28/08/2017 12:30, Ferruh Yigit: > On 8/28/2017 10:55 AM, Gaëtan Rivet wrote: > > Hi Raslan, > > > > Redoing the review with the remarks from Thomas in mind. > > > > On Wed, Aug 23, 2017 at 05:37:04PM +0300, Raslan Darawsheh wrote: > >> Added hotplug in testpmd, to be able to test hotplug function > >> in the PMD's. > >> > >> Signed-off-by: Raslan Darawsheh <rasl...@mellanox.com> > >> --- > >> app/test-pmd/cmdline.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ > >> app/test-pmd/testpmd.c | 18 ++++++++++++++++++ > >> app/test-pmd/testpmd.h | 1 + > >> 3 files changed, 63 insertions(+) > >> > >> diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c > >> index cd8c358..b32a368 100644 > >> --- a/app/test-pmd/cmdline.c > >> +++ b/app/test-pmd/cmdline.c > >> @@ -716,6 +716,9 @@ static void cmd_help_long_parsed(void *parsed_result, > >> "port config (port_id|all) l2-tunnel E-tag" > >> " (enable|disable)\n" > >> " Enable/disable the E-tag support.\n\n" > >> + > >> + " device remove (device)\n" > > > > I think (device) might be hard to understand for a user. > > Maybe (device name)? > > I am suspicious on adding new root level command to testpmd, it is > getting bigger, each command looks OK on its own context, but as a whole > app getting more confusing [1].
We can keep the root level "port" if we make clear that the impact is beyond the port. > Since dealing with device is kind of new, it can be OK to create new > command tree, but there are already hotplug commands per port: > "port attach #PCI|#VDEV_NAME" > "port detach #P" > > perhaps it can be good to keep "attach", "detach" keywords for device to > be consistent? Not sure. Anyway ethdev attach and detach should be deprecated. We are moving to a more correct design where hotplug is done at EAL level. > "device attach #name" > "device detach #name" > > Also a show equivalent can be added to work in device level: > "show device info" > > > [1] > http://dpdk.org/ml/archives/dev/2017-August/072764.html
