Re: [dpdk-dev] [PATCH v2 2/2] librte_ip_frag: add mbuf counter

["Ananyev, Konstantin" <konstantin.ananyev@intel.com>]

Hi Alex,

> 
> Hi Konstantin.
> 
> Could we please make a final decision about counting mbufs, since It still feels to me like an unfinished business?
> Below are my final argumens. if they are not sound to you, just nack ;)

Sorry, but as I said before - not sure that it really worth it.
Still think some sort of wrapper might be a better approach.
Konstantin

> 
> > Hi Konstantin.
> >> Hi Alex,
> >> Sorry for delay in reply.
> 
> 
> >>> >> There might be situations (kind of attack when a lot of
> >>> >> fragmented packets are sent to a dpdk application in order
> >>> >> to flood the fragmentation table) when no additional mbufs
> >>> >> must be added to the fragmentations table since it already
> >>> >> contains to many of them. Currently there is no way to
> >>> >> determine the number of mbufs holded int the fragmentation
> >>> >> table. This patch allows to keep track of the number of mbufs
> >>> >> holded in the fragmentation table.
> 
> >>> > I understand your intention, but still not sure it is worth it.
> >>> > My thought was that you can estimate by upper limit (num_entries * entries_per_bucket) or so.
> >>> No, I can't. The estimation error might be so big that there would be no difference at all.
> 
> >> Not sure why? If you'll use upper limit, then worst thing could happen -
> >> you would start your table cleanup a bit earlier.
> > Since bucket size is 4, an estimation error might be 400%.
> > So, for example, if I want to setup the upper limit (max number mbufs that
> > can be stored in frag table) to 20% of all my available mbufs
> > I have to be ready that 80% of all mbufs might end up in a frag table
> > (every bucket is full). Or if I take into account bucket size, and devide 20%
> > by 4 in order the number mbufs to be exactly 20% in the worse case when every bucket is full,
> > I could end up in the opposite border situation when exactly single mbuf is stored
> > in every bucket, so upper limit of mbufs would be 20 / 4 = 5%. Both ways are not
> > good since either you have to reserve extra mbufs just to correct estimation error
> > or you upper limit would to small and you will be dropping good fragments.
> 
> 
> >>> > Probably another way to account number of mbufs without changes in the lib -
> >>> > apply something like that(assuming that your fragmets are not multisegs):
> 
> >>> > uint32_t mbuf_in_frag_table = 0;
> >>> > ....
> 
> >>> n= dr->>cnt;
> >>> > mb = rte_ipv4_frag_reassemble_packet(...);
> >>> > if (mb != NULL)
> >>> >    mbuf_in_frag_table += mb->nb_segs;
> >>> > mbuf_in_frag_table += dr->cnt - n + 1;
> 
> >> Sorry, my bad, I think it should be
> >> mbuf_in_frag_table  -= dr->cnt - n + 1;
> 
> 
> >>> > In theory that could be applied even if fragments might be multisegs, but for that,
> >>> > we'll need to change rte_ip_frag_free_death_row() to return total number of freed segments.
> 
> >>> That should be a little bit more complicated wrapper code:
> 
> >>> uint32_t mbuf_in_frag_table = 0;
> >>> ....
> 
> >>> n= dr->cnt;
> >>> reassembled_mbuf = rte_ipv4_frag_reassemble_packet(..., fragmented_mbuf, ...);
> >>> if (reassembled_mbuf == NULL)
> >>>     mbuf_in_frag_table += fragmented_mbuf->nb_segs;
> 
> >> We don't know for sure here.
> >> fragmented_mbuf could be in death row by now.
> > Yes. That's exactly why you have to keep track of
> > mbufs here and later after rte_ip_frag_free_death_row().
> 
> > User have to think about frag table and death row as a single entity,
> > kind of a black box, since it's impossible to say where
> > (in the frag table or in the death row) your mbuf will be
> > after you call rte_ipv4_frag_reassemble_packet(). So, a caller/user should
> > keep track of mbuf on every border/interface of that black box.
> > One interface is rte_ipv4_frag_reassemble_packet and the other is
> > rte_ip_frag_free_death_row.
> 
> > So, that's why it's easier to keep track of mbufs inside the library.
> >
> 
> >>> else
> >>>     mbuf_in_frag_table -= reassembled_mbuf->nb_segs;
> >>> mbuf_in_frag_table += dr->cnt - n;
> 
> 
> >>> Also, in that case every rte_ip_frag_free_death_row() needs a wrapper code too.
> 
> >>> n= dr->cnt;
> >>> rte_ip_frag_free_death_row(..)
> >>> mbuf_in_frag_table += dr->cnt - n;
> 
> >> I don't think it is necessary.
> >> After packet is put in the death-row it is no longer in the table.
