> -----Original Message----- > From: Ferruh Yigit [mailto:[email protected]] > Sent: Wednesday, April 15, 2020 11:15 PM > To: wangyunjian <[email protected]>; [email protected] > Cc: [email protected]; Lilijun (Jerry) <[email protected]>; > xudingke > <[email protected]>; [email protected] > Subject: Re: [dpdk-dev] [PATCH v4 5/5] net/tap: fix leak of fds on failure > > On 4/11/2020 12:24 PM, wangyunjian wrote: > > From: Yunjian Wang <[email protected]> > > > > When eth_dev_tap_create() is failed, nlsk_fd and ka_fd won't be closed > > thus leading fds leak. Zero is a valid fd. Ultimately leads to a valid > > fd was closed by mistake. > > > > Fixes: bf7b7f437b49 ("net/tap: create netdevice during probing") > > Fixes: cb7e68da630a ("net/tap: fix cleanup on allocation failure") > > CC: [email protected] > > > > Signed-off-by: Yunjian Wang <[email protected]> > > --- > > drivers/net/tap/rte_eth_tap.c | 9 +++++++-- > > 1 file changed, 7 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/net/tap/rte_eth_tap.c > > b/drivers/net/tap/rte_eth_tap.c index 829a9e9b4..9bea5d9d0 100644 > > --- a/drivers/net/tap/rte_eth_tap.c > > +++ b/drivers/net/tap/rte_eth_tap.c > > @@ -1820,6 +1820,8 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, > const char *tap_name, > > pmd->dev = dev; > > strlcpy(pmd->name, tap_name, sizeof(pmd->name)); > > pmd->type = type; > > + pmd->ka_fd = -1; > > + pmd->nlsk_fd = -1; > > > > pmd->ioctl_sock = socket(AF_INET, SOCK_DGRAM, 0); > > if (pmd->ioctl_sock == -1) { > > @@ -1850,7 +1852,6 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, > const char *tap_name, > > dev->intr_handle = &pmd->intr_handle; > > > > /* Presetup the fds to -1 as being not valid */ > > - pmd->ka_fd = -1; > > for (i = 0; i < RTE_PMD_TAP_MAX_QUEUES; i++) { > > process_private->rxq_fds[i] = -1; > > process_private->txq_fds[i] = -1; > > @@ -1990,7 +1991,11 @@ eth_dev_tap_create(struct rte_vdev_device > *vdev, const char *tap_name, > > tap_flow_implicit_flush(pmd, NULL); > > > > error_exit: > > - if (pmd->ioctl_sock > 0) > > + if (pmd->nlsk_fd == -1) > > + close(pmd->nlsk_fd); > > Shouldn't this be '!=', "pmd->nlsk_fd != -1"?
I'm sorry for my mistakes. I will fix it in next version. Thanks, Yunjian
