From f5db59067c3130d186ddb619bcbdc6bea0ffb704 Mon Sep 17 00:00:00 2001 From: ganglxie Date: Thu, 24 Apr 2025 17:11:47 +0800 Subject: [PATCH 3619/5403] Refine RAS bad page records counting and parsing in eeprom V3 there is only MCA records in V3, no need to care about PA records. recalculate the value of ras_num_bad_pages when parsing failed and go on with the left records instead of quit. Signed-off-by: ganglxie Reviewed-by: Tao Zhou Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 62 ++++++++++++++----------- 1 file changed, 36 insertions(+), 26 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index f40b35f7f679..babc3c9cad65 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -2889,6 +2889,7 @@ static int __amdgpu_ras_convert_rec_from_rom(struct amdgpu_device *adev, if (amdgpu_ras_mca2pa_by_idx(adev, bps, err_data)) return -EINVAL; } + return __amdgpu_ras_restore_bad_pages(adev, err_data->err_addr, adev->umc.retire_unit); } @@ -2903,7 +2904,7 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, &adev->psp.ras_context.ras->eeprom_control; enum amdgpu_memory_partition nps = AMDGPU_NPS1_PARTITION_MODE; int ret = 0; - uint32_t i; + uint32_t i = 0; if (!con || !con->eh_data || !bps || pages <= 0) return 0; @@ -2924,34 +2925,36 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev, mutex_lock(&con->recovery_lock); if (from_rom) { - for (i = 0; i < pages; i++) { - if (control->ras_num_recs - i >= adev->umc.retire_unit) { - if ((bps[i].address == bps[i + 1].address) && - (bps[i].mem_channel == bps[i + 1].mem_channel)) { - //deal with retire_unit records a time - ret = __amdgpu_ras_convert_rec_array_from_rom(adev, - &bps[i], &err_data, nps); - if (ret) - goto free; - i += (adev->umc.retire_unit - 1); + /* there is no pa recs in V3, so skip pa recs processing */ + if (control->tbl_hdr.version < RAS_TABLE_VER_V3) { + for (i = 0; i < pages; i++) { + if (control->ras_num_recs - i >= adev->umc.retire_unit) { + if ((bps[i].address == bps[i + 1].address) && + (bps[i].mem_channel == bps[i + 1].mem_channel)) { + /* deal with retire_unit records a time */ + ret = __amdgpu_ras_convert_rec_array_from_rom(adev, + &bps[i], &err_data, nps); + if (ret) + control->ras_num_bad_pages -= adev->umc.retire_unit; + i += (adev->umc.retire_unit - 1); + } else { + break; + } } else { break; } - } else { - break; } } for (; i < pages; i++) { ret = __amdgpu_ras_convert_rec_from_rom(adev, &bps[i], &err_data, nps); if (ret) - goto free; + control->ras_num_bad_pages -= adev->umc.retire_unit; } } else { ret = __amdgpu_ras_restore_bad_pages(adev, bps, pages); } -free: if (from_rom) kfree(err_data.err_addr); mutex_unlock(&con->recovery_lock); @@ -3040,21 +3043,28 @@ static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev) dev_err(adev->dev, "Failed to load EEPROM table records!"); } else { if (adev->umc.ras && adev->umc.ras->convert_ras_err_addr) { - for (i = 0; i < control->ras_num_recs; i++) { - if ((control->ras_num_recs - i) >= adev->umc.retire_unit) { - if ((bps[i].address == bps[i + 1].address) && - (bps[i].mem_channel == bps[i + 1].mem_channel)) { - control->ras_num_pa_recs += adev->umc.retire_unit; - i += (adev->umc.retire_unit - 1); + /*In V3, there is no pa recs, and some cases(when address==0) may be parsed + as pa recs, so add verion check to avoid it. + */ + if (control->tbl_hdr.version < RAS_TABLE_VER_V3) { + for (i = 0; i < control->ras_num_recs; i++) { + if ((control->ras_num_recs - i) >= adev->umc.retire_unit) { + if ((bps[i].address == bps[i + 1].address) && + (bps[i].mem_channel == bps[i + 1].mem_channel)) { + control->ras_num_pa_recs += adev->umc.retire_unit; + i += (adev->umc.retire_unit - 1); + } else { + control->ras_num_mca_recs += + (control->ras_num_recs - i); + break; + } } else { - control->ras_num_mca_recs += - (control->ras_num_recs - i); + control->ras_num_mca_recs += (control->ras_num_recs - i); break; } - } else { - control->ras_num_mca_recs += (control->ras_num_recs - i); - break; } + } else { + control->ras_num_mca_recs = control->ras_num_recs; } } -- 2.54.0