> > It's critical, since from a user point of view death row and frag table
> > is a black box due rte_ipv4_frag_reassemble_packet() doesn't indicate a caller
> > where his packet has been stored (in the frag table or death row).
> 
> >> Konstantin
> 
> 
> 
> >>> I think my approach is simplier.
> 
> >>> > Konstantin
> 
> 
> >>> >> Signed-off-by: Alex Kiselev <alex@therouter.net>
> >>> >> ---
> >>> >>  lib/librte_ip_frag/ip_frag_common.h        | 16 +++++++++-------
> >>> >>  lib/librte_ip_frag/ip_frag_internal.c      | 16 +++++++++-------
> >>> >>  lib/librte_ip_frag/rte_ip_frag.h           | 18 +++++++++++++++++-
> >>> >>  lib/librte_ip_frag/rte_ip_frag_common.c    |  1 +
> >>> >>  lib/librte_ip_frag/rte_ip_frag_version.map |  1 +
> >>> >>  lib/librte_ip_frag/rte_ipv4_reassembly.c   |  2 +-
> >>> >>  lib/librte_ip_frag/rte_ipv6_reassembly.c   |  2 +-
> >>> >>  7 files changed, 39 insertions(+), 17 deletions(-)
> 
> >>> >> diff --git a/lib/librte_ip_frag/ip_frag_common.h b/lib/librte_ip_frag/ip_frag_common.h
> >>> >> index 0fdcc7d0f..9fe5c0559 100644
> >>> >> --- a/lib/librte_ip_frag/ip_frag_common.h
> >>> >> +++ b/lib/librte_ip_frag/ip_frag_common.h
> >>> >> @@ -32,15 +32,15 @@
> >>> >>  #endif /* IP_FRAG_TBL_STAT */
> 
> >>> >>  /* internal functions declarations */
> >>> >> -struct rte_mbuf * ip_frag_process(struct ip_frag_pkt *fp,
> >>> >> -             struct rte_ip_frag_death_row *dr, struct rte_mbuf *mb,
> >>> >> -             uint16_t ofs, uint16_t len, uint16_t more_frags);
> >>> >> +struct rte_mbuf *ip_frag_process(struct rte_ip_frag_tbl *tbl,
> >>> >> +     struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
> >>> >> +     struct rte_mbuf *mb, uint16_t ofs, uint16_t len, uint16_t more_frags);
> 
> >>> >> -struct ip_frag_pkt * ip_frag_find(struct rte_ip_frag_tbl *tbl,
> >>> >> +struct ip_frag_pkt *ip_frag_find(struct rte_ip_frag_tbl *tbl,
> >>> >>               struct rte_ip_frag_death_row *dr,
> >>> >>               const struct ip_frag_key *key, uint64_t tms);
> 
> >>> >> -struct ip_frag_pkt * ip_frag_lookup(struct rte_ip_frag_tbl *tbl,
> >>> >> +struct ip_frag_pkt *ip_frag_lookup(struct rte_ip_frag_tbl *tbl,
> >>> >>       const struct ip_frag_key *key, uint64_t tms,
> >>> >>       struct ip_frag_pkt **free, struct ip_frag_pkt **stale);
> 
> >>> >> @@ -91,7 +91,8 @@ ip_frag_key_cmp(const struct ip_frag_key * k1, const struct ip_frag_key * k2)
> 
> >>> >>  /* put fragment on death row */
> >>> >>  static inline void
> >>> >> -ip_frag_free(struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr)
> >>> >> +ip_frag_free(struct rte_ip_frag_tbl *tbl, struct ip_frag_pkt *fp,
> >>> >> +     struct rte_ip_frag_death_row *dr)
> >>> >>  {
> >>> >>       uint32_t i, k;
> 
> >>> >> @@ -100,6 +101,7 @@ ip_frag_free(struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr)
> >>> >>               if (fp->frags[i].mb != NULL) {
> >>> >>                       dr->row[k++] = fp->frags[i].mb;
> >>> >>                       fp->frags[i].mb = NULL;
> >>> >> +                     tbl->nb_mbufs--;
> >>> >>               }
> >>> >>       }
> 
> >>> >> @@ -160,7 +162,7 @@ static inline void
> >>> >>  ip_frag_tbl_del(struct rte_ip_frag_tbl *tbl, struct rte_ip_frag_death_row *dr,
> >>> >>       struct ip_frag_pkt *fp)
> >>> >>  {
> >>> >> -     ip_frag_free(fp, dr);
> >>> >> +     ip_frag_free(tbl, fp, dr);
> >>> >>       ip_frag_key_invalidate(&fp->key);
> >>> >>       TAILQ_REMOVE(&tbl->lru, fp, lru);
> >>> >>       tbl->use_entries--;
> >>> >> diff --git a/lib/librte_ip_frag/ip_frag_internal.c b/lib/librte_ip_frag/ip_frag_internal.c
> >>> >> index 97470a872..4c47d3fb4 100644
> >>> >> --- a/lib/librte_ip_frag/ip_frag_internal.c
