123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999100010011002100310041005100610071008100910101011101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047104810491050105110521053105410551056105710581059106010611062106310641065106610671068106910701071107210731074107510761077107810791080108110821083108410851086108710881089109010911092109310941095109610971098109911001101110211031104110511061107110811091110111111121113111411151116111711181119112011211122112311241125112611271128112911301131113211331134113511361137113811391140114111421143114411451146114711481149115011511152115311541155115611571158115911601161116211631164116511661167116811691170117111721173117411751176117711781179118011811182118311841185118611871188118911901191119211931194119511961197119811991200120112021203120412051206120712081209121012111212121312141215121612171218121912201221122212231224122512261227122812291230123112321233123412351236123712381239124012411242124312441245124612471248124912501251125212531254125512561257125812591260126112621263126412651266126712681269127012711272127312741275127612771278127912801281128212831284128512861287128812891290129112921293129412951296129712981299130013011302130313041305130613071308130913101311131213131314131513161317131813191320132113221323132413251326132713281329133013311332133313341335133613371338133913401341134213431344134513461347134813491350135113521353135413551356135713581359136013611362136313641365136613671368136913701371137213731374137513761377137813791380138113821383138413851386138713881389139013911392139313941395139613971398139914001401140214031404140514061407140814091410141114121413141414151416141714181419142014211422142314241425142614271428142914301431143214331434143514361437143814391440144114421443144414451446144714481449145014511452145314541455145614571458145914601461146214631464146514661467146814691470147114721473147414751476147714781479148014811482148314841485148614871488148914901491149214931494149514961497149814991500150115021503150415051506150715081509151015111512151315141515151615171518151915201521152215231524152515261527152815291530153115321533153415351536153715381539154015411542154315441545154615471548154915501551155215531554155515561557155815591560156115621563156415651566156715681569157015711572157315741575157615771578157915801581158215831584158515861587158815891590159115921593159415951596159715981599160016011602160316041605160616071608160916101611161216131614161516161617161816191620162116221623162416251626162716281629163016311632163316341635163616371638163916401641164216431644164516461647164816491650165116521653165416551656165716581659166016611662166316641665166616671668166916701671167216731674167516761677167816791680168116821683168416851686168716881689169016911692169316941695169616971698169917001701170217031704170517061707170817091710171117121713171417151716171717181719172017211722172317241725172617271728172917301731173217331734173517361737173817391740174117421743174417451746174717481749175017511752175317541755175617571758175917601761176217631764176517661767176817691770177117721773177417751776177717781779178017811782178317841785178617871788178917901791179217931794179517961797179817991800180118021803180418051806180718081809181018111812181318141815181618171818181918201821182218231824182518261827182818291830183118321833183418351836183718381839184018411842184318441845184618471848184918501851185218531854185518561857185818591860186118621863186418651866186718681869187018711872187318741875187618771878187918801881188218831884188518861887188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911191219131914191519161917191819191920192119221923192419251926192719281929193019311932193319341935193619371938193919401941194219431944194519461947194819491950195119521953195419551956195719581959196019611962196319641965196619671968196919701971197219731974197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010201120122013201420152016201720182019202020212022202320242025202620272028202920302031203220332034203520362037203820392040204120422043204420452046204720482049205020512052205320542055205620572058205920602061206220632064206520662067206820692070207120722073207420752076207720782079208020812082208320842085208620872088208920902091209220932094209520962097209820992100210121022103210421052106210721082109211021112112211321142115211621172118211921202121212221232124212521262127212821292130213121322133213421352136213721382139214021412142214321442145214621472148214921502151215221532154215521562157215821592160216121622163216421652166216721682169217021712172217321742175217621772178217921802181218221832184218521862187218821892190219121922193219421952196219721982199220022012202220322042205220622072208220922102211221222132214221522162217221822192220222122222223222422252226222722282229223022312232223322342235223622372238223922402241224222432244224522462247224822492250225122522253225422552256225722582259226022612262226322642265226622672268226922702271227222732274227522762277227822792280228122822283228422852286228722882289229022912292229322942295229622972298229923002301230223032304230523062307230823092310231123122313231423152316231723182319232023212322232323242325232623272328232923302331233223332334233523362337233823392340234123422343234423452346234723482349235023512352235323542355235623572358235923602361236223632364236523662367236823692370237123722373237423752376237723782379238023812382238323842385238623872388238923902391239223932394239523962397239823992400240124022403240424052406240724082409241024112412241324142415241624172418241924202421242224232424242524262427242824292430243124322433243424352436243724382439244024412442244324442445244624472448244924502451245224532454245524562457245824592460246124622463246424652466246724682469247024712472247324742475247624772478247924802481248224832484248524862487248824892490249124922493249424952496249724982499250025012502250325042505250625072508250925102511251225132514251525162517251825192520252125222523252425252526252725282529253025312532253325342535253625372538253925402541254225432544254525462547254825492550255125522553255425552556255725582559256025612562256325642565256625672568256925702571257225732574257525762577257825792580258125822583258425852586258725882589259025912592259325942595259625972598259926002601260226032604260526062607260826092610261126122613261426152616261726182619262026212622262326242625262626272628262926302631263226332634263526362637263826392640264126422643264426452646264726482649265026512652265326542655265626572658265926602661266226632664266526662667266826692670267126722673267426752676267726782679268026812682268326842685268626872688268926902691269226932694269526962697269826992700270127022703270427052706270727082709271027112712271327142715271627172718271927202721272227232724272527262727272827292730273127322733273427352736273727382739274027412742274327442745274627472748274927502751275227532754275527562757275827592760276127622763276427652766276727682769277027712772277327742775277627772778277927802781278227832784278527862787278827892790279127922793279427952796279727982799280028012802280328042805280628072808280928102811281228132814281528162817281828192820282128222823282428252826282728282829283028312832283328342835283628372838283928402841284228432844284528462847284828492850285128522853285428552856285728582859286028612862286328642865286628672868286928702871287228732874287528762877287828792880288128822883288428852886288728882889289028912892289328942895289628972898289929002901290229032904290529062907290829092910291129122913291429152916291729182919292029212922292329242925292629272928292929302931293229332934293529362937293829392940294129422943294429452946294729482949295029512952295329542955295629572958295929602961296229632964296529662967296829692970297129722973297429752976297729782979298029812982298329842985298629872988298929902991299229932994299529962997299829993000300130023003300430053006300730083009301030113012301330143015301630173018301930203021302230233024302530263027302830293030303130323033303430353036303730383039304030413042304330443045304630473048304930503051305230533054305530563057305830593060306130623063306430653066306730683069307030713072307330743075307630773078307930803081308230833084308530863087308830893090309130923093309430953096309730983099310031013102310331043105310631073108310931103111311231133114311531163117311831193120312131223123312431253126312731283129313031313132313331343135313631373138313931403141314231433144314531463147314831493150315131523153315431553156315731583159316031613162316331643165316631673168316931703171317231733174317531763177317831793180318131823183318431853186318731883189319031913192319331943195319631973198319932003201320232033204320532063207320832093210321132123213321432153216321732183219322032213222322332243225322632273228322932303231323232333234323532363237323832393240324132423243324432453246324732483249325032513252325332543255325632573258325932603261326232633264326532663267326832693270327132723273327432753276327732783279328032813282328332843285328632873288328932903291329232933294329532963297329832993300330133023303330433053306330733083309331033113312331333143315331633173318331933203321332233233324332533263327332833293330333133323333333433353336333733383339334033413342334333443345334633473348334933503351335233533354335533563357335833593360336133623363336433653366336733683369337033713372337333743375337633773378337933803381338233833384338533863387338833893390339133923393339433953396339733983399340034013402340334043405340634073408340934103411341234133414341534163417341834193420342134223423342434253426342734283429343034313432343334343435343634373438343934403441344234433444344534463447344834493450345134523453345434553456345734583459346034613462346334643465346634673468346934703471347234733474347534763477347834793480348134823483348434853486348734883489349034913492349334943495349634973498349935003501350235033504350535063507350835093510351135123513351435153516351735183519352035213522352335243525352635273528352935303531353235333534353535363537353835393540354135423543354435453546354735483549355035513552355335543555355635573558355935603561356235633564356535663567356835693570357135723573357435753576357735783579358035813582358335843585358635873588358935903591359235933594359535963597359835993600360136023603360436053606360736083609361036113612361336143615361636173618361936203621362236233624362536263627362836293630363136323633363436353636363736383639364036413642364336443645364636473648364936503651365236533654365536563657365836593660366136623663366436653666366736683669367036713672367336743675367636773678367936803681368236833684368536863687368836893690369136923693369436953696369736983699370037013702370337043705370637073708370937103711371237133714371537163717371837193720372137223723372437253726372737283729373037313732373337343735373637373738373937403741374237433744374537463747374837493750375137523753375437553756375737583759376037613762376337643765376637673768376937703771377237733774377537763777377837793780378137823783378437853786378737883789379037913792379337943795379637973798379938003801380238033804380538063807380838093810381138123813381438153816381738183819382038213822382338243825382638273828382938303831383238333834383538363837383838393840384138423843384438453846384738483849385038513852385338543855385638573858385938603861386238633864386538663867386838693870387138723873387438753876387738783879388038813882388338843885388638873888388938903891389238933894389538963897389838993900390139023903390439053906390739083909391039113912391339143915391639173918391939203921392239233924392539263927392839293930393139323933393439353936393739383939394039413942394339443945394639473948394939503951395239533954395539563957395839593960396139623963396439653966396739683969397039713972397339743975397639773978397939803981398239833984398539863987398839893990399139923993399439953996399739983999400040014002400340044005400640074008400940104011401240134014401540164017401840194020402140224023402440254026402740284029403040314032403340344035403640374038403940404041404240434044404540464047404840494050405140524053405440554056405740584059406040614062406340644065406640674068406940704071407240734074407540764077407840794080408140824083408440854086408740884089409040914092409340944095409640974098409941004101410241034104410541064107410841094110411141124113411441154116411741184119412041214122412341244125412641274128412941304131413241334134413541364137413841394140414141424143414441454146414741484149415041514152415341544155415641574158415941604161416241634164416541664167416841694170417141724173417441754176417741784179418041814182418341844185418641874188418941904191419241934194419541964197419841994200420142024203420442054206420742084209421042114212421342144215421642174218421942204221422242234224422542264227422842294230423142324233423442354236423742384239424042414242424342444245424642474248424942504251425242534254425542564257425842594260426142624263426442654266426742684269427042714272427342744275427642774278427942804281428242834284428542864287428842894290429142924293429442954296429742984299430043014302430343044305430643074308430943104311431243134314431543164317431843194320432143224323432443254326432743284329433043314332433343344335433643374338433943404341434243434344434543464347434843494350435143524353435443554356435743584359436043614362436343644365436643674368436943704371437243734374437543764377437843794380438143824383438443854386438743884389439043914392439343944395439643974398439944004401440244034404440544064407440844094410441144124413441444154416441744184419442044214422442344244425442644274428442944304431443244334434443544364437443844394440444144424443444444454446444744484449445044514452445344544455445644574458445944604461446244634464446544664467446844694470447144724473447444754476447744784479448044814482448344844485448644874488448944904491449244934494449544964497449844994500450145024503450445054506450745084509451045114512451345144515451645174518451945204521452245234524452545264527452845294530453145324533453445354536453745384539454045414542454345444545454645474548454945504551455245534554455545564557455845594560456145624563456445654566456745684569457045714572457345744575457645774578457945804581458245834584458545864587458845894590459145924593459445954596459745984599460046014602460346044605460646074608460946104611461246134614461546164617461846194620462146224623462446254626462746284629463046314632463346344635463646374638463946404641464246434644464546464647464846494650465146524653465446554656465746584659466046614662466346644665466646674668466946704671467246734674467546764677467846794680468146824683468446854686468746884689469046914692469346944695469646974698469947004701470247034704470547064707470847094710471147124713471447154716471747184719472047214722472347244725472647274728472947304731473247334734473547364737473847394740474147424743474447454746474747484749475047514752475347544755475647574758475947604761476247634764476547664767476847694770477147724773477447754776477747784779478047814782478347844785478647874788478947904791479247934794479547964797479847994800480148024803480448054806480748084809481048114812481348144815481648174818481948204821482248234824482548264827482848294830483148324833483448354836483748384839484048414842484348444845484648474848484948504851485248534854485548564857485848594860486148624863486448654866486748684869487048714872487348744875487648774878487948804881488248834884488548864887488848894890489148924893489448954896489748984899490049014902490349044905490649074908490949104911491249134914491549164917491849194920492149224923492449254926492749284929493049314932493349344935493649374938493949404941494249434944494549464947494849494950495149524953495449554956495749584959496049614962496349644965496649674968496949704971497249734974497549764977497849794980498149824983498449854986498749884989499049914992499349944995499649974998499950005001500250035004500550065007500850095010501150125013501450155016501750185019502050215022502350245025502650275028502950305031503250335034503550365037503850395040504150425043504450455046504750485049505050515052505350545055505650575058505950605061506250635064506550665067506850695070507150725073507450755076507750785079508050815082508350845085508650875088508950905091509250935094509550965097509850995100510151025103510451055106510751085109511051115112511351145115511651175118511951205121512251235124512551265127512851295130513151325133513451355136513751385139514051415142514351445145514651475148514951505151515251535154515551565157515851595160516151625163516451655166516751685169517051715172517351745175517651775178517951805181518251835184518551865187518851895190519151925193519451955196519751985199520052015202520352045205520652075208520952105211521252135214521552165217521852195220522152225223522452255226522752285229523052315232523352345235523652375238523952405241524252435244524552465247524852495250525152525253525452555256525752585259526052615262526352645265526652675268526952705271527252735274527552765277527852795280528152825283528452855286528752885289529052915292529352945295529652975298529953005301530253035304530553065307530853095310531153125313531453155316531753185319532053215322532353245325532653275328532953305331533253335334533553365337533853395340534153425343534453455346534753485349535053515352535353545355535653575358535953605361536253635364536553665367536853695370537153725373537453755376537753785379538053815382538353845385538653875388538953905391539253935394539553965397539853995400540154025403540454055406540754085409541054115412541354145415541654175418541954205421542254235424542554265427542854295430543154325433543454355436543754385439544054415442544354445445544654475448544954505451545254535454545554565457545854595460546154625463546454655466546754685469547054715472547354745475547654775478547954805481548254835484548554865487548854895490549154925493549454955496549754985499550055015502550355045505550655075508550955105511551255135514551555165517551855195520552155225523552455255526552755285529553055315532553355345535553655375538553955405541554255435544554555465547554855495550555155525553555455555556555755585559556055615562556355645565556655675568556955705571557255735574557555765577557855795580558155825583558455855586558755885589559055915592559355945595559655975598559956005601560256035604560556065607560856095610561156125613561456155616561756185619562056215622562356245625562656275628562956305631563256335634563556365637563856395640564156425643564456455646564756485649565056515652565356545655565656575658565956605661566256635664566556665667566856695670567156725673567456755676567756785679568056815682568356845685568656875688568956905691569256935694569556965697569856995700570157025703570457055706570757085709571057115712571357145715571657175718571957205721572257235724572557265727572857295730573157325733573457355736573757385739574057415742574357445745574657475748574957505751575257535754575557565757575857595760576157625763576457655766576757685769577057715772577357745775577657775778577957805781578257835784578557865787578857895790579157925793579457955796579757985799580058015802580358045805580658075808580958105811581258135814581558165817581858195820582158225823582458255826582758285829583058315832583358345835583658375838583958405841584258435844584558465847584858495850585158525853585458555856585758585859586058615862586358645865586658675868586958705871587258735874587558765877587858795880588158825883588458855886588758885889589058915892589358945895589658975898589959005901590259035904590559065907590859095910591159125913591459155916591759185919592059215922592359245925592659275928592959305931593259335934593559365937593859395940594159425943594459455946594759485949595059515952595359545955595659575958595959605961596259635964596559665967596859695970597159725973597459755976597759785979598059815982598359845985598659875988598959905991599259935994599559965997599859996000600160026003600460056006600760086009601060116012601360146015601660176018601960206021602260236024602560266027602860296030603160326033603460356036603760386039604060416042604360446045604660476048604960506051605260536054605560566057605860596060606160626063606460656066606760686069607060716072607360746075607660776078607960806081608260836084608560866087608860896090609160926093609460956096609760986099610061016102610361046105610661076108610961106111611261136114611561166117611861196120612161226123612461256126612761286129613061316132613361346135613661376138613961406141614261436144614561466147614861496150615161526153615461556156615761586159616061616162616361646165616661676168616961706171617261736174617561766177617861796180618161826183618461856186618761886189619061916192619361946195619661976198619962006201620262036204620562066207620862096210621162126213621462156216621762186219622062216222622362246225622662276228622962306231623262336234623562366237623862396240624162426243624462456246624762486249625062516252625362546255625662576258625962606261626262636264626562666267626862696270627162726273627462756276627762786279628062816282628362846285628662876288628962906291629262936294629562966297629862996300630163026303630463056306630763086309631063116312631363146315631663176318631963206321632263236324632563266327632863296330633163326333633463356336633763386339634063416342634363446345634663476348634963506351635263536354635563566357635863596360636163626363636463656366636763686369637063716372637363746375637663776378637963806381638263836384638563866387638863896390639163926393639463956396639763986399640064016402640364046405640664076408640964106411641264136414641564166417641864196420642164226423642464256426642764286429643064316432643364346435643664376438643964406441644264436444644564466447644864496450645164526453645464556456645764586459646064616462646364646465646664676468646964706471647264736474647564766477647864796480648164826483648464856486648764886489649064916492649364946495649664976498649965006501650265036504650565066507650865096510651165126513651465156516651765186519652065216522652365246525652665276528652965306531653265336534653565366537653865396540654165426543654465456546654765486549655065516552655365546555655665576558655965606561656265636564656565666567656865696570657165726573657465756576657765786579658065816582658365846585658665876588658965906591659265936594659565966597659865996600660166026603660466056606660766086609661066116612661366146615661666176618661966206621662266236624662566266627662866296630663166326633663466356636663766386639664066416642664366446645664666476648664966506651665266536654665566566657665866596660666166626663666466656666666766686669667066716672667366746675667666776678667966806681668266836684668566866687668866896690669166926693669466956696669766986699670067016702670367046705670667076708670967106711671267136714671567166717671867196720672167226723672467256726672767286729673067316732673367346735673667376738673967406741674267436744674567466747674867496750675167526753675467556756675767586759676067616762676367646765676667676768676967706771677267736774677567766777677867796780678167826783678467856786678767886789679067916792679367946795679667976798679968006801680268036804680568066807680868096810681168126813681468156816681768186819682068216822682368246825682668276828682968306831683268336834683568366837683868396840684168426843684468456846684768486849685068516852685368546855685668576858685968606861686268636864686568666867686868696870687168726873687468756876687768786879688068816882688368846885688668876888688968906891689268936894689568966897689868996900690169026903690469056906690769086909691069116912691369146915691669176918691969206921692269236924692569266927692869296930693169326933693469356936693769386939694069416942694369446945694669476948694969506951695269536954695569566957695869596960696169626963696469656966696769686969697069716972697369746975697669776978697969806981698269836984698569866987698869896990699169926993699469956996699769986999700070017002700370047005700670077008700970107011701270137014701570167017701870197020702170227023702470257026702770287029703070317032703370347035703670377038703970407041704270437044704570467047704870497050705170527053705470557056705770587059706070617062706370647065706670677068706970707071707270737074707570767077707870797080708170827083708470857086708770887089709070917092709370947095709670977098709971007101710271037104710571067107710871097110711171127113711471157116711771187119712071217122712371247125712671277128712971307131713271337134713571367137713871397140714171427143714471457146714771487149715071517152715371547155715671577158715971607161716271637164716571667167716871697170717171727173717471757176717771787179718071817182718371847185718671877188718971907191719271937194719571967197719871997200720172027203720472057206720772087209721072117212721372147215721672177218721972207221722272237224722572267227722872297230723172327233723472357236723772387239724072417242724372447245724672477248724972507251725272537254725572567257725872597260726172627263726472657266726772687269727072717272727372747275727672777278727972807281728272837284728572867287728872897290729172927293729472957296729772987299730073017302730373047305730673077308730973107311731273137314/*
 * drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
 * Copyright (c) 2016-2018 Mellanox Technologies. All rights reserved.
 * Copyright (c) 2016 Jiri Pirko <jiri@mellanox.com>
 * Copyright (c) 2016 Ido Schimmel <idosch@mellanox.com>
 * Copyright (c) 2016 Yotam Gigi <yotamg@mellanox.com>
 * Copyright (c) 2017-2018 Petr Machata <petrm@mellanox.com>
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the names of the copyright holders nor the names of its
 *    contributors may be used to endorse or promote products derived from
 *    this software without specific prior written permission.
 *
 * Alternatively, this software may be distributed under the terms of the
 * GNU General Public License ("GPL") version 2 as published by the Free
 * Software Foundation.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
 * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/rhashtable.h>
#include <linux/bitops.h>
#include <linux/in6.h>
#include <linux/notifier.h>
#include <linux/inetdevice.h>
#include <linux/netdevice.h>
#include <linux/if_bridge.h>
#include <linux/socket.h>
#include <linux/route.h>
#include <linux/gcd.h>
#include <linux/random.h>
#include <net/netevent.h>
#include <net/neighbour.h>
#include <net/arp.h>
#include <net/ip_fib.h>
#include <net/ip6_fib.h>
#include <net/fib_rules.h>
#include <net/ip_tunnels.h>
#include <net/l3mdev.h>
#include <net/addrconf.h>
#include <net/ndisc.h>
#include <net/ipv6.h>
#include <net/fib_notifier.h>

#include "spectrum.h"
#include "core.h"
#include "reg.h"
#include "spectrum_cnt.h"
#include "spectrum_dpipe.h"
#include "spectrum_ipip.h"
#include "spectrum_mr.h"
#include "spectrum_mr_tcam.h"
#include "spectrum_router.h"
#include "spectrum_span.h"

struct mlxsw_sp_fib;
struct mlxsw_sp_vr;
struct mlxsw_sp_lpm_tree;
struct mlxsw_sp_rif_ops;

struct mlxsw_sp_router {
	struct mlxsw_sp *mlxsw_sp;
	struct mlxsw_sp_rif **rifs;
	struct mlxsw_sp_vr *vrs;
	struct rhashtable neigh_ht;
	struct rhashtable nexthop_group_ht;
	struct rhashtable nexthop_ht;
	struct list_head nexthop_list;
	struct {
		/* One tree for each protocol: IPv4 and IPv6 */
		struct mlxsw_sp_lpm_tree *proto_trees[2];
		struct mlxsw_sp_lpm_tree *trees;
		unsigned int tree_count;
	} lpm;
	struct {
		struct delayed_work dw;
		unsigned long interval;	/* ms */
	} neighs_update;
	struct delayed_work nexthop_probe_dw;
#define MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL 5000 /* ms */
	struct list_head nexthop_neighs_list;
	struct list_head ipip_list;
	bool aborted;
	struct notifier_block fib_nb;
	struct notifier_block netevent_nb;
	const struct mlxsw_sp_rif_ops **rif_ops_arr;
	const struct mlxsw_sp_ipip_ops **ipip_ops_arr;
};

struct mlxsw_sp_rif {
	struct list_head nexthop_list;
	struct list_head neigh_list;
	struct net_device *dev;
	struct mlxsw_sp_fid *fid;
	unsigned char addr[ETH_ALEN];
	int mtu;
	u16 rif_index;
	u16 vr_id;
	const struct mlxsw_sp_rif_ops *ops;
	struct mlxsw_sp *mlxsw_sp;

	unsigned int counter_ingress;
	bool counter_ingress_valid;
	unsigned int counter_egress;
	bool counter_egress_valid;
};

struct mlxsw_sp_rif_params {
	struct net_device *dev;
	union {
		u16 system_port;
		u16 lag_id;
	};
	u16 vid;
	bool lag;
};

struct mlxsw_sp_rif_subport {
	struct mlxsw_sp_rif common;
	union {
		u16 system_port;
		u16 lag_id;
	};
	u16 vid;
	bool lag;
};

struct mlxsw_sp_rif_ipip_lb {
	struct mlxsw_sp_rif common;
	struct mlxsw_sp_rif_ipip_lb_config lb_config;
	u16 ul_vr_id; /* Reserved for Spectrum-2. */
};

struct mlxsw_sp_rif_params_ipip_lb {
	struct mlxsw_sp_rif_params common;
	struct mlxsw_sp_rif_ipip_lb_config lb_config;
};

struct mlxsw_sp_rif_ops {
	enum mlxsw_sp_rif_type type;
	size_t rif_size;

	void (*setup)(struct mlxsw_sp_rif *rif,
		      const struct mlxsw_sp_rif_params *params);
	int (*configure)(struct mlxsw_sp_rif *rif);
	void (*deconfigure)(struct mlxsw_sp_rif *rif);
	struct mlxsw_sp_fid * (*fid_get)(struct mlxsw_sp_rif *rif);
};

static void mlxsw_sp_lpm_tree_hold(struct mlxsw_sp_lpm_tree *lpm_tree);
static void mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_lpm_tree *lpm_tree);
static int mlxsw_sp_vr_lpm_tree_bind(struct mlxsw_sp *mlxsw_sp,
				     const struct mlxsw_sp_fib *fib,
				     u8 tree_id);
static int mlxsw_sp_vr_lpm_tree_unbind(struct mlxsw_sp *mlxsw_sp,
				       const struct mlxsw_sp_fib *fib);

static unsigned int *
mlxsw_sp_rif_p_counter_get(struct mlxsw_sp_rif *rif,
			   enum mlxsw_sp_rif_counter_dir dir)
{
	switch (dir) {
	case MLXSW_SP_RIF_COUNTER_EGRESS:
		return &rif->counter_egress;
	case MLXSW_SP_RIF_COUNTER_INGRESS:
		return &rif->counter_ingress;
	}
	return NULL;
}

static bool
mlxsw_sp_rif_counter_valid_get(struct mlxsw_sp_rif *rif,
			       enum mlxsw_sp_rif_counter_dir dir)
{
	switch (dir) {
	case MLXSW_SP_RIF_COUNTER_EGRESS:
		return rif->counter_egress_valid;
	case MLXSW_SP_RIF_COUNTER_INGRESS:
		return rif->counter_ingress_valid;
	}
	return false;
}

static void
mlxsw_sp_rif_counter_valid_set(struct mlxsw_sp_rif *rif,
			       enum mlxsw_sp_rif_counter_dir dir,
			       bool valid)
{
	switch (dir) {
	case MLXSW_SP_RIF_COUNTER_EGRESS:
		rif->counter_egress_valid = valid;
		break;
	case MLXSW_SP_RIF_COUNTER_INGRESS:
		rif->counter_ingress_valid = valid;
		break;
	}
}

static int mlxsw_sp_rif_counter_edit(struct mlxsw_sp *mlxsw_sp, u16 rif_index,
				     unsigned int counter_index, bool enable,
				     enum mlxsw_sp_rif_counter_dir dir)
{
	char ritr_pl[MLXSW_REG_RITR_LEN];
	bool is_egress = false;
	int err;

	if (dir == MLXSW_SP_RIF_COUNTER_EGRESS)
		is_egress = true;
	mlxsw_reg_ritr_rif_pack(ritr_pl, rif_index);
	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl);
	if (err)
		return err;

	mlxsw_reg_ritr_counter_pack(ritr_pl, counter_index, enable,
				    is_egress);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl);
}

int mlxsw_sp_rif_counter_value_get(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_rif *rif,
				   enum mlxsw_sp_rif_counter_dir dir, u64 *cnt)
{
	char ricnt_pl[MLXSW_REG_RICNT_LEN];
	unsigned int *p_counter_index;
	bool valid;
	int err;

	valid = mlxsw_sp_rif_counter_valid_get(rif, dir);
	if (!valid)
		return -EINVAL;

	p_counter_index = mlxsw_sp_rif_p_counter_get(rif, dir);
	if (!p_counter_index)
		return -EINVAL;
	mlxsw_reg_ricnt_pack(ricnt_pl, *p_counter_index,
			     MLXSW_REG_RICNT_OPCODE_NOP);
	err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(ricnt), ricnt_pl);
	if (err)
		return err;
	*cnt = mlxsw_reg_ricnt_good_unicast_packets_get(ricnt_pl);
	return 0;
}

static int mlxsw_sp_rif_counter_clear(struct mlxsw_sp *mlxsw_sp,
				      unsigned int counter_index)
{
	char ricnt_pl[MLXSW_REG_RICNT_LEN];

	mlxsw_reg_ricnt_pack(ricnt_pl, counter_index,
			     MLXSW_REG_RICNT_OPCODE_CLEAR);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ricnt), ricnt_pl);
}

int mlxsw_sp_rif_counter_alloc(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_rif *rif,
			       enum mlxsw_sp_rif_counter_dir dir)
{
	unsigned int *p_counter_index;
	int err;

	p_counter_index = mlxsw_sp_rif_p_counter_get(rif, dir);
	if (!p_counter_index)
		return -EINVAL;
	err = mlxsw_sp_counter_alloc(mlxsw_sp, MLXSW_SP_COUNTER_SUB_POOL_RIF,
				     p_counter_index);
	if (err)
		return err;

	err = mlxsw_sp_rif_counter_clear(mlxsw_sp, *p_counter_index);
	if (err)
		goto err_counter_clear;

	err = mlxsw_sp_rif_counter_edit(mlxsw_sp, rif->rif_index,
					*p_counter_index, true, dir);
	if (err)
		goto err_counter_edit;
	mlxsw_sp_rif_counter_valid_set(rif, dir, true);
	return 0;

err_counter_edit:
err_counter_clear:
	mlxsw_sp_counter_free(mlxsw_sp, MLXSW_SP_COUNTER_SUB_POOL_RIF,
			      *p_counter_index);
	return err;
}

void mlxsw_sp_rif_counter_free(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_rif *rif,
			       enum mlxsw_sp_rif_counter_dir dir)
{
	unsigned int *p_counter_index;

	if (!mlxsw_sp_rif_counter_valid_get(rif, dir))
		return;

	p_counter_index = mlxsw_sp_rif_p_counter_get(rif, dir);
	if (WARN_ON(!p_counter_index))
		return;
	mlxsw_sp_rif_counter_edit(mlxsw_sp, rif->rif_index,
				  *p_counter_index, false, dir);
	mlxsw_sp_counter_free(mlxsw_sp, MLXSW_SP_COUNTER_SUB_POOL_RIF,
			      *p_counter_index);
	mlxsw_sp_rif_counter_valid_set(rif, dir, false);
}

static void mlxsw_sp_rif_counters_alloc(struct mlxsw_sp_rif *rif)
{
	struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp;
	struct devlink *devlink;

	devlink = priv_to_devlink(mlxsw_sp->core);
	if (!devlink_dpipe_table_counter_enabled(devlink,
						 MLXSW_SP_DPIPE_TABLE_NAME_ERIF))
		return;
	mlxsw_sp_rif_counter_alloc(mlxsw_sp, rif, MLXSW_SP_RIF_COUNTER_EGRESS);
}

static void mlxsw_sp_rif_counters_free(struct mlxsw_sp_rif *rif)
{
	struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp;

	mlxsw_sp_rif_counter_free(mlxsw_sp, rif, MLXSW_SP_RIF_COUNTER_EGRESS);
}

static struct mlxsw_sp_rif *
mlxsw_sp_rif_find_by_dev(const struct mlxsw_sp *mlxsw_sp,
			 const struct net_device *dev);

