drm/amdgpu: mark soft recovered fences with -ENODATA

Set the fence error code before trying to soft-recover it.

It gets overwritten when a hard recovery is required.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 9fd55a10da191fae46a41dbf30a54f169b4fe713..80d6e132e4095d6b30811dca266bdfa88f77d240 100644 (file)
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -433,11 +433,18 @@ void amdgpu_ring_emit_reg_write_reg_wait_helper(struct amdgpu_ring *ring,
 bool amdgpu_ring_soft_recovery(struct amdgpu_ring *ring, unsigned int vmid,
                               struct dma_fence *fence)
 {
+       unsigned long flags;
+
        ktime_t deadline = ktime_add_us(ktime_get(), 10000);
 
        if (amdgpu_sriov_vf(ring->adev) || !ring->funcs->soft_recovery || !fence)
                return false;
 
+       spin_lock_irqsave(fence->lock, flags);
+       if (!dma_fence_is_signaled_locked(fence))
+               dma_fence_set_error(fence, -ENODATA);
+       spin_unlock_irqrestore(fence->lock, flags);
+
        atomic_inc(&ring->adev->gpu_reset_counter);
        while (!dma_fence_is_signaled(fence) &&
               ktime_to_ns(ktime_sub(deadline, ktime_get())) > 0)