diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 111 |
1 files changed, 84 insertions, 27 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index cef94e2169fe..43055a01f35e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -31,6 +31,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" #include "amdgpu_atomfirmware.h" +#include "amdgpu_xgmi.h" #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h" const char *ras_error_string[] = { @@ -720,6 +721,9 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, if (adev->nbio.funcs->query_ras_error_count) adev->nbio.funcs->query_ras_error_count(adev, &err_data); break; + case AMDGPU_RAS_BLOCK__XGMI_WAFL: + amdgpu_xgmi_query_ras_error_count(adev, &err_data); + break; default: break; } @@ -742,20 +746,6 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, return 0; } -uint64_t get_xgmi_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr) -{ - uint32_t df_inst_id; - - if ((!adev->df.funcs) || - (!adev->df.funcs->get_df_inst_id) || - (!adev->df.funcs->get_dram_base_addr)) - return addr; - - df_inst_id = adev->df.funcs->get_df_inst_id(adev); - - return addr + adev->df.funcs->get_dram_base_addr(adev, df_inst_id); -} - /* wrapper of psp_ras_trigger_error */ int amdgpu_ras_error_inject(struct amdgpu_device *adev, struct ras_inject_if *info) @@ -775,8 +765,9 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, /* Calculate XGMI relative offset */ if (adev->gmc.xgmi.num_physical_nodes > 1) { - block_info.address = get_xgmi_relative_phy_addr(adev, - block_info.address); + block_info.address = + amdgpu_xgmi_get_relative_phy_addr(adev, + block_info.address); } switch (info->head.block) { @@ -1122,6 +1113,32 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, &amdgpu_ras_debugfs_ops); } +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj; + struct ras_fs_if fs_info; + + /* + * it won't be called in resume path, no need to check + * suspend and gpu reset status + */ + if (!con) + return; + + amdgpu_ras_debugfs_create_ctrl_node(adev); + + list_for_each_entry(obj, &con->head, node) { + if (amdgpu_ras_is_supported(adev, obj->head.block) && + (obj->attr_inuse == 1)) { + sprintf(fs_info.debugfs_name, "%s_err_inject", + ras_block_str(obj->head.block)); + fs_info.head = obj->head; + amdgpu_ras_debugfs_create(adev, &fs_info); + } + } +} + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head) { @@ -1154,7 +1171,6 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev) static int amdgpu_ras_fs_init(struct amdgpu_device *adev) { amdgpu_ras_sysfs_create_feature_node(adev); - amdgpu_ras_debugfs_create_ctrl_node(adev); return 0; } @@ -1319,6 +1335,33 @@ static int amdgpu_ras_interrupt_remove_all(struct amdgpu_device *adev) } /* ih end */ +/* traversal all IPs except NBIO to query error counter */ +static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj; + + if (!con) + return; + + list_for_each_entry(obj, &con->head, node) { + struct ras_query_if info = { + .head = obj->head, + }; + + /* + * PCIE_BIF IP has one different isr by ras controller + * interrupt, the specific ras counter query will be + * done in that isr. So skip such block from common + * sync flood interrupt isr calling. + */ + if (info.head.block == AMDGPU_RAS_BLOCK__PCIE_BIF) + continue; + + amdgpu_ras_error_query(adev, &info); + } +} + /* recovery begin */ /* return 0 on success. @@ -1373,6 +1416,12 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) struct amdgpu_ras *ras = container_of(work, struct amdgpu_ras, recovery_work); + /* + * Query and print non zero error counter per IP block for + * awareness before recovering GPU. + */ + amdgpu_ras_log_on_err_counter(ras->adev); + if (amdgpu_device_should_recover_gpu(ras->adev)) amdgpu_device_gpu_recover(ras->adev, 0); atomic_set(&ras->in_recovery, 0); @@ -1713,18 +1762,30 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev, *hw_supported = 0; *supported = 0; - if (amdgpu_sriov_vf(adev) || + if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw || (adev->asic_type != CHIP_VEGA20 && adev->asic_type != CHIP_ARCTURUS)) return; - if (adev->is_atom_fw && - (amdgpu_atomfirmware_mem_ecc_supported(adev) || - amdgpu_atomfirmware_sram_ecc_supported(adev))) - *hw_supported = AMDGPU_RAS_BLOCK_MASK; + if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { + DRM_INFO("HBM ECC is active.\n"); + *hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("HBM ECC is not presented.\n"); + + if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { + DRM_INFO("SRAM ECC is active.\n"); + *hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("SRAM ECC is not presented.\n"); + + /* hw_supported needs to be aligned with RAS block mask. */ + *hw_supported &= AMDGPU_RAS_BLOCK_MASK; *supported = amdgpu_ras_enable == 0 ? - 0 : *hw_supported & amdgpu_ras_mask; + 0 : *hw_supported & amdgpu_ras_mask; } int amdgpu_ras_init(struct amdgpu_device *adev) @@ -1825,8 +1886,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, goto interrupt; } - amdgpu_ras_debugfs_create(adev, fs_info); - r = amdgpu_ras_sysfs_create(adev, fs_info); if (r) goto sysfs; @@ -1835,7 +1894,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, cleanup: amdgpu_ras_sysfs_remove(adev, ras_block); sysfs: - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); interrupt: @@ -1852,7 +1910,6 @@ void amdgpu_ras_late_fini(struct amdgpu_device *adev, return; amdgpu_ras_sysfs_remove(adev, ras_block); - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); amdgpu_ras_feature_enable(adev, ras_block, 0); |