#define MLXSW_SP_PREFIX_COUNT (sizeof(struct in6_addr) * BITS_PER_BYTE + 1)

struct mlxsw_sp_prefix_usage {
	DECLARE_BITMAP(b, MLXSW_SP_PREFIX_COUNT);
};

#define mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) \
	for_each_set_bit(prefix, (prefix_usage)->b, MLXSW_SP_PREFIX_COUNT)

static bool
mlxsw_sp_prefix_usage_eq(struct mlxsw_sp_prefix_usage *prefix_usage1,
			 struct mlxsw_sp_prefix_usage *prefix_usage2)
{
	return !memcmp(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1));
}

static void
mlxsw_sp_prefix_usage_cpy(struct mlxsw_sp_prefix_usage *prefix_usage1,
			  struct mlxsw_sp_prefix_usage *prefix_usage2)
{
	memcpy(prefix_usage1, prefix_usage2, sizeof(*prefix_usage1));
}

static void
mlxsw_sp_prefix_usage_set(struct mlxsw_sp_prefix_usage *prefix_usage,
			  unsigned char prefix_len)
{
	set_bit(prefix_len, prefix_usage->b);
}

static void
mlxsw_sp_prefix_usage_clear(struct mlxsw_sp_prefix_usage *prefix_usage,
			    unsigned char prefix_len)
{
	clear_bit(prefix_len, prefix_usage->b);
}

struct mlxsw_sp_fib_key {
	unsigned char addr[sizeof(struct in6_addr)];
	unsigned char prefix_len;
};

enum mlxsw_sp_fib_entry_type {
	MLXSW_SP_FIB_ENTRY_TYPE_REMOTE,
	MLXSW_SP_FIB_ENTRY_TYPE_LOCAL,
	MLXSW_SP_FIB_ENTRY_TYPE_TRAP,

	/* This is a special case of local delivery, where a packet should be
	 * decapsulated on reception. Note that there is no corresponding ENCAP,
	 * because that's a type of next hop, not of FIB entry. (There can be
	 * several next hops in a REMOTE entry, and some of them may be
	 * encapsulating entries.)
	 */
	MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP,
};

struct mlxsw_sp_nexthop_group;

struct mlxsw_sp_fib_node {
	struct list_head entry_list;
	struct list_head list;
	struct rhash_head ht_node;
	struct mlxsw_sp_fib *fib;
	struct mlxsw_sp_fib_key key;
};

struct mlxsw_sp_fib_entry_decap {
	struct mlxsw_sp_ipip_entry *ipip_entry;
	u32 tunnel_index;
};

struct mlxsw_sp_fib_entry {
	struct list_head list;
	struct mlxsw_sp_fib_node *fib_node;
	enum mlxsw_sp_fib_entry_type type;
	struct list_head nexthop_group_node;
	struct mlxsw_sp_nexthop_group *nh_group;
	struct mlxsw_sp_fib_entry_decap decap; /* Valid for decap entries. */
};

struct mlxsw_sp_fib4_entry {
	struct mlxsw_sp_fib_entry common;
	u32 tb_id;
	u32 prio;
	u8 tos;
	u8 type;
};

struct mlxsw_sp_fib6_entry {
	struct mlxsw_sp_fib_entry common;
	struct list_head rt6_list;
	unsigned int nrt6;
};

struct mlxsw_sp_rt6 {
	struct list_head list;
	struct fib6_info *rt;
};

struct mlxsw_sp_lpm_tree {
	u8 id; /* tree ID */
	unsigned int ref_count;
	enum mlxsw_sp_l3proto proto;
	unsigned long prefix_ref_count[MLXSW_SP_PREFIX_COUNT];
	struct mlxsw_sp_prefix_usage prefix_usage;
};

struct mlxsw_sp_fib {
	struct rhashtable ht;
	struct list_head node_list;
	struct mlxsw_sp_vr *vr;
	struct mlxsw_sp_lpm_tree *lpm_tree;
	enum mlxsw_sp_l3proto proto;
};

struct mlxsw_sp_vr {
	u16 id; /* virtual router ID */
	u32 tb_id; /* kernel fib table id */
	unsigned int rif_count;
	struct mlxsw_sp_fib *fib4;
	struct mlxsw_sp_fib *fib6;
	struct mlxsw_sp_mr_table *mr_table[MLXSW_SP_L3_PROTO_MAX];
};

static const struct rhashtable_params mlxsw_sp_fib_ht_params;

static struct mlxsw_sp_fib *mlxsw_sp_fib_create(struct mlxsw_sp *mlxsw_sp,
						struct mlxsw_sp_vr *vr,
						enum mlxsw_sp_l3proto proto)
{
	struct mlxsw_sp_lpm_tree *lpm_tree;
	struct mlxsw_sp_fib *fib;
	int err;

	lpm_tree = mlxsw_sp->router->lpm.proto_trees[proto];
	fib = kzalloc(sizeof(*fib), GFP_KERNEL);
	if (!fib)
		return ERR_PTR(-ENOMEM);
	err = rhashtable_init(&fib->ht, &mlxsw_sp_fib_ht_params);
	if (err)
		goto err_rhashtable_init;
	INIT_LIST_HEAD(&fib->node_list);
	fib->proto = proto;
	fib->vr = vr;
	fib->lpm_tree = lpm_tree;
	mlxsw_sp_lpm_tree_hold(lpm_tree);
	err = mlxsw_sp_vr_lpm_tree_bind(mlxsw_sp, fib, lpm_tree->id);
	if (err)
		goto err_lpm_tree_bind;
	return fib;

err_lpm_tree_bind:
	mlxsw_sp_lpm_tree_put(mlxsw_sp, lpm_tree);
err_rhashtable_init:
	kfree(fib);
	return ERR_PTR(err);
}

static void mlxsw_sp_fib_destroy(struct mlxsw_sp *mlxsw_sp,
				 struct mlxsw_sp_fib *fib)
{
	mlxsw_sp_vr_lpm_tree_unbind(mlxsw_sp, fib);
	mlxsw_sp_lpm_tree_put(mlxsw_sp, fib->lpm_tree);
	WARN_ON(!list_empty(&fib->node_list));
	rhashtable_destroy(&fib->ht);
	kfree(fib);
}

static struct mlxsw_sp_lpm_tree *
mlxsw_sp_lpm_tree_find_unused(struct mlxsw_sp *mlxsw_sp)
{
	static struct mlxsw_sp_lpm_tree *lpm_tree;
	int i;

	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
		if (lpm_tree->ref_count == 0)
			return lpm_tree;
	}
	return NULL;
}

static int mlxsw_sp_lpm_tree_alloc(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_lpm_tree *lpm_tree)
{
	char ralta_pl[MLXSW_REG_RALTA_LEN];

	mlxsw_reg_ralta_pack(ralta_pl, true,
			     (enum mlxsw_reg_ralxx_protocol) lpm_tree->proto,
			     lpm_tree->id);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl);
}

static void mlxsw_sp_lpm_tree_free(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_lpm_tree *lpm_tree)
{
	char ralta_pl[MLXSW_REG_RALTA_LEN];

	mlxsw_reg_ralta_pack(ralta_pl, false,
			     (enum mlxsw_reg_ralxx_protocol) lpm_tree->proto,
			     lpm_tree->id);
	mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl);
}

static int
mlxsw_sp_lpm_tree_left_struct_set(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_prefix_usage *prefix_usage,
				  struct mlxsw_sp_lpm_tree *lpm_tree)
{
	char ralst_pl[MLXSW_REG_RALST_LEN];
	u8 root_bin = 0;
	u8 prefix;
	u8 last_prefix = MLXSW_REG_RALST_BIN_NO_CHILD;

	mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage)
		root_bin = prefix;

	mlxsw_reg_ralst_pack(ralst_pl, root_bin, lpm_tree->id);
	mlxsw_sp_prefix_usage_for_each(prefix, prefix_usage) {
		if (prefix == 0)
			continue;
		mlxsw_reg_ralst_bin_pack(ralst_pl, prefix, last_prefix,
					 MLXSW_REG_RALST_BIN_NO_CHILD);
		last_prefix = prefix;
	}
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralst), ralst_pl);
}

static struct mlxsw_sp_lpm_tree *
mlxsw_sp_lpm_tree_create(struct mlxsw_sp *mlxsw_sp,
			 struct mlxsw_sp_prefix_usage *prefix_usage,
			 enum mlxsw_sp_l3proto proto)
{
	struct mlxsw_sp_lpm_tree *lpm_tree;
	int err;

	lpm_tree = mlxsw_sp_lpm_tree_find_unused(mlxsw_sp);
	if (!lpm_tree)
		return ERR_PTR(-EBUSY);
	lpm_tree->proto = proto;
	err = mlxsw_sp_lpm_tree_alloc(mlxsw_sp, lpm_tree);
	if (err)
		return ERR_PTR(err);

	err = mlxsw_sp_lpm_tree_left_struct_set(mlxsw_sp, prefix_usage,
						lpm_tree);
	if (err)
		goto err_left_struct_set;
	memcpy(&lpm_tree->prefix_usage, prefix_usage,
	       sizeof(lpm_tree->prefix_usage));
	memset(&lpm_tree->prefix_ref_count, 0,
	       sizeof(lpm_tree->prefix_ref_count));
	lpm_tree->ref_count = 1;
	return lpm_tree;

err_left_struct_set:
	mlxsw_sp_lpm_tree_free(mlxsw_sp, lpm_tree);
	return ERR_PTR(err);
}

static void mlxsw_sp_lpm_tree_destroy(struct mlxsw_sp *mlxsw_sp,
				      struct mlxsw_sp_lpm_tree *lpm_tree)
{
	mlxsw_sp_lpm_tree_free(mlxsw_sp, lpm_tree);
}

static struct mlxsw_sp_lpm_tree *
mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp,
		      struct mlxsw_sp_prefix_usage *prefix_usage,
		      enum mlxsw_sp_l3proto proto)
{
	struct mlxsw_sp_lpm_tree *lpm_tree;
	int i;

	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
		if (lpm_tree->ref_count != 0 &&
		    lpm_tree->proto == proto &&
		    mlxsw_sp_prefix_usage_eq(&lpm_tree->prefix_usage,
					     prefix_usage)) {
			mlxsw_sp_lpm_tree_hold(lpm_tree);
			return lpm_tree;
		}
	}
	return mlxsw_sp_lpm_tree_create(mlxsw_sp, prefix_usage, proto);
}

static void mlxsw_sp_lpm_tree_hold(struct mlxsw_sp_lpm_tree *lpm_tree)
{
	lpm_tree->ref_count++;
}

static void mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_lpm_tree *lpm_tree)
{
	if (--lpm_tree->ref_count == 0)
		mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree);
}

#define MLXSW_SP_LPM_TREE_MIN 1 /* tree 0 is reserved */

static int mlxsw_sp_lpm_init(struct mlxsw_sp *mlxsw_sp)
{
	struct mlxsw_sp_prefix_usage req_prefix_usage = {{ 0 } };
	struct mlxsw_sp_lpm_tree *lpm_tree;
	u64 max_trees;
	int err, i;

	if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, MAX_LPM_TREES))
		return -EIO;

	max_trees = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_LPM_TREES);
	mlxsw_sp->router->lpm.tree_count = max_trees - MLXSW_SP_LPM_TREE_MIN;
	mlxsw_sp->router->lpm.trees = kcalloc(mlxsw_sp->router->lpm.tree_count,
					     sizeof(struct mlxsw_sp_lpm_tree),
					     GFP_KERNEL);
	if (!mlxsw_sp->router->lpm.trees)
		return -ENOMEM;

	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
		lpm_tree->id = i + MLXSW_SP_LPM_TREE_MIN;
	}

	lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, &req_prefix_usage,
					 MLXSW_SP_L3_PROTO_IPV4);
	if (IS_ERR(lpm_tree)) {
		err = PTR_ERR(lpm_tree);
		goto err_ipv4_tree_get;
	}
	mlxsw_sp->router->lpm.proto_trees[MLXSW_SP_L3_PROTO_IPV4] = lpm_tree;

	lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, &req_prefix_usage,
					 MLXSW_SP_L3_PROTO_IPV6);
	if (IS_ERR(lpm_tree)) {
		err = PTR_ERR(lpm_tree);
		goto err_ipv6_tree_get;
	}
	mlxsw_sp->router->lpm.proto_trees[MLXSW_SP_L3_PROTO_IPV6] = lpm_tree;

	return 0;

err_ipv6_tree_get:
	lpm_tree = mlxsw_sp->router->lpm.proto_trees[MLXSW_SP_L3_PROTO_IPV4];
	mlxsw_sp_lpm_tree_put(mlxsw_sp, lpm_tree);
err_ipv4_tree_get:
	kfree(mlxsw_sp->router->lpm.trees);
	return err;
}

static void mlxsw_sp_lpm_fini(struct mlxsw_sp *mlxsw_sp)
{
	struct mlxsw_sp_lpm_tree *lpm_tree;

	lpm_tree = mlxsw_sp->router->lpm.proto_trees[MLXSW_SP_L3_PROTO_IPV6];
	mlxsw_sp_lpm_tree_put(mlxsw_sp, lpm_tree);

	lpm_tree = mlxsw_sp->router->lpm.proto_trees[MLXSW_SP_L3_PROTO_IPV4];
	mlxsw_sp_lpm_tree_put(mlxsw_sp, lpm_tree);

	kfree(mlxsw_sp->router->lpm.trees);
}

static bool mlxsw_sp_vr_is_used(const struct mlxsw_sp_vr *vr)
{
	return !!vr->fib4 || !!vr->fib6 ||
	       !!vr->mr_table[MLXSW_SP_L3_PROTO_IPV4] ||
	       !!vr->mr_table[MLXSW_SP_L3_PROTO_IPV6];
}

static struct mlxsw_sp_vr *mlxsw_sp_vr_find_unused(struct mlxsw_sp *mlxsw_sp)
{
	struct mlxsw_sp_vr *vr;
	int i;

	for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) {
		vr = &mlxsw_sp->router->vrs[i];
		if (!mlxsw_sp_vr_is_used(vr))
			return vr;
	}
	return NULL;
}

static int mlxsw_sp_vr_lpm_tree_bind(struct mlxsw_sp *mlxsw_sp,
				     const struct mlxsw_sp_fib *fib, u8 tree_id)
{
	char raltb_pl[MLXSW_REG_RALTB_LEN];

	mlxsw_reg_raltb_pack(raltb_pl, fib->vr->id,
			     (enum mlxsw_reg_ralxx_protocol) fib->proto,
			     tree_id);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb), raltb_pl);
}

static int mlxsw_sp_vr_lpm_tree_unbind(struct mlxsw_sp *mlxsw_sp,
				       const struct mlxsw_sp_fib *fib)
{
	char raltb_pl[MLXSW_REG_RALTB_LEN];

	/* Bind to tree 0 which is default */
	mlxsw_reg_raltb_pack(raltb_pl, fib->vr->id,
			     (enum mlxsw_reg_ralxx_protocol) fib->proto, 0);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb), raltb_pl);
}

static u32 mlxsw_sp_fix_tb_id(u32 tb_id)
{
	/* For our purpose, squash main, default and local tables into one */
	if (tb_id == RT_TABLE_LOCAL || tb_id == RT_TABLE_DEFAULT)
		tb_id = RT_TABLE_MAIN;
	return tb_id;
}

static struct mlxsw_sp_vr *mlxsw_sp_vr_find(struct mlxsw_sp *mlxsw_sp,
					    u32 tb_id)
{
	struct mlxsw_sp_vr *vr;
	int i;

	tb_id = mlxsw_sp_fix_tb_id(tb_id);

	for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) {
		vr = &mlxsw_sp->router->vrs[i];
		if (mlxsw_sp_vr_is_used(vr) && vr->tb_id == tb_id)
			return vr;
	}
	return NULL;
}

static struct mlxsw_sp_fib *mlxsw_sp_vr_fib(const struct mlxsw_sp_vr *vr,
					    enum mlxsw_sp_l3proto proto)
{
	switch (proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		return vr->fib4;
	case MLXSW_SP_L3_PROTO_IPV6:
		return vr->fib6;
	}
	return NULL;
}

static struct mlxsw_sp_vr *mlxsw_sp_vr_create(struct mlxsw_sp *mlxsw_sp,
					      u32 tb_id,
					      struct netlink_ext_ack *extack)
{
	struct mlxsw_sp_mr_table *mr4_table, *mr6_table;
	struct mlxsw_sp_fib *fib4;
	struct mlxsw_sp_fib *fib6;
	struct mlxsw_sp_vr *vr;
	int err;

	vr = mlxsw_sp_vr_find_unused(mlxsw_sp);
	if (!vr) {
		NL_SET_ERR_MSG_MOD(extack, "Exceeded number of supported virtual routers");
		return ERR_PTR(-EBUSY);
	}
	fib4 = mlxsw_sp_fib_create(mlxsw_sp, vr, MLXSW_SP_L3_PROTO_IPV4);
	if (IS_ERR(fib4))
		return ERR_CAST(fib4);
	fib6 = mlxsw_sp_fib_create(mlxsw_sp, vr, MLXSW_SP_L3_PROTO_IPV6);
	if (IS_ERR(fib6)) {
		err = PTR_ERR(fib6);
		goto err_fib6_create;
	}
	mr4_table = mlxsw_sp_mr_table_create(mlxsw_sp, vr->id,
					     MLXSW_SP_L3_PROTO_IPV4);
	if (IS_ERR(mr4_table)) {
		err = PTR_ERR(mr4_table);
		goto err_mr4_table_create;
	}
	mr6_table = mlxsw_sp_mr_table_create(mlxsw_sp, vr->id,
					     MLXSW_SP_L3_PROTO_IPV6);
	if (IS_ERR(mr6_table)) {
		err = PTR_ERR(mr6_table);
		goto err_mr6_table_create;
	}

	vr->fib4 = fib4;
	vr->fib6 = fib6;
	vr->mr_table[MLXSW_SP_L3_PROTO_IPV4] = mr4_table;
	vr->mr_table[MLXSW_SP_L3_PROTO_IPV6] = mr6_table;
	vr->tb_id = tb_id;
	return vr;

err_mr6_table_create:
	mlxsw_sp_mr_table_destroy(mr4_table);
err_mr4_table_create:
	mlxsw_sp_fib_destroy(mlxsw_sp, fib6);
err_fib6_create:
	mlxsw_sp_fib_destroy(mlxsw_sp, fib4);
	return ERR_PTR(err);
}

static void mlxsw_sp_vr_destroy(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_vr *vr)
{
	mlxsw_sp_mr_table_destroy(vr->mr_table[MLXSW_SP_L3_PROTO_IPV6]);
	vr->mr_table[MLXSW_SP_L3_PROTO_IPV6] = NULL;
	mlxsw_sp_mr_table_destroy(vr->mr_table[MLXSW_SP_L3_PROTO_IPV4]);
	vr->mr_table[MLXSW_SP_L3_PROTO_IPV4] = NULL;
	mlxsw_sp_fib_destroy(mlxsw_sp, vr->fib6);
	vr->fib6 = NULL;
	mlxsw_sp_fib_destroy(mlxsw_sp, vr->fib4);
	vr->fib4 = NULL;
}

static struct mlxsw_sp_vr *mlxsw_sp_vr_get(struct mlxsw_sp *mlxsw_sp, u32 tb_id,
					   struct netlink_ext_ack *extack)
{
	struct mlxsw_sp_vr *vr;

	tb_id = mlxsw_sp_fix_tb_id(tb_id);
	vr = mlxsw_sp_vr_find(mlxsw_sp, tb_id);
	if (!vr)
		vr = mlxsw_sp_vr_create(mlxsw_sp, tb_id, extack);
	return vr;
}

static void mlxsw_sp_vr_put(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_vr *vr)
{
	if (!vr->rif_count && list_empty(&vr->fib4->node_list) &&
	    list_empty(&vr->fib6->node_list) &&
	    mlxsw_sp_mr_table_empty(vr->mr_table[MLXSW_SP_L3_PROTO_IPV4]) &&
	    mlxsw_sp_mr_table_empty(vr->mr_table[MLXSW_SP_L3_PROTO_IPV6]))
		mlxsw_sp_vr_destroy(mlxsw_sp, vr);
}

static bool
mlxsw_sp_vr_lpm_tree_should_replace(struct mlxsw_sp_vr *vr,
				    enum mlxsw_sp_l3proto proto, u8 tree_id)
{
	struct mlxsw_sp_fib *fib = mlxsw_sp_vr_fib(vr, proto);

	if (!mlxsw_sp_vr_is_used(vr))
		return false;
	if (fib->lpm_tree->id == tree_id)
		return true;
	return false;
}

static int mlxsw_sp_vr_lpm_tree_replace(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib *fib,
					struct mlxsw_sp_lpm_tree *new_tree)
{
	struct mlxsw_sp_lpm_tree *old_tree = fib->lpm_tree;
	int err;

	fib->lpm_tree = new_tree;
	mlxsw_sp_lpm_tree_hold(new_tree);
	err = mlxsw_sp_vr_lpm_tree_bind(mlxsw_sp, fib, new_tree->id);
	if (err)
		goto err_tree_bind;
	mlxsw_sp_lpm_tree_put(mlxsw_sp, old_tree);
	return 0;

err_tree_bind:
	mlxsw_sp_lpm_tree_put(mlxsw_sp, new_tree);
	fib->lpm_tree = old_tree;
	return err;
}

static int mlxsw_sp_vrs_lpm_tree_replace(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_fib *fib,
					 struct mlxsw_sp_lpm_tree *new_tree)
{
	enum mlxsw_sp_l3proto proto = fib->proto;
	struct mlxsw_sp_lpm_tree *old_tree;
	u8 old_id, new_id = new_tree->id;
	struct mlxsw_sp_vr *vr;
	int i, err;

	old_tree = mlxsw_sp->router->lpm.proto_trees[proto];
	old_id = old_tree->id;

	for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) {
		vr = &mlxsw_sp->router->vrs[i];
		if (!mlxsw_sp_vr_lpm_tree_should_replace(vr, proto, old_id))
			continue;
		err = mlxsw_sp_vr_lpm_tree_replace(mlxsw_sp,
						   mlxsw_sp_vr_fib(vr, proto),
						   new_tree);
		if (err)
			goto err_tree_replace;
	}

	memcpy(new_tree->prefix_ref_count, old_tree->prefix_ref_count,
	       sizeof(new_tree->prefix_ref_count));
	mlxsw_sp->router->lpm.proto_trees[proto] = new_tree;
	mlxsw_sp_lpm_tree_put(mlxsw_sp, old_tree);

	return 0;

err_tree_replace:
	for (i--; i >= 0; i--) {
		if (!mlxsw_sp_vr_lpm_tree_should_replace(vr, proto, new_id))
			continue;
		mlxsw_sp_vr_lpm_tree_replace(mlxsw_sp,
					     mlxsw_sp_vr_fib(vr, proto),
					     old_tree);
	}
	return err;
}

static int mlxsw_sp_vrs_init(struct mlxsw_sp *mlxsw_sp)
{
	struct mlxsw_sp_vr *vr;
	u64 max_vrs;
	int i;

	if (!MLXSW_CORE_RES_VALID(mlxsw_sp->core, MAX_VRS))
		return -EIO;

	max_vrs = MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS);
	mlxsw_sp->router->vrs = kcalloc(max_vrs, sizeof(struct mlxsw_sp_vr),
					GFP_KERNEL);
	if (!mlxsw_sp->router->vrs)
		return -ENOMEM;

	for (i = 0; i < max_vrs; i++) {
		vr = &mlxsw_sp->router->vrs[i];
		vr->id = i;
	}

	return 0;
}

static void mlxsw_sp_router_fib_flush(struct mlxsw_sp *mlxsw_sp);

static void mlxsw_sp_vrs_fini(struct mlxsw_sp *mlxsw_sp)
{
	/* At this stage we're guaranteed not to have new incoming
	 * FIB notifications and the work queue is free from FIBs
	 * sitting on top of mlxsw netdevs. However, we can still
	 * have other FIBs queued. Flush the queue before flushing
	 * the device's tables. No need for locks, as we're the only
	 * writer.
	 */
	mlxsw_core_flush_owq();
	mlxsw_sp_router_fib_flush(mlxsw_sp);
	kfree(mlxsw_sp->router->vrs);
}

static struct net_device *
__mlxsw_sp_ipip_netdev_ul_dev_get(const struct net_device *ol_dev)
{
	struct ip_tunnel *tun = netdev_priv(ol_dev);
	struct net *net = dev_net(ol_dev);

	return __dev_get_by_index(net, tun->parms.link);
}

u32 mlxsw_sp_ipip_dev_ul_tb_id(const struct net_device *ol_dev)
{
	struct net_device *d = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev);

	if (d)
		return l3mdev_fib_table(d) ? : RT_TABLE_MAIN;
	else
		return l3mdev_fib_table(ol_dev) ? : RT_TABLE_MAIN;
}

static struct mlxsw_sp_rif *
mlxsw_sp_rif_create(struct mlxsw_sp *mlxsw_sp,
		    const struct mlxsw_sp_rif_params *params,
		    struct netlink_ext_ack *extack);

static struct mlxsw_sp_rif_ipip_lb *
mlxsw_sp_ipip_ol_ipip_lb_create(struct mlxsw_sp *mlxsw_sp,
				enum mlxsw_sp_ipip_type ipipt,
				struct net_device *ol_dev,
				struct netlink_ext_ack *extack)
{
	struct mlxsw_sp_rif_params_ipip_lb lb_params;
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	struct mlxsw_sp_rif *rif;

	ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipipt];
	lb_params = (struct mlxsw_sp_rif_params_ipip_lb) {
		.common.dev = ol_dev,
		.common.lag = false,
		.lb_config = ipip_ops->ol_loopback_config(mlxsw_sp, ol_dev),
	};

	rif = mlxsw_sp_rif_create(mlxsw_sp, &lb_params.common, extack);
	if (IS_ERR(rif))
		return ERR_CAST(rif);
	return container_of(rif, struct mlxsw_sp_rif_ipip_lb, common);
}

static struct mlxsw_sp_ipip_entry *
mlxsw_sp_ipip_entry_alloc(struct mlxsw_sp *mlxsw_sp,
			  enum mlxsw_sp_ipip_type ipipt,
			  struct net_device *ol_dev)
{
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	struct mlxsw_sp_ipip_entry *ipip_entry;
	struct mlxsw_sp_ipip_entry *ret = NULL;

	ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipipt];
	ipip_entry = kzalloc(sizeof(*ipip_entry), GFP_KERNEL);
	if (!ipip_entry)
		return ERR_PTR(-ENOMEM);

	ipip_entry->ol_lb = mlxsw_sp_ipip_ol_ipip_lb_create(mlxsw_sp, ipipt,
							    ol_dev, NULL);
	if (IS_ERR(ipip_entry->ol_lb)) {
		ret = ERR_CAST(ipip_entry->ol_lb);
		goto err_ol_ipip_lb_create;
	}

	ipip_entry->ipipt = ipipt;
	ipip_entry->ol_dev = ol_dev;

	switch (ipip_ops->ul_proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		ipip_entry->parms4 = mlxsw_sp_ipip_netdev_parms4(ol_dev);
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		WARN_ON(1);
		break;
	}

	return ipip_entry;

err_ol_ipip_lb_create:
	kfree(ipip_entry);
	return ret;
}

static void
mlxsw_sp_ipip_entry_dealloc(struct mlxsw_sp_ipip_entry *ipip_entry)
{
	mlxsw_sp_rif_destroy(&ipip_entry->ol_lb->common);
	kfree(ipip_entry);
}

static bool
mlxsw_sp_ipip_entry_saddr_matches(struct mlxsw_sp *mlxsw_sp,
				  const enum mlxsw_sp_l3proto ul_proto,
				  union mlxsw_sp_l3addr saddr,
				  u32 ul_tb_id,
				  struct mlxsw_sp_ipip_entry *ipip_entry)
{
	u32 tun_ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(ipip_entry->ol_dev);
	enum mlxsw_sp_ipip_type ipipt = ipip_entry->ipipt;
	union mlxsw_sp_l3addr tun_saddr;

	if (mlxsw_sp->router->ipip_ops_arr[ipipt]->ul_proto != ul_proto)
		return false;

	tun_saddr = mlxsw_sp_ipip_netdev_saddr(ul_proto, ipip_entry->ol_dev);
	return tun_ul_tb_id == ul_tb_id &&
	       mlxsw_sp_l3addr_eq(&tun_saddr, &saddr);
}

static int
mlxsw_sp_fib_entry_decap_init(struct mlxsw_sp *mlxsw_sp,
			      struct mlxsw_sp_fib_entry *fib_entry,
			      struct mlxsw_sp_ipip_entry *ipip_entry)
{
	u32 tunnel_index;
	int err;

	err = mlxsw_sp_kvdl_alloc(mlxsw_sp, 1, &tunnel_index);
	if (err)
		return err;

	ipip_entry->decap_fib_entry = fib_entry;
	fib_entry->decap.ipip_entry = ipip_entry;
	fib_entry->decap.tunnel_index = tunnel_index;
	return 0;
}

static void mlxsw_sp_fib_entry_decap_fini(struct mlxsw_sp *mlxsw_sp,
					  struct mlxsw_sp_fib_entry *fib_entry)
{
	/* Unlink this node from the IPIP entry that it's the decap entry of. */
	fib_entry->decap.ipip_entry->decap_fib_entry = NULL;
	fib_entry->decap.ipip_entry = NULL;
	mlxsw_sp_kvdl_free(mlxsw_sp, fib_entry->decap.tunnel_index);
}

static struct mlxsw_sp_fib_node *
mlxsw_sp_fib_node_lookup(struct mlxsw_sp_fib *fib, const void *addr,
			 size_t addr_len, unsigned char prefix_len);
static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp,
				     struct mlxsw_sp_fib_entry *fib_entry);

static void
mlxsw_sp_ipip_entry_demote_decap(struct mlxsw_sp *mlxsw_sp,
				 struct mlxsw_sp_ipip_entry *ipip_entry)
{
	struct mlxsw_sp_fib_entry *fib_entry = ipip_entry->decap_fib_entry;

	mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, fib_entry);
	fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP;

	mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
}

static void
mlxsw_sp_ipip_entry_promote_decap(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_ipip_entry *ipip_entry,
				  struct mlxsw_sp_fib_entry *decap_fib_entry)
{
	if (mlxsw_sp_fib_entry_decap_init(mlxsw_sp, decap_fib_entry,
					  ipip_entry))
		return;
	decap_fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP;

	if (mlxsw_sp_fib_entry_update(mlxsw_sp, decap_fib_entry))
		mlxsw_sp_ipip_entry_demote_decap(mlxsw_sp, ipip_entry);
}

/* Given an IPIP entry, find the corresponding decap route. */
static struct mlxsw_sp_fib_entry *
mlxsw_sp_ipip_entry_find_decap(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_ipip_entry *ipip_entry)
{
	static struct mlxsw_sp_fib_node *fib_node;
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	struct mlxsw_sp_fib_entry *fib_entry;
	unsigned char saddr_prefix_len;
	union mlxsw_sp_l3addr saddr;
	struct mlxsw_sp_fib *ul_fib;
	struct mlxsw_sp_vr *ul_vr;
	const void *saddrp;
	size_t saddr_len;
	u32 ul_tb_id;
	u32 saddr4;

	ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipip_entry->ipipt];

	ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(ipip_entry->ol_dev);
	ul_vr = mlxsw_sp_vr_find(mlxsw_sp, ul_tb_id);
	if (!ul_vr)
		return NULL;

	ul_fib = mlxsw_sp_vr_fib(ul_vr, ipip_ops->ul_proto);
	saddr = mlxsw_sp_ipip_netdev_saddr(ipip_ops->ul_proto,
					   ipip_entry->ol_dev);

	switch (ipip_ops->ul_proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		saddr4 = be32_to_cpu(saddr.addr4);
		saddrp = &saddr4;
		saddr_len = 4;
		saddr_prefix_len = 32;
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		WARN_ON(1);
		return NULL;
	}

	fib_node = mlxsw_sp_fib_node_lookup(ul_fib, saddrp, saddr_len,
					    saddr_prefix_len);
	if (!fib_node || list_empty(&fib_node->entry_list))
		return NULL;

	fib_entry = list_first_entry(&fib_node->entry_list,
				     struct mlxsw_sp_fib_entry, list);
	if (fib_entry->type != MLXSW_SP_FIB_ENTRY_TYPE_TRAP)
		return NULL;

	return fib_entry;
}

static struct mlxsw_sp_ipip_entry *
mlxsw_sp_ipip_entry_create(struct mlxsw_sp *mlxsw_sp,
			   enum mlxsw_sp_ipip_type ipipt,
			   struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	ipip_entry = mlxsw_sp_ipip_entry_alloc(mlxsw_sp, ipipt, ol_dev);
	if (IS_ERR(ipip_entry))
		return ipip_entry;

	list_add_tail(&ipip_entry->ipip_list_node,
		      &mlxsw_sp->router->ipip_list);

	return ipip_entry;
}

static void
mlxsw_sp_ipip_entry_destroy(struct mlxsw_sp *mlxsw_sp,
			    struct mlxsw_sp_ipip_entry *ipip_entry)
{
	list_del(&ipip_entry->ipip_list_node);
	mlxsw_sp_ipip_entry_dealloc(ipip_entry);
}

static bool
mlxsw_sp_ipip_entry_matches_decap(struct mlxsw_sp *mlxsw_sp,
				  const struct net_device *ul_dev,
				  enum mlxsw_sp_l3proto ul_proto,
				  union mlxsw_sp_l3addr ul_dip,
				  struct mlxsw_sp_ipip_entry *ipip_entry)
{
	u32 ul_tb_id = l3mdev_fib_table(ul_dev) ? : RT_TABLE_MAIN;
	enum mlxsw_sp_ipip_type ipipt = ipip_entry->ipipt;
	struct net_device *ipip_ul_dev;

	if (mlxsw_sp->router->ipip_ops_arr[ipipt]->ul_proto != ul_proto)
		return false;

	ipip_ul_dev = __mlxsw_sp_ipip_netdev_ul_dev_get(ipip_entry->ol_dev);
	return mlxsw_sp_ipip_entry_saddr_matches(mlxsw_sp, ul_proto, ul_dip,
						 ul_tb_id, ipip_entry) &&
	       (!ipip_ul_dev || ipip_ul_dev == ul_dev);
}

/* Given decap parameters, find the corresponding IPIP entry. */
static struct mlxsw_sp_ipip_entry *
mlxsw_sp_ipip_entry_find_by_decap(struct mlxsw_sp *mlxsw_sp,
				  const struct net_device *ul_dev,
				  enum mlxsw_sp_l3proto ul_proto,
				  union mlxsw_sp_l3addr ul_dip)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	list_for_each_entry(ipip_entry, &mlxsw_sp->router->ipip_list,
			    ipip_list_node)
		if (mlxsw_sp_ipip_entry_matches_decap(mlxsw_sp, ul_dev,
						      ul_proto, ul_dip,
						      ipip_entry))
			return ipip_entry;

	return NULL;
}

static bool mlxsw_sp_netdev_ipip_type(const struct mlxsw_sp *mlxsw_sp,
				      const struct net_device *dev,
				      enum mlxsw_sp_ipip_type *p_type)
{
	struct mlxsw_sp_router *router = mlxsw_sp->router;
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	enum mlxsw_sp_ipip_type ipipt;

	for (ipipt = 0; ipipt < MLXSW_SP_IPIP_TYPE_MAX; ++ipipt) {
		ipip_ops = router->ipip_ops_arr[ipipt];
		if (dev->type == ipip_ops->dev_type) {
			if (p_type)
				*p_type = ipipt;
			return true;
		}
	}
	return false;
}

bool mlxsw_sp_netdev_is_ipip_ol(const struct mlxsw_sp *mlxsw_sp,
				const struct net_device *dev)
{
	return mlxsw_sp_netdev_ipip_type(mlxsw_sp, dev, NULL);
}

static struct mlxsw_sp_ipip_entry *
mlxsw_sp_ipip_entry_find_by_ol_dev(struct mlxsw_sp *mlxsw_sp,
				   const struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	list_for_each_entry(ipip_entry, &mlxsw_sp->router->ipip_list,
			    ipip_list_node)
		if (ipip_entry->ol_dev == ol_dev)
			return ipip_entry;

	return NULL;
}

static struct mlxsw_sp_ipip_entry *
mlxsw_sp_ipip_entry_find_by_ul_dev(const struct mlxsw_sp *mlxsw_sp,
				   const struct net_device *ul_dev,
				   struct mlxsw_sp_ipip_entry *start)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	ipip_entry = list_prepare_entry(start, &mlxsw_sp->router->ipip_list,
					ipip_list_node);
	list_for_each_entry_continue(ipip_entry, &mlxsw_sp->router->ipip_list,
				     ipip_list_node) {
		struct net_device *ipip_ul_dev =
			__mlxsw_sp_ipip_netdev_ul_dev_get(ipip_entry->ol_dev);

		if (ipip_ul_dev == ul_dev)
			return ipip_entry;
	}

	return NULL;
}

bool mlxsw_sp_netdev_is_ipip_ul(const struct mlxsw_sp *mlxsw_sp,
				const struct net_device *dev)
{
	return mlxsw_sp_ipip_entry_find_by_ul_dev(mlxsw_sp, dev, NULL);
}

static bool mlxsw_sp_netdevice_ipip_can_offload(struct mlxsw_sp *mlxsw_sp,
						const struct net_device *ol_dev,
						enum mlxsw_sp_ipip_type ipipt)
{
	const struct mlxsw_sp_ipip_ops *ops
		= mlxsw_sp->router->ipip_ops_arr[ipipt];

	/* For deciding whether decap should be offloaded, we don't care about
	 * overlay protocol, so ask whether either one is supported.
	 */
	return ops->can_offload(mlxsw_sp, ol_dev, MLXSW_SP_L3_PROTO_IPV4) ||
	       ops->can_offload(mlxsw_sp, ol_dev, MLXSW_SP_L3_PROTO_IPV6);
}

static int mlxsw_sp_netdevice_ipip_ol_reg_event(struct mlxsw_sp *mlxsw_sp,
						struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;
	enum mlxsw_sp_l3proto ul_proto;
	enum mlxsw_sp_ipip_type ipipt;
	union mlxsw_sp_l3addr saddr;
	u32 ul_tb_id;

	mlxsw_sp_netdev_ipip_type(mlxsw_sp, ol_dev, &ipipt);
	if (mlxsw_sp_netdevice_ipip_can_offload(mlxsw_sp, ol_dev, ipipt)) {
		ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(ol_dev);
		ul_proto = mlxsw_sp->router->ipip_ops_arr[ipipt]->ul_proto;
		saddr = mlxsw_sp_ipip_netdev_saddr(ul_proto, ol_dev);
		if (!mlxsw_sp_ipip_demote_tunnel_by_saddr(mlxsw_sp, ul_proto,
							  saddr, ul_tb_id,
							  NULL)) {
			ipip_entry = mlxsw_sp_ipip_entry_create(mlxsw_sp, ipipt,
								ol_dev);
			if (IS_ERR(ipip_entry))
				return PTR_ERR(ipip_entry);
		}
	}

	return 0;
}

static void mlxsw_sp_netdevice_ipip_ol_unreg_event(struct mlxsw_sp *mlxsw_sp,
						   struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, ol_dev);
	if (ipip_entry)
		mlxsw_sp_ipip_entry_destroy(mlxsw_sp, ipip_entry);
}

static void
mlxsw_sp_ipip_entry_ol_up_event(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_ipip_entry *ipip_entry)
{
	struct mlxsw_sp_fib_entry *decap_fib_entry;

	decap_fib_entry = mlxsw_sp_ipip_entry_find_decap(mlxsw_sp, ipip_entry);
	if (decap_fib_entry)
		mlxsw_sp_ipip_entry_promote_decap(mlxsw_sp, ipip_entry,
						  decap_fib_entry);
}

static int
mlxsw_sp_rif_ipip_lb_op(struct mlxsw_sp_rif_ipip_lb *lb_rif,
			struct mlxsw_sp_vr *ul_vr, bool enable)
{
	struct mlxsw_sp_rif_ipip_lb_config lb_cf = lb_rif->lb_config;
	struct mlxsw_sp_rif *rif = &lb_rif->common;
	struct mlxsw_sp *mlxsw_sp = rif->mlxsw_sp;
	char ritr_pl[MLXSW_REG_RITR_LEN];
	u32 saddr4;

	switch (lb_cf.ul_protocol) {
	case MLXSW_SP_L3_PROTO_IPV4:
		saddr4 = be32_to_cpu(lb_cf.saddr.addr4);
		mlxsw_reg_ritr_pack(ritr_pl, enable, MLXSW_REG_RITR_LOOPBACK_IF,
				    rif->rif_index, rif->vr_id, rif->dev->mtu);
		mlxsw_reg_ritr_loopback_ipip4_pack(ritr_pl, lb_cf.lb_ipipt,
			    MLXSW_REG_RITR_LOOPBACK_IPIP_OPTIONS_GRE_KEY_PRESET,
			    ul_vr->id, saddr4, lb_cf.okey);
		break;

	case MLXSW_SP_L3_PROTO_IPV6:
		return -EAFNOSUPPORT;
	}

	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ritr), ritr_pl);
}

static int mlxsw_sp_netdevice_ipip_ol_update_mtu(struct mlxsw_sp *mlxsw_sp,
						 struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;
	struct mlxsw_sp_rif_ipip_lb *lb_rif;
	struct mlxsw_sp_vr *ul_vr;
	int err = 0;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, ol_dev);
	if (ipip_entry) {
		lb_rif = ipip_entry->ol_lb;
		ul_vr = &mlxsw_sp->router->vrs[lb_rif->ul_vr_id];
		err = mlxsw_sp_rif_ipip_lb_op(lb_rif, ul_vr, true);
		if (err)
			goto out;
		lb_rif->common.mtu = ol_dev->mtu;
	}

out:
	return err;
}

static void mlxsw_sp_netdevice_ipip_ol_up_event(struct mlxsw_sp *mlxsw_sp,
						struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, ol_dev);
	if (ipip_entry)
		mlxsw_sp_ipip_entry_ol_up_event(mlxsw_sp, ipip_entry);
}

static void
mlxsw_sp_ipip_entry_ol_down_event(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_ipip_entry *ipip_entry)
{
	if (ipip_entry->decap_fib_entry)
		mlxsw_sp_ipip_entry_demote_decap(mlxsw_sp, ipip_entry);
}

static void mlxsw_sp_netdevice_ipip_ol_down_event(struct mlxsw_sp *mlxsw_sp,
						  struct net_device *ol_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, ol_dev);
	if (ipip_entry)
		mlxsw_sp_ipip_entry_ol_down_event(mlxsw_sp, ipip_entry);
}

static void mlxsw_sp_nexthop_rif_migrate(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_rif *old_rif,
					 struct mlxsw_sp_rif *new_rif);
static int
mlxsw_sp_ipip_entry_ol_lb_update(struct mlxsw_sp *mlxsw_sp,
				 struct mlxsw_sp_ipip_entry *ipip_entry,
				 bool keep_encap,
				 struct netlink_ext_ack *extack)
{
	struct mlxsw_sp_rif_ipip_lb *old_lb_rif = ipip_entry->ol_lb;
	struct mlxsw_sp_rif_ipip_lb *new_lb_rif;

	new_lb_rif = mlxsw_sp_ipip_ol_ipip_lb_create(mlxsw_sp,
						     ipip_entry->ipipt,
						     ipip_entry->ol_dev,
						     extack);
	if (IS_ERR(new_lb_rif))
		return PTR_ERR(new_lb_rif);
	ipip_entry->ol_lb = new_lb_rif;

	if (keep_encap)
		mlxsw_sp_nexthop_rif_migrate(mlxsw_sp, &old_lb_rif->common,
					     &new_lb_rif->common);

	mlxsw_sp_rif_destroy(&old_lb_rif->common);

	return 0;
}

static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_rif *rif);

/**
 * Update the offload related to an IPIP entry. This always updates decap, and
 * in addition to that it also:
 * @recreate_loopback: recreates the associated loopback RIF
 * @keep_encap: updates next hops that use the tunnel netdevice. This is only
 *              relevant when recreate_loopback is true.
 * @update_nexthops: updates next hops, keeping the current loopback RIF. This
 *                   is only relevant when recreate_loopback is false.
 */
int __mlxsw_sp_ipip_entry_update_tunnel(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_ipip_entry *ipip_entry,
					bool recreate_loopback,
					bool keep_encap,
					bool update_nexthops,
					struct netlink_ext_ack *extack)
{
	int err;

	/* RIFs can't be edited, so to update loopback, we need to destroy and
	 * recreate it. That creates a window of opportunity where RALUE and
	 * RATR registers end up referencing a RIF that's already gone. RATRs
	 * are handled in mlxsw_sp_ipip_entry_ol_lb_update(), and to take care
	 * of RALUE, demote the decap route back.
	 */
	if (ipip_entry->decap_fib_entry)
		mlxsw_sp_ipip_entry_demote_decap(mlxsw_sp, ipip_entry);

	if (recreate_loopback) {
		err = mlxsw_sp_ipip_entry_ol_lb_update(mlxsw_sp, ipip_entry,
						       keep_encap, extack);
		if (err)
			return err;
	} else if (update_nexthops) {
		mlxsw_sp_nexthop_rif_update(mlxsw_sp,
					    &ipip_entry->ol_lb->common);
	}

	if (ipip_entry->ol_dev->flags & IFF_UP)
		mlxsw_sp_ipip_entry_ol_up_event(mlxsw_sp, ipip_entry);

	return 0;
}

static int mlxsw_sp_netdevice_ipip_ol_vrf_event(struct mlxsw_sp *mlxsw_sp,
						struct net_device *ol_dev,
						struct netlink_ext_ack *extack)
{
	struct mlxsw_sp_ipip_entry *ipip_entry =
		mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, ol_dev);
	enum mlxsw_sp_l3proto ul_proto;
	union mlxsw_sp_l3addr saddr;
	u32 ul_tb_id;

	if (!ipip_entry)
		return 0;

	/* For flat configuration cases, moving overlay to a different VRF might
	 * cause local address conflict, and the conflicting tunnels need to be
	 * demoted.
	 */
	ul_tb_id = mlxsw_sp_ipip_dev_ul_tb_id(ol_dev);
	ul_proto = mlxsw_sp->router->ipip_ops_arr[ipip_entry->ipipt]->ul_proto;
	saddr = mlxsw_sp_ipip_netdev_saddr(ul_proto, ol_dev);
	if (mlxsw_sp_ipip_demote_tunnel_by_saddr(mlxsw_sp, ul_proto,
						 saddr, ul_tb_id,
						 ipip_entry)) {
		mlxsw_sp_ipip_entry_demote_tunnel(mlxsw_sp, ipip_entry);
		return 0;
	}

	return __mlxsw_sp_ipip_entry_update_tunnel(mlxsw_sp, ipip_entry,
						   true, false, false, extack);
}

static int
mlxsw_sp_netdevice_ipip_ul_vrf_event(struct mlxsw_sp *mlxsw_sp,
				     struct mlxsw_sp_ipip_entry *ipip_entry,
				     struct net_device *ul_dev,
				     struct netlink_ext_ack *extack)
{
	return __mlxsw_sp_ipip_entry_update_tunnel(mlxsw_sp, ipip_entry,
						   true, true, false, extack);
}

static int
mlxsw_sp_netdevice_ipip_ul_up_event(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_ipip_entry *ipip_entry,
				    struct net_device *ul_dev)
{
	return __mlxsw_sp_ipip_entry_update_tunnel(mlxsw_sp, ipip_entry,
						   false, false, true, NULL);
}

static int
mlxsw_sp_netdevice_ipip_ul_down_event(struct mlxsw_sp *mlxsw_sp,
				      struct mlxsw_sp_ipip_entry *ipip_entry,
				      struct net_device *ul_dev)
{
	/* A down underlay device causes encapsulated packets to not be
	 * forwarded, but decap still works. So refresh next hops without
	 * touching anything else.
	 */
	return __mlxsw_sp_ipip_entry_update_tunnel(mlxsw_sp, ipip_entry,
						   false, false, true, NULL);
}

static int
mlxsw_sp_netdevice_ipip_ol_change_event(struct mlxsw_sp *mlxsw_sp,
					struct net_device *ol_dev,
					struct netlink_ext_ack *extack)
{
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	struct mlxsw_sp_ipip_entry *ipip_entry;
	int err;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, ol_dev);
	if (!ipip_entry)
		/* A change might make a tunnel eligible for offloading, but
		 * that is currently not implemented. What falls to slow path
		 * stays there.
		 */
		return 0;

	/* A change might make a tunnel not eligible for offloading. */
	if (!mlxsw_sp_netdevice_ipip_can_offload(mlxsw_sp, ol_dev,
						 ipip_entry->ipipt)) {
		mlxsw_sp_ipip_entry_demote_tunnel(mlxsw_sp, ipip_entry);
		return 0;
	}

	ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipip_entry->ipipt];
	err = ipip_ops->ol_netdev_change(mlxsw_sp, ipip_entry, extack);
	return err;
}

void mlxsw_sp_ipip_entry_demote_tunnel(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_ipip_entry *ipip_entry)
{
	struct net_device *ol_dev = ipip_entry->ol_dev;

	if (ol_dev->flags & IFF_UP)
		mlxsw_sp_ipip_entry_ol_down_event(mlxsw_sp, ipip_entry);
	mlxsw_sp_ipip_entry_destroy(mlxsw_sp, ipip_entry);
}

