On 2019-07-04 15:43, Zhenyu Wang wrote:
On 2019.07.04 15:06:13 +0800, Colin Xu wrote:
Windows guest can't run after force-TDR with host log:
...
gvt: vgpu 1: workload shadow ppgtt isn't ready
gvt: vgpu 1: fail to dispatch workload, skip
...
The error is raised by set_context_ppgtt_from_shadow(), when it checks
and found the shadow_mm isn't marked as shadowed.
In work thread before each submission, a shadow_mm is set to shadowed in:
shadow_ppgtt_mm()
<-intel_vgpu_pin_mm()
<-prepare_workload()
<-dispatch_workload()
<-workload_thread()
However checking whether or not shadow_mm is shadowed is prior to it:
set_context_ppgtt_from_shadow()
<-dispatch_workload()
<-workload_thread()
In normal case, create workload will check the existence of shadow_mm,
if not it will create a new one and marked as shadowed. If already exist
it will reuse the old one. Since shadow_mm is reused, checking of shadowed
in set_context_ppgtt_from_shadow() actually always see the state set in
creation, but not the state set in intel_vgpu_pin_mm().
When force-TDR, all engines are reset, since it's not dmlr level, all
ppgtt_mm are invalidated but not destroyed. Invalidation will mark all
reused shadow_mm as not shadowed but still keeps in ppgtt_mm_list_head.
If workload submission phase those shadow_mm are reused with shadowed
not set, then set_context_ppgtt_from_shadow() will report error.
Fixes: 4f15665ccbba (drm/i915: Add ppgtt to GVT GEM context)
v2:
Move set_context_ppgtt_from_shadow() after prepare_workload(). (zhenyu)
Cc: stable@xxxxxxxxxxxxxxx
Signed-off-by: Colin Xu <colin.xu@xxxxxxxxx>
---
drivers/gpu/drm/i915/gvt/scheduler.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c
index 196b4155a309..100040209188 100644
--- a/drivers/gpu/drm/i915/gvt/scheduler.c
+++ b/drivers/gpu/drm/i915/gvt/scheduler.c
@@ -685,13 +685,6 @@ static int dispatch_workload(struct intel_vgpu_workload *workload)
mutex_lock(&vgpu->vgpu_lock);
mutex_lock(&dev_priv->drm.struct_mutex);
- ret = set_context_ppgtt_from_shadow(workload,
- s->shadow[ring_id]->gem_context);
- if (ret < 0) {
- gvt_vgpu_err("workload shadow ppgtt isn't ready\n");
- goto err_req;
- }
-
ret = intel_gvt_workload_req_alloc(workload);
if (ret)
goto err_req;
@@ -707,6 +700,13 @@ static int dispatch_workload(struct intel_vgpu_workload *workload)
}
ret = prepare_workload(workload);
+ if (ret)
+ goto out;
+
+ ret = set_context_ppgtt_from_shadow(workload,
+ s->shadow[ring_id]->gem_context);
+ if (ret)
+ gvt_vgpu_err("workload shadow ppgtt isn't ready\n");
As workload's shadow_mm should always be for ppgtt, so we don't need return
for set_context_ppgtt_from_shadow, can just be void. Then how about do that
in prepare_workload after we settle down shadow pdp?
If so, is checking mm->type and shadowed flag stil necessary? since intel_vgpu_pin_mm()
in prepare_workload will guaranee that, if intel_vgpu_pin_mm() fails, shadow pdp won't
get updated and no need to pin. Am I right?
out:
if (ret) {
/* We might still need to add request with
--
2.22.0
--
Best Regards,
Colin Xu
