i].fw, fw_name, adev->dev); - if (err) - goto out; - err = amdgpu_ucode_validate(adev->sdma.instance[i].fw); + err = amdgpu_ucode_request(adev, &adev->sdma.instance[i].fw, + AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_sdma1.bin", chip_name); if (err) goto out; hdr = (const struct sdma_firmware_header_v1_0 *)adev->sdma.instance[i].fw->data; @@ -175,11 +171,10 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) out: if (err) { - pr_err("sdma_v2_4: Failed to load firmware \"%s\"\n", fw_name); - for (i = 0; i < adev->sdma.num_instances; i++) { - release_firmware(adev->sdma.instance[i].fw); - adev->sdma.instance[i].fw = NULL; - } + pr_err("sdma_v2_4: Failed to load firmware \"%s_sdma%s.bin\"\n", + chip_name, i == 0 ? "" : "1"); + for (i = 0; i < adev->sdma.num_instances; i++) + amdgpu_ucode_release(&adev->sdma.instance[i].fw); } return err; } @@ -194,7 +189,7 @@ out: static uint64_t sdma_v2_4_ring_get_rptr(struct amdgpu_ring *ring) { /* XXX check if swapping is necessary on BE */ - return ring->adev->wb.wb[ring->rptr_offs] >> 2; + return *ring->rptr_cpu_addr >> 2; } /** @@ -223,7 +218,7 @@ static void sdma_v2_4_ring_set_wptr(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[ring->me], lower_32_bits(ring->wptr) << 2); + WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[ring->me], ring->wptr << 2); } static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) @@ -342,15 +337,9 @@ static void sdma_v2_4_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 se */ static void sdma_v2_4_gfx_stop(struct amdgpu_device *adev) { - struct amdgpu_ring *sdma0 = &adev->sdma.instance[0].ring; - struct amdgpu_ring *sdma1 = &adev->sdma.instance[1].ring; u32 rb_cntl, ib_cntl; int i; - if ((adev->mman.buffer_funcs_ring == sdma0) || - (adev->mman.buffer_funcs_ring == sdma1)) - amdgpu_ttm_set_buffer_funcs_status(adev, false); - for (i = 0; i < adev->sdma.num_instances; i++) { rb_cntl = RREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i]); rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0); @@ -414,12 +403,10 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) struct amdgpu_ring *ring; u32 rb_cntl, ib_cntl; u32 rb_bufsz; - u32 wb_offset; int i, j, r; for (i = 0; i < adev->sdma.num_instances; i++) { ring = &adev->sdma.instance[i].ring; - wb_offset = (ring->rptr_offs * 4); mutex_lock(&adev->srbm_mutex); for (j = 0; j < 16; j++) { @@ -455,9 +442,9 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) /* set the wb address whether it's enabled or not */ WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_HI + sdma_offsets[i], - upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF); + upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); WREG32(mmSDMA0_GFX_RB_RPTR_ADDR_LO + sdma_offsets[i], - lower_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFC); + lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); @@ -465,7 +452,7 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) WREG32(mmSDMA0_GFX_RB_BASE_HI + sdma_offsets[i], ring->gpu_addr >> 40); ring->wptr = 0; - WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], lower_32_bits(ring->wptr) << 2); + WREG32(mmSDMA0_GFX_RB_WPTR + sdma_offsets[i], ring->wptr << 2); /* enable DMA RB */ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1); @@ -478,8 +465,6 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) #endif /* enable DMA IBs */ WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl); - - ring->sched.ready = true; } sdma_v2_4_enable(adev, true); @@ -488,9 +473,6 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev) r = amdgpu_ring_test_helper(ring); if (r) return r; - - if (adev->mman.buffer_funcs_ring == ring) - amdgpu_ttm_set_buffer_funcs_status(adev, true); } return 0; @@ -651,7 +633,7 @@ static int sdma_v2_4_ring_test_ib(struct amdgpu_ring *ring, long timeout) r = -EINVAL; err1: - amdgpu_ib_free(adev, &ib, NULL); + amdgpu_ib_free(&ib, NULL); dma_fence_put(f); err0: amdgpu_device_wb_free(adev, index); @@ -827,12 +809,17 @@ static void sdma_v2_4_ring_emit_wreg(struct amdgpu_ring *ring, amdgpu_ring_write(ring, val); } -static int sdma_v2_4_early_init(void *handle) +static int sdma_v2_4_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; + int r; adev->sdma.num_instances = SDMA_MAX_INSTANCE; + r = sdma_v2_4_init_microcode(adev); + if (r) + return r; + sdma_v2_4_set_ring_funcs(adev); sdma_v2_4_set_buffer_funcs(adev); sdma_v2_4_set_vm_pte_funcs(adev); @@ -841,11 +828,11 @@ static int sdma_v2_4_early_init(void *handle) return 0; } -static int sdma_v2_4_sw_init(void *handle) +static int sdma_v2_4_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* SDMA trap event */ r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, VISLANDS30_IV_SRCID_SDMA_TRAP, @@ -865,12 +852,6 @@ static int sdma_v2_4_sw_init(void *handle) if (r) return r; - r = sdma_v2_4_init_microcode(adev); - if (r) { - DRM_ERROR("Failed to load sdma firmware!