/* The configuration where several tunnels have the same local address in the
 * same underlay table needs special treatment in the HW. That is currently not
 * implemented in the driver. This function finds and demotes the first tunnel
 * with a given source address, except the one passed in in the argument
 * `except'.
 */
bool
mlxsw_sp_ipip_demote_tunnel_by_saddr(struct mlxsw_sp *mlxsw_sp,
				     enum mlxsw_sp_l3proto ul_proto,
				     union mlxsw_sp_l3addr saddr,
				     u32 ul_tb_id,
				     const struct mlxsw_sp_ipip_entry *except)
{
	struct mlxsw_sp_ipip_entry *ipip_entry, *tmp;

	list_for_each_entry_safe(ipip_entry, tmp, &mlxsw_sp->router->ipip_list,
				 ipip_list_node) {
		if (ipip_entry != except &&
		    mlxsw_sp_ipip_entry_saddr_matches(mlxsw_sp, ul_proto, saddr,
						      ul_tb_id, ipip_entry)) {
			mlxsw_sp_ipip_entry_demote_tunnel(mlxsw_sp, ipip_entry);
			return true;
		}
	}

	return false;
}

static void mlxsw_sp_ipip_demote_tunnel_by_ul_netdev(struct mlxsw_sp *mlxsw_sp,
						     struct net_device *ul_dev)
{
	struct mlxsw_sp_ipip_entry *ipip_entry, *tmp;

	list_for_each_entry_safe(ipip_entry, tmp, &mlxsw_sp->router->ipip_list,
				 ipip_list_node) {
		struct net_device *ipip_ul_dev =
			__mlxsw_sp_ipip_netdev_ul_dev_get(ipip_entry->ol_dev);

		if (ipip_ul_dev == ul_dev)
			mlxsw_sp_ipip_entry_demote_tunnel(mlxsw_sp, ipip_entry);
	}
}

int mlxsw_sp_netdevice_ipip_ol_event(struct mlxsw_sp *mlxsw_sp,
				     struct net_device *ol_dev,
				     unsigned long event,
				     struct netdev_notifier_info *info)
{
	struct netdev_notifier_changeupper_info *chup;
	struct netlink_ext_ack *extack;

	switch (event) {
	case NETDEV_REGISTER:
		return mlxsw_sp_netdevice_ipip_ol_reg_event(mlxsw_sp, ol_dev);
	case NETDEV_UNREGISTER:
		mlxsw_sp_netdevice_ipip_ol_unreg_event(mlxsw_sp, ol_dev);
		return 0;
	case NETDEV_UP:
		mlxsw_sp_netdevice_ipip_ol_up_event(mlxsw_sp, ol_dev);
		return 0;
	case NETDEV_DOWN:
		mlxsw_sp_netdevice_ipip_ol_down_event(mlxsw_sp, ol_dev);
		return 0;
	case NETDEV_CHANGEUPPER:
		chup = container_of(info, typeof(*chup), info);
		extack = info->extack;
		if (netif_is_l3_master(chup->upper_dev))
			return mlxsw_sp_netdevice_ipip_ol_vrf_event(mlxsw_sp,
								    ol_dev,
								    extack);
		return 0;
	case NETDEV_CHANGE:
		extack = info->extack;
		return mlxsw_sp_netdevice_ipip_ol_change_event(mlxsw_sp,
							       ol_dev, extack);
	case NETDEV_CHANGEMTU:
		return mlxsw_sp_netdevice_ipip_ol_update_mtu(mlxsw_sp, ol_dev);
	}
	return 0;
}

static int
__mlxsw_sp_netdevice_ipip_ul_event(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_ipip_entry *ipip_entry,
				   struct net_device *ul_dev,
				   unsigned long event,
				   struct netdev_notifier_info *info)
{
	struct netdev_notifier_changeupper_info *chup;
	struct netlink_ext_ack *extack;

	switch (event) {
	case NETDEV_CHANGEUPPER:
		chup = container_of(info, typeof(*chup), info);
		extack = info->extack;
		if (netif_is_l3_master(chup->upper_dev))
			return mlxsw_sp_netdevice_ipip_ul_vrf_event(mlxsw_sp,
								    ipip_entry,
								    ul_dev,
								    extack);
		break;

	case NETDEV_UP:
		return mlxsw_sp_netdevice_ipip_ul_up_event(mlxsw_sp, ipip_entry,
							   ul_dev);
	case NETDEV_DOWN:
		return mlxsw_sp_netdevice_ipip_ul_down_event(mlxsw_sp,
							     ipip_entry,
							     ul_dev);
	}
	return 0;
}

int
mlxsw_sp_netdevice_ipip_ul_event(struct mlxsw_sp *mlxsw_sp,
				 struct net_device *ul_dev,
				 unsigned long event,
				 struct netdev_notifier_info *info)
{
	struct mlxsw_sp_ipip_entry *ipip_entry = NULL;
	int err;

	while ((ipip_entry = mlxsw_sp_ipip_entry_find_by_ul_dev(mlxsw_sp,
								ul_dev,
								ipip_entry))) {
		err = __mlxsw_sp_netdevice_ipip_ul_event(mlxsw_sp, ipip_entry,
							 ul_dev, event, info);
		if (err) {
			mlxsw_sp_ipip_demote_tunnel_by_ul_netdev(mlxsw_sp,
								 ul_dev);
			return err;
		}
	}

	return 0;
}

struct mlxsw_sp_neigh_key {
	struct neighbour *n;
};

struct mlxsw_sp_neigh_entry {
	struct list_head rif_list_node;
	struct rhash_head ht_node;
	struct mlxsw_sp_neigh_key key;
	u16 rif;
	bool connected;
	unsigned char ha[ETH_ALEN];
	struct list_head nexthop_list; /* list of nexthops using
					* this neigh entry
					*/
	struct list_head nexthop_neighs_list_node;
	unsigned int counter_index;
	bool counter_valid;
};

static const struct rhashtable_params mlxsw_sp_neigh_ht_params = {
	.key_offset = offsetof(struct mlxsw_sp_neigh_entry, key),
	.head_offset = offsetof(struct mlxsw_sp_neigh_entry, ht_node),
	.key_len = sizeof(struct mlxsw_sp_neigh_key),
};

struct mlxsw_sp_neigh_entry *
mlxsw_sp_rif_neigh_next(struct mlxsw_sp_rif *rif,
			struct mlxsw_sp_neigh_entry *neigh_entry)
{
	if (!neigh_entry) {
		if (list_empty(&rif->neigh_list))
			return NULL;
		else
			return list_first_entry(&rif->neigh_list,
						typeof(*neigh_entry),
						rif_list_node);
	}
	if (list_is_last(&neigh_entry->rif_list_node, &rif->neigh_list))
		return NULL;
	return list_next_entry(neigh_entry, rif_list_node);
}

int mlxsw_sp_neigh_entry_type(struct mlxsw_sp_neigh_entry *neigh_entry)
{
	return neigh_entry->key.n->tbl->family;
}

unsigned char *
mlxsw_sp_neigh_entry_ha(struct mlxsw_sp_neigh_entry *neigh_entry)
{
	return neigh_entry->ha;
}

u32 mlxsw_sp_neigh4_entry_dip(struct mlxsw_sp_neigh_entry *neigh_entry)
{
	struct neighbour *n;

	n = neigh_entry->key.n;
	return ntohl(*((__be32 *) n->primary_key));
}

struct in6_addr *
mlxsw_sp_neigh6_entry_dip(struct mlxsw_sp_neigh_entry *neigh_entry)
{
	struct neighbour *n;

	n = neigh_entry->key.n;
	return (struct in6_addr *) &n->primary_key;
}

int mlxsw_sp_neigh_counter_get(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_neigh_entry *neigh_entry,
			       u64 *p_counter)
{
	if (!neigh_entry->counter_valid)
		return -EINVAL;

	return mlxsw_sp_flow_counter_get(mlxsw_sp, neigh_entry->counter_index,
					 p_counter, NULL);
}

static struct mlxsw_sp_neigh_entry *
mlxsw_sp_neigh_entry_alloc(struct mlxsw_sp *mlxsw_sp, struct neighbour *n,
			   u16 rif)
{
	struct mlxsw_sp_neigh_entry *neigh_entry;

	neigh_entry = kzalloc(sizeof(*neigh_entry), GFP_KERNEL);
	if (!neigh_entry)
		return NULL;

	neigh_entry->key.n = n;
	neigh_entry->rif = rif;
	INIT_LIST_HEAD(&neigh_entry->nexthop_list);

	return neigh_entry;
}

static void mlxsw_sp_neigh_entry_free(struct mlxsw_sp_neigh_entry *neigh_entry)
{
	kfree(neigh_entry);
}

static int
mlxsw_sp_neigh_entry_insert(struct mlxsw_sp *mlxsw_sp,
			    struct mlxsw_sp_neigh_entry *neigh_entry)
{
	return rhashtable_insert_fast(&mlxsw_sp->router->neigh_ht,
				      &neigh_entry->ht_node,
				      mlxsw_sp_neigh_ht_params);
}

static void
mlxsw_sp_neigh_entry_remove(struct mlxsw_sp *mlxsw_sp,
			    struct mlxsw_sp_neigh_entry *neigh_entry)
{
	rhashtable_remove_fast(&mlxsw_sp->router->neigh_ht,
			       &neigh_entry->ht_node,
			       mlxsw_sp_neigh_ht_params);
}

static bool
mlxsw_sp_neigh_counter_should_alloc(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_neigh_entry *neigh_entry)
{
	struct devlink *devlink;
	const char *table_name;

	switch (mlxsw_sp_neigh_entry_type(neigh_entry)) {
	case AF_INET:
		table_name = MLXSW_SP_DPIPE_TABLE_NAME_HOST4;
		break;
	case AF_INET6:
		table_name = MLXSW_SP_DPIPE_TABLE_NAME_HOST6;
		break;
	default:
		WARN_ON(1);
		return false;
	}

	devlink = priv_to_devlink(mlxsw_sp->core);
	return devlink_dpipe_table_counter_enabled(devlink, table_name);
}

static void
mlxsw_sp_neigh_counter_alloc(struct mlxsw_sp *mlxsw_sp,
			     struct mlxsw_sp_neigh_entry *neigh_entry)
{
	if (!mlxsw_sp_neigh_counter_should_alloc(mlxsw_sp, neigh_entry))
		return;

	if (mlxsw_sp_flow_counter_alloc(mlxsw_sp, &neigh_entry->counter_index))
		return;

	neigh_entry->counter_valid = true;
}

static void
mlxsw_sp_neigh_counter_free(struct mlxsw_sp *mlxsw_sp,
			    struct mlxsw_sp_neigh_entry *neigh_entry)
{
	if (!neigh_entry->counter_valid)
		return;
	mlxsw_sp_flow_counter_free(mlxsw_sp,
				   neigh_entry->counter_index);
	neigh_entry->counter_valid = false;
}

static struct mlxsw_sp_neigh_entry *
mlxsw_sp_neigh_entry_create(struct mlxsw_sp *mlxsw_sp, struct neighbour *n)
{
	struct mlxsw_sp_neigh_entry *neigh_entry;
	struct mlxsw_sp_rif *rif;
	int err;

	rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, n->dev);
	if (!rif)
		return ERR_PTR(-EINVAL);

	neigh_entry = mlxsw_sp_neigh_entry_alloc(mlxsw_sp, n, rif->rif_index);
	if (!neigh_entry)
		return ERR_PTR(-ENOMEM);

	err = mlxsw_sp_neigh_entry_insert(mlxsw_sp, neigh_entry);
	if (err)
		goto err_neigh_entry_insert;

	mlxsw_sp_neigh_counter_alloc(mlxsw_sp, neigh_entry);
	list_add(&neigh_entry->rif_list_node, &rif->neigh_list);

	return neigh_entry;

err_neigh_entry_insert:
	mlxsw_sp_neigh_entry_free(neigh_entry);
	return ERR_PTR(err);
}

static void
mlxsw_sp_neigh_entry_destroy(struct mlxsw_sp *mlxsw_sp,
			     struct mlxsw_sp_neigh_entry *neigh_entry)
{
	list_del(&neigh_entry->rif_list_node);
	mlxsw_sp_neigh_counter_free(mlxsw_sp, neigh_entry);
	mlxsw_sp_neigh_entry_remove(mlxsw_sp, neigh_entry);
	mlxsw_sp_neigh_entry_free(neigh_entry);
}

static struct mlxsw_sp_neigh_entry *
mlxsw_sp_neigh_entry_lookup(struct mlxsw_sp *mlxsw_sp, struct neighbour *n)
{
	struct mlxsw_sp_neigh_key key;

	key.n = n;
	return rhashtable_lookup_fast(&mlxsw_sp->router->neigh_ht,
				      &key, mlxsw_sp_neigh_ht_params);
}

static void
mlxsw_sp_router_neighs_update_interval_init(struct mlxsw_sp *mlxsw_sp)
{
	unsigned long interval;

#if IS_ENABLED(CONFIG_IPV6)
	interval = min_t(unsigned long,
			 NEIGH_VAR(&arp_tbl.parms, DELAY_PROBE_TIME),
			 NEIGH_VAR(&nd_tbl.parms, DELAY_PROBE_TIME));
#else
	interval = NEIGH_VAR(&arp_tbl.parms, DELAY_PROBE_TIME);
#endif
	mlxsw_sp->router->neighs_update.interval = jiffies_to_msecs(interval);
}

static void mlxsw_sp_router_neigh_ent_ipv4_process(struct mlxsw_sp *mlxsw_sp,
						   char *rauhtd_pl,
						   int ent_index)
{
	struct net_device *dev;
	struct neighbour *n;
	__be32 dipn;
	u32 dip;
	u16 rif;

	mlxsw_reg_rauhtd_ent_ipv4_unpack(rauhtd_pl, ent_index, &rif, &dip);

	if (!mlxsw_sp->router->rifs[rif]) {
		dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Incorrect RIF in neighbour entry\n");
		return;
	}

	dipn = htonl(dip);
	dev = mlxsw_sp->router->rifs[rif]->dev;
	n = neigh_lookup(&arp_tbl, &dipn, dev);
	if (!n)
		return;

	netdev_dbg(dev, "Updating neighbour with IP=%pI4h\n", &dip);
	neigh_event_send(n, NULL);
	neigh_release(n);
}

#if IS_ENABLED(CONFIG_IPV6)
static void mlxsw_sp_router_neigh_ent_ipv6_process(struct mlxsw_sp *mlxsw_sp,
						   char *rauhtd_pl,
						   int rec_index)
{
	struct net_device *dev;
	struct neighbour *n;
	struct in6_addr dip;
	u16 rif;

	mlxsw_reg_rauhtd_ent_ipv6_unpack(rauhtd_pl, rec_index, &rif,
					 (char *) &dip);

	if (!mlxsw_sp->router->rifs[rif]) {
		dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Incorrect RIF in neighbour entry\n");
		return;
	}

	dev = mlxsw_sp->router->rifs[rif]->dev;
	n = neigh_lookup(&nd_tbl, &dip, dev);
	if (!n)
		return;

	netdev_dbg(dev, "Updating neighbour with IP=%pI6c\n", &dip);
	neigh_event_send(n, NULL);
	neigh_release(n);
}
#else
static void mlxsw_sp_router_neigh_ent_ipv6_process(struct mlxsw_sp *mlxsw_sp,
						   char *rauhtd_pl,
						   int rec_index)
{
}
#endif

static void mlxsw_sp_router_neigh_rec_ipv4_process(struct mlxsw_sp *mlxsw_sp,
						   char *rauhtd_pl,
						   int rec_index)
{
	u8 num_entries;
	int i;

	num_entries = mlxsw_reg_rauhtd_ipv4_rec_num_entries_get(rauhtd_pl,
								rec_index);
	/* Hardware starts counting at 0, so add 1. */
	num_entries++;

	/* Each record consists of several neighbour entries. */
	for (i = 0; i < num_entries; i++) {
		int ent_index;

		ent_index = rec_index * MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC + i;
		mlxsw_sp_router_neigh_ent_ipv4_process(mlxsw_sp, rauhtd_pl,
						       ent_index);
	}

}

static void mlxsw_sp_router_neigh_rec_ipv6_process(struct mlxsw_sp *mlxsw_sp,
						   char *rauhtd_pl,
						   int rec_index)
{
	/* One record contains one entry. */
	mlxsw_sp_router_neigh_ent_ipv6_process(mlxsw_sp, rauhtd_pl,
					       rec_index);
}

static void mlxsw_sp_router_neigh_rec_process(struct mlxsw_sp *mlxsw_sp,
					      char *rauhtd_pl, int rec_index)
{
	switch (mlxsw_reg_rauhtd_rec_type_get(rauhtd_pl, rec_index)) {
	case MLXSW_REG_RAUHTD_TYPE_IPV4:
		mlxsw_sp_router_neigh_rec_ipv4_process(mlxsw_sp, rauhtd_pl,
						       rec_index);
		break;
	case MLXSW_REG_RAUHTD_TYPE_IPV6:
		mlxsw_sp_router_neigh_rec_ipv6_process(mlxsw_sp, rauhtd_pl,
						       rec_index);
		break;
	}
}

static bool mlxsw_sp_router_rauhtd_is_full(char *rauhtd_pl)
{
	u8 num_rec, last_rec_index, num_entries;

	num_rec = mlxsw_reg_rauhtd_num_rec_get(rauhtd_pl);
	last_rec_index = num_rec - 1;

	if (num_rec < MLXSW_REG_RAUHTD_REC_MAX_NUM)
		return false;
	if (mlxsw_reg_rauhtd_rec_type_get(rauhtd_pl, last_rec_index) ==
	    MLXSW_REG_RAUHTD_TYPE_IPV6)
		return true;

	num_entries = mlxsw_reg_rauhtd_ipv4_rec_num_entries_get(rauhtd_pl,
								last_rec_index);
	if (++num_entries == MLXSW_REG_RAUHTD_IPV4_ENT_PER_REC)
		return true;
	return false;
}

static int
__mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp,
				       char *rauhtd_pl,
				       enum mlxsw_reg_rauhtd_type type)
{
	int i, num_rec;
	int err;

	/* Make sure the neighbour's netdev isn't removed in the
	 * process.
	 */
	rtnl_lock();
	do {
		mlxsw_reg_rauhtd_pack(rauhtd_pl, type);
		err = mlxsw_reg_query(mlxsw_sp->core, MLXSW_REG(rauhtd),
				      rauhtd_pl);
		if (err) {
			dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to dump neighbour table\n");
			break;
		}
		num_rec = mlxsw_reg_rauhtd_num_rec_get(rauhtd_pl);
		for (i = 0; i < num_rec; i++)
			mlxsw_sp_router_neigh_rec_process(mlxsw_sp, rauhtd_pl,
							  i);
	} while (mlxsw_sp_router_rauhtd_is_full(rauhtd_pl));
	rtnl_unlock();

	return err;
}

static int mlxsw_sp_router_neighs_update_rauhtd(struct mlxsw_sp *mlxsw_sp)
{
	enum mlxsw_reg_rauhtd_type type;
	char *rauhtd_pl;
	int err;

	rauhtd_pl = kmalloc(MLXSW_REG_RAUHTD_LEN, GFP_KERNEL);
	if (!rauhtd_pl)
		return -ENOMEM;

	type = MLXSW_REG_RAUHTD_TYPE_IPV4;
	err = __mlxsw_sp_router_neighs_update_rauhtd(mlxsw_sp, rauhtd_pl, type);
	if (err)
		goto out;

	type = MLXSW_REG_RAUHTD_TYPE_IPV6;
	err = __mlxsw_sp_router_neighs_update_rauhtd(mlxsw_sp, rauhtd_pl, type);
out:
	kfree(rauhtd_pl);
	return err;
}

static void mlxsw_sp_router_neighs_update_nh(struct mlxsw_sp *mlxsw_sp)
{
	struct mlxsw_sp_neigh_entry *neigh_entry;

	/* Take RTNL mutex here to prevent lists from changes */
	rtnl_lock();
	list_for_each_entry(neigh_entry, &mlxsw_sp->router->nexthop_neighs_list,
			    nexthop_neighs_list_node)
		/* If this neigh have nexthops, make the kernel think this neigh
		 * is active regardless of the traffic.
		 */
		neigh_event_send(neigh_entry->key.n, NULL);
	rtnl_unlock();
}

static void
mlxsw_sp_router_neighs_update_work_schedule(struct mlxsw_sp *mlxsw_sp)
{
	unsigned long interval = mlxsw_sp->router->neighs_update.interval;

	mlxsw_core_schedule_dw(&mlxsw_sp->router->neighs_update.dw,
			       msecs_to_jiffies(interval));
}

static void mlxsw_sp_router_neighs_update_work(struct work_struct *work)
{
	struct mlxsw_sp_router *router;
	int err;

	router = container_of(work, struct mlxsw_sp_router,
			      neighs_update.dw.work);
	err = mlxsw_sp_router_neighs_update_rauhtd(router->mlxsw_sp);
	if (err)
		dev_err(router->mlxsw_sp->bus_info->dev, "Could not update kernel for neigh activity");

	mlxsw_sp_router_neighs_update_nh(router->mlxsw_sp);

	mlxsw_sp_router_neighs_update_work_schedule(router->mlxsw_sp);
}

static void mlxsw_sp_router_probe_unresolved_nexthops(struct work_struct *work)
{
	struct mlxsw_sp_neigh_entry *neigh_entry;
	struct mlxsw_sp_router *router;

	router = container_of(work, struct mlxsw_sp_router,
			      nexthop_probe_dw.work);
	/* Iterate over nexthop neighbours, find those who are unresolved and
	 * send arp on them. This solves the chicken-egg problem when
	 * the nexthop wouldn't get offloaded until the neighbor is resolved
	 * but it wouldn't get resolved ever in case traffic is flowing in HW
	 * using different nexthop.
	 *
	 * Take RTNL mutex here to prevent lists from changes.
	 */
	rtnl_lock();
	list_for_each_entry(neigh_entry, &router->nexthop_neighs_list,
			    nexthop_neighs_list_node)
		if (!neigh_entry->connected)
			neigh_event_send(neigh_entry->key.n, NULL);
	rtnl_unlock();

	mlxsw_core_schedule_dw(&router->nexthop_probe_dw,
			       MLXSW_SP_UNRESOLVED_NH_PROBE_INTERVAL);
}

static void
mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp,
			      struct mlxsw_sp_neigh_entry *neigh_entry,
			      bool removing);

static enum mlxsw_reg_rauht_op mlxsw_sp_rauht_op(bool adding)
{
	return adding ? MLXSW_REG_RAUHT_OP_WRITE_ADD :
			MLXSW_REG_RAUHT_OP_WRITE_DELETE;
}

static void
mlxsw_sp_router_neigh_entry_op4(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_neigh_entry *neigh_entry,
				enum mlxsw_reg_rauht_op op)
{
	struct neighbour *n = neigh_entry->key.n;
	u32 dip = ntohl(*((__be32 *) n->primary_key));
	char rauht_pl[MLXSW_REG_RAUHT_LEN];

	mlxsw_reg_rauht_pack4(rauht_pl, op, neigh_entry->rif, neigh_entry->ha,
			      dip);
	if (neigh_entry->counter_valid)
		mlxsw_reg_rauht_pack_counter(rauht_pl,
					     neigh_entry->counter_index);
	mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rauht), rauht_pl);
}

static void
mlxsw_sp_router_neigh_entry_op6(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_neigh_entry *neigh_entry,
				enum mlxsw_reg_rauht_op op)
{
	struct neighbour *n = neigh_entry->key.n;
	char rauht_pl[MLXSW_REG_RAUHT_LEN];
	const char *dip = n->primary_key;

	mlxsw_reg_rauht_pack6(rauht_pl, op, neigh_entry->rif, neigh_entry->ha,
			      dip);
	if (neigh_entry->counter_valid)
		mlxsw_reg_rauht_pack_counter(rauht_pl,
					     neigh_entry->counter_index);
	mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(rauht), rauht_pl);
}

bool mlxsw_sp_neigh_ipv6_ignore(struct mlxsw_sp_neigh_entry *neigh_entry)
{
	struct neighbour *n = neigh_entry->key.n;

	/* Packets with a link-local destination address are trapped
	 * after LPM lookup and never reach the neighbour table, so
	 * there is no need to program such neighbours to the device.
	 */
	if (ipv6_addr_type((struct in6_addr *) &n->primary_key) &
	    IPV6_ADDR_LINKLOCAL)
		return true;
	return false;
}

static void
mlxsw_sp_neigh_entry_update(struct mlxsw_sp *mlxsw_sp,
			    struct mlxsw_sp_neigh_entry *neigh_entry,
			    bool adding)
{
	if (!adding && !neigh_entry->connected)
		return;
	neigh_entry->connected = adding;
	if (neigh_entry->key.n->tbl->family == AF_INET) {
		mlxsw_sp_router_neigh_entry_op4(mlxsw_sp, neigh_entry,
						mlxsw_sp_rauht_op(adding));
	} else if (neigh_entry->key.n->tbl->family == AF_INET6) {
		if (mlxsw_sp_neigh_ipv6_ignore(neigh_entry))
			return;
		mlxsw_sp_router_neigh_entry_op6(mlxsw_sp, neigh_entry,
						mlxsw_sp_rauht_op(adding));
	} else {
		WARN_ON_ONCE(1);
	}
}

void
mlxsw_sp_neigh_entry_counter_update(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_neigh_entry *neigh_entry,
				    bool adding)
{
	if (adding)
		mlxsw_sp_neigh_counter_alloc(mlxsw_sp, neigh_entry);
	else
		mlxsw_sp_neigh_counter_free(mlxsw_sp, neigh_entry);
	mlxsw_sp_neigh_entry_update(mlxsw_sp, neigh_entry, true);
}

struct mlxsw_sp_netevent_work {
	struct work_struct work;
	struct mlxsw_sp *mlxsw_sp;
	struct neighbour *n;
};

static void mlxsw_sp_router_neigh_event_work(struct work_struct *work)
{
	struct mlxsw_sp_netevent_work *net_work =
		container_of(work, struct mlxsw_sp_netevent_work, work);
	struct mlxsw_sp *mlxsw_sp = net_work->mlxsw_sp;
	struct mlxsw_sp_neigh_entry *neigh_entry;
	struct neighbour *n = net_work->n;
	unsigned char ha[ETH_ALEN];
	bool entry_connected;
	u8 nud_state, dead;

	/* If these parameters are changed after we release the lock,
	 * then we are guaranteed to receive another event letting us
	 * know about it.
	 */
	read_lock_bh(&n->lock);
	memcpy(ha, n->ha, ETH_ALEN);
	nud_state = n->nud_state;
	dead = n->dead;
	read_unlock_bh(&n->lock);

	rtnl_lock();
	mlxsw_sp_span_respin(mlxsw_sp);

	entry_connected = nud_state & NUD_VALID && !dead;
	neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, n);
	if (!entry_connected && !neigh_entry)
		goto out;
	if (!neigh_entry) {
		neigh_entry = mlxsw_sp_neigh_entry_create(mlxsw_sp, n);
		if (IS_ERR(neigh_entry))
			goto out;
	}

	memcpy(neigh_entry->ha, ha, ETH_ALEN);
	mlxsw_sp_neigh_entry_update(mlxsw_sp, neigh_entry, entry_connected);
	mlxsw_sp_nexthop_neigh_update(mlxsw_sp, neigh_entry, !entry_connected);

	if (!neigh_entry->connected && list_empty(&neigh_entry->nexthop_list))
		mlxsw_sp_neigh_entry_destroy(mlxsw_sp, neigh_entry);

out:
	rtnl_unlock();
	neigh_release(n);
	kfree(net_work);
}

static int mlxsw_sp_mp_hash_init(struct mlxsw_sp *mlxsw_sp);

static void mlxsw_sp_router_mp_hash_event_work(struct work_struct *work)
{
	struct mlxsw_sp_netevent_work *net_work =
		container_of(work, struct mlxsw_sp_netevent_work, work);
	struct mlxsw_sp *mlxsw_sp = net_work->mlxsw_sp;

	mlxsw_sp_mp_hash_init(mlxsw_sp);
	kfree(net_work);
}

static int mlxsw_sp_router_netevent_event(struct notifier_block *nb,
					  unsigned long event, void *ptr)
{
	struct mlxsw_sp_netevent_work *net_work;
	struct mlxsw_sp_port *mlxsw_sp_port;
	struct mlxsw_sp_router *router;
	struct mlxsw_sp *mlxsw_sp;
	unsigned long interval;
	struct neigh_parms *p;
	struct neighbour *n;
	struct net *net;

	switch (event) {
	case NETEVENT_DELAY_PROBE_TIME_UPDATE:
		p = ptr;

		/* We don't care about changes in the default table. */
		if (!p->dev || (p->tbl->family != AF_INET &&
				p->tbl->family != AF_INET6))
			return NOTIFY_DONE;

