Re: [dpdk-dev] Memory leak in rte_pci_scan

[David Marchand <david.marchand@redhat.com>]

On Tue, Jun 8, 2021 at 8:48 PM Owen Hilyard <ohilyard@iol.unh.edu> wrote:
>
> Hello All,
>
> As part of the community lab's work to deploy static analysis tools, we've
> been doing test runs of the various tools. One of the problems we found is
> that rte_pci_scan leaks 214368 Bytes every time it is run. There are also

I suspect the "leak" is on pci device objects that are not released
unless hot(un)plugging.
Cc: Gaetan.



> numerous (180856) instances of the error message "EAL: recvmsg failed, Bad
> file descriptor". Attached is the log from running the fast-tests suite
> with ASAN enabled, and a 10-second timeout. The real timeout seems to be
> the timeout argument multiplied by 10, making this really a 100-second
> timeout. It is attached as a google drive link since the log itself is too
> large for the mailing list.

Those logs come from the mp channel.
At cleanup the rte_mp_handle control thread is sleeping on the mp socket.
The thread serving rte_eal_cleanup shutdowns this socket fd, and the
control thread will spew the log messages you saw.

I'll send a patch.


-- 
David Marchand