> >>> >> +++ b/lib/librte_ip_frag/ip_frag_internal.c
> >>> >> @@ -29,14 +29,13 @@ static inline void
> >>> >>  ip_frag_tbl_reuse(struct rte_ip_frag_tbl *tbl, struct rte_ip_frag_death_row *dr,
> >>> >>       struct ip_frag_pkt *fp, uint64_t tms)
> >>> >>  {
> >>> >> -     ip_frag_free(fp, dr);
> >>> >> +     ip_frag_free(tbl, fp, dr);
> >>> >>       ip_frag_reset(fp, tms);
> >>> >>       TAILQ_REMOVE(&tbl->lru, fp, lru);
> >>> >>       TAILQ_INSERT_TAIL(&tbl->lru, fp, lru);
> >>> >>       IP_FRAG_TBL_STAT_UPDATE(&tbl->stat, reuse_num, 1);
> >>> >>  }
> 
> >>> >> -
> >>> >>  static inline void
> >>> >>  ipv4_frag_hash(const struct ip_frag_key *key, uint32_t *v1, uint32_t *v2)
> >>> >>  {
> >>> >> @@ -88,8 +87,9 @@ ipv6_frag_hash(const struct ip_frag_key *key, uint32_t *v1, uint32_t *v2)
> >>> >>  }
> 
> >>> >>  struct rte_mbuf *
> >>> >> -ip_frag_process(struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
> >>> >> -     struct rte_mbuf *mb, uint16_t ofs, uint16_t len, uint16_t more_frags)
> >>> >> +ip_frag_process(struct rte_ip_frag_tbl *tbl, struct ip_frag_pkt *fp,
> >>> >> +     struct rte_ip_frag_death_row *dr, struct rte_mbuf *mb, uint16_t ofs,
> >>> >> +     uint16_t len, uint16_t more_frags)
> >>> >>  {
> >>> >>       uint32_t idx;
> 
> >>> >> @@ -147,7 +147,7 @@ ip_frag_process(struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
> >>> >>                               fp->frags[IP_LAST_FRAG_IDX].len);
> 
> >>> >>               /* free all fragments, invalidate the entry. */
> >>> >> -             ip_frag_free(fp, dr);
> >>> >> +             ip_frag_free(tbl, fp, dr);
> >>> >>               ip_frag_key_invalidate(&fp->key);
> >>> >>               IP_FRAG_MBUF2DR(dr, mb);
> 
> >>> >> @@ -157,6 +157,7 @@ ip_frag_process(struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
> >>> >>       fp->frags[idx].ofs = ofs;
> >>> >>       fp->frags[idx].len = len;
> >>> >>       fp->frags[idx].mb = mb;
> >>> >> +     tbl->nb_mbufs++;
> 
> >>> >>       mb = NULL;
> 
> >>> >> @@ -205,8 +206,9 @@ ip_frag_process(struct ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
> >>> >>                               fp->frags[IP_LAST_FRAG_IDX].len);
> 
> >>> >>               /* free associated resources. */
> >>> >> -             ip_frag_free(fp, dr);
> >>> >> -     }
> >>> >> +             ip_frag_free(tbl, fp, dr);
> >>> >> +     } else
> >>> >> +             tbl->nb_mbufs -= fp->last_idx;
> 
> >>> >>       /* we are done with that entry, invalidate it. */
> >>> >>       ip_frag_key_invalidate(&fp->key);
> >>> >> diff --git a/lib/librte_ip_frag/rte_ip_frag.h b/lib/librte_ip_frag/rte_ip_frag.h
> >>> >> index 7f425f610..623934d87 100644
> >>> >> --- a/lib/librte_ip_frag/rte_ip_frag.h
> >>> >> +++ b/lib/librte_ip_frag/rte_ip_frag.h
> >>> >> @@ -96,6 +96,7 @@ struct rte_ip_frag_tbl {
> >>> >>       uint32_t             bucket_entries;  /**< hash associativity. */
> >>> >>       uint32_t             nb_entries;      /**< total size of the table. */
> >>> >>       uint32_t             nb_buckets;      /**< num of associativity lines. */
> >>> >> +     uint32_t             nb_mbufs;        /**< num of mbufs holded in the tbl. */
> >>> >>       struct ip_frag_pkt *last;         /**< last used entry. */
> >>> >>       struct ip_pkt_list lru;           /**< LRU list for table entries. */
> >>> >>       struct ip_frag_tbl_stat stat;     /**< statistics counters. */
> >>> >> @@ -329,8 +330,23 @@ void
> >>> >>  rte_ip_frag_table_statistics_dump(FILE * f, const struct rte_ip_frag_tbl *tbl);
> 
> >>> >>  /**
> >>> >> - * Delete expired fragments
> >>> >> + * Number of mbufs holded in the fragmentation table.