		/* We are in atomic context and can't take RTNL mutex,
		 * so use RCU variant to walk the device chain.
		 */
		mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(p->dev);
		if (!mlxsw_sp_port)
			return NOTIFY_DONE;

		mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
		interval = jiffies_to_msecs(NEIGH_VAR(p, DELAY_PROBE_TIME));
		mlxsw_sp->router->neighs_update.interval = interval;

		mlxsw_sp_port_dev_put(mlxsw_sp_port);
		break;
	case NETEVENT_NEIGH_UPDATE:
		n = ptr;

		if (n->tbl->family != AF_INET && n->tbl->family != AF_INET6)
			return NOTIFY_DONE;

		mlxsw_sp_port = mlxsw_sp_port_lower_dev_hold(n->dev);
		if (!mlxsw_sp_port)
			return NOTIFY_DONE;

		net_work = kzalloc(sizeof(*net_work), GFP_ATOMIC);
		if (!net_work) {
			mlxsw_sp_port_dev_put(mlxsw_sp_port);
			return NOTIFY_BAD;
		}

		INIT_WORK(&net_work->work, mlxsw_sp_router_neigh_event_work);
		net_work->mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
		net_work->n = n;

		/* Take a reference to ensure the neighbour won't be
		 * destructed until we drop the reference in delayed
		 * work.
		 */
		neigh_clone(n);
		mlxsw_core_schedule_work(&net_work->work);
		mlxsw_sp_port_dev_put(mlxsw_sp_port);
		break;
	case NETEVENT_IPV4_MPATH_HASH_UPDATE:
	case NETEVENT_IPV6_MPATH_HASH_UPDATE:
		net = ptr;

		if (!net_eq(net, &init_net))
			return NOTIFY_DONE;

		net_work = kzalloc(sizeof(*net_work), GFP_ATOMIC);
		if (!net_work)
			return NOTIFY_BAD;

		router = container_of(nb, struct mlxsw_sp_router, netevent_nb);
		INIT_WORK(&net_work->work, mlxsw_sp_router_mp_hash_event_work);
		net_work->mlxsw_sp = router->mlxsw_sp;
		mlxsw_core_schedule_work(&net_work->work);
		break;
	}

	return NOTIFY_DONE;
}

static int mlxsw_sp_neigh_init(struct mlxsw_sp *mlxsw_sp)
{
	int err;

	err = rhashtable_init(&mlxsw_sp->router->neigh_ht,
			      &mlxsw_sp_neigh_ht_params);
	if (err)
		return err;

	/* Initialize the polling interval according to the default
	 * table.
	 */
	mlxsw_sp_router_neighs_update_interval_init(mlxsw_sp);

	/* Create the delayed works for the activity_update */
	INIT_DELAYED_WORK(&mlxsw_sp->router->neighs_update.dw,
			  mlxsw_sp_router_neighs_update_work);
	INIT_DELAYED_WORK(&mlxsw_sp->router->nexthop_probe_dw,
			  mlxsw_sp_router_probe_unresolved_nexthops);
	mlxsw_core_schedule_dw(&mlxsw_sp->router->neighs_update.dw, 0);
	mlxsw_core_schedule_dw(&mlxsw_sp->router->nexthop_probe_dw, 0);
	return 0;
}

static void mlxsw_sp_neigh_fini(struct mlxsw_sp *mlxsw_sp)
{
	cancel_delayed_work_sync(&mlxsw_sp->router->neighs_update.dw);
	cancel_delayed_work_sync(&mlxsw_sp->router->nexthop_probe_dw);
	rhashtable_destroy(&mlxsw_sp->router->neigh_ht);
}

static void mlxsw_sp_neigh_rif_gone_sync(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_rif *rif)
{
	struct mlxsw_sp_neigh_entry *neigh_entry, *tmp;

	list_for_each_entry_safe(neigh_entry, tmp, &rif->neigh_list,
				 rif_list_node) {
		mlxsw_sp_neigh_entry_update(mlxsw_sp, neigh_entry, false);
		mlxsw_sp_neigh_entry_destroy(mlxsw_sp, neigh_entry);
	}
}

enum mlxsw_sp_nexthop_type {
	MLXSW_SP_NEXTHOP_TYPE_ETH,
	MLXSW_SP_NEXTHOP_TYPE_IPIP,
};

struct mlxsw_sp_nexthop_key {
	struct fib_nh *fib_nh;
};

struct mlxsw_sp_nexthop {
	struct list_head neigh_list_node; /* member of neigh entry list */
	struct list_head rif_list_node;
	struct list_head router_list_node;
	struct mlxsw_sp_nexthop_group *nh_grp; /* pointer back to the group
						* this belongs to
						*/
	struct rhash_head ht_node;
	struct mlxsw_sp_nexthop_key key;
	unsigned char gw_addr[sizeof(struct in6_addr)];
	int ifindex;
	int nh_weight;
	int norm_nh_weight;
	int num_adj_entries;
	struct mlxsw_sp_rif *rif;
	u8 should_offload:1, /* set indicates this neigh is connected and
			      * should be put to KVD linear area of this group.
			      */
	   offloaded:1, /* set in case the neigh is actually put into
			 * KVD linear area of this group.
			 */
	   update:1; /* set indicates that MAC of this neigh should be
		      * updated in HW
		      */
	enum mlxsw_sp_nexthop_type type;
	union {
		struct mlxsw_sp_neigh_entry *neigh_entry;
		struct mlxsw_sp_ipip_entry *ipip_entry;
	};
	unsigned int counter_index;
	bool counter_valid;
};

struct mlxsw_sp_nexthop_group {
	void *priv;
	struct rhash_head ht_node;
	struct list_head fib_list; /* list of fib entries that use this group */
	struct neigh_table *neigh_tbl;
	u8 adj_index_valid:1,
	   gateway:1; /* routes using the group use a gateway */
	u32 adj_index;
	u16 ecmp_size;
	u16 count;
	int sum_norm_weight;
	struct mlxsw_sp_nexthop nexthops[0];
#define nh_rif	nexthops[0].rif
};

void mlxsw_sp_nexthop_counter_alloc(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_nexthop *nh)
{
	struct devlink *devlink;

	devlink = priv_to_devlink(mlxsw_sp->core);
	if (!devlink_dpipe_table_counter_enabled(devlink,
						 MLXSW_SP_DPIPE_TABLE_NAME_ADJ))
		return;

	if (mlxsw_sp_flow_counter_alloc(mlxsw_sp, &nh->counter_index))
		return;

	nh->counter_valid = true;
}

void mlxsw_sp_nexthop_counter_free(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_nexthop *nh)
{
	if (!nh->counter_valid)
		return;
	mlxsw_sp_flow_counter_free(mlxsw_sp, nh->counter_index);
	nh->counter_valid = false;
}

int mlxsw_sp_nexthop_counter_get(struct mlxsw_sp *mlxsw_sp,
				 struct mlxsw_sp_nexthop *nh, u64 *p_counter)
{
	if (!nh->counter_valid)
		return -EINVAL;

	return mlxsw_sp_flow_counter_get(mlxsw_sp, nh->counter_index,
					 p_counter, NULL);
}

struct mlxsw_sp_nexthop *mlxsw_sp_nexthop_next(struct mlxsw_sp_router *router,
					       struct mlxsw_sp_nexthop *nh)
{
	if (!nh) {
		if (list_empty(&router->nexthop_list))
			return NULL;
		else
			return list_first_entry(&router->nexthop_list,
						typeof(*nh), router_list_node);
	}
	if (list_is_last(&nh->router_list_node, &router->nexthop_list))
		return NULL;
	return list_next_entry(nh, router_list_node);
}

bool mlxsw_sp_nexthop_offload(struct mlxsw_sp_nexthop *nh)
{
	return nh->offloaded;
}

unsigned char *mlxsw_sp_nexthop_ha(struct mlxsw_sp_nexthop *nh)
{
	if (!nh->offloaded)
		return NULL;
	return nh->neigh_entry->ha;
}

int mlxsw_sp_nexthop_indexes(struct mlxsw_sp_nexthop *nh, u32 *p_adj_index,
			     u32 *p_adj_size, u32 *p_adj_hash_index)
{
	struct mlxsw_sp_nexthop_group *nh_grp = nh->nh_grp;
	u32 adj_hash_index = 0;
	int i;

	if (!nh->offloaded || !nh_grp->adj_index_valid)
		return -EINVAL;

	*p_adj_index = nh_grp->adj_index;
	*p_adj_size = nh_grp->ecmp_size;

	for (i = 0; i < nh_grp->count; i++) {
		struct mlxsw_sp_nexthop *nh_iter = &nh_grp->nexthops[i];

		if (nh_iter == nh)
			break;
		if (nh_iter->offloaded)
			adj_hash_index += nh_iter->num_adj_entries;
	}

	*p_adj_hash_index = adj_hash_index;
	return 0;
}

struct mlxsw_sp_rif *mlxsw_sp_nexthop_rif(struct mlxsw_sp_nexthop *nh)
{
	return nh->rif;
}

bool mlxsw_sp_nexthop_group_has_ipip(struct mlxsw_sp_nexthop *nh)
{
	struct mlxsw_sp_nexthop_group *nh_grp = nh->nh_grp;
	int i;

	for (i = 0; i < nh_grp->count; i++) {
		struct mlxsw_sp_nexthop *nh_iter = &nh_grp->nexthops[i];

		if (nh_iter->type == MLXSW_SP_NEXTHOP_TYPE_IPIP)
			return true;
	}
	return false;
}

static struct fib_info *
mlxsw_sp_nexthop4_group_fi(const struct mlxsw_sp_nexthop_group *nh_grp)
{
	return nh_grp->priv;
}

struct mlxsw_sp_nexthop_group_cmp_arg {
	enum mlxsw_sp_l3proto proto;
	union {
		struct fib_info *fi;
		struct mlxsw_sp_fib6_entry *fib6_entry;
	};
};

static bool
mlxsw_sp_nexthop6_group_has_nexthop(const struct mlxsw_sp_nexthop_group *nh_grp,
				    const struct in6_addr *gw, int ifindex,
				    int weight)
{
	int i;

	for (i = 0; i < nh_grp->count; i++) {
		const struct mlxsw_sp_nexthop *nh;

		nh = &nh_grp->nexthops[i];
		if (nh->ifindex == ifindex && nh->nh_weight == weight &&
		    ipv6_addr_equal(gw, (struct in6_addr *) nh->gw_addr))
			return true;
	}

	return false;
}

static bool
mlxsw_sp_nexthop6_group_cmp(const struct mlxsw_sp_nexthop_group *nh_grp,
			    const struct mlxsw_sp_fib6_entry *fib6_entry)
{
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;

	if (nh_grp->count != fib6_entry->nrt6)
		return false;

	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
		struct in6_addr *gw;
		int ifindex, weight;

		ifindex = mlxsw_sp_rt6->rt->fib6_nh.nh_dev->ifindex;
		weight = mlxsw_sp_rt6->rt->fib6_nh.nh_weight;
		gw = &mlxsw_sp_rt6->rt->fib6_nh.nh_gw;
		if (!mlxsw_sp_nexthop6_group_has_nexthop(nh_grp, gw, ifindex,
							 weight))
			return false;
	}

	return true;
}

static int
mlxsw_sp_nexthop_group_cmp(struct rhashtable_compare_arg *arg, const void *ptr)
{
	const struct mlxsw_sp_nexthop_group_cmp_arg *cmp_arg = arg->key;
	const struct mlxsw_sp_nexthop_group *nh_grp = ptr;

	switch (cmp_arg->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		return cmp_arg->fi != mlxsw_sp_nexthop4_group_fi(nh_grp);
	case MLXSW_SP_L3_PROTO_IPV6:
		return !mlxsw_sp_nexthop6_group_cmp(nh_grp,
						    cmp_arg->fib6_entry);
	default:
		WARN_ON(1);
		return 1;
	}
}

static int
mlxsw_sp_nexthop_group_type(const struct mlxsw_sp_nexthop_group *nh_grp)
{
	return nh_grp->neigh_tbl->family;
}

static u32 mlxsw_sp_nexthop_group_hash_obj(const void *data, u32 len, u32 seed)
{
	const struct mlxsw_sp_nexthop_group *nh_grp = data;
	const struct mlxsw_sp_nexthop *nh;
	struct fib_info *fi;
	unsigned int val;
	int i;

	switch (mlxsw_sp_nexthop_group_type(nh_grp)) {
	case AF_INET:
		fi = mlxsw_sp_nexthop4_group_fi(nh_grp);
		return jhash(&fi, sizeof(fi), seed);
	case AF_INET6:
		val = nh_grp->count;
		for (i = 0; i < nh_grp->count; i++) {
			nh = &nh_grp->nexthops[i];
			val ^= nh->ifindex;
		}
		return jhash(&val, sizeof(val), seed);
	default:
		WARN_ON(1);
		return 0;
	}
}

static u32
mlxsw_sp_nexthop6_group_hash(struct mlxsw_sp_fib6_entry *fib6_entry, u32 seed)
{
	unsigned int val = fib6_entry->nrt6;
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
	struct net_device *dev;

	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
		dev = mlxsw_sp_rt6->rt->fib6_nh.nh_dev;
		val ^= dev->ifindex;
	}

	return jhash(&val, sizeof(val), seed);
}

static u32
mlxsw_sp_nexthop_group_hash(const void *data, u32 len, u32 seed)
{
	const struct mlxsw_sp_nexthop_group_cmp_arg *cmp_arg = data;

	switch (cmp_arg->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		return jhash(&cmp_arg->fi, sizeof(cmp_arg->fi), seed);
	case MLXSW_SP_L3_PROTO_IPV6:
		return mlxsw_sp_nexthop6_group_hash(cmp_arg->fib6_entry, seed);
	default:
		WARN_ON(1);
		return 0;
	}
}

static const struct rhashtable_params mlxsw_sp_nexthop_group_ht_params = {
	.head_offset = offsetof(struct mlxsw_sp_nexthop_group, ht_node),
	.hashfn	     = mlxsw_sp_nexthop_group_hash,
	.obj_hashfn  = mlxsw_sp_nexthop_group_hash_obj,
	.obj_cmpfn   = mlxsw_sp_nexthop_group_cmp,
};

static int mlxsw_sp_nexthop_group_insert(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_nexthop_group *nh_grp)
{
	if (mlxsw_sp_nexthop_group_type(nh_grp) == AF_INET6 &&
	    !nh_grp->gateway)
		return 0;

	return rhashtable_insert_fast(&mlxsw_sp->router->nexthop_group_ht,
				      &nh_grp->ht_node,
				      mlxsw_sp_nexthop_group_ht_params);
}

static void mlxsw_sp_nexthop_group_remove(struct mlxsw_sp *mlxsw_sp,
					  struct mlxsw_sp_nexthop_group *nh_grp)
{
	if (mlxsw_sp_nexthop_group_type(nh_grp) == AF_INET6 &&
	    !nh_grp->gateway)
		return;

	rhashtable_remove_fast(&mlxsw_sp->router->nexthop_group_ht,
			       &nh_grp->ht_node,
			       mlxsw_sp_nexthop_group_ht_params);
}

static struct mlxsw_sp_nexthop_group *
mlxsw_sp_nexthop4_group_lookup(struct mlxsw_sp *mlxsw_sp,
			       struct fib_info *fi)
{
	struct mlxsw_sp_nexthop_group_cmp_arg cmp_arg;

	cmp_arg.proto = MLXSW_SP_L3_PROTO_IPV4;
	cmp_arg.fi = fi;
	return rhashtable_lookup_fast(&mlxsw_sp->router->nexthop_group_ht,
				      &cmp_arg,
				      mlxsw_sp_nexthop_group_ht_params);
}

static struct mlxsw_sp_nexthop_group *
mlxsw_sp_nexthop6_group_lookup(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_fib6_entry *fib6_entry)
{
	struct mlxsw_sp_nexthop_group_cmp_arg cmp_arg;

	cmp_arg.proto = MLXSW_SP_L3_PROTO_IPV6;
	cmp_arg.fib6_entry = fib6_entry;
	return rhashtable_lookup_fast(&mlxsw_sp->router->nexthop_group_ht,
				      &cmp_arg,
				      mlxsw_sp_nexthop_group_ht_params);
}

static const struct rhashtable_params mlxsw_sp_nexthop_ht_params = {
	.key_offset = offsetof(struct mlxsw_sp_nexthop, key),
	.head_offset = offsetof(struct mlxsw_sp_nexthop, ht_node),
	.key_len = sizeof(struct mlxsw_sp_nexthop_key),
};

static int mlxsw_sp_nexthop_insert(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_nexthop *nh)
{
	return rhashtable_insert_fast(&mlxsw_sp->router->nexthop_ht,
				      &nh->ht_node, mlxsw_sp_nexthop_ht_params);
}

static void mlxsw_sp_nexthop_remove(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_nexthop *nh)
{
	rhashtable_remove_fast(&mlxsw_sp->router->nexthop_ht, &nh->ht_node,
			       mlxsw_sp_nexthop_ht_params);
}

static struct mlxsw_sp_nexthop *
mlxsw_sp_nexthop_lookup(struct mlxsw_sp *mlxsw_sp,
			struct mlxsw_sp_nexthop_key key)
{
	return rhashtable_lookup_fast(&mlxsw_sp->router->nexthop_ht, &key,
				      mlxsw_sp_nexthop_ht_params);
}

static int mlxsw_sp_adj_index_mass_update_vr(struct mlxsw_sp *mlxsw_sp,
					     const struct mlxsw_sp_fib *fib,
					     u32 adj_index, u16 ecmp_size,
					     u32 new_adj_index,
					     u16 new_ecmp_size)
{
	char raleu_pl[MLXSW_REG_RALEU_LEN];

	mlxsw_reg_raleu_pack(raleu_pl,
			     (enum mlxsw_reg_ralxx_protocol) fib->proto,
			     fib->vr->id, adj_index, ecmp_size, new_adj_index,
			     new_ecmp_size);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raleu), raleu_pl);
}

static int mlxsw_sp_adj_index_mass_update(struct mlxsw_sp *mlxsw_sp,
					  struct mlxsw_sp_nexthop_group *nh_grp,
					  u32 old_adj_index, u16 old_ecmp_size)
{
	struct mlxsw_sp_fib_entry *fib_entry;
	struct mlxsw_sp_fib *fib = NULL;
	int err;

	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) {
		if (fib == fib_entry->fib_node->fib)
			continue;
		fib = fib_entry->fib_node->fib;
		err = mlxsw_sp_adj_index_mass_update_vr(mlxsw_sp, fib,
							old_adj_index,
							old_ecmp_size,
							nh_grp->adj_index,
							nh_grp->ecmp_size);
		if (err)
			return err;
	}
	return 0;
}

static int __mlxsw_sp_nexthop_update(struct mlxsw_sp *mlxsw_sp, u32 adj_index,
				     struct mlxsw_sp_nexthop *nh)
{
	struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry;
	char ratr_pl[MLXSW_REG_RATR_LEN];

	mlxsw_reg_ratr_pack(ratr_pl, MLXSW_REG_RATR_OP_WRITE_WRITE_ENTRY,
			    true, MLXSW_REG_RATR_TYPE_ETHERNET,
			    adj_index, neigh_entry->rif);
	mlxsw_reg_ratr_eth_entry_pack(ratr_pl, neigh_entry->ha);
	if (nh->counter_valid)
		mlxsw_reg_ratr_counter_pack(ratr_pl, nh->counter_index, true);
	else
		mlxsw_reg_ratr_counter_pack(ratr_pl, 0, false);

	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ratr), ratr_pl);
}

int mlxsw_sp_nexthop_update(struct mlxsw_sp *mlxsw_sp, u32 adj_index,
			    struct mlxsw_sp_nexthop *nh)
{
	int i;

	for (i = 0; i < nh->num_adj_entries; i++) {
		int err;

		err = __mlxsw_sp_nexthop_update(mlxsw_sp, adj_index + i, nh);
		if (err)
			return err;
	}

	return 0;
}

static int __mlxsw_sp_nexthop_ipip_update(struct mlxsw_sp *mlxsw_sp,
					  u32 adj_index,
					  struct mlxsw_sp_nexthop *nh)
{
	const struct mlxsw_sp_ipip_ops *ipip_ops;

	ipip_ops = mlxsw_sp->router->ipip_ops_arr[nh->ipip_entry->ipipt];
	return ipip_ops->nexthop_update(mlxsw_sp, adj_index, nh->ipip_entry);
}

static int mlxsw_sp_nexthop_ipip_update(struct mlxsw_sp *mlxsw_sp,
					u32 adj_index,
					struct mlxsw_sp_nexthop *nh)
{
	int i;

	for (i = 0; i < nh->num_adj_entries; i++) {
		int err;

		err = __mlxsw_sp_nexthop_ipip_update(mlxsw_sp, adj_index + i,
						     nh);
		if (err)
			return err;
	}

	return 0;
}

static int
mlxsw_sp_nexthop_group_update(struct mlxsw_sp *mlxsw_sp,
			      struct mlxsw_sp_nexthop_group *nh_grp,
			      bool reallocate)
{
	u32 adj_index = nh_grp->adj_index; /* base */
	struct mlxsw_sp_nexthop *nh;
	int i;
	int err;

	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];

		if (!nh->should_offload) {
			nh->offloaded = 0;
			continue;
		}

		if (nh->update || reallocate) {
			switch (nh->type) {
			case MLXSW_SP_NEXTHOP_TYPE_ETH:
				err = mlxsw_sp_nexthop_update
					    (mlxsw_sp, adj_index, nh);
				break;
			case MLXSW_SP_NEXTHOP_TYPE_IPIP:
				err = mlxsw_sp_nexthop_ipip_update
					    (mlxsw_sp, adj_index, nh);
				break;
			}
			if (err)
				return err;
			nh->update = 0;
			nh->offloaded = 1;
		}
		adj_index += nh->num_adj_entries;
	}
	return 0;
}

static bool
mlxsw_sp_fib_node_entry_is_first(const struct mlxsw_sp_fib_node *fib_node,
				 const struct mlxsw_sp_fib_entry *fib_entry);

static int
mlxsw_sp_nexthop_fib_entries_update(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_nexthop_group *nh_grp)
{
	struct mlxsw_sp_fib_entry *fib_entry;
	int err;

	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) {
		if (!mlxsw_sp_fib_node_entry_is_first(fib_entry->fib_node,
						      fib_entry))
			continue;
		err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
		if (err)
			return err;
	}
	return 0;
}

static void
mlxsw_sp_fib_entry_offload_refresh(struct mlxsw_sp_fib_entry *fib_entry,
				   enum mlxsw_reg_ralue_op op, int err);

static void
mlxsw_sp_nexthop_fib_entries_refresh(struct mlxsw_sp_nexthop_group *nh_grp)
{
	enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_WRITE;
	struct mlxsw_sp_fib_entry *fib_entry;

	list_for_each_entry(fib_entry, &nh_grp->fib_list, nexthop_group_node) {
		if (!mlxsw_sp_fib_node_entry_is_first(fib_entry->fib_node,
						      fib_entry))
			continue;
		mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0);
	}
}

static void mlxsw_sp_adj_grp_size_round_up(u16 *p_adj_grp_size)
{
	/* Valid sizes for an adjacency group are:
	 * 1-64, 512, 1024, 2048 and 4096.
	 */
	if (*p_adj_grp_size <= 64)
		return;
	else if (*p_adj_grp_size <= 512)
		*p_adj_grp_size = 512;
	else if (*p_adj_grp_size <= 1024)
		*p_adj_grp_size = 1024;
	else if (*p_adj_grp_size <= 2048)
		*p_adj_grp_size = 2048;
	else
		*p_adj_grp_size = 4096;
}

static void mlxsw_sp_adj_grp_size_round_down(u16 *p_adj_grp_size,
					     unsigned int alloc_size)
{
	if (alloc_size >= 4096)
		*p_adj_grp_size = 4096;
	else if (alloc_size >= 2048)
		*p_adj_grp_size = 2048;
	else if (alloc_size >= 1024)
		*p_adj_grp_size = 1024;
	else if (alloc_size >= 512)
		*p_adj_grp_size = 512;
}

static int mlxsw_sp_fix_adj_grp_size(struct mlxsw_sp *mlxsw_sp,
				     u16 *p_adj_grp_size)
{
	unsigned int alloc_size;
	int err;

	/* Round up the requested group size to the next size supported
	 * by the device and make sure the request can be satisfied.
	 */
	mlxsw_sp_adj_grp_size_round_up(p_adj_grp_size);
	err = mlxsw_sp_kvdl_alloc_size_query(mlxsw_sp, *p_adj_grp_size,
					     &alloc_size);
	if (err)
		return err;
	/* It is possible the allocation results in more allocated
	 * entries than requested. Try to use as much of them as
	 * possible.
	 */
	mlxsw_sp_adj_grp_size_round_down(p_adj_grp_size, alloc_size);

	return 0;
}

static void
mlxsw_sp_nexthop_group_normalize(struct mlxsw_sp_nexthop_group *nh_grp)
{
	int i, g = 0, sum_norm_weight = 0;
	struct mlxsw_sp_nexthop *nh;

	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];

		if (!nh->should_offload)
			continue;
		if (g > 0)
			g = gcd(nh->nh_weight, g);
		else
			g = nh->nh_weight;
	}

	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];

		if (!nh->should_offload)
			continue;
		nh->norm_nh_weight = nh->nh_weight / g;
		sum_norm_weight += nh->norm_nh_weight;
	}

	nh_grp->sum_norm_weight = sum_norm_weight;
}

static void
mlxsw_sp_nexthop_group_rebalance(struct mlxsw_sp_nexthop_group *nh_grp)
{
	int total = nh_grp->sum_norm_weight;
	u16 ecmp_size = nh_grp->ecmp_size;
	int i, weight = 0, lower_bound = 0;

	for (i = 0; i < nh_grp->count; i++) {
		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
		int upper_bound;

		if (!nh->should_offload)
			continue;
		weight += nh->norm_nh_weight;
		upper_bound = DIV_ROUND_CLOSEST(ecmp_size * weight, total);
		nh->num_adj_entries = upper_bound - lower_bound;
		lower_bound = upper_bound;
	}
}

static void
mlxsw_sp_nexthop_group_refresh(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_nexthop_group *nh_grp)
{
	u16 ecmp_size, old_ecmp_size;
	struct mlxsw_sp_nexthop *nh;
	bool offload_change = false;
	u32 adj_index;
	bool old_adj_index_valid;
	u32 old_adj_index;
	int i;
	int err;

	if (!nh_grp->gateway) {
		mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp);
		return;
	}

	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];

		if (nh->should_offload != nh->offloaded) {
			offload_change = true;
			if (nh->should_offload)
				nh->update = 1;
		}
	}
	if (!offload_change) {
		/* Nothing was added or removed, so no need to reallocate. Just
		 * update MAC on existing adjacency indexes.
		 */
		err = mlxsw_sp_nexthop_group_update(mlxsw_sp, nh_grp, false);
		if (err) {
			dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n");
			goto set_trap;
		}
		return;
	}
	mlxsw_sp_nexthop_group_normalize(nh_grp);
	if (!nh_grp->sum_norm_weight)
		/* No neigh of this group is connected so we just set
		 * the trap and let everthing flow through kernel.
		 */
		goto set_trap;

	ecmp_size = nh_grp->sum_norm_weight;
	err = mlxsw_sp_fix_adj_grp_size(mlxsw_sp, &ecmp_size);
	if (err)
		/* No valid allocation size available. */
		goto set_trap;

	err = mlxsw_sp_kvdl_alloc(mlxsw_sp, ecmp_size, &adj_index);
	if (err) {
		/* We ran out of KVD linear space, just set the
		 * trap and let everything flow through kernel.
		 */
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to allocate KVD linear area for nexthop group.\n");
		goto set_trap;
	}
	old_adj_index_valid = nh_grp->adj_index_valid;
	old_adj_index = nh_grp->adj_index;
	old_ecmp_size = nh_grp->ecmp_size;
	nh_grp->adj_index_valid = 1;
	nh_grp->adj_index = adj_index;
	nh_grp->ecmp_size = ecmp_size;
	mlxsw_sp_nexthop_group_rebalance(nh_grp);
	err = mlxsw_sp_nexthop_group_update(mlxsw_sp, nh_grp, true);
	if (err) {
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to update neigh MAC in adjacency table.\n");
		goto set_trap;
	}

	if (!old_adj_index_valid) {
		/* The trap was set for fib entries, so we have to call
		 * fib entry update to unset it and use adjacency index.
		 */
		err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp);
		if (err) {
			dev_warn(mlxsw_sp->bus_info->dev, "Failed to add adjacency index to fib entries.\n");
			goto set_trap;
		}
		return;
	}

	err = mlxsw_sp_adj_index_mass_update(mlxsw_sp, nh_grp,
					     old_adj_index, old_ecmp_size);
	mlxsw_sp_kvdl_free(mlxsw_sp, old_adj_index);
	if (err) {
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to mass-update adjacency index for nexthop group.\n");
		goto set_trap;
	}

