[dpdk-dev] [PATCH v5 1/3] lib/lpm: integrate RCU QSBR

[Ruifeng Wang]

Hi David,

Sorry, I missed tracking of this thread.

> -----Original Message-----
> From: David Marchand <david.marchand at redhat.com>
> Sent: Monday, June 29, 2020 7:56 PM
> To: Ruifeng Wang <Ruifeng.Wang at arm.com>; Vladimir Medvedkin
> <vladimir.medvedkin at intel.com>; Bruce Richardson
> <bruce.richardson at intel.com>
> Cc: John McNamara <john.mcnamara at intel.com>; Marko Kovacevic
> <marko.kovacevic at intel.com>; Ray Kinsella <mdr at ashroe.eu>; Neil Horman
> <nhorman at tuxdriver.com>; dev <dev at dpdk.org>; Ananyev, Konstantin
> <konstantin.ananyev at intel.com>; Honnappa Nagarahalli
> <Honnappa.Nagarahalli at arm.com>; nd <nd at arm.com>
> Subject: Re: [dpdk-dev] [PATCH v5 1/3] lib/lpm: integrate RCU QSBR
> 
> On Mon, Jun 29, 2020 at 10:03 AM Ruifeng Wang <ruifeng.wang at arm.com>
> wrote:
> >
> > Currently, the tbl8 group is freed even though the readers might be
> > using the tbl8 group entries. The freed tbl8 group can be reallocated
> > quickly. This results in incorrect lookup results.
> >
> > RCU QSBR process is integrated for safe tbl8 group reclaim.
> > Refer to RCU documentation to understand various aspects of
> > integrating RCU library into other libraries.
> >
> > Signed-off-by: Ruifeng Wang <ruifeng.wang at arm.com>
> > Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli at arm.com>
> > ---
> >  doc/guides/prog_guide/lpm_lib.rst  |  32 +++++++
> >  lib/librte_lpm/Makefile            |   2 +-
> >  lib/librte_lpm/meson.build         |   1 +
> >  lib/librte_lpm/rte_lpm.c           | 129 ++++++++++++++++++++++++++---
> >  lib/librte_lpm/rte_lpm.h           |  59 +++++++++++++
> >  lib/librte_lpm/rte_lpm_version.map |   6 ++
> >  6 files changed, 216 insertions(+), 13 deletions(-)
> >
> > diff --git a/doc/guides/prog_guide/lpm_lib.rst
> > b/doc/guides/prog_guide/lpm_lib.rst
> > index 1609a57d0..7cc99044a 100644
> > --- a/doc/guides/prog_guide/lpm_lib.rst
> > +++ b/doc/guides/prog_guide/lpm_lib.rst
> > @@ -145,6 +145,38 @@ depending on whether we need to move to the
> next table or not.
> >  Prefix expansion is one of the keys of this algorithm,  since it
> > improves the speed dramatically by adding redundancy.
> >
> > +Deletion
> > +~~~~~~~~
> > +
> > +When deleting a rule, a replacement rule is searched for. Replacement
> > +rule is an existing rule that has the longest prefix match with the rule to be
> deleted, but has smaller depth.
> > +
> > +If a replacement rule is found, target tbl24 and tbl8 entries are
> > +updated to have the same depth and next hop value with the
> replacement rule.
> > +
> > +If no replacement rule can be found, target tbl24 and tbl8 entries will be
> cleared.
> > +
> > +Prefix expansion is performed if the rule's depth is not exactly 24 bits or
> 32 bits.
> > +
> > +After deleting a rule, a group of tbl8s that belongs to the same tbl24 entry
> are freed in following cases:
> > +
> > +*   All tbl8s in the group are empty .
> > +
> > +*   All tbl8s in the group have the same values and with depth no greater
> than 24.
> > +
> > +Free of tbl8s have different behaviors:
> > +
> > +*   If RCU is not used, tbl8s are cleared and reclaimed immediately.
> > +
> > +*   If RCU is used, tbl8s are reclaimed when readers are in quiescent state.
> > +
> > +When the LPM is not using RCU, tbl8 group can be freed immediately
> > +even though the readers might be using the tbl8 group entries. This might
> result in incorrect lookup results.
> > +
> > +RCU QSBR process is integrated for safe tbl8 group reclaimation.
> > +Application has certain responsibilities while using this feature.
> > +Please refer to resource reclaimation framework of :ref:`RCU library
> <RCU_Library>` for more details.
> > +
> 
> Would the lpm6 library benefit from the same?
> Asking as I do not see much code shared between lpm and lpm6.
> 
Didn't look into lpm6. It may need separate integration with RCU since no shared code between lpm and lpm6 as you mentioned.

