From: Mukul Joshi <[email protected]> When two threads import the same UALINK handle concurrently, the first xa_insert() succeeds and the second gets -EBUSY. Convert -EBUSY to -EAGAIN so user-space retries instead of seeing a spurious error.
Signed-off-by: Mukul Joshi <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]> --- drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c index a5dc6a13970f2..68c156c1d281b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ualink.c @@ -3852,9 +3852,15 @@ int amdgpu_ualink_import_handle(struct drm_device *dev, imp_xa_node, GFP_KERNEL); if (r) { kfree(imp_xa_node); - dev_err(adev->dev, - "IMPORT: XA insert failed for handle:%llx:%llx err:%d\n", - handle.handle_hi, handle.handle_lo, r); + /* -EBUSY means another thread raced us and inserted a + * node for the same handle. Ask user-space to retry. + */ + if (r == -EBUSY) + r = -EAGAIN; + else + dev_err(adev->dev, + "IMPORT: XA insert failed for handle:%llx:%llx err:%d\n", + handle.handle_hi, handle.handle_lo, r); goto out; } -- 2.55.0
