Hello Ryder Lee, The patch 57b9df6fa5f5: "mt76: mt7915: fix possible deadlock in mt7915_stop" from May 12, 2020, leads to the following static checker warning: drivers/net/wireless/mediatek/mt76/mt7915/init.c:595 mt7915_register_ext_phy() error: we previously assumed 'phy' could be null (see line 592) drivers/net/wireless/mediatek/mt76/mt7615/init.c:371 mt7615_register_ext_phy() error: we previously assumed 'phy' could be null (see line 368) drivers/net/wireless/mediatek/mt76/mt7915/init.c 589 if (test_bit(MT76_STATE_RUNNING, &dev->mphy.state)) 590 return -EINVAL; 591 592 if (phy) ^^^ 593 return 0; 594 595 INIT_DELAYED_WORK(&phy->mac_work, mt7915_mac_work); ^^^^^^^^^^^^^^ This will Oops. 596 mt7915_cap_dbdc_enable(dev); 597 mphy = mt76_alloc_phy(&dev->mt76, sizeof(*phy), &mt7915_ops); 598 if (!mphy) 599 return -ENOMEM; 600 601 phy = mphy->priv; ^^^^^^^^^^^^^^^^ phy is set here again. 602 phy->dev = dev; 603 phy->mt76 = mphy; 604 phy->chainmask = dev->chainmask & ~dev->phy.chainmask; 605 mphy->antenna_mask = BIT(hweight8(phy->chainmask)) - 1; 606 mt7915_init_wiphy(mphy->hw); See also: drivers/net/wireless/mediatek/mt76/mt7615/init.c:371 mt7615_register_ext_phy() error: we previously assumed 'phy' could be null (see line 368) regards, dan carpenter