\n"); - return r; - } - for (i = 0; i < adev->sdma.num_instances; i++) { ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; @@ -887,9 +868,9 @@ static int sdma_v2_4_sw_init(void *handle) return r; } -static int sdma_v2_4_sw_fini(void *handle) +static int sdma_v2_4_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -899,10 +880,10 @@ static int sdma_v2_4_sw_fini(void *handle) return 0; } -static int sdma_v2_4_hw_init(void *handle) +static int sdma_v2_4_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; sdma_v2_4_init_golden_registers(adev); @@ -913,32 +894,26 @@ static int sdma_v2_4_hw_init(void *handle) return r; } -static int sdma_v2_4_hw_fini(void *handle) +static int sdma_v2_4_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - sdma_v2_4_enable(adev, false); + sdma_v2_4_enable(ip_block->adev, false); return 0; } -static int sdma_v2_4_suspend(void *handle) +static int sdma_v2_4_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v2_4_hw_fini(adev); + return sdma_v2_4_hw_fini(ip_block); } -static int sdma_v2_4_resume(void *handle) +static int sdma_v2_4_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v2_4_hw_init(adev); + return sdma_v2_4_hw_init(ip_block); } -static bool sdma_v2_4_is_idle(void *handle) +static bool sdma_v2_4_is_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp = RREG32(mmSRBM_STATUS2); if (tmp & (SRBM_STATUS2__SDMA_BUSY_MASK | @@ -948,11 +923,11 @@ static bool sdma_v2_4_is_idle(void *handle) return true; } -static int sdma_v2_4_wait_for_idle(void *handle) +static int sdma_v2_4_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { tmp = RREG32(mmSRBM_STATUS2) & (SRBM_STATUS2__SDMA_BUSY_MASK | @@ -965,10 +940,10 @@ static int sdma_v2_4_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int sdma_v2_4_soft_reset(void *handle) +static int sdma_v2_4_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp = RREG32(mmSRBM_STATUS2); if (tmp & SRBM_STATUS2__SDMA_BUSY_MASK) { @@ -1107,14 +1082,14 @@ static int sdma_v2_4_process_illegal_inst_irq(struct amdgpu_device *adev, return 0; } -static int sdma_v2_4_set_clockgating_state(void *handle, +static int sdma_v2_4_set_clockgating_state(struct amdgpu_ip_block *ip_block, enum amd_clockgating_state state) { /* XXX handled via the smc on VI */ return 0; } -static int sdma_v2_4_set_powergating_state(void *handle, +static int sdma_v2_4_set_powergating_state(struct amdgpu_ip_block *ip_block, enum amd_powergating_state state) { return 0; @@ -1123,7 +1098,6 @@ static int sdma_v2_4_set_powergating_state(void *handle, static const struct amd_ip_funcs sdma_v2_4_ip_funcs = { .name = "sdma_v2_4", .early_init = sdma_v2_4_early_init, - .late_init = NULL, .sw_init = sdma_v2_4_sw_init, .sw_fini = sdma_v2_4_sw_fini, .hw_init = sdma_v2_4_hw_init, @@ -1142,6 +1116,7 @@ static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = { .align_mask = 0xf, .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP), .support_64bit_ptrs = false, + .secure_submission_supported = true, .get_rptr = sdma_v2_4_ring_get_rptr, .get_wptr = sdma_v2_4_ring_get_wptr, .set_wptr = sdma_v2_4_ring_set_wptr, @@ -1197,7 +1172,7 @@ static void sdma_v2_4_set_irq_funcs(struct amdgpu_device *adev) * @src_offset: src GPU address * @dst_offset: dst GPU address * @byte_count: number of bytes to xfer - * @tmz: unused + * @copy_flags: unused * * Copy GPU buffers using the DMA engine (VI). * Used by the amdgpu ttm implementation to move pages if @@ -1207,7 +1182,7 @@ static void sdma_v2_4_emit_copy_buffer(struct amdgpu_ib *ib, uint64_t src_offset, uint64_t dst_offset, uint32_t byte_count, - bool tmz) + uint32_t copy_flags) { ib->ptr[ib->length_dw++] = SDMA_PKT_HEADER_OP(SDMA_OP_COPY) | SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_COPY_LINEAR); @@ -1277,8 +1252,7 @@ static void sdma_v2_4_set_vm_pte_funcs(struct amdgpu_device *adev) adev->vm_manager.vm_pte_num_scheds = adev->sdma.num_instances; } -const struct amdgpu_ip_block_version sdma_v2_4_ip_block = -{ +const struct amdgpu_ip_block_version sdma_v2_4_ip_block = { .type = AMD_IP_BLOCK_TYPE_SDMA, .major = 2, .minor = 4, |