[PATCH] drm/amdgpu: Move mca debug mode decision to ras

[Wang, Yang(Kevin)]

Hi Lijo,

after our private discussion, in driver polling mode,  the RAS will return 0 error count in driver probe stage (at that time the debug mode is in off state).
In the case of driver polling mode, debug mode needs to be turned on in advance.

Best Regards,
Kevin

-----Original Message-----
From: Lazar, Lijo <Lijo.Lazar at amd.com> 
Sent: Thursday, November 9, 2023 1:26 PM
To: amd-gfx at lists.freedesktop.org
Cc: Zhang, Hawking <Hawking.Zhang at amd.com>; Deucher, Alexander <Alexander.Deucher at amd.com>; Wang, Yang(Kevin) <KevinYang.Wang at amd.com>; Zhou1, Tao <Tao.Zhou1 at amd.com>
Subject: [PATCH] drm/amdgpu: Move mca debug mode decision to ras

Refactor code such that ras block decides the default mca debug mode, and not swsmu block.

By default mca debug mode is set to true for now.

Signed-off-by: Lijo Lazar <lijo.lazar at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c            |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c            | 14 +++++++++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h            |  2 +-
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c   | 12 ------------
 4 files changed, 13 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
index 5a828c175e3a..cb51a46e8535 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c
@@ -218,7 +218,7 @@ static int amdgpu_mca_smu_debug_mode_set(void *data, u64 val)
 	struct amdgpu_device *adev = (struct amdgpu_device *)data;
 	int ret;
 
-	ret = amdgpu_mca_smu_set_debug_mode(adev, val ? true : false);
+	ret = amdgpu_ras_set_mca_debug_mode(adev, val ? true : false);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index b7fe5951b166..b2e9ed65a061 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -3120,6 +3120,8 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev)
 			amdgpu_ras_block_late_init_default(adev, &obj->ras_comm);
 	}
 
+	amdgpu_ras_set_mca_debug_mode(adev, true);
+
 	return 0;
 }
 
@@ -3375,12 +3377,18 @@ int amdgpu_ras_reset_gpu(struct amdgpu_device *adev)
 	return 0;
 }
 
-void amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool enable)
+int amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool 
+enable)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	int ret;
 
-	if (con)
-		con->is_mca_debug_mode = enable;
+	if (con) {
+		ret = amdgpu_mca_smu_set_debug_mode(adev, enable);
+		if (!ret)
+			con->is_mca_debug_mode = enable;
+	}
+
+	return ret;
 }
 
 bool amdgpu_ras_get_mca_debug_mode(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 665414c22ca9..e1e0fc66ea49 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -767,7 +767,7 @@ struct amdgpu_ras* amdgpu_ras_get_context(struct amdgpu_device *adev);
 
 int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_con);
 
-void amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool enable);
+int amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool 
+enable);
 bool amdgpu_ras_get_mca_debug_mode(struct amdgpu_device *adev);
 
 int amdgpu_ras_register_ras_block(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index cecd210397d6..404db2b0fb71 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -1524,7 +1524,6 @@ static int smu_v13_0_6_mca_set_debug_mode(struct smu_context *smu, bool enable)
 	if (smu->smc_fw_version < 0x554800)
 		return 0;
 
-	amdgpu_ras_set_mca_debug_mode(smu->adev, enable);
 	return smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_ClearMcaOnRead,
 					       enable ? 0 : ClearMcaOnRead_UE_FLAG_MASK | ClearMcaOnRead_CE_POLL_MASK,
 					       NULL);
@@ -2346,16 +2345,6 @@ static int smu_v13_0_6_smu_send_hbm_bad_page_num(struct smu_context *smu,
 	return ret;
 }
 
-static int smu_v13_0_6_post_init(struct smu_context *smu) -{
-	struct amdgpu_device *adev = smu->adev;
-
-	if (!amdgpu_sriov_vf(adev) && adev->ras_enabled)
-		return smu_v13_0_6_mca_set_debug_mode(smu, true);
-
-	return 0;
-}
-
 static int mca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable)  {
 	struct smu_context *smu = adev->powerplay.pp_handle; @@ -2945,7 +2934,6 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.i2c_init = smu_v13_0_6_i2c_control_init,
 	.i2c_fini = smu_v13_0_6_i2c_control_fini,
 	.send_hbm_bad_pages_num = smu_v13_0_6_smu_send_hbm_bad_page_num,
-	.post_init = smu_v13_0_6_post_init,
 };
 
 void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
--
2.25.1