> [...]
> 
> > diff --git a/lib/librte_lpm/rte_lpm.c b/lib/librte_lpm/rte_lpm.c index
> > 38ab512a4..41e9c49b8 100644
> > --- a/lib/librte_lpm/rte_lpm.c
> > +++ b/lib/librte_lpm/rte_lpm.c
> > @@ -1,5 +1,6 @@
> >  /* SPDX-License-Identifier: BSD-3-Clause
> >   * Copyright(c) 2010-2014 Intel Corporation
> > + * Copyright(c) 2020 Arm Limited
> >   */
> >
> >  #include <string.h>
> > @@ -245,13 +246,84 @@ rte_lpm_free(struct rte_lpm *lpm)
> >                 TAILQ_REMOVE(lpm_list, te, next);
> >
> >         rte_mcfg_tailq_write_unlock();
> > -
> > +#ifdef ALLOW_EXPERIMENTAL_API
> > +       if (lpm->dq)
> > +               rte_rcu_qsbr_dq_delete(lpm->dq); #endif
> 
> All DPDK code under lib/ is compiled with the ALLOW_EXPERIMENTAL_API
> flag set.
> There is no need to protect against this flag in rte_lpm.c.
> 
OK, I see. So DPDK libraries will always be compiled with the ALLOW_EXPERIMENTAL_API. It is application's 
choice to use experimental APIs. 
Will update in next version to remove the ALLOW_EXPERIMENTAL_API flag from rte_lpm.c and only keep the one in rte_lpm.h.

> [...]
> 
> > diff --git a/lib/librte_lpm/rte_lpm.h b/lib/librte_lpm/rte_lpm.h index
> > b9d49ac87..7889f21b3 100644
> > --- a/lib/librte_lpm/rte_lpm.h
> > +++ b/lib/librte_lpm/rte_lpm.h
> 
> > @@ -130,6 +143,28 @@ struct rte_lpm {
> >                         __rte_cache_aligned; /**< LPM tbl24 table. */
> >         struct rte_lpm_tbl_entry *tbl8; /**< LPM tbl8 table. */
> >         struct rte_lpm_rule *rules_tbl; /**< LPM rules. */
> > +#ifdef ALLOW_EXPERIMENTAL_API
> > +       /* RCU config. */
> > +       struct rte_rcu_qsbr *v;         /* RCU QSBR variable. */
> > +       enum rte_lpm_qsbr_mode rcu_mode;/* Blocking, defer queue. */
> > +       struct rte_rcu_qsbr_dq *dq;     /* RCU QSBR defer queue. */
> > +#endif
> > +};
> 
> This is more a comment/question for the lpm maintainers.
> 
> Afaics, the rte_lpm structure is exported/public because of lookup which is
> inlined.
> But most of the structure can be hidden and stored in a private structure that
> would embed the exposed rte_lpm.
> The slowpath functions would only have to translate from publicly exposed
> to internal representation (via container_of).
> 
> This patch could do this and be the first step to hide the unneeded exposure
> of other fields (later/in 20.11 ?).
> 
To hide most of the structure is reasonable. 
Since it will break ABI, I can do that in 20.11.

> Thoughts?
> 
> 
> --
> David Marchand