[dpdk-dev] [PATCH] doc: flow rule removal on port stop

Gregory Etelson getelson at nvidia.com
Wed Nov 18 10:06:23 CET 2020


> >> On 11/17/20 10:18 PM, Gregory Etelson wrote:
> >>> There is a discrepancy between RTE ETHDEV API and flow rules guide
> >>> regarding flow rules maintenance after port stop.  RTE ETHDEV API in
> >>> librte_ethdev.h declares that flow rules will not be stored in PMD
> >>> after port stop:
> >>>   >>>>> Quite start
> >>>   Please note that some configuration is not stored between calls to
> >>>   rte_eth_dev_stop()/rte_eth_dev_start(). The following configuration
> >>>   will be retained:
> >>>
> >>>   - MTU
> >>>   - flow control settings
> >>>   - receive mode configuration (promiscuous mode, all-multicast mode,
> >>>     hardware checksum mode, RSS/VMDQ settings etc.)
> >>>   - VLAN filtering configuration
> >>>   - default MAC address
> >>>   - MAC addresses supplied to MAC address array
> >>>   - flow director filtering mode (but not filtering rules)
> >>>   - NIC queue statistics mappings
> >>>   <<<< Quote end
> >>>
> >>> PMD cannot always correctly restore flow rules after port stop /
> >>> port start because application may alter port configuration after
> >>> port stop without PMD knowledge about undergoing changes.  Consider
> >>> the following scenario:
> >>> application configures 2 queues 0 and 1 and creates a flow rule with
> >>> 'queue index 1' action. After that application stops the port and
> >>> removes queue 1.
> >>> Although PMD can implement flow rule shadow copy to be used for
> >>> restore after port start, attempt to restore flow rule from shadow
> >>> will fail in example above and PMD could not notify application
> >>> about that failure.  As the result, flow rules map in HW will differ
> >>> from what application expects.  In addition, flow rules shadow copy
> >>> used for port start restore consumes considerable amount of system
> >>> memory, especially in systems with millions of flow rules.
> >>>
> >>> Signed-off-by: Gregory Etelson <getelson at nvidia.com>
> >>> Acked-by: Ori Kam <orika at nvidia.com>
> >>> ---
> >>>   doc/guides/prog_guide/rte_flow.rst | 5 ++---
> >>>   1 file changed, 2 insertions(+), 3 deletions(-)
> >>>
> >>> diff --git a/doc/guides/prog_guide/rte_flow.rst
> >>> b/doc/guides/prog_guide/rte_flow.rst
> >>> index 944e8242d6..dfe5a40f8e 100644
> >>> --- a/doc/guides/prog_guide/rte_flow.rst
> >>> +++ b/doc/guides/prog_guide/rte_flow.rst
> >>> @@ -3055,10 +3055,9 @@ Caveats
> >>>     temporarily replacing the burst function pointers), an
> >>> appropriate
> >> error
> >>>     code must be returned (``EBUSY``).
> >>>
> >>> -- PMDs, not applications, are responsible for maintaining flow
> >>> rules
> >>> +- Applications, not PMDs, are responsible for maintaining flow
> >>> +rules
> >>>     configuration when stopping and restarting a port or performing
> >>> other
> >>> -  actions which may affect them. They can only be destroyed
> >>> explicitly by
> >>> -  applications.
> >>> +  actions which may affect them.
> >>>
> >>>   For devices exposing multiple ports sharing global settings
> >>> affected
> >> by flow
> >>>   rules:
> >>>
> >>
> >> Re-reading it, it still looks vague. What happens on:
> >>   - port stop without removal of flow rule before
> >>   - port close without removal of flow rules before
> >>   - port reset (which could be stop/start, e.g. to recover from error
> >> condition)
> >
> > PMD should remove all flows related to hardware resource that was
> invalidated.
> 
> Stop? Close? I agree and documentation should say so in a bit clear way.

I'll post updated document patch.


More information about the dev mailing list