Hi Jiange
You need to send the patch to amd-gfx for review if it intend to go drm-next branch
For gibraltar branch your patch is -1 by CI
_____________________________________
Monk Liu|GPU Virtualization Team |AMD
From: Zhao, Jiange <Jiange.Zhao@xxxxxxx>
Sent: Wednesday, October 30, 2019 10:26 AM
To: Zhao, Jiange <Jiange.Zhao@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Cc: Deng, Emily <Emily.Deng@xxxxxxx>; Liu, Monk <Monk.Liu@xxxxxxx>; Chang, HaiJun <HaiJun.Chang@xxxxxxx>
Subject: Re: [PATCH] drm/amdgpu/SRIOV: Only reset hw.status for target IP
From: Jiange Zhao <Jiange.Zhao@xxxxxxx>
In the old way, when doing IH hw_init, PSP, nv_common
and GMC hw.status would be reset to false, even though
their hw_init have been done. In the next step, fw_loading,
PSP would do hw_init again.
In the new way, only reset hw.status to false for the target
IP in the list. In this way, PSP will only do hw_init once.
Signed-off-by: Jiange Zhao <Jiange.Zhao@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 4eee40b9d0b0..ad6d2452fed9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2352,11 +2352,11 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev)
for (j = 0; j < adev->num_ip_blocks; j++) {
block = &adev->ip_blocks[j];
- block->status.hw = false;
if (block->version->type != ip_order[i] ||
!block->status.valid)
continue;
+ block->status.hw = false;
r = block->version->funcs->hw_init(adev);
DRM_INFO("RE-INIT-early: %s %s\n", block->version->funcs->name, r?"failed":"succeeded");
if (r)
--
2.20.1
|
