RE: [PATCH] drm/amdgpu: skip reset other device in the same hive if it's sriov vf

[Liu, Shaoyun]

[AMD Official Use Only]

I think you need to describe more details on why the  hive reset on guest side 
is not necessary and  how host and guest driver will work  together to handle 
the hive  reset . You should have  2 patches together as a serials  to handle 
the FLR and  mode 1 reset on XGMI configuration.  
The describe is something  like :
  On SRIOV, host driver can support FLR(function level reset) on individual VF 
within the hive which might bring the individual device back to normal  without 
the necessary to execute the  hive reset.  If the FLR failed , host driver will 
trigger the hive reset , each guest VF will get reset notification before the 
real hive  reset been  executed . The VF device can handle the reset request 
individually in it's reset work handler . 

-----Original Message-----
From: amd-gfx <amd-gfx-boun...@lists.freedesktop.org> On Behalf Of Zhigang Luo
Sent: Friday, December 3, 2021 5:06 PM
To: amd-gfx@lists.freedesktop.org
Cc: Luo, Zhigang <zhigang....@amd.com>
Subject: [PATCH] drm/amdgpu: skip reset other device in the same hive if it's 
sriov vf

For sriov vf hang, vf flr will be triggered. Hive reset is not needed.

Signed-off-by: Zhigang Luo <zhigang....@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 3c5afa45173c..474f8ea58aa5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -4746,7 +4746,7 @@ static int amdgpu_device_lock_hive_adev(struct 
amdgpu_device *adev, struct amdgp  {
        struct amdgpu_device *tmp_adev = NULL;
 
-       if (adev->gmc.xgmi.num_physical_nodes > 1) {
+       if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1)) 
+{
                if (!hive) {
                        dev_err(adev->dev, "Hive is NULL while device has 
multiple xgmi nodes");
                        return -ENODEV;
@@ -4958,7 +4958,8 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
         * We always reset all schedulers for device and all devices for XGMI
         * hive so that should take care of them too.
         */
-       hive = amdgpu_get_xgmi_hive(adev);
+       if (!amdgpu_sriov_vf(adev))
+               hive = amdgpu_get_xgmi_hive(adev);
        if (hive) {
                if (atomic_cmpxchg(&hive->in_reset, 0, 1) != 0) {
                        DRM_INFO("Bailing on TDR for s_job:%llx, hive: %llx as 
another already in progress", @@ -4999,7 +5000,7 @@ int 
amdgpu_device_gpu_recover(struct amdgpu_device *adev,
         * to put adev in the 1st position.
         */
        INIT_LIST_HEAD(&device_list);
-       if (adev->gmc.xgmi.num_physical_nodes > 1) {
+       if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1)) 
+{
                list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head)
                        list_add_tail(&tmp_adev->reset_list, &device_list);
                if (!list_is_first(&adev->reset_list, &device_list))
--
2.17.1