Re: [PATCH net-next v7 1/2] net: refactor set_rx_mode into snapshot and deferred I/O
From: I Viswanath <hidden>
Date: 2026-01-06 10:28:30
Also in:
lkml, virtualization
quoted hunk ↗ jump to hunk
+static void netif_free_cleanup_work(struct net_device *dev) +{ + if (!dev->cleanup_work) + return; + + cancel_work_sync(&dev->cleanup_work->work); + kfree(dev->cleanup_work); + dev->cleanup_work = NULL; +} +@@ -1682,6 +1882,16 @@ static int __dev_open(struct net_device *dev, struct netlink_ext_ack *extack) if (!ret && ops->ndo_open) ret = ops->ndo_open(dev); + if (!ret && dev->needs_cleanup_work) { + if (!dev->cleanup_work) + ret = netif_alloc_cleanup_work(dev); + else + cancel_work_sync(&dev->cleanup_work->work); + } + + if (!ret && ops->ndo_write_rx_mode) + ret = netif_alloc_rx_mode_ctx(dev); + netpoll_poll_enable(dev);
This is the response to the AI review. Honestly impressed by how good the AI review is My bad, It should be flush_work() not cancel_work_sync() in __dev_open() and also in netif_free_cleanup_work(). These are the only places where execution needs to wait for completion of the cleanup work It's ok to just cancel rx_mode work so this issue is only with the cleanup work