> >>> >> + *
> >>> >> + * @param tbl
> >>> >> + *   Fragmentation table
> >>> >>   *
> >>> >> + * @return
> >>> >> + *   Number of mbufs holded in the fragmentation table.
> >>> >> + */
> >>> >> +static inline uint32_t __rte_experimental
> >>> >> +rte_frag_table_mbuf_count(const struct rte_ip_frag_tbl *tbl)
> >>> >> +{
> >>> >> +     return tbl->nb_mbufs;
> >>> >> +}
> >>> >> +
> >>> >> +/**
> >>> >> + * Delete expired fragments
> >>> >> + *
> >>> >>   * @param tbl
> >>> >>   *   Table to delete expired fragments from
> >>> >>   * @param dr
> >>> >> diff --git a/lib/librte_ip_frag/rte_ip_frag_common.c b/lib/librte_ip_frag/rte_ip_frag_common.c
> >>> >> index a23f6f24f..46c2df84a 100644
> >>> >> --- a/lib/librte_ip_frag/rte_ip_frag_common.c
> >>> >> +++ b/lib/librte_ip_frag/rte_ip_frag_common.c
> >>> >> @@ -75,6 +75,7 @@ rte_ip_frag_table_create(uint32_t bucket_num, uint32_t bucket_entries,
> >>> >>       tbl->nb_buckets = bucket_num;
> >>> >>       tbl->bucket_entries = bucket_entries;
> >>> >>       tbl->entry_mask = (tbl->nb_entries - 1) & ~(tbl->bucket_entries  - 1);
> >>> >> +     tbl->nb_mbufs = 0;
> 
> >>> >>       TAILQ_INIT(&(tbl->lru));
> >>> >>       return tbl;
> >>> >> diff --git a/lib/librte_ip_frag/rte_ip_frag_version.map b/lib/librte_ip_frag/rte_ip_frag_version.map
> >>> >> index d40d5515f..f4700f460 100644
> >>> >> --- a/lib/librte_ip_frag/rte_ip_frag_version.map
> >>> >> +++ b/lib/librte_ip_frag/rte_ip_frag_version.map
> >>> >> @@ -23,4 +23,5 @@ EXPERIMENTAL {
> >>> >>       global:
> 
> >>> >>       rte_frag_table_del_expired_entries;
> >>> >> +     rte_frag_table_mbuf_count;
> >>> >>  };
> >>> >> diff --git a/lib/librte_ip_frag/rte_ipv4_reassembly.c b/lib/librte_ip_frag/rte_ipv4_reassembly.c
> >>> >> index 4956b99ea..fbdfd860a 100644
> >>> >> --- a/lib/librte_ip_frag/rte_ipv4_reassembly.c
> >>> >> +++ b/lib/librte_ip_frag/rte_ipv4_reassembly.c
> >>> >> @@ -146,7 +146,7 @@ rte_ipv4_frag_reassemble_packet(struct rte_ip_frag_tbl *tbl,
> 
> 
> >>> >>       /* process the fragmented packet. */
> >>> >> -     mb = ip_frag_process(fp, dr, mb, ip_ofs, ip_len, ip_flag);
> >>> >> +     mb = ip_frag_process(tbl, fp, dr, mb, ip_ofs, ip_len, ip_flag);
> >>> >>       ip_frag_inuse(tbl, fp);
> 
> >>> >>       IP_FRAG_LOG(DEBUG, "%s:%d:\n"
> >>> >> diff --git a/lib/librte_ip_frag/rte_ipv6_reassembly.c b/lib/librte_ip_frag/rte_ipv6_reassembly.c
> >>> >> index db249fe60..dda5a57b7 100644
> >>> >> --- a/lib/librte_ip_frag/rte_ipv6_reassembly.c
> >>> >> +++ b/lib/librte_ip_frag/rte_ipv6_reassembly.c
> >>> >> @@ -186,7 +186,7 @@ rte_ipv6_frag_reassemble_packet(struct rte_ip_frag_tbl *tbl,
> 
> 
> >>> >>       /* process the fragmented packet. */
> >>> >> -     mb = ip_frag_process(fp, dr, mb, ip_ofs, ip_len,
> >>> >> +     mb = ip_frag_process(tbl, fp, dr, mb, ip_ofs, ip_len,
> >>> >>                       MORE_FRAGS(frag_hdr->frag_data));
> >>> >>       ip_frag_inuse(tbl, fp);
> 
> >>> >> --
> >>> >> 2.16.1.windows.1
> 
> 
> 
> 
> >>> --
> >>> Alex
> 
> 
> 
> 
> 
> 
> 
> --
> Alex