[AMD Official Use Only - AMD Internal Distribution Only]
Reviewed-by: Feifei Xu <Feifei.Xu@xxxxxxx>
-----Original Message-----
From: amd-gfx <amd-gfx-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of Lijo Lazar
Sent: Tuesday, September 24, 2024 1:57 PM
To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Cc: Zhang, Hawking <Hawking.Zhang@xxxxxxx>; Deucher, Alexander <Alexander.Deucher@xxxxxxx>; Koenig, Christian <Christian.Koenig@xxxxxxx>; Bhardwaj, Rajneesh <Rajneesh.Bhardwaj@xxxxxxx>; Errabolu, Ramesh <Ramesh.Errabolu@xxxxxxx>
Subject: [PATCH 5/7] drm/amdgpu: Place NPS mode request on unload
If a user has requested NPS mode switch, place the request through PSP during unload of the driver. For devices which are part of a hive, all requests are placed together. If one of them fails, revert back to the current NPS mode.
Signed-off-by: Lijo Lazar <lijo.lazar@xxxxxxx>
Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 47 ++++++++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h | 2 + drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 38 +++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h | 4 ++
5 files changed, 92 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 95331294509c..d16cdcdb2114 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2428,6 +2428,7 @@ amdgpu_pci_remove(struct pci_dev *pdev)
struct amdgpu_device *adev = drm_to_adev(dev);
amdgpu_xcp_dev_unplug(adev);
+ amdgpu_gmc_prepare_nps_mode_change(adev);
drm_dev_unplug(dev);
if (adev->pm.rpm_mode != AMDGPU_RUNPM_NONE) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 24a1f931d9ed..21f1e65c9dc9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -1345,3 +1345,50 @@ int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev,
return psp_memory_partition(&adev->psp, nps_mode); }
+
+static inline bool amdgpu_gmc_need_nps_switch_req(struct amdgpu_device *adev,
+ int req_nps_mode,
+ int cur_nps_mode)
+{
+ return (((BIT(req_nps_mode) & adev->gmc.supported_nps_modes) ==
+ BIT(req_nps_mode)) &&
+ req_nps_mode != cur_nps_mode);
+}
+
+void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev) {
+ int req_nps_mode, cur_nps_mode, r;
+ struct amdgpu_hive_info *hive;
+
+ if (amdgpu_sriov_vf(adev) || !adev->gmc.supported_nps_modes ||
+ !adev->gmc.gmc_funcs->request_mem_partition_mode)
+ return;
+
+ cur_nps_mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev);
+ hive = amdgpu_get_xgmi_hive(adev);
+ if (hive) {
+ req_nps_mode = atomic_read(&hive->requested_nps_mode);
+ if (!amdgpu_gmc_need_nps_switch_req(adev, req_nps_mode,
+ cur_nps_mode)) {
+ amdgpu_put_xgmi_hive(hive);
+ return;
+ }
+ r = amdgpu_xgmi_request_nps_change(adev, hive, req_nps_mode);
+ amdgpu_put_xgmi_hive(hive);
+ goto out;
+ }
+
+ req_nps_mode = adev->gmc.requested_nps_mode;
+ if (!amdgpu_gmc_need_nps_switch_req(adev, req_nps_mode, cur_nps_mode))
+ return;
+
+ /* even if this fails, we should let driver unload w/o blocking */
+ r = adev->gmc.gmc_funcs->request_mem_partition_mode(adev,
+req_nps_mode);
+out:
+ if (r)
+ dev_err(adev->dev, "NPS mode change request failed\n");
+ else
+ dev_info(
+ adev->dev,
+ "NPS mode change request done, reload driver to complete the
+change\n"); }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 1a50639a003a..b13d6adb5efd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -467,4 +467,6 @@ int amdgpu_gmc_get_nps_memranges(struct amdgpu_device *adev,
int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev,
int nps_mode);
+void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev);
+
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 5d721ccb9dfd..db2c1b11b813 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -1564,3 +1564,41 @@ int amdgpu_xgmi_reset_on_init(struct amdgpu_device *adev)
return 0;
}
+
+int amdgpu_xgmi_request_nps_change(struct amdgpu_device *adev,
+ struct amdgpu_hive_info *hive,
+ int req_nps_mode)
+{
+ struct amdgpu_device *tmp_adev;
+ int cur_nps_mode, r;
+
+ /* This is expected to be called only during unload of driver. The
+ * request needs to be placed only once for all devices in the hive. If
+ * one of them fail, revert the request for previous successful devices.
+ * After placing the request, make hive mode as UNKNOWN so that other
+ * devices don't request anymore.
+ */
+ mutex_lock(&hive->hive_lock);
+ list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+ r = adev->gmc.gmc_funcs->request_mem_partition_mode(
+ tmp_adev, req_nps_mode);
+ if (r)
+ goto err;
+ }
+ /* Set to UNKNOWN so that other devices don't request anymore */
+ atomic_set(&hive->requested_nps_mode, UNKNOWN_MEMORY_PARTITION_MODE);
+
+ mutex_unlock(&hive->hive_lock);
+
+ return 0;
+err:
+ /* Request back current mode if one of the requests failed */
+ cur_nps_mode = adev->gmc.gmc_funcs->query_mem_partition_mode(tmp_adev);
+ list_for_each_entry_continue_reverse(tmp_adev, &hive->device_list,
+ gmc.xgmi.head)
+ adev->gmc.gmc_funcs->request_mem_partition_mode(tmp_adev,
+ cur_nps_mode);
+ mutex_lock(&hive->hive_lock);
+
+ return r;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
index 67abadb4f298..41d5f97fc77a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
@@ -79,4 +79,8 @@ static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, int amdgpu_xgmi_ras_sw_init(struct amdgpu_device *adev); int amdgpu_xgmi_reset_on_init(struct amdgpu_device *adev);
+int amdgpu_xgmi_request_nps_change(struct amdgpu_device *adev,
+ struct amdgpu_hive_info *hive,
+ int req_nps_mode);
+
#endif
--
2.25.1
