2.6.29 patch has introduced flexible route cache rebuilding. Unfortunately the patch has at least one critical flaw, and another problem. rt_intern_hash calculates rthi pointer, which is later used for new entry insertion. The same loop calculates cand pointer which is used to clean the list. If the pointers are the same, rtable leak occurs, as first the cand is removed then the new entry is appended to it. This leak leads to unregister_netdevice problem (usage count > 0). Another problem of the patch is that it tries to insert the entries in certain order, to facilitate counting of entries distinct by all but QoS parameters. Unfortunately, referencing an existing rtable entry moves it to list beginning, to speed up further lookups, so the carefully built order is destroyed. For the first problem the simplest patch it to set rthi=0 when rthi==cand, but it will also destroy the ordering.
On Monday 25 May 2009, Eric Dumazet wrote: > Rafael J. Wysocki a écrit : > > This message has been generated automatically as a part of a report > > of recent regressions. > > > > The following bug entry is on the current list of known regressions > > from 2.6.29. Please verify if it still should be listed and let me know > > (either way). > > > > > > Bug-Entry : http://bugzilla.kernel.org/show_bug.cgi?id=13339 > > Subject : rtable leak in ipv4/route.c > > Submitter : Alexander V. Lukyanov <lav@yar.ru> > > Date : 2009-05-18 14:10 (7 days old) > > > > Bug was present in 2.6.29, so its a regression from 2.6.28 > > It is solved and available in David tree (net-2.6), and scheduled for stable > submission > > commit 1ddbcb005c395518c2cd0df504cff3d4b5c85853 > net: fix rtable leak in net/ipv4/route.c > > Alexander V. Lukyanov found a regression in 2.6.29 and made a complete > analysis found in http://bugzilla.kernel.org/show_bug.cgi?id=13339 > Quoted here because its a perfect one : > > begin_of_quotation > 2.6.29 patch has introduced flexible route cache rebuilding. Unfortunately > the > patch has at least one critical flaw, and another problem. > > rt_intern_hash calculates rthi pointer, which is later used for new entry > insertion. The same loop calculates cand pointer which is used to clean the > list. If the pointers are the same, rtable leak occurs, as first the cand is > removed then the new entry is appended to it. > > This leak leads to unregister_netdevice problem (usage count > 0). > > Another problem of the patch is that it tries to insert the entries in > certain > order, to facilitate counting of entries distinct by all but QoS parameters. > Unfortunately, referencing an existing rtable entry moves it to list > beginning, > to speed up further lookups, so the carefully built order is destroyed. > > For the first problem the simplest patch it to set rthi=0 when rthi==cand, > but > it will also destroy the ordering. > end_of_quotation > > Problematic commit is 1080d709fb9d8cd4392f93476ee46a9d6ea05a5b > (net: implement emergency route cache rebulds when gc_elasticity is exceeded) > > Trying to keep dst_entries ordered is too complex and breaks the fact that > order should depend on the frequency of use for garbage collection. > > A possible fix is to make rt_intern_hash() simpler, and only makes > rt_check_expire() a litle bit smarter, being able to cope with an arbitrary > entries order. The added loop is running on cache hot data, while cpu > is prefetching next object, so should be unnoticied. > > Reported-and-analyzed-by: Alexander V. Lukyanov <lav@yar.ru> > Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> > Acked-by: Neil Horman <nhorman@tuxdriver.com> > Signed-off-by: David S. Miller <davem@davemloft.net>