	/* Offload state within the group changed, so update the flags. */
	mlxsw_sp_nexthop_fib_entries_refresh(nh_grp);

	return;

set_trap:
	old_adj_index_valid = nh_grp->adj_index_valid;
	nh_grp->adj_index_valid = 0;
	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];
		nh->offloaded = 0;
	}
	err = mlxsw_sp_nexthop_fib_entries_update(mlxsw_sp, nh_grp);
	if (err)
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to set traps for fib entries.\n");
	if (old_adj_index_valid)
		mlxsw_sp_kvdl_free(mlxsw_sp, nh_grp->adj_index);
}

static void __mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp_nexthop *nh,
					    bool removing)
{
	if (!removing)
		nh->should_offload = 1;
	else
		nh->should_offload = 0;
	nh->update = 1;
}

static void
mlxsw_sp_nexthop_neigh_update(struct mlxsw_sp *mlxsw_sp,
			      struct mlxsw_sp_neigh_entry *neigh_entry,
			      bool removing)
{
	struct mlxsw_sp_nexthop *nh;

	list_for_each_entry(nh, &neigh_entry->nexthop_list,
			    neigh_list_node) {
		__mlxsw_sp_nexthop_neigh_update(nh, removing);
		mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nh_grp);
	}
}

static void mlxsw_sp_nexthop_rif_init(struct mlxsw_sp_nexthop *nh,
				      struct mlxsw_sp_rif *rif)
{
	if (nh->rif)
		return;

	nh->rif = rif;
	list_add(&nh->rif_list_node, &rif->nexthop_list);
}

static void mlxsw_sp_nexthop_rif_fini(struct mlxsw_sp_nexthop *nh)
{
	if (!nh->rif)
		return;

	list_del(&nh->rif_list_node);
	nh->rif = NULL;
}

static int mlxsw_sp_nexthop_neigh_init(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_nexthop *nh)
{
	struct mlxsw_sp_neigh_entry *neigh_entry;
	struct neighbour *n;
	u8 nud_state, dead;
	int err;

	if (!nh->nh_grp->gateway || nh->neigh_entry)
		return 0;

	/* Take a reference of neigh here ensuring that neigh would
	 * not be destructed before the nexthop entry is finished.
	 * The reference is taken either in neigh_lookup() or
	 * in neigh_create() in case n is not found.
	 */
	n = neigh_lookup(nh->nh_grp->neigh_tbl, &nh->gw_addr, nh->rif->dev);
	if (!n) {
		n = neigh_create(nh->nh_grp->neigh_tbl, &nh->gw_addr,
				 nh->rif->dev);
		if (IS_ERR(n))
			return PTR_ERR(n);
		neigh_event_send(n, NULL);
	}
	neigh_entry = mlxsw_sp_neigh_entry_lookup(mlxsw_sp, n);
	if (!neigh_entry) {
		neigh_entry = mlxsw_sp_neigh_entry_create(mlxsw_sp, n);
		if (IS_ERR(neigh_entry)) {
			err = -EINVAL;
			goto err_neigh_entry_create;
		}
	}

	/* If that is the first nexthop connected to that neigh, add to
	 * nexthop_neighs_list
	 */
	if (list_empty(&neigh_entry->nexthop_list))
		list_add_tail(&neigh_entry->nexthop_neighs_list_node,
			      &mlxsw_sp->router->nexthop_neighs_list);

	nh->neigh_entry = neigh_entry;
	list_add_tail(&nh->neigh_list_node, &neigh_entry->nexthop_list);
	read_lock_bh(&n->lock);
	nud_state = n->nud_state;
	dead = n->dead;
	read_unlock_bh(&n->lock);
	__mlxsw_sp_nexthop_neigh_update(nh, !(nud_state & NUD_VALID && !dead));

	return 0;

err_neigh_entry_create:
	neigh_release(n);
	return err;
}

static void mlxsw_sp_nexthop_neigh_fini(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_nexthop *nh)
{
	struct mlxsw_sp_neigh_entry *neigh_entry = nh->neigh_entry;
	struct neighbour *n;

	if (!neigh_entry)
		return;
	n = neigh_entry->key.n;

	__mlxsw_sp_nexthop_neigh_update(nh, true);
	list_del(&nh->neigh_list_node);
	nh->neigh_entry = NULL;

	/* If that is the last nexthop connected to that neigh, remove from
	 * nexthop_neighs_list
	 */
	if (list_empty(&neigh_entry->nexthop_list))
		list_del(&neigh_entry->nexthop_neighs_list_node);

	if (!neigh_entry->connected && list_empty(&neigh_entry->nexthop_list))
		mlxsw_sp_neigh_entry_destroy(mlxsw_sp, neigh_entry);

	neigh_release(n);
}

static bool mlxsw_sp_ipip_netdev_ul_up(struct net_device *ol_dev)
{
	struct net_device *ul_dev = __mlxsw_sp_ipip_netdev_ul_dev_get(ol_dev);

	return ul_dev ? (ul_dev->flags & IFF_UP) : true;
}

static void mlxsw_sp_nexthop_ipip_init(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_nexthop *nh,
				       struct mlxsw_sp_ipip_entry *ipip_entry)
{
	bool removing;

	if (!nh->nh_grp->gateway || nh->ipip_entry)
		return;

	nh->ipip_entry = ipip_entry;
	removing = !mlxsw_sp_ipip_netdev_ul_up(ipip_entry->ol_dev);
	__mlxsw_sp_nexthop_neigh_update(nh, removing);
	mlxsw_sp_nexthop_rif_init(nh, &ipip_entry->ol_lb->common);
}

static void mlxsw_sp_nexthop_ipip_fini(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_nexthop *nh)
{
	struct mlxsw_sp_ipip_entry *ipip_entry = nh->ipip_entry;

	if (!ipip_entry)
		return;

	__mlxsw_sp_nexthop_neigh_update(nh, true);
	nh->ipip_entry = NULL;
}

static bool mlxsw_sp_nexthop4_ipip_type(const struct mlxsw_sp *mlxsw_sp,
					const struct fib_nh *fib_nh,
					enum mlxsw_sp_ipip_type *p_ipipt)
{
	struct net_device *dev = fib_nh->nh_dev;

	return dev &&
	       fib_nh->nh_parent->fib_type == RTN_UNICAST &&
	       mlxsw_sp_netdev_ipip_type(mlxsw_sp, dev, p_ipipt);
}

static void mlxsw_sp_nexthop_type_fini(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_nexthop *nh)
{
	switch (nh->type) {
	case MLXSW_SP_NEXTHOP_TYPE_ETH:
		mlxsw_sp_nexthop_neigh_fini(mlxsw_sp, nh);
		mlxsw_sp_nexthop_rif_fini(nh);
		break;
	case MLXSW_SP_NEXTHOP_TYPE_IPIP:
		mlxsw_sp_nexthop_rif_fini(nh);
		mlxsw_sp_nexthop_ipip_fini(mlxsw_sp, nh);
		break;
	}
}

static int mlxsw_sp_nexthop4_type_init(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_nexthop *nh,
				       struct fib_nh *fib_nh)
{
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	struct net_device *dev = fib_nh->nh_dev;
	struct mlxsw_sp_ipip_entry *ipip_entry;
	struct mlxsw_sp_rif *rif;
	int err;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, dev);
	if (ipip_entry) {
		ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipip_entry->ipipt];
		if (ipip_ops->can_offload(mlxsw_sp, dev,
					  MLXSW_SP_L3_PROTO_IPV4)) {
			nh->type = MLXSW_SP_NEXTHOP_TYPE_IPIP;
			mlxsw_sp_nexthop_ipip_init(mlxsw_sp, nh, ipip_entry);
			return 0;
		}
	}

	nh->type = MLXSW_SP_NEXTHOP_TYPE_ETH;
	rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev);
	if (!rif)
		return 0;

	mlxsw_sp_nexthop_rif_init(nh, rif);
	err = mlxsw_sp_nexthop_neigh_init(mlxsw_sp, nh);
	if (err)
		goto err_neigh_init;

	return 0;

err_neigh_init:
	mlxsw_sp_nexthop_rif_fini(nh);
	return err;
}

static void mlxsw_sp_nexthop4_type_fini(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_nexthop *nh)
{
	mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh);
}

static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_nexthop_group *nh_grp,
				  struct mlxsw_sp_nexthop *nh,
				  struct fib_nh *fib_nh)
{
	struct net_device *dev = fib_nh->nh_dev;
	struct in_device *in_dev;
	int err;

	nh->nh_grp = nh_grp;
	nh->key.fib_nh = fib_nh;
#ifdef CONFIG_IP_ROUTE_MULTIPATH
	nh->nh_weight = fib_nh->nh_weight;
#else
	nh->nh_weight = 1;
#endif
	memcpy(&nh->gw_addr, &fib_nh->nh_gw, sizeof(fib_nh->nh_gw));
	err = mlxsw_sp_nexthop_insert(mlxsw_sp, nh);
	if (err)
		return err;

	mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh);
	list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list);

	if (!dev)
		return 0;

	in_dev = __in_dev_get_rtnl(dev);
	if (in_dev && IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
	    fib_nh->nh_flags & RTNH_F_LINKDOWN)
		return 0;

	err = mlxsw_sp_nexthop4_type_init(mlxsw_sp, nh, fib_nh);
	if (err)
		goto err_nexthop_neigh_init;

	return 0;

err_nexthop_neigh_init:
	mlxsw_sp_nexthop_remove(mlxsw_sp, nh);
	return err;
}

static void mlxsw_sp_nexthop4_fini(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_nexthop *nh)
{
	mlxsw_sp_nexthop4_type_fini(mlxsw_sp, nh);
	list_del(&nh->router_list_node);
	mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh);
	mlxsw_sp_nexthop_remove(mlxsw_sp, nh);
}

static void mlxsw_sp_nexthop4_event(struct mlxsw_sp *mlxsw_sp,
				    unsigned long event, struct fib_nh *fib_nh)
{
	struct mlxsw_sp_nexthop_key key;
	struct mlxsw_sp_nexthop *nh;

	if (mlxsw_sp->router->aborted)
		return;

	key.fib_nh = fib_nh;
	nh = mlxsw_sp_nexthop_lookup(mlxsw_sp, key);
	if (WARN_ON_ONCE(!nh))
		return;

	switch (event) {
	case FIB_EVENT_NH_ADD:
		mlxsw_sp_nexthop4_type_init(mlxsw_sp, nh, fib_nh);
		break;
	case FIB_EVENT_NH_DEL:
		mlxsw_sp_nexthop4_type_fini(mlxsw_sp, nh);
		break;
	}

	mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nh_grp);
}

static void mlxsw_sp_nexthop_rif_update(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_rif *rif)
{
	struct mlxsw_sp_nexthop *nh;
	bool removing;

	list_for_each_entry(nh, &rif->nexthop_list, rif_list_node) {
		switch (nh->type) {
		case MLXSW_SP_NEXTHOP_TYPE_ETH:
			removing = false;
			break;
		case MLXSW_SP_NEXTHOP_TYPE_IPIP:
			removing = !mlxsw_sp_ipip_netdev_ul_up(rif->dev);
			break;
		default:
			WARN_ON(1);
			continue;
		}

		__mlxsw_sp_nexthop_neigh_update(nh, removing);
		mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nh_grp);
	}
}

static void mlxsw_sp_nexthop_rif_migrate(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_rif *old_rif,
					 struct mlxsw_sp_rif *new_rif)
{
	struct mlxsw_sp_nexthop *nh;

	list_splice_init(&old_rif->nexthop_list, &new_rif->nexthop_list);
	list_for_each_entry(nh, &new_rif->nexthop_list, rif_list_node)
		nh->rif = new_rif;
	mlxsw_sp_nexthop_rif_update(mlxsw_sp, new_rif);
}

static void mlxsw_sp_nexthop_rif_gone_sync(struct mlxsw_sp *mlxsw_sp,
					   struct mlxsw_sp_rif *rif)
{
	struct mlxsw_sp_nexthop *nh, *tmp;

	list_for_each_entry_safe(nh, tmp, &rif->nexthop_list, rif_list_node) {
		mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh);
		mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh->nh_grp);
	}
}

static bool mlxsw_sp_fi_is_gateway(const struct mlxsw_sp *mlxsw_sp,
				   const struct fib_info *fi)
{
	return fi->fib_nh->nh_scope == RT_SCOPE_LINK ||
	       mlxsw_sp_nexthop4_ipip_type(mlxsw_sp, fi->fib_nh, NULL);
}

static struct mlxsw_sp_nexthop_group *
mlxsw_sp_nexthop4_group_create(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi)
{
	struct mlxsw_sp_nexthop_group *nh_grp;
	struct mlxsw_sp_nexthop *nh;
	struct fib_nh *fib_nh;
	size_t alloc_size;
	int i;
	int err;

	alloc_size = sizeof(*nh_grp) +
		     fi->fib_nhs * sizeof(struct mlxsw_sp_nexthop);
	nh_grp = kzalloc(alloc_size, GFP_KERNEL);
	if (!nh_grp)
		return ERR_PTR(-ENOMEM);
	nh_grp->priv = fi;
	INIT_LIST_HEAD(&nh_grp->fib_list);
	nh_grp->neigh_tbl = &arp_tbl;

	nh_grp->gateway = mlxsw_sp_fi_is_gateway(mlxsw_sp, fi);
	nh_grp->count = fi->fib_nhs;
	fib_info_hold(fi);
	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];
		fib_nh = &fi->fib_nh[i];
		err = mlxsw_sp_nexthop4_init(mlxsw_sp, nh_grp, nh, fib_nh);
		if (err)
			goto err_nexthop4_init;
	}
	err = mlxsw_sp_nexthop_group_insert(mlxsw_sp, nh_grp);
	if (err)
		goto err_nexthop_group_insert;
	mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp);
	return nh_grp;

err_nexthop_group_insert:
err_nexthop4_init:
	for (i--; i >= 0; i--) {
		nh = &nh_grp->nexthops[i];
		mlxsw_sp_nexthop4_fini(mlxsw_sp, nh);
	}
	fib_info_put(fi);
	kfree(nh_grp);
	return ERR_PTR(err);
}

static void
mlxsw_sp_nexthop4_group_destroy(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_nexthop_group *nh_grp)
{
	struct mlxsw_sp_nexthop *nh;
	int i;

	mlxsw_sp_nexthop_group_remove(mlxsw_sp, nh_grp);
	for (i = 0; i < nh_grp->count; i++) {
		nh = &nh_grp->nexthops[i];
		mlxsw_sp_nexthop4_fini(mlxsw_sp, nh);
	}
	mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp);
	WARN_ON_ONCE(nh_grp->adj_index_valid);
	fib_info_put(mlxsw_sp_nexthop4_group_fi(nh_grp));
	kfree(nh_grp);
}

static int mlxsw_sp_nexthop4_group_get(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_fib_entry *fib_entry,
				       struct fib_info *fi)
{
	struct mlxsw_sp_nexthop_group *nh_grp;

	nh_grp = mlxsw_sp_nexthop4_group_lookup(mlxsw_sp, fi);
	if (!nh_grp) {
		nh_grp = mlxsw_sp_nexthop4_group_create(mlxsw_sp, fi);
		if (IS_ERR(nh_grp))
			return PTR_ERR(nh_grp);
	}
	list_add_tail(&fib_entry->nexthop_group_node, &nh_grp->fib_list);
	fib_entry->nh_group = nh_grp;
	return 0;
}

static void mlxsw_sp_nexthop4_group_put(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;

	list_del(&fib_entry->nexthop_group_node);
	if (!list_empty(&nh_grp->fib_list))
		return;
	mlxsw_sp_nexthop4_group_destroy(mlxsw_sp, nh_grp);
}

static bool
mlxsw_sp_fib4_entry_should_offload(const struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_fib4_entry *fib4_entry;

	fib4_entry = container_of(fib_entry, struct mlxsw_sp_fib4_entry,
				  common);
	return !fib4_entry->tos;
}

static bool
mlxsw_sp_fib_entry_should_offload(const struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_nexthop_group *nh_group = fib_entry->nh_group;

	switch (fib_entry->fib_node->fib->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		if (!mlxsw_sp_fib4_entry_should_offload(fib_entry))
			return false;
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		break;
	}

	switch (fib_entry->type) {
	case MLXSW_SP_FIB_ENTRY_TYPE_REMOTE:
		return !!nh_group->adj_index_valid;
	case MLXSW_SP_FIB_ENTRY_TYPE_LOCAL:
		return !!nh_group->nh_rif;
	case MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP:
		return true;
	default:
		return false;
	}
}

static struct mlxsw_sp_nexthop *
mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp,
		     const struct mlxsw_sp_rt6 *mlxsw_sp_rt6)
{
	int i;

	for (i = 0; i < nh_grp->count; i++) {
		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
		struct fib6_info *rt = mlxsw_sp_rt6->rt;

		if (nh->rif && nh->rif->dev == rt->fib6_nh.nh_dev &&
		    ipv6_addr_equal((const struct in6_addr *) &nh->gw_addr,
				    &rt->fib6_nh.nh_gw))
			return nh;
		continue;
	}

	return NULL;
}

static void
mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
	int i;

	if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL ||
	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP) {
		nh_grp->nexthops->key.fib_nh->nh_flags |= RTNH_F_OFFLOAD;
		return;
	}

	for (i = 0; i < nh_grp->count; i++) {
		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];

		if (nh->offloaded)
			nh->key.fib_nh->nh_flags |= RTNH_F_OFFLOAD;
		else
			nh->key.fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
	}
}

static void
mlxsw_sp_fib4_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
	int i;

	if (!list_is_singular(&nh_grp->fib_list))
		return;

	for (i = 0; i < nh_grp->count; i++) {
		struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];

		nh->key.fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
	}
}

static void
mlxsw_sp_fib6_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;

	fib6_entry = container_of(fib_entry, struct mlxsw_sp_fib6_entry,
				  common);

	if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL) {
		list_first_entry(&fib6_entry->rt6_list, struct mlxsw_sp_rt6,
				 list)->rt->fib6_nh.nh_flags |= RTNH_F_OFFLOAD;
		return;
	}

	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
		struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
		struct mlxsw_sp_nexthop *nh;

		nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6);
		if (nh && nh->offloaded)
			mlxsw_sp_rt6->rt->fib6_nh.nh_flags |= RTNH_F_OFFLOAD;
		else
			mlxsw_sp_rt6->rt->fib6_nh.nh_flags &= ~RTNH_F_OFFLOAD;
	}
}

static void
mlxsw_sp_fib6_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;

	fib6_entry = container_of(fib_entry, struct mlxsw_sp_fib6_entry,
				  common);
	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
		struct fib6_info *rt = mlxsw_sp_rt6->rt;

		rt->fib6_nh.nh_flags &= ~RTNH_F_OFFLOAD;
	}
}

static void mlxsw_sp_fib_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
{
	switch (fib_entry->fib_node->fib->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		mlxsw_sp_fib4_entry_offload_set(fib_entry);
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		mlxsw_sp_fib6_entry_offload_set(fib_entry);
		break;
	}
}

static void
mlxsw_sp_fib_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
{
	switch (fib_entry->fib_node->fib->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		mlxsw_sp_fib4_entry_offload_unset(fib_entry);
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		mlxsw_sp_fib6_entry_offload_unset(fib_entry);
		break;
	}
}

static void
mlxsw_sp_fib_entry_offload_refresh(struct mlxsw_sp_fib_entry *fib_entry,
				   enum mlxsw_reg_ralue_op op, int err)
{
	switch (op) {
	case MLXSW_REG_RALUE_OP_WRITE_DELETE:
		return mlxsw_sp_fib_entry_offload_unset(fib_entry);
	case MLXSW_REG_RALUE_OP_WRITE_WRITE:
		if (err)
			return;
		if (mlxsw_sp_fib_entry_should_offload(fib_entry))
			mlxsw_sp_fib_entry_offload_set(fib_entry);
		else
			mlxsw_sp_fib_entry_offload_unset(fib_entry);
		return;
	default:
		return;
	}
}

static void
mlxsw_sp_fib_entry_ralue_pack(char *ralue_pl,
			      const struct mlxsw_sp_fib_entry *fib_entry,
			      enum mlxsw_reg_ralue_op op)
{
	struct mlxsw_sp_fib *fib = fib_entry->fib_node->fib;
	enum mlxsw_reg_ralxx_protocol proto;
	u32 *p_dip;

	proto = (enum mlxsw_reg_ralxx_protocol) fib->proto;

	switch (fib->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		p_dip = (u32 *) fib_entry->fib_node->key.addr;
		mlxsw_reg_ralue_pack4(ralue_pl, proto, op, fib->vr->id,
				      fib_entry->fib_node->key.prefix_len,
				      *p_dip);
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		mlxsw_reg_ralue_pack6(ralue_pl, proto, op, fib->vr->id,
				      fib_entry->fib_node->key.prefix_len,
				      fib_entry->fib_node->key.addr);
		break;
	}
}

static int mlxsw_sp_fib_entry_op_remote(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib_entry *fib_entry,
					enum mlxsw_reg_ralue_op op)
{
	char ralue_pl[MLXSW_REG_RALUE_LEN];
	enum mlxsw_reg_ralue_trap_action trap_action;
	u16 trap_id = 0;
	u32 adjacency_index = 0;
	u16 ecmp_size = 0;

	/* In case the nexthop group adjacency index is valid, use it
	 * with provided ECMP size. Otherwise, setup trap and pass
	 * traffic to kernel.
	 */
	if (mlxsw_sp_fib_entry_should_offload(fib_entry)) {
		trap_action = MLXSW_REG_RALUE_TRAP_ACTION_NOP;
		adjacency_index = fib_entry->nh_group->adj_index;
		ecmp_size = fib_entry->nh_group->ecmp_size;
	} else {
		trap_action = MLXSW_REG_RALUE_TRAP_ACTION_TRAP;
		trap_id = MLXSW_TRAP_ID_RTR_INGRESS0;
	}

	mlxsw_sp_fib_entry_ralue_pack(ralue_pl, fib_entry, op);
	mlxsw_reg_ralue_act_remote_pack(ralue_pl, trap_action, trap_id,
					adjacency_index, ecmp_size);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl);
}

static int mlxsw_sp_fib_entry_op_local(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_fib_entry *fib_entry,
				       enum mlxsw_reg_ralue_op op)
{
	struct mlxsw_sp_rif *rif = fib_entry->nh_group->nh_rif;
	enum mlxsw_reg_ralue_trap_action trap_action;
	char ralue_pl[MLXSW_REG_RALUE_LEN];
	u16 trap_id = 0;
	u16 rif_index = 0;

	if (mlxsw_sp_fib_entry_should_offload(fib_entry)) {
		trap_action = MLXSW_REG_RALUE_TRAP_ACTION_NOP;
		rif_index = rif->rif_index;
	} else {
		trap_action = MLXSW_REG_RALUE_TRAP_ACTION_TRAP;
		trap_id = MLXSW_TRAP_ID_RTR_INGRESS0;
	}

	mlxsw_sp_fib_entry_ralue_pack(ralue_pl, fib_entry, op);
	mlxsw_reg_ralue_act_local_pack(ralue_pl, trap_action, trap_id,
				       rif_index);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl);
}

static int mlxsw_sp_fib_entry_op_trap(struct mlxsw_sp *mlxsw_sp,
				      struct mlxsw_sp_fib_entry *fib_entry,
				      enum mlxsw_reg_ralue_op op)
{
	char ralue_pl[MLXSW_REG_RALUE_LEN];

	mlxsw_sp_fib_entry_ralue_pack(ralue_pl, fib_entry, op);
	mlxsw_reg_ralue_act_ip2me_pack(ralue_pl);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl);
}

static int
mlxsw_sp_fib_entry_op_ipip_decap(struct mlxsw_sp *mlxsw_sp,
				 struct mlxsw_sp_fib_entry *fib_entry,
				 enum mlxsw_reg_ralue_op op)
{
	struct mlxsw_sp_ipip_entry *ipip_entry = fib_entry->decap.ipip_entry;
	const struct mlxsw_sp_ipip_ops *ipip_ops;

	if (WARN_ON(!ipip_entry))
		return -EINVAL;

	ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipip_entry->ipipt];
	return ipip_ops->fib_entry_op(mlxsw_sp, ipip_entry, op,
				      fib_entry->decap.tunnel_index);
}

static int __mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_fib_entry *fib_entry,
				   enum mlxsw_reg_ralue_op op)
{
	switch (fib_entry->type) {
	case MLXSW_SP_FIB_ENTRY_TYPE_REMOTE:
		return mlxsw_sp_fib_entry_op_remote(mlxsw_sp, fib_entry, op);
	case MLXSW_SP_FIB_ENTRY_TYPE_LOCAL:
		return mlxsw_sp_fib_entry_op_local(mlxsw_sp, fib_entry, op);
	case MLXSW_SP_FIB_ENTRY_TYPE_TRAP:
		return mlxsw_sp_fib_entry_op_trap(mlxsw_sp, fib_entry, op);
	case MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP:
		return mlxsw_sp_fib_entry_op_ipip_decap(mlxsw_sp,
							fib_entry, op);
	}
	return -EINVAL;
}

static int mlxsw_sp_fib_entry_op(struct mlxsw_sp *mlxsw_sp,
				 struct mlxsw_sp_fib_entry *fib_entry,
				 enum mlxsw_reg_ralue_op op)
{
	int err = __mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry, op);

	mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, err);

	return err;
}

static int mlxsw_sp_fib_entry_update(struct mlxsw_sp *mlxsw_sp,
				     struct mlxsw_sp_fib_entry *fib_entry)
{
	return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry,
				     MLXSW_REG_RALUE_OP_WRITE_WRITE);
}

static int mlxsw_sp_fib_entry_del(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_fib_entry *fib_entry)
{
	return mlxsw_sp_fib_entry_op(mlxsw_sp, fib_entry,
				     MLXSW_REG_RALUE_OP_WRITE_DELETE);
}

static int
mlxsw_sp_fib4_entry_type_set(struct mlxsw_sp *mlxsw_sp,
			     const struct fib_entry_notifier_info *fen_info,
			     struct mlxsw_sp_fib_entry *fib_entry)
{
	union mlxsw_sp_l3addr dip = { .addr4 = htonl(fen_info->dst) };
	struct net_device *dev = fen_info->fi->fib_dev;
	struct mlxsw_sp_ipip_entry *ipip_entry;
	struct fib_info *fi = fen_info->fi;

	switch (fen_info->type) {
	case RTN_LOCAL:
		ipip_entry = mlxsw_sp_ipip_entry_find_by_decap(mlxsw_sp, dev,
						 MLXSW_SP_L3_PROTO_IPV4, dip);
		if (ipip_entry && ipip_entry->ol_dev->flags & IFF_UP) {
			fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP;
			return mlxsw_sp_fib_entry_decap_init(mlxsw_sp,
							     fib_entry,
							     ipip_entry);
		}
		/* fall through */
	case RTN_BROADCAST:
		fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP;
		return 0;
	case RTN_UNREACHABLE: /* fall through */
	case RTN_BLACKHOLE: /* fall through */
	case RTN_PROHIBIT:
		/* Packets hitting these routes need to be trapped, but
		 * can do so with a lower priority than packets directed
		 * at the host, so use action type local instead of trap.
		 */
		fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_LOCAL;
		return 0;
	case RTN_UNICAST:
		if (mlxsw_sp_fi_is_gateway(mlxsw_sp, fi))
			fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_REMOTE;
		else
			fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_LOCAL;
		return 0;
	default:
		return -EINVAL;
	}
}

static struct mlxsw_sp_fib4_entry *
mlxsw_sp_fib4_entry_create(struct mlxsw_sp *mlxsw_sp,
			   struct mlxsw_sp_fib_node *fib_node,
			   const struct fib_entry_notifier_info *fen_info)
{
	struct mlxsw_sp_fib4_entry *fib4_entry;
	struct mlxsw_sp_fib_entry *fib_entry;
	int err;

	fib4_entry = kzalloc(sizeof(*fib4_entry), GFP_KERNEL);
	if (!fib4_entry)
		return ERR_PTR(-ENOMEM);
	fib_entry = &fib4_entry->common;

	err = mlxsw_sp_fib4_entry_type_set(mlxsw_sp, fen_info, fib_entry);
	if (err)
		goto err_fib4_entry_type_set;

	err = mlxsw_sp_nexthop4_group_get(mlxsw_sp, fib_entry, fen_info->fi);
	if (err)
		goto err_nexthop4_group_get;

	fib4_entry->prio = fen_info->fi->fib_priority;
	fib4_entry->tb_id = fen_info->tb_id;
	fib4_entry->type = fen_info->type;
	fib4_entry->tos = fen_info->tos;

	fib_entry->fib_node = fib_node;

	return fib4_entry;

err_nexthop4_group_get:
err_fib4_entry_type_set:
	kfree(fib4_entry);
	return ERR_PTR(err);
}

static void mlxsw_sp_fib4_entry_destroy(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib4_entry *fib4_entry)
{
	mlxsw_sp_nexthop4_group_put(mlxsw_sp, &fib4_entry->common);
	kfree(fib4_entry);
}

static struct mlxsw_sp_fib4_entry *
mlxsw_sp_fib4_entry_lookup(struct mlxsw_sp *mlxsw_sp,
			   const struct fib_entry_notifier_info *fen_info)
{
	struct mlxsw_sp_fib4_entry *fib4_entry;
	struct mlxsw_sp_fib_node *fib_node;
	struct mlxsw_sp_fib *fib;
	struct mlxsw_sp_vr *vr;

	vr = mlxsw_sp_vr_find(mlxsw_sp, fen_info->tb_id);
	if (!vr)
		return NULL;
	fib = mlxsw_sp_vr_fib(vr, MLXSW_SP_L3_PROTO_IPV4);

	fib_node = mlxsw_sp_fib_node_lookup(fib, &fen_info->dst,
					    sizeof(fen_info->dst),
					    fen_info->dst_len);
	if (!fib_node)
		return NULL;

	list_for_each_entry(fib4_entry, &fib_node->entry_list, common.list) {
		if (fib4_entry->tb_id == fen_info->tb_id &&
		    fib4_entry->tos == fen_info->tos &&
		    fib4_entry->type == fen_info->type &&
		    mlxsw_sp_nexthop4_group_fi(fib4_entry->common.nh_group) ==
		    fen_info->fi) {
			return fib4_entry;
		}
	}

	return NULL;
}

static const struct rhashtable_params mlxsw_sp_fib_ht_params = {
	.key_offset = offsetof(struct mlxsw_sp_fib_node, key),
	.head_offset = offsetof(struct mlxsw_sp_fib_node, ht_node),
	.key_len = sizeof(struct mlxsw_sp_fib_key),
	.automatic_shrinking = true,
};

static int mlxsw_sp_fib_node_insert(struct mlxsw_sp_fib *fib,
				    struct mlxsw_sp_fib_node *fib_node)
{
	return rhashtable_insert_fast(&fib->ht, &fib_node->ht_node,
				      mlxsw_sp_fib_ht_params);
}

static void mlxsw_sp_fib_node_remove(struct mlxsw_sp_fib *fib,
				     struct mlxsw_sp_fib_node *fib_node)
{
	rhashtable_remove_fast(&fib->ht, &fib_node->ht_node,
			       mlxsw_sp_fib_ht_params);
}

static struct mlxsw_sp_fib_node *
mlxsw_sp_fib_node_lookup(struct mlxsw_sp_fib *fib, const void *addr,
			 size_t addr_len, unsigned char prefix_len)
{
	struct mlxsw_sp_fib_key key;

	memset(&key, 0, sizeof(key));
	memcpy(key.addr, addr, addr_len);
	key.prefix_len = prefix_len;
	return rhashtable_lookup_fast(&fib->ht, &key, mlxsw_sp_fib_ht_params);
}

static struct mlxsw_sp_fib_node *
mlxsw_sp_fib_node_create(struct mlxsw_sp_fib *fib, const void *addr,
			 size_t addr_len, unsigned char prefix_len)
{
	struct mlxsw_sp_fib_node *fib_node;

	fib_node = kzalloc(sizeof(*fib_node), GFP_KERNEL);
	if (!fib_node)
		return NULL;

	INIT_LIST_HEAD(&fib_node->entry_list);
	list_add(&fib_node->list, &fib->node_list);
	memcpy(fib_node->key.addr, addr, addr_len);
	fib_node->key.prefix_len = prefix_len;

	return fib_node;
}

static void mlxsw_sp_fib_node_destroy(struct mlxsw_sp_fib_node *fib_node)
{
	list_del(&fib_node->list);
	WARN_ON(!list_empty(&fib_node->entry_list));
	kfree(fib_node);
}

static bool
mlxsw_sp_fib_node_entry_is_first(const struct mlxsw_sp_fib_node *fib_node,
				 const struct mlxsw_sp_fib_entry *fib_entry)
{
	return list_first_entry(&fib_node->entry_list,
				struct mlxsw_sp_fib_entry, list) == fib_entry;
}

static int mlxsw_sp_fib_lpm_tree_link(struct mlxsw_sp *mlxsw_sp,
				      struct mlxsw_sp_fib_node *fib_node)
{
	struct mlxsw_sp_prefix_usage req_prefix_usage;
	struct mlxsw_sp_fib *fib = fib_node->fib;
	struct mlxsw_sp_lpm_tree *lpm_tree;
	int err;

	lpm_tree = mlxsw_sp->router->lpm.proto_trees[fib->proto];
	if (lpm_tree->prefix_ref_count[fib_node->key.prefix_len] != 0)
		goto out;

	mlxsw_sp_prefix_usage_cpy(&req_prefix_usage, &lpm_tree->prefix_usage);
	mlxsw_sp_prefix_usage_set(&req_prefix_usage, fib_node->key.prefix_len);
	lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, &req_prefix_usage,
					 fib->proto);
	if (IS_ERR(lpm_tree))
		return PTR_ERR(lpm_tree);

	err = mlxsw_sp_vrs_lpm_tree_replace(mlxsw_sp, fib, lpm_tree);
	if (err)
		goto err_lpm_tree_replace;

out:
	lpm_tree->prefix_ref_count[fib_node->key.prefix_len]++;
	return 0;

err_lpm_tree_replace:
	mlxsw_sp_lpm_tree_put(mlxsw_sp, lpm_tree);
	return err;
}

static void mlxsw_sp_fib_lpm_tree_unlink(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_fib_node *fib_node)
{
	struct mlxsw_sp_lpm_tree *lpm_tree = fib_node->fib->lpm_tree;
	struct mlxsw_sp_prefix_usage req_prefix_usage;
	struct mlxsw_sp_fib *fib = fib_node->fib;
	int err;

	if (--lpm_tree->prefix_ref_count[fib_node->key.prefix_len] != 0)
		return;
	/* Try to construct a new LPM tree from the current prefix usage
	 * minus the unused one. If we fail, continue using the old one.
	 */
	mlxsw_sp_prefix_usage_cpy(&req_prefix_usage, &lpm_tree->prefix_usage);
	mlxsw_sp_prefix_usage_clear(&req_prefix_usage,
				    fib_node->key.prefix_len);
	lpm_tree = mlxsw_sp_lpm_tree_get(mlxsw_sp, &req_prefix_usage,
					 fib->proto);
	if (IS_ERR(lpm_tree))
		return;

	err = mlxsw_sp_vrs_lpm_tree_replace(mlxsw_sp, fib, lpm_tree);
	if (err)
		goto err_lpm_tree_replace;

	return;

err_lpm_tree_replace:
	mlxsw_sp_lpm_tree_put(mlxsw_sp, lpm_tree);
}

static int mlxsw_sp_fib_node_init(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_fib_node *fib_node,
				  struct mlxsw_sp_fib *fib)
{
	int err;

	err = mlxsw_sp_fib_node_insert(fib, fib_node);
	if (err)
		return err;
	fib_node->fib = fib;

	err = mlxsw_sp_fib_lpm_tree_link(mlxsw_sp, fib_node);
	if (err)
		goto err_fib_lpm_tree_link;

	return 0;

err_fib_lpm_tree_link:
	fib_node->fib = NULL;
	mlxsw_sp_fib_node_remove(fib, fib_node);
	return err;
}

static void mlxsw_sp_fib_node_fini(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_fib_node *fib_node)
{
	struct mlxsw_sp_fib *fib = fib_node->fib;

	mlxsw_sp_fib_lpm_tree_unlink(mlxsw_sp, fib_node);
	fib_node->fib = NULL;
	mlxsw_sp_fib_node_remove(fib, fib_node);
}

static struct mlxsw_sp_fib_node *
mlxsw_sp_fib_node_get(struct mlxsw_sp *mlxsw_sp, u32 tb_id, const void *addr,
		      size_t addr_len, unsigned char prefix_len,
		      enum mlxsw_sp_l3proto proto)
{
	struct mlxsw_sp_fib_node *fib_node;
	struct mlxsw_sp_fib *fib;
	struct mlxsw_sp_vr *vr;
	int err;

	vr = mlxsw_sp_vr_get(mlxsw_sp, tb_id, NULL);
	if (IS_ERR(vr))
		return ERR_CAST(vr);
	fib = mlxsw_sp_vr_fib(vr, proto);

	fib_node = mlxsw_sp_fib_node_lookup(fib, addr, addr_len, prefix_len);
	if (fib_node)
		return fib_node;

	fib_node = mlxsw_sp_fib_node_create(fib, addr, addr_len, prefix_len);
	if (!fib_node) {
		err = -ENOMEM;
		goto err_fib_node_create;
	}

	err = mlxsw_sp_fib_node_init(mlxsw_sp, fib_node, fib);
	if (err)
		goto err_fib_node_init;

	return fib_node;

err_fib_node_init:
	mlxsw_sp_fib_node_destroy(fib_node);
err_fib_node_create:
	mlxsw_sp_vr_put(mlxsw_sp, vr);
	return ERR_PTR(err);
}

static void mlxsw_sp_fib_node_put(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_fib_node *fib_node)
{
	struct mlxsw_sp_vr *vr = fib_node->fib->vr;

	if (!list_empty(&fib_node->entry_list))
		return;
	mlxsw_sp_fib_node_fini(mlxsw_sp, fib_node);
	mlxsw_sp_fib_node_destroy(fib_node);
	mlxsw_sp_vr_put(mlxsw_sp, vr);
}

static struct mlxsw_sp_fib4_entry *
mlxsw_sp_fib4_node_entry_find(const struct mlxsw_sp_fib_node *fib_node,
			      const struct mlxsw_sp_fib4_entry *new4_entry)
{
	struct mlxsw_sp_fib4_entry *fib4_entry;

	list_for_each_entry(fib4_entry, &fib_node->entry_list, common.list) {
		if (fib4_entry->tb_id > new4_entry->tb_id)
			continue;
		if (fib4_entry->tb_id != new4_entry->tb_id)
			break;
		if (fib4_entry->tos > new4_entry->tos)
			continue;
		if (fib4_entry->prio >= new4_entry->prio ||
		    fib4_entry->tos < new4_entry->tos)
			return fib4_entry;
	}

	return NULL;
}

static int
mlxsw_sp_fib4_node_list_append(struct mlxsw_sp_fib4_entry *fib4_entry,
			       struct mlxsw_sp_fib4_entry *new4_entry)
{
	struct mlxsw_sp_fib_node *fib_node;

	if (WARN_ON(!fib4_entry))
		return -EINVAL;

	fib_node = fib4_entry->common.fib_node;
	list_for_each_entry_from(fib4_entry, &fib_node->entry_list,
				 common.list) {
		if (fib4_entry->tb_id != new4_entry->tb_id ||
		    fib4_entry->tos != new4_entry->tos ||
		    fib4_entry->prio != new4_entry->prio)
			break;
	}

	list_add_tail(&new4_entry->common.list, &fib4_entry->common.list);
	return 0;
}

static int
mlxsw_sp_fib4_node_list_insert(struct mlxsw_sp_fib4_entry *new4_entry,
			       bool replace, bool append)
{
	struct mlxsw_sp_fib_node *fib_node = new4_entry->common.fib_node;
	struct mlxsw_sp_fib4_entry *fib4_entry;

	fib4_entry = mlxsw_sp_fib4_node_entry_find(fib_node, new4_entry);

	if (append)
		return mlxsw_sp_fib4_node_list_append(fib4_entry, new4_entry);
	if (replace && WARN_ON(!fib4_entry))
		return -EINVAL;

	/* Insert new entry before replaced one, so that we can later
	 * remove the second.
	 */
	if (fib4_entry) {
		list_add_tail(&new4_entry->common.list,
			      &fib4_entry->common.list);
	} else {
		struct mlxsw_sp_fib4_entry *last;

		list_for_each_entry(last, &fib_node->entry_list, common.list) {
			if (new4_entry->tb_id > last->tb_id)
				break;
			fib4_entry = last;
		}

		if (fib4_entry)
			list_add(&new4_entry->common.list,
				 &fib4_entry->common.list);
		else
			list_add(&new4_entry->common.list,
				 &fib_node->entry_list);
	}

	return 0;
}

static void
mlxsw_sp_fib4_node_list_remove(struct mlxsw_sp_fib4_entry *fib4_entry)
{
	list_del(&fib4_entry->common.list);
}

static int mlxsw_sp_fib_node_entry_add(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;

	if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry))
		return 0;

	/* To prevent packet loss, overwrite the previously offloaded
	 * entry.
	 */
	if (!list_is_singular(&fib_node->entry_list)) {
		enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_DELETE;
		struct mlxsw_sp_fib_entry *n = list_next_entry(fib_entry, list);

		mlxsw_sp_fib_entry_offload_refresh(n, op, 0);
	}

	return mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
}

static void mlxsw_sp_fib_node_entry_del(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_fib_node *fib_node = fib_entry->fib_node;

	if (!mlxsw_sp_fib_node_entry_is_first(fib_node, fib_entry))
		return;

	/* Promote the next entry by overwriting the deleted entry */
	if (!list_is_singular(&fib_node->entry_list)) {
		struct mlxsw_sp_fib_entry *n = list_next_entry(fib_entry, list);
		enum mlxsw_reg_ralue_op op = MLXSW_REG_RALUE_OP_WRITE_DELETE;

		mlxsw_sp_fib_entry_update(mlxsw_sp, n);
		mlxsw_sp_fib_entry_offload_refresh(fib_entry, op, 0);
		return;
	}

	mlxsw_sp_fib_entry_del(mlxsw_sp, fib_entry);
}

static int mlxsw_sp_fib4_node_entry_link(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_fib4_entry *fib4_entry,
					 bool replace, bool append)
{
	int err;

	err = mlxsw_sp_fib4_node_list_insert(fib4_entry, replace, append);
	if (err)
		return err;

	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib4_entry->common);
	if (err)
		goto err_fib_node_entry_add;

	return 0;

err_fib_node_entry_add:
	mlxsw_sp_fib4_node_list_remove(fib4_entry);
	return err;
}

static void
mlxsw_sp_fib4_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_fib4_entry *fib4_entry)
{
	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib4_entry->common);
	mlxsw_sp_fib4_node_list_remove(fib4_entry);

	if (fib4_entry->common.type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP)
		mlxsw_sp_fib_entry_decap_fini(mlxsw_sp, &fib4_entry->common);
}

static void mlxsw_sp_fib4_entry_replace(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib4_entry *fib4_entry,
					bool replace)
{
	struct mlxsw_sp_fib_node *fib_node = fib4_entry->common.fib_node;
	struct mlxsw_sp_fib4_entry *replaced;

	if (!replace)
		return;

	/* We inserted the new entry before replaced one */
	replaced = list_next_entry(fib4_entry, common.list);

	mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, replaced);
	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, replaced);
	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
}

static int
mlxsw_sp_router_fib4_add(struct mlxsw_sp *mlxsw_sp,
			 const struct fib_entry_notifier_info *fen_info,
			 bool replace, bool append)
{
	struct mlxsw_sp_fib4_entry *fib4_entry;
	struct mlxsw_sp_fib_node *fib_node;
	int err;

	if (mlxsw_sp->router->aborted)
		return 0;

	fib_node = mlxsw_sp_fib_node_get(mlxsw_sp, fen_info->tb_id,
					 &fen_info->dst, sizeof(fen_info->dst),
					 fen_info->dst_len,
					 MLXSW_SP_L3_PROTO_IPV4);
	if (IS_ERR(fib_node)) {
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to get FIB node\n");
		return PTR_ERR(fib_node);
	}

	fib4_entry = mlxsw_sp_fib4_entry_create(mlxsw_sp, fib_node, fen_info);
	if (IS_ERR(fib4_entry)) {
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to create FIB entry\n");
		err = PTR_ERR(fib4_entry);
		goto err_fib4_entry_create;
	}

	err = mlxsw_sp_fib4_node_entry_link(mlxsw_sp, fib4_entry, replace,
					    append);
	if (err) {
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to link FIB entry to node\n");
		goto err_fib4_node_entry_link;
	}

	mlxsw_sp_fib4_entry_replace(mlxsw_sp, fib4_entry, replace);

	return 0;

err_fib4_node_entry_link:
	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
err_fib4_entry_create:
	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
	return err;
}

static void mlxsw_sp_router_fib4_del(struct mlxsw_sp *mlxsw_sp,
				     struct fib_entry_notifier_info *fen_info)
{
	struct mlxsw_sp_fib4_entry *fib4_entry;
	struct mlxsw_sp_fib_node *fib_node;

	if (mlxsw_sp->router->aborted)
		return;

	fib4_entry = mlxsw_sp_fib4_entry_lookup(mlxsw_sp, fen_info);
	if (WARN_ON(!fib4_entry))
		return;
	fib_node = fib4_entry->common.fib_node;

	mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, fib4_entry);
	mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
}

static bool mlxsw_sp_fib6_rt_should_ignore(const struct fib6_info *rt)
{
	/* Packets with link-local destination IP arriving to the router
	 * are trapped to the CPU, so no need to program specific routes
	 * for them.
	 */
	if (ipv6_addr_type(&rt->fib6_dst.addr) & IPV6_ADDR_LINKLOCAL)
		return true;

	/* Multicast routes aren't supported, so ignore them. Neighbour
	 * Discovery packets are specifically trapped.
	 */
	if (ipv6_addr_type(&rt->fib6_dst.addr) & IPV6_ADDR_MULTICAST)
		return true;

	/* Cloned routes are irrelevant in the forwarding path. */
	if (rt->fib6_flags & RTF_CACHE)
		return true;

	return false;
}

static struct mlxsw_sp_rt6 *mlxsw_sp_rt6_create(struct fib6_info *rt)
{
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;

	mlxsw_sp_rt6 = kzalloc(sizeof(*mlxsw_sp_rt6), GFP_KERNEL);
	if (!mlxsw_sp_rt6)
		return ERR_PTR(-ENOMEM);

	/* In case of route replace, replaced route is deleted with
	 * no notification. Take reference to prevent accessing freed
	 * memory.
	 */
	mlxsw_sp_rt6->rt = rt;
	fib6_info_hold(rt);

	return mlxsw_sp_rt6;
}

#if IS_ENABLED(CONFIG_IPV6)
static void mlxsw_sp_rt6_release(struct fib6_info *rt)
{
	fib6_info_release(rt);
}
#else
static void mlxsw_sp_rt6_release(struct fib6_info *rt)
{
}
#endif

static void mlxsw_sp_rt6_destroy(struct mlxsw_sp_rt6 *mlxsw_sp_rt6)
{
	mlxsw_sp_rt6_release(mlxsw_sp_rt6->rt);
	kfree(mlxsw_sp_rt6);
}

static struct fib6_info *
mlxsw_sp_fib6_entry_rt(const struct mlxsw_sp_fib6_entry *fib6_entry)
{
	return list_first_entry(&fib6_entry->rt6_list, struct mlxsw_sp_rt6,
				list)->rt;
}

static struct mlxsw_sp_fib6_entry *
mlxsw_sp_fib6_node_mp_entry_find(const struct mlxsw_sp_fib_node *fib_node,
				 const struct fib6_info *nrt, bool append)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;

	if (!append)
		return NULL;

	list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) {
		struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry);

		/* RT6_TABLE_LOCAL and RT6_TABLE_MAIN share the same
		 * virtual router.
		 */
		if (rt->fib6_table->tb6_id > nrt->fib6_table->tb6_id)
			continue;
		if (rt->fib6_table->tb6_id != nrt->fib6_table->tb6_id)
			break;
		if (rt->fib6_metric < nrt->fib6_metric)
			continue;
		if (rt->fib6_metric == nrt->fib6_metric)
			return fib6_entry;
		if (rt->fib6_metric > nrt->fib6_metric)
			break;
	}

	return NULL;
}

static struct mlxsw_sp_rt6 *
mlxsw_sp_fib6_entry_rt_find(const struct mlxsw_sp_fib6_entry *fib6_entry,
			    const struct fib6_info *rt)
{
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;

	list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
		if (mlxsw_sp_rt6->rt == rt)
			return mlxsw_sp_rt6;
	}

	return NULL;
}

static bool mlxsw_sp_nexthop6_ipip_type(const struct mlxsw_sp *mlxsw_sp,
					const struct fib6_info *rt,
					enum mlxsw_sp_ipip_type *ret)
{
	return rt->fib6_nh.nh_dev &&
	       mlxsw_sp_netdev_ipip_type(mlxsw_sp, rt->fib6_nh.nh_dev, ret);
}

static int mlxsw_sp_nexthop6_type_init(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_nexthop_group *nh_grp,
				       struct mlxsw_sp_nexthop *nh,
				       const struct fib6_info *rt)
{
	const struct mlxsw_sp_ipip_ops *ipip_ops;
	struct mlxsw_sp_ipip_entry *ipip_entry;
	struct net_device *dev = rt->fib6_nh.nh_dev;
	struct mlxsw_sp_rif *rif;
	int err;

	ipip_entry = mlxsw_sp_ipip_entry_find_by_ol_dev(mlxsw_sp, dev);
	if (ipip_entry) {
		ipip_ops = mlxsw_sp->router->ipip_ops_arr[ipip_entry->ipipt];
		if (ipip_ops->can_offload(mlxsw_sp, dev,
					  MLXSW_SP_L3_PROTO_IPV6)) {
			nh->type = MLXSW_SP_NEXTHOP_TYPE_IPIP;
			mlxsw_sp_nexthop_ipip_init(mlxsw_sp, nh, ipip_entry);
			return 0;
		}
	}

	nh->type = MLXSW_SP_NEXTHOP_TYPE_ETH;
	rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, dev);
	if (!rif)
		return 0;
	mlxsw_sp_nexthop_rif_init(nh, rif);

	err = mlxsw_sp_nexthop_neigh_init(mlxsw_sp, nh);
	if (err)
		goto err_nexthop_neigh_init;

	return 0;

err_nexthop_neigh_init:
	mlxsw_sp_nexthop_rif_fini(nh);
	return err;
}

static void mlxsw_sp_nexthop6_type_fini(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_nexthop *nh)
{
	mlxsw_sp_nexthop_type_fini(mlxsw_sp, nh);
}

static int mlxsw_sp_nexthop6_init(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_nexthop_group *nh_grp,
				  struct mlxsw_sp_nexthop *nh,
				  const struct fib6_info *rt)
{
	struct net_device *dev = rt->fib6_nh.nh_dev;

	nh->nh_grp = nh_grp;
	nh->nh_weight = rt->fib6_nh.nh_weight;
	memcpy(&nh->gw_addr, &rt->fib6_nh.nh_gw, sizeof(nh->gw_addr));
	mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh);

	list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list);

	if (!dev)
		return 0;
	nh->ifindex = dev->ifindex;

	return mlxsw_sp_nexthop6_type_init(mlxsw_sp, nh_grp, nh, rt);
}

static void mlxsw_sp_nexthop6_fini(struct mlxsw_sp *mlxsw_sp,
				   struct mlxsw_sp_nexthop *nh)
{
	mlxsw_sp_nexthop6_type_fini(mlxsw_sp, nh);
	list_del(&nh->router_list_node);
	mlxsw_sp_nexthop_counter_free(mlxsw_sp, nh);
}

static bool mlxsw_sp_rt6_is_gateway(const struct mlxsw_sp *mlxsw_sp,
				    const struct fib6_info *rt)
{
	return rt->fib6_flags & RTF_GATEWAY ||
	       mlxsw_sp_nexthop6_ipip_type(mlxsw_sp, rt, NULL);
}

static struct mlxsw_sp_nexthop_group *
mlxsw_sp_nexthop6_group_create(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_fib6_entry *fib6_entry)
{
	struct mlxsw_sp_nexthop_group *nh_grp;
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
	struct mlxsw_sp_nexthop *nh;
	size_t alloc_size;
	int i = 0;
	int err;

	alloc_size = sizeof(*nh_grp) +
		     fib6_entry->nrt6 * sizeof(struct mlxsw_sp_nexthop);
	nh_grp = kzalloc(alloc_size, GFP_KERNEL);
	if (!nh_grp)
		return ERR_PTR(-ENOMEM);
	INIT_LIST_HEAD(&nh_grp->fib_list);
#if IS_ENABLED(CONFIG_IPV6)
	nh_grp->neigh_tbl = &nd_tbl;
#endif
	mlxsw_sp_rt6 = list_first_entry(&fib6_entry->rt6_list,
					struct mlxsw_sp_rt6, list);
	nh_grp->gateway = mlxsw_sp_rt6_is_gateway(mlxsw_sp, mlxsw_sp_rt6->rt);
	nh_grp->count = fib6_entry->nrt6;
	for (i = 0; i < nh_grp->count; i++) {
		struct fib6_info *rt = mlxsw_sp_rt6->rt;

		nh = &nh_grp->nexthops[i];
		err = mlxsw_sp_nexthop6_init(mlxsw_sp, nh_grp, nh, rt);
		if (err)
			goto err_nexthop6_init;
		mlxsw_sp_rt6 = list_next_entry(mlxsw_sp_rt6, list);
	}

	err = mlxsw_sp_nexthop_group_insert(mlxsw_sp, nh_grp);
	if (err)
		goto err_nexthop_group_insert;

	mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp);
	return nh_grp;

err_nexthop_group_insert:
err_nexthop6_init:
	for (i--; i >= 0; i--) {
		nh = &nh_grp->nexthops[i];
		mlxsw_sp_nexthop6_fini(mlxsw_sp, nh);
	}
	kfree(nh_grp);
	return ERR_PTR(err);
}

static void
mlxsw_sp_nexthop6_group_destroy(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_nexthop_group *nh_grp)
{
	struct mlxsw_sp_nexthop *nh;
	int i = nh_grp->count;

	mlxsw_sp_nexthop_group_remove(mlxsw_sp, nh_grp);
	for (i--; i >= 0; i--) {
		nh = &nh_grp->nexthops[i];
		mlxsw_sp_nexthop6_fini(mlxsw_sp, nh);
	}
	mlxsw_sp_nexthop_group_refresh(mlxsw_sp, nh_grp);
	WARN_ON(nh_grp->adj_index_valid);
	kfree(nh_grp);
}

static int mlxsw_sp_nexthop6_group_get(struct mlxsw_sp *mlxsw_sp,
				       struct mlxsw_sp_fib6_entry *fib6_entry)
{
	struct mlxsw_sp_nexthop_group *nh_grp;

	nh_grp = mlxsw_sp_nexthop6_group_lookup(mlxsw_sp, fib6_entry);
	if (!nh_grp) {
		nh_grp = mlxsw_sp_nexthop6_group_create(mlxsw_sp, fib6_entry);
		if (IS_ERR(nh_grp))
			return PTR_ERR(nh_grp);
	}

	list_add_tail(&fib6_entry->common.nexthop_group_node,
		      &nh_grp->fib_list);
	fib6_entry->common.nh_group = nh_grp;

	return 0;
}

static void mlxsw_sp_nexthop6_group_put(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib_entry *fib_entry)
{
	struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;

	list_del(&fib_entry->nexthop_group_node);
	if (!list_empty(&nh_grp->fib_list))
		return;
	mlxsw_sp_nexthop6_group_destroy(mlxsw_sp, nh_grp);
}

static int
mlxsw_sp_nexthop6_group_update(struct mlxsw_sp *mlxsw_sp,
			       struct mlxsw_sp_fib6_entry *fib6_entry)
{
	struct mlxsw_sp_nexthop_group *old_nh_grp = fib6_entry->common.nh_group;
	int err;

	fib6_entry->common.nh_group = NULL;
	list_del(&fib6_entry->common.nexthop_group_node);

	err = mlxsw_sp_nexthop6_group_get(mlxsw_sp, fib6_entry);
	if (err)
		goto err_nexthop6_group_get;

