diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 72 |
1 files changed, 67 insertions, 5 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 54ab51a4ada7..b6397d3229e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -33,6 +33,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" #include "amdgpu_reset.h" +#include "amdgpu_dpm.h" #include "vi.h" #include "soc15.h" #include "nv.h" @@ -86,8 +87,10 @@ int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init) if (virt->ops && virt->ops->req_full_gpu) { r = virt->ops->req_full_gpu(adev, init); - if (r) + if (r) { + adev->no_hw_access = true; return r; + } adev->virt.caps &= ~AMDGPU_SRIOV_CAPS_RUNTIME; } @@ -153,6 +156,20 @@ void amdgpu_virt_request_init_data(struct amdgpu_device *adev) } /** + * amdgpu_virt_ready_to_reset() - send ready to reset to host + * @adev: amdgpu device. + * Send ready to reset message to GPU hypervisor to signal we have stopped GPU + * activity and is ready for host FLR + */ +void amdgpu_virt_ready_to_reset(struct amdgpu_device *adev) +{ + struct amdgpu_virt *virt = &adev->virt; + + if (virt->ops && virt->ops->reset_gpu) + virt->ops->ready_to_reset(adev); +} + +/** * amdgpu_virt_wait_reset() - wait for reset gpu completed * @adev: amdgpu device. * Wait for GPU reset completed. @@ -215,6 +232,22 @@ void amdgpu_virt_free_mm_table(struct amdgpu_device *adev) adev->virt.mm_table.gpu_addr = 0; } +/** + * amdgpu_virt_rcvd_ras_interrupt() - receive ras interrupt + * @adev: amdgpu device. + * Check whether host sent RAS error message + * Return: true if found, otherwise false + */ +bool amdgpu_virt_rcvd_ras_interrupt(struct amdgpu_device *adev) +{ + struct amdgpu_virt *virt = &adev->virt; + + if (!virt->ops || !virt->ops->rcvd_ras_intr) + return false; + + return virt->ops->rcvd_ras_intr(adev); +} + unsigned int amd_sriov_msg_checksum(void *obj, unsigned long obj_size, @@ -395,6 +428,8 @@ static void amdgpu_virt_add_bad_page(struct amdgpu_device *adev, else vram_usage_va = adev->mman.drv_vram_usage_va; + memset(&bp, 0, sizeof(bp)); + if (bp_block_size) { bp_cnt = bp_block_size / sizeof(uint64_t); for (bp_idx = 0; bp_idx < bp_cnt; bp_idx++) { @@ -583,7 +618,7 @@ static int amdgpu_virt_write_vf2pf_data(struct amdgpu_device *adev) } vf2pf_info->checksum = amd_sriov_msg_checksum( - vf2pf_info, vf2pf_info->header.size, 0, 0); + vf2pf_info, sizeof(*vf2pf_info), 0, 0); return 0; } @@ -596,11 +631,14 @@ static void amdgpu_virt_update_vf2pf_work_item(struct work_struct *work) ret = amdgpu_virt_read_pf2vf_data(adev); if (ret) { adev->virt.vf2pf_update_retry_cnt++; - if ((adev->virt.vf2pf_update_retry_cnt >= AMDGPU_VF2PF_UPDATE_MAX_RETRY_LIMIT) && - amdgpu_sriov_runtime(adev) && !amdgpu_in_reset(adev)) { + + if ((amdgpu_virt_rcvd_ras_interrupt(adev) || + adev->virt.vf2pf_update_retry_cnt >= AMDGPU_VF2PF_UPDATE_MAX_RETRY_LIMIT) && + amdgpu_sriov_runtime(adev)) { + amdgpu_ras_set_fed(adev, true); if (amdgpu_reset_domain_schedule(adev->reset_domain, - &adev->virt.flr_work)) + &adev->kfd.reset_work)) return; else dev_err(adev->dev, "Failed to queue work! at %s", __func__); @@ -812,6 +850,13 @@ enum amdgpu_sriov_vf_mode amdgpu_virt_get_sriov_vf_mode(struct amdgpu_device *ad return mode; } +void amdgpu_virt_pre_reset(struct amdgpu_device *adev) +{ + /* stop the data exchange thread */ + amdgpu_virt_fini_data_exchange(adev); + amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_FLR); +} + void amdgpu_virt_post_reset(struct amdgpu_device *adev) { if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(11, 0, 3)) { @@ -820,6 +865,8 @@ void amdgpu_virt_post_reset(struct amdgpu_device *adev) */ adev->gfx.is_poweron = false; } + + adev->mes.ring[0].sched.ready = false; } bool amdgpu_virt_fw_load_skip_check(struct amdgpu_device *adev, uint32_t ucode_id) @@ -975,11 +1022,17 @@ u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 f return 0; } + if (amdgpu_device_skip_hw_access(adev)) + return 0; + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[xcc_id]; scratch_reg0 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg0; scratch_reg1 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg1; scratch_reg2 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg2; scratch_reg3 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg3; + + mutex_lock(&adev->virt.rlcg_reg_lock); + if (reg_access_ctrl->spare_int) spare_int = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->spare_int; @@ -1036,6 +1089,9 @@ u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 f } ret = readl(scratch_reg0); + + mutex_unlock(&adev->virt.rlcg_reg_lock); + return ret; } @@ -1045,6 +1101,9 @@ void amdgpu_sriov_wreg(struct amdgpu_device *adev, { u32 rlcg_flag; + if (amdgpu_device_skip_hw_access(adev)) + return; + if (!amdgpu_sriov_runtime(adev) && amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, true, &rlcg_flag)) { amdgpu_virt_rlcg_reg_rw(adev, offset, value, rlcg_flag, xcc_id); @@ -1062,6 +1121,9 @@ u32 amdgpu_sriov_rreg(struct amdgpu_device *adev, { u32 rlcg_flag; + if (amdgpu_device_skip_hw_access(adev)) + return 0; + if (!amdgpu_sriov_runtime(adev) && amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, false, &rlcg_flag)) return amdgpu_virt_rlcg_reg_rw(adev, offset, 0, rlcg_flag, xcc_id); |