On 10/15/2021 4:44 AM, Huisong Li wrote:
In secondary process, rte_eth_dev_close() doesn't clear eth_dev->data. If calling rte_dev_remove() after rte_eth_dev_close(), in rte_eth_dev_pci_generic_remove() function, the released eth device still can be found by its name in shared memory. As a result, the eth device will be released repeatedly. The state of the eth device is modified to RTE_ETH_DEV_UNUSED after rte_eth_dev_close(). So this state can be used to avoid this problem.
Hi Huisong, Can you please add Fixes line, and stable tag if the change is requested for backport?
Signed-off-by: Huisong Li <lihuis...@huawei.com> --- v2 -> v3: * fix the commit log description and the comment inside the code. v1 -> v2: * fix the commit log description. RFC -> v1: * fix commit log and add a judgment for secondary process. --- lib/ethdev/ethdev_pci.h | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/lib/ethdev/ethdev_pci.h b/lib/ethdev/ethdev_pci.h index 8edca82ce8..fcabae02fa 100644 --- a/lib/ethdev/ethdev_pci.h +++ b/lib/ethdev/ethdev_pci.h @@ -151,6 +151,16 @@ rte_eth_dev_pci_generic_remove(struct rte_pci_device *pci_dev, if (!eth_dev) return 0;+ /*+ * In secondary process, a released eth device can be found by its name + * in shared memory. + * If the state of the eth device is RTE_ETH_DEV_UNUSED, it means the + * eth device has been released. + */ + if (rte_eal_process_type() == RTE_PROC_SECONDARY && + eth_dev->state == RTE_ETH_DEV_UNUSED) + return 0; + if (dev_uninit) { ret = dev_uninit(eth_dev); if (ret)