	/* In case this entry is offloaded, then the adjacency index
	 * currently associated with it in the device's table is that
	 * of the old group. Start using the new one instead.
	 */
	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common);
	if (err)
		goto err_fib_node_entry_add;

	if (list_empty(&old_nh_grp->fib_list))
		mlxsw_sp_nexthop6_group_destroy(mlxsw_sp, old_nh_grp);

	return 0;

err_fib_node_entry_add:
	mlxsw_sp_nexthop6_group_put(mlxsw_sp, &fib6_entry->common);
err_nexthop6_group_get:
	list_add_tail(&fib6_entry->common.nexthop_group_node,
		      &old_nh_grp->fib_list);
	fib6_entry->common.nh_group = old_nh_grp;
	return err;
}

static int
mlxsw_sp_fib6_entry_nexthop_add(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_fib6_entry *fib6_entry,
				struct fib6_info *rt)
{
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
	int err;

	mlxsw_sp_rt6 = mlxsw_sp_rt6_create(rt);
	if (IS_ERR(mlxsw_sp_rt6))
		return PTR_ERR(mlxsw_sp_rt6);

	list_add_tail(&mlxsw_sp_rt6->list, &fib6_entry->rt6_list);
	fib6_entry->nrt6++;

	err = mlxsw_sp_nexthop6_group_update(mlxsw_sp, fib6_entry);
	if (err)
		goto err_nexthop6_group_update;

	return 0;

err_nexthop6_group_update:
	fib6_entry->nrt6--;
	list_del(&mlxsw_sp_rt6->list);
	mlxsw_sp_rt6_destroy(mlxsw_sp_rt6);
	return err;
}

static void
mlxsw_sp_fib6_entry_nexthop_del(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_fib6_entry *fib6_entry,
				struct fib6_info *rt)
{
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;

	mlxsw_sp_rt6 = mlxsw_sp_fib6_entry_rt_find(fib6_entry, rt);
	if (WARN_ON(!mlxsw_sp_rt6))
		return;

	fib6_entry->nrt6--;
	list_del(&mlxsw_sp_rt6->list);
	mlxsw_sp_nexthop6_group_update(mlxsw_sp, fib6_entry);
	mlxsw_sp_rt6_destroy(mlxsw_sp_rt6);
}

static void mlxsw_sp_fib6_entry_type_set(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_fib_entry *fib_entry,
					 const struct fib6_info *rt)
{
	/* Packets hitting RTF_REJECT routes need to be discarded by the
	 * stack. We can rely on their destination device not having a
	 * RIF (it's the loopback device) and can thus use action type
	 * local, which will cause them to be trapped with a lower
	 * priority than packets that need to be locally received.
	 */
	if (rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST))
		fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP;
	else if (rt->fib6_flags & RTF_REJECT)
		fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_LOCAL;
	else if (mlxsw_sp_rt6_is_gateway(mlxsw_sp, rt))
		fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_REMOTE;
	else
		fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_LOCAL;
}

static void
mlxsw_sp_fib6_entry_rt_destroy_all(struct mlxsw_sp_fib6_entry *fib6_entry)
{
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6, *tmp;

	list_for_each_entry_safe(mlxsw_sp_rt6, tmp, &fib6_entry->rt6_list,
				 list) {
		fib6_entry->nrt6--;
		list_del(&mlxsw_sp_rt6->list);
		mlxsw_sp_rt6_destroy(mlxsw_sp_rt6);
	}
}

static struct mlxsw_sp_fib6_entry *
mlxsw_sp_fib6_entry_create(struct mlxsw_sp *mlxsw_sp,
			   struct mlxsw_sp_fib_node *fib_node,
			   struct fib6_info *rt)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;
	struct mlxsw_sp_fib_entry *fib_entry;
	struct mlxsw_sp_rt6 *mlxsw_sp_rt6;
	int err;

	fib6_entry = kzalloc(sizeof(*fib6_entry), GFP_KERNEL);
	if (!fib6_entry)
		return ERR_PTR(-ENOMEM);
	fib_entry = &fib6_entry->common;

	mlxsw_sp_rt6 = mlxsw_sp_rt6_create(rt);
	if (IS_ERR(mlxsw_sp_rt6)) {
		err = PTR_ERR(mlxsw_sp_rt6);
		goto err_rt6_create;
	}

	mlxsw_sp_fib6_entry_type_set(mlxsw_sp, fib_entry, mlxsw_sp_rt6->rt);

	INIT_LIST_HEAD(&fib6_entry->rt6_list);
	list_add_tail(&mlxsw_sp_rt6->list, &fib6_entry->rt6_list);
	fib6_entry->nrt6 = 1;
	err = mlxsw_sp_nexthop6_group_get(mlxsw_sp, fib6_entry);
	if (err)
		goto err_nexthop6_group_get;

	fib_entry->fib_node = fib_node;

	return fib6_entry;

err_nexthop6_group_get:
	list_del(&mlxsw_sp_rt6->list);
	mlxsw_sp_rt6_destroy(mlxsw_sp_rt6);
err_rt6_create:
	kfree(fib6_entry);
	return ERR_PTR(err);
}

static void mlxsw_sp_fib6_entry_destroy(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib6_entry *fib6_entry)
{
	mlxsw_sp_nexthop6_group_put(mlxsw_sp, &fib6_entry->common);
	mlxsw_sp_fib6_entry_rt_destroy_all(fib6_entry);
	WARN_ON(fib6_entry->nrt6);
	kfree(fib6_entry);
}

static struct mlxsw_sp_fib6_entry *
mlxsw_sp_fib6_node_entry_find(const struct mlxsw_sp_fib_node *fib_node,
			      const struct fib6_info *nrt, bool replace)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;

	list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) {
		struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(fib6_entry);

		if (rt->fib6_table->tb6_id > nrt->fib6_table->tb6_id)
			continue;
		if (rt->fib6_table->tb6_id != nrt->fib6_table->tb6_id)
			break;
		if (replace && rt->fib6_metric == nrt->fib6_metric)
			return fib6_entry;
		if (rt->fib6_metric > nrt->fib6_metric)
			return fib6_entry;
	}

	return NULL;
}

static int
mlxsw_sp_fib6_node_list_insert(struct mlxsw_sp_fib6_entry *new6_entry,
			       bool replace)
{
	struct mlxsw_sp_fib_node *fib_node = new6_entry->common.fib_node;
	struct fib6_info *nrt = mlxsw_sp_fib6_entry_rt(new6_entry);
	struct mlxsw_sp_fib6_entry *fib6_entry;

	fib6_entry = mlxsw_sp_fib6_node_entry_find(fib_node, nrt, replace);

	if (replace && WARN_ON(!fib6_entry))
		return -EINVAL;

	if (fib6_entry) {
		list_add_tail(&new6_entry->common.list,
			      &fib6_entry->common.list);
	} else {
		struct mlxsw_sp_fib6_entry *last;

		list_for_each_entry(last, &fib_node->entry_list, common.list) {
			struct fib6_info *rt = mlxsw_sp_fib6_entry_rt(last);

			if (nrt->fib6_table->tb6_id > rt->fib6_table->tb6_id)
				break;
			fib6_entry = last;
		}

		if (fib6_entry)
			list_add(&new6_entry->common.list,
				 &fib6_entry->common.list);
		else
			list_add(&new6_entry->common.list,
				 &fib_node->entry_list);
	}

	return 0;
}

static void
mlxsw_sp_fib6_node_list_remove(struct mlxsw_sp_fib6_entry *fib6_entry)
{
	list_del(&fib6_entry->common.list);
}

static int mlxsw_sp_fib6_node_entry_link(struct mlxsw_sp *mlxsw_sp,
					 struct mlxsw_sp_fib6_entry *fib6_entry,
					 bool replace)
{
	int err;

	err = mlxsw_sp_fib6_node_list_insert(fib6_entry, replace);
	if (err)
		return err;

	err = mlxsw_sp_fib_node_entry_add(mlxsw_sp, &fib6_entry->common);
	if (err)
		goto err_fib_node_entry_add;

	return 0;

err_fib_node_entry_add:
	mlxsw_sp_fib6_node_list_remove(fib6_entry);
	return err;
}

static void
mlxsw_sp_fib6_node_entry_unlink(struct mlxsw_sp *mlxsw_sp,
				struct mlxsw_sp_fib6_entry *fib6_entry)
{
	mlxsw_sp_fib_node_entry_del(mlxsw_sp, &fib6_entry->common);
	mlxsw_sp_fib6_node_list_remove(fib6_entry);
}

static struct mlxsw_sp_fib6_entry *
mlxsw_sp_fib6_entry_lookup(struct mlxsw_sp *mlxsw_sp,
			   const struct fib6_info *rt)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;
	struct mlxsw_sp_fib_node *fib_node;
	struct mlxsw_sp_fib *fib;
	struct mlxsw_sp_vr *vr;

	vr = mlxsw_sp_vr_find(mlxsw_sp, rt->fib6_table->tb6_id);
	if (!vr)
		return NULL;
	fib = mlxsw_sp_vr_fib(vr, MLXSW_SP_L3_PROTO_IPV6);

	fib_node = mlxsw_sp_fib_node_lookup(fib, &rt->fib6_dst.addr,
					    sizeof(rt->fib6_dst.addr),
					    rt->fib6_dst.plen);
	if (!fib_node)
		return NULL;

	list_for_each_entry(fib6_entry, &fib_node->entry_list, common.list) {
		struct fib6_info *iter_rt = mlxsw_sp_fib6_entry_rt(fib6_entry);

		if (rt->fib6_table->tb6_id == iter_rt->fib6_table->tb6_id &&
		    rt->fib6_metric == iter_rt->fib6_metric &&
		    mlxsw_sp_fib6_entry_rt_find(fib6_entry, rt))
			return fib6_entry;
	}

	return NULL;
}

static void mlxsw_sp_fib6_entry_replace(struct mlxsw_sp *mlxsw_sp,
					struct mlxsw_sp_fib6_entry *fib6_entry,
					bool replace)
{
	struct mlxsw_sp_fib_node *fib_node = fib6_entry->common.fib_node;
	struct mlxsw_sp_fib6_entry *replaced;

	if (!replace)
		return;

	replaced = list_next_entry(fib6_entry, common.list);

	mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, replaced);
	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, replaced);
	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
}

static int mlxsw_sp_router_fib6_add(struct mlxsw_sp *mlxsw_sp,
				    struct fib6_info *rt, bool replace,
				    bool append)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;
	struct mlxsw_sp_fib_node *fib_node;
	int err;

	if (mlxsw_sp->router->aborted)
		return 0;

	if (rt->fib6_src.plen)
		return -EINVAL;

	if (mlxsw_sp_fib6_rt_should_ignore(rt))
		return 0;

	fib_node = mlxsw_sp_fib_node_get(mlxsw_sp, rt->fib6_table->tb6_id,
					 &rt->fib6_dst.addr,
					 sizeof(rt->fib6_dst.addr),
					 rt->fib6_dst.plen,
					 MLXSW_SP_L3_PROTO_IPV6);
	if (IS_ERR(fib_node))
		return PTR_ERR(fib_node);

	/* Before creating a new entry, try to append route to an existing
	 * multipath entry.
	 */
	fib6_entry = mlxsw_sp_fib6_node_mp_entry_find(fib_node, rt, append);
	if (fib6_entry) {
		err = mlxsw_sp_fib6_entry_nexthop_add(mlxsw_sp, fib6_entry, rt);
		if (err)
			goto err_fib6_entry_nexthop_add;
		return 0;
	}

	/* We received an append event, yet did not find any route to
	 * append to.
	 */
	if (WARN_ON(append)) {
		err = -EINVAL;
		goto err_fib6_entry_append;
	}

	fib6_entry = mlxsw_sp_fib6_entry_create(mlxsw_sp, fib_node, rt);
	if (IS_ERR(fib6_entry)) {
		err = PTR_ERR(fib6_entry);
		goto err_fib6_entry_create;
	}

	err = mlxsw_sp_fib6_node_entry_link(mlxsw_sp, fib6_entry, replace);
	if (err)
		goto err_fib6_node_entry_link;

	mlxsw_sp_fib6_entry_replace(mlxsw_sp, fib6_entry, replace);

	return 0;

err_fib6_node_entry_link:
	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
err_fib6_entry_create:
err_fib6_entry_append:
err_fib6_entry_nexthop_add:
	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
	return err;
}

static void mlxsw_sp_router_fib6_del(struct mlxsw_sp *mlxsw_sp,
				     struct fib6_info *rt)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;
	struct mlxsw_sp_fib_node *fib_node;

	if (mlxsw_sp->router->aborted)
		return;

	if (mlxsw_sp_fib6_rt_should_ignore(rt))
		return;

	fib6_entry = mlxsw_sp_fib6_entry_lookup(mlxsw_sp, rt);
	if (WARN_ON(!fib6_entry))
		return;

	/* If route is part of a multipath entry, but not the last one
	 * removed, then only reduce its nexthop group.
	 */
	if (!list_is_singular(&fib6_entry->rt6_list)) {
		mlxsw_sp_fib6_entry_nexthop_del(mlxsw_sp, fib6_entry, rt);
		return;
	}

	fib_node = fib6_entry->common.fib_node;

	mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, fib6_entry);
	mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
	mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
}

static int __mlxsw_sp_router_set_abort_trap(struct mlxsw_sp *mlxsw_sp,
					    enum mlxsw_reg_ralxx_protocol proto,
					    u8 tree_id)
{
	char ralta_pl[MLXSW_REG_RALTA_LEN];
	char ralst_pl[MLXSW_REG_RALST_LEN];
	int i, err;

	mlxsw_reg_ralta_pack(ralta_pl, true, proto, tree_id);
	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralta), ralta_pl);
	if (err)
		return err;

	mlxsw_reg_ralst_pack(ralst_pl, 0xff, tree_id);
	err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralst), ralst_pl);
	if (err)
		return err;

	for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) {
		struct mlxsw_sp_vr *vr = &mlxsw_sp->router->vrs[i];
		char raltb_pl[MLXSW_REG_RALTB_LEN];
		char ralue_pl[MLXSW_REG_RALUE_LEN];

		mlxsw_reg_raltb_pack(raltb_pl, vr->id, proto, tree_id);
		err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(raltb),
				      raltb_pl);
		if (err)
			return err;

		mlxsw_reg_ralue_pack(ralue_pl, proto,
				     MLXSW_REG_RALUE_OP_WRITE_WRITE, vr->id, 0);
		mlxsw_reg_ralue_act_ip2me_pack(ralue_pl);
		err = mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue),
				      ralue_pl);
		if (err)
			return err;
	}

	return 0;
}

static struct mlxsw_sp_mr_table *
mlxsw_sp_router_fibmr_family_to_table(struct mlxsw_sp_vr *vr, int family)
{
	if (family == RTNL_FAMILY_IPMR)
		return vr->mr_table[MLXSW_SP_L3_PROTO_IPV4];
	else
		return vr->mr_table[MLXSW_SP_L3_PROTO_IPV6];
}

static int mlxsw_sp_router_fibmr_add(struct mlxsw_sp *mlxsw_sp,
				     struct mfc_entry_notifier_info *men_info,
				     bool replace)
{
	struct mlxsw_sp_mr_table *mrt;
	struct mlxsw_sp_vr *vr;

	if (mlxsw_sp->router->aborted)
		return 0;

	vr = mlxsw_sp_vr_get(mlxsw_sp, men_info->tb_id, NULL);
	if (IS_ERR(vr))
		return PTR_ERR(vr);

	mrt = mlxsw_sp_router_fibmr_family_to_table(vr, men_info->info.family);
	return mlxsw_sp_mr_route_add(mrt, men_info->mfc, replace);
}

static void mlxsw_sp_router_fibmr_del(struct mlxsw_sp *mlxsw_sp,
				      struct mfc_entry_notifier_info *men_info)
{
	struct mlxsw_sp_mr_table *mrt;
	struct mlxsw_sp_vr *vr;

	if (mlxsw_sp->router->aborted)
		return;

	vr = mlxsw_sp_vr_find(mlxsw_sp, men_info->tb_id);
	if (WARN_ON(!vr))
		return;

	mrt = mlxsw_sp_router_fibmr_family_to_table(vr, men_info->info.family);
	mlxsw_sp_mr_route_del(mrt, men_info->mfc);
	mlxsw_sp_vr_put(mlxsw_sp, vr);
}

static int
mlxsw_sp_router_fibmr_vif_add(struct mlxsw_sp *mlxsw_sp,
			      struct vif_entry_notifier_info *ven_info)
{
	struct mlxsw_sp_mr_table *mrt;
	struct mlxsw_sp_rif *rif;
	struct mlxsw_sp_vr *vr;

	if (mlxsw_sp->router->aborted)
		return 0;

	vr = mlxsw_sp_vr_get(mlxsw_sp, ven_info->tb_id, NULL);
	if (IS_ERR(vr))
		return PTR_ERR(vr);

	mrt = mlxsw_sp_router_fibmr_family_to_table(vr, ven_info->info.family);
	rif = mlxsw_sp_rif_find_by_dev(mlxsw_sp, ven_info->dev);
	return mlxsw_sp_mr_vif_add(mrt, ven_info->dev,
				   ven_info->vif_index,
				   ven_info->vif_flags, rif);
}

static void
mlxsw_sp_router_fibmr_vif_del(struct mlxsw_sp *mlxsw_sp,
			      struct vif_entry_notifier_info *ven_info)
{
	struct mlxsw_sp_mr_table *mrt;
	struct mlxsw_sp_vr *vr;

	if (mlxsw_sp->router->aborted)
		return;

	vr = mlxsw_sp_vr_find(mlxsw_sp, ven_info->tb_id);
	if (WARN_ON(!vr))
		return;

	mrt = mlxsw_sp_router_fibmr_family_to_table(vr, ven_info->info.family);
	mlxsw_sp_mr_vif_del(mrt, ven_info->vif_index);
	mlxsw_sp_vr_put(mlxsw_sp, vr);
}

static int mlxsw_sp_router_set_abort_trap(struct mlxsw_sp *mlxsw_sp)
{
	enum mlxsw_reg_ralxx_protocol proto = MLXSW_REG_RALXX_PROTOCOL_IPV4;
	int err;

	err = __mlxsw_sp_router_set_abort_trap(mlxsw_sp, proto,
					       MLXSW_SP_LPM_TREE_MIN);
	if (err)
		return err;

	/* The multicast router code does not need an abort trap as by default,
	 * packets that don't match any routes are trapped to the CPU.
	 */

	proto = MLXSW_REG_RALXX_PROTOCOL_IPV6;
	return __mlxsw_sp_router_set_abort_trap(mlxsw_sp, proto,
						MLXSW_SP_LPM_TREE_MIN + 1);
}

static void mlxsw_sp_fib4_node_flush(struct mlxsw_sp *mlxsw_sp,
				     struct mlxsw_sp_fib_node *fib_node)
{
	struct mlxsw_sp_fib4_entry *fib4_entry, *tmp;

	list_for_each_entry_safe(fib4_entry, tmp, &fib_node->entry_list,
				 common.list) {
		bool do_break = &tmp->common.list == &fib_node->entry_list;

		mlxsw_sp_fib4_node_entry_unlink(mlxsw_sp, fib4_entry);
		mlxsw_sp_fib4_entry_destroy(mlxsw_sp, fib4_entry);
		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
		/* Break when entry list is empty and node was freed.
		 * Otherwise, we'll access freed memory in the next
		 * iteration.
		 */
		if (do_break)
			break;
	}
}

static void mlxsw_sp_fib6_node_flush(struct mlxsw_sp *mlxsw_sp,
				     struct mlxsw_sp_fib_node *fib_node)
{
	struct mlxsw_sp_fib6_entry *fib6_entry, *tmp;

	list_for_each_entry_safe(fib6_entry, tmp, &fib_node->entry_list,
				 common.list) {
		bool do_break = &tmp->common.list == &fib_node->entry_list;

		mlxsw_sp_fib6_node_entry_unlink(mlxsw_sp, fib6_entry);
		mlxsw_sp_fib6_entry_destroy(mlxsw_sp, fib6_entry);
		mlxsw_sp_fib_node_put(mlxsw_sp, fib_node);
		if (do_break)
			break;
	}
}

static void mlxsw_sp_fib_node_flush(struct mlxsw_sp *mlxsw_sp,
				    struct mlxsw_sp_fib_node *fib_node)
{
	switch (fib_node->fib->proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		mlxsw_sp_fib4_node_flush(mlxsw_sp, fib_node);
		break;
	case MLXSW_SP_L3_PROTO_IPV6:
		mlxsw_sp_fib6_node_flush(mlxsw_sp, fib_node);
		break;
	}
}

static void mlxsw_sp_vr_fib_flush(struct mlxsw_sp *mlxsw_sp,
				  struct mlxsw_sp_vr *vr,
				  enum mlxsw_sp_l3proto proto)
{
	struct mlxsw_sp_fib *fib = mlxsw_sp_vr_fib(vr, proto);
	struct mlxsw_sp_fib_node *fib_node, *tmp;

	list_for_each_entry_safe(fib_node, tmp, &fib->node_list, list) {
		bool do_break = &tmp->list == &fib->node_list;

		mlxsw_sp_fib_node_flush(mlxsw_sp, fib_node);
		if (do_break)
			break;
	}
}

static void mlxsw_sp_router_fib_flush(struct mlxsw_sp *mlxsw_sp)
{
	int i, j;

	for (i = 0; i < MLXSW_CORE_RES_GET(mlxsw_sp->core, MAX_VRS); i++) {
		struct mlxsw_sp_vr *vr = &mlxsw_sp->router->vrs[i];

		if (!mlxsw_sp_vr_is_used(vr))
			continue;

		for (j = 0; j < MLXSW_SP_L3_PROTO_MAX; j++)
			mlxsw_sp_mr_table_flush(vr->mr_table[j]);
		mlxsw_sp_vr_fib_flush(mlxsw_sp, vr, MLXSW_SP_L3_PROTO_IPV4);

		/* If virtual router was only used for IPv4, then it's no
		 * longer used.
		 */
		if (!mlxsw_sp_vr_is_used(vr))
			continue;
		mlxsw_sp_vr_fib_flush(mlxsw_sp, vr, MLXSW_SP_L3_PROTO_IPV6);
	}
}

static void mlxsw_sp_router_fib_abort(struct mlxsw_sp *mlxsw_sp)
{
	int err;

	if (mlxsw_sp->router->aborted)
		return;
	dev_warn(mlxsw_sp->bus_info->dev, "FIB abort triggered. Note that FIB entries are no longer being offloaded to this device.\n");
	mlxsw_sp_router_fib_flush(mlxsw_sp);
	mlxsw_sp->router->aborted = true;
	err = mlxsw_sp_router_set_abort_trap(mlxsw_sp);
	if (err)
		dev_warn(mlxsw_sp->bus_info->dev, "Failed to set abort trap.\n");
}

struct mlxsw_sp_fib_event_work {
	struct work_struct work;
	union {
		struct fib6_entry_notifier_info fen6_info;
		struct fib_entry_notifier_info fen_info;
		struct fib_rule_notifier_info fr_info;
		struct fib_nh_notifier_info fnh_info;
		struct mfc_entry_notifier_info men_info;
		struct vif_entry_notifier_info ven_info;
	};
	struct mlxsw_sp *mlxsw_sp;
	unsigned long event;
};

static void mlxsw_sp_router_fib4_event_work(struct work_struct *work)
{
	struct mlxsw_sp_fib_event_work *fib_work =
		container_of(work, struct mlxsw_sp_fib_event_work, work);
	struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp;
	bool replace, append;
	int err;

	/* Protect internal structures from changes */
	rtnl_lock();
	mlxsw_sp_span_respin(mlxsw_sp);

	switch (fib_work->event) {
	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
	case FIB_EVENT_ENTRY_APPEND: /* fall through */
	case FIB_EVENT_ENTRY_ADD:
		replace = fib_work->event == FIB_EVENT_ENTRY_REPLACE;
		append = fib_work->event == FIB_EVENT_ENTRY_APPEND;
		err = mlxsw_sp_router_fib4_add(mlxsw_sp, &fib_work->fen_info,
					       replace, append);
		if (err)
			mlxsw_sp_router_fib_abort(mlxsw_sp);
		fib_info_put(fib_work->fen_info.fi);
		break;
	case FIB_EVENT_ENTRY_DEL:
		mlxsw_sp_router_fib4_del(mlxsw_sp, &fib_work->fen_info);
		fib_info_put(fib_work->fen_info.fi);
		break;
	case FIB_EVENT_RULE_ADD:
		/* if we get here, a rule was added that we do not support.
		 * just do the fib_abort
		 */
		mlxsw_sp_router_fib_abort(mlxsw_sp);
		break;
	case FIB_EVENT_NH_ADD: /* fall through */
	case FIB_EVENT_NH_DEL:
		mlxsw_sp_nexthop4_event(mlxsw_sp, fib_work->event,
					fib_work->fnh_info.fib_nh);
		fib_info_put(fib_work->fnh_info.fib_nh->nh_parent);
		break;
	}
	rtnl_unlock();
	kfree(fib_work);
}

static void mlxsw_sp_router_fib6_event_work(struct work_struct *work)
{
	struct mlxsw_sp_fib_event_work *fib_work =
		container_of(work, struct mlxsw_sp_fib_event_work, work);