[dpdk-stable] patch 'app/testpmd: fix exit for virtio-user' has been queued to stable release 18.02.2

[Yuanhan Liu]

On Wed, May 23, 2018 at 01:09:54PM +0100, luca.boccassi at gmail.com wrote:
> Hi,
> 
> FYI, your patch has been queued to stable release 18.02.2
> 
> Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
> It will be pushed if I get no objections before 05/25/18. So please
> shout if anyone has objections.

I'm late, but I'd suggest you to not include this patch, for

- it's a workaround

- it tries to "fix" an issue introduced in v18.05, for adding virtio-user
  server mode.

	--yliu
> 
> Thanks.
> 
> Luca Boccassi
> 
> ---
> >From 8e96cf1276094a75ad0d3fd32912c672489e9811 Mon Sep 17 00:00:00 2001
> From: Zhiyong Yang <zhiyong.yang at intel.com>
> Date: Fri, 18 May 2018 17:59:37 +0800
> Subject: [PATCH] app/testpmd: fix exit for virtio-user
> 
> [ upstream commit 124909d7e1bf299c959b58c2db7c5fbd7abbb7c0 ]
> 
> For vdev, just calling rte_eth_dev_close() isn't enough to free all
> the resources allocated during device probe, e.g. for virtio-user,
> virtio_user_pmd_remove(), i.e. the remove() method of a vdev driver,
> needs to be called to unlink the socket file created during device
> probe. So this patch calls the rte_eth_dev_detach() for vdev when
> quitting testpmd.
> 
> vdevs detach on testpmd exit implemented as workaround to fix
> a virtio-user issue. The issue was virtio-user cleanup is not
> called and existing socket file not cleaned up which will fail
> next run.
> 
> Added a comment that this workaround should be converted to a proper
> cleanup, not something specific to virtio-user, and not something
> specific to vdev and testpmd.
> 
> Fixes: af75078fece3 ("first public release")
> Fixes: bd8f50a45d0f ("net/virtio-user: support server mode")
> 
> Signed-off-by: Zhiyong Yang <zhiyong.yang at intel.com>
> Tested-by: Lei Yao <lei.a.yao at intel.com>
> Acked-by: Bernard Iremonger <bernard.iremonger at intel.com>
> ---
>  app/test-pmd/testpmd.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
> 
> diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
> index 05a68a068..98e0a9dc6 100644
> --- a/app/test-pmd/testpmd.c
> +++ b/app/test-pmd/testpmd.c
> @@ -1951,6 +1951,7 @@ detach_port(portid_t port_id)
>  void
>  pmd_test_exit(void)
>  {
> +	struct rte_device *device;
>  	portid_t pt_id;
>  
>  	if (test_done == 0)
> @@ -1963,6 +1964,18 @@ pmd_test_exit(void)
>  			fflush(stdout);
>  			stop_port(pt_id);
>  			close_port(pt_id);
> +
> +			/*
> +			 * This is a workaround to fix a virtio-user issue that
> +			 * requires to call clean-up routine to remove existing
> +			 * socket.
> +			 * This workaround valid only for testpmd, needs a fix
> +			 * valid for all applications.
> +			 * TODO: Implement proper resource cleanup
> +			 */
> +			device = rte_eth_devices[pt_id].device;
> +			if (device && !strcmp(device->driver->name, "net_virtio_user"))
> +				detach_port(pt_id);
>  		}
>  	}
>  	printf("\nBye...\n");
> -- 
> 2.14.2