From ab66c832847fcdffc97d4591ba5547e3990d9d33 Mon Sep 17 00:00:00 2001 From: Zhigang Luo Date: Thu, 29 Feb 2024 16:04:35 -0500 Subject: drm/amdgpu: trigger flr_work if reading pf2vf data failed if reading pf2vf data failed 30 times continuously, it means something is wrong. Need to trigger flr_work to recover the issue. also use dev_err to print the error message to get which device has issue and add warning message if waiting IDH_FLR_NOTIFICATION_CMPL timeout. Signed-off-by: Zhigang Luo Acked-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c | 29 ++++++++++++++++++++++++----- 1 file changed, 24 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 7a4eae36778a..aed60aaf1a55 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -32,6 +32,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" +#include "amdgpu_reset.h" #include "vi.h" #include "soc15.h" #include "nv.h" @@ -424,7 +425,7 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev) return -EINVAL; if (pf2vf_info->size > 1024) { - DRM_ERROR("invalid pf2vf message size\n"); + dev_err(adev->dev, "invalid pf2vf message size: 0x%x\n", pf2vf_info->size); return -EINVAL; } @@ -435,7 +436,9 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev) adev->virt.fw_reserve.p_pf2vf, pf2vf_info->size, adev->virt.fw_reserve.checksum_key, checksum); if (checksum != checkval) { - DRM_ERROR("invalid pf2vf message\n"); + dev_err(adev->dev, + "invalid pf2vf message: header checksum=0x%x calculated checksum=0x%x\n", + checksum, checkval); return -EINVAL; } @@ -449,7 +452,9 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev) adev->virt.fw_reserve.p_pf2vf, pf2vf_info->size, 0, checksum); if (checksum != checkval) { - DRM_ERROR("invalid pf2vf message\n"); + dev_err(adev->dev, + "invalid pf2vf message: header checksum=0x%x calculated checksum=0x%x\n", + checksum, checkval); return -EINVAL; } @@ -485,7 +490,7 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev) ((struct amd_sriov_msg_pf2vf_info *)pf2vf_info)->uuid; break; default: - DRM_ERROR("invalid pf2vf version\n"); + dev_err(adev->dev, "invalid pf2vf version: 0x%x\n", pf2vf_info->version); return -EINVAL; } @@ -584,8 +589,21 @@ static void amdgpu_virt_update_vf2pf_work_item(struct work_struct *work) int ret; ret = amdgpu_virt_read_pf2vf_data(adev); - if (ret) + 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_reset_domain_schedule(adev->reset_domain, + &adev->virt.flr_work)) + return; + else + dev_err(adev->dev, "Failed to queue work! at %s", __func__); + } + goto out; + } + + adev->virt.vf2pf_update_retry_cnt = 0; amdgpu_virt_write_vf2pf_data(adev); out: @@ -606,6 +624,7 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev) adev->virt.fw_reserve.p_pf2vf = NULL; adev->virt.fw_reserve.p_vf2pf = NULL; adev->virt.vf2pf_update_interval_ms = 0; + adev->virt.vf2pf_update_retry_cnt = 0; if (adev->mman.fw_vram_usage_va && adev->mman.drv_vram_usage_va) { DRM_WARN("Currently fw_vram and drv_vram should not have values at the same time!"); -- cgit v1.2.3-70-g09d2