On 2023/12/28 14:17, Tian, Kevin wrote:
From: Liu, Yi L <yi.l....@intel.com> Sent: Thursday, December 28, 2023 12:14 AM@@ -1376,6 +1383,8 @@ int qi_submit_sync(struct intel_iommu *iommu, struct qi_desc *desc, restart: rc = 0; + if (fault) + *fault = 0;move it to right before the loop of qi_check_fault()
ok.
raw_spin_lock_irqsave(&qi->q_lock, flags); /* @@ -1430,7 +1439,7 @@ int qi_submit_sync(struct intel_iommu *iommu, struct qi_desc *desc, * a deadlock where the interrupt context can wait indefinitely * for free slots in the queue. */ - rc = qi_check_fault(iommu, index, wait_index); + rc = qi_check_fault(iommu, index, wait_index, fault); if (rc) break;and as replied in another thread let's change qi_check_fault to return -ETIMEDOUT to break the restart loop when fault pointer is valid.
sure. -- Regards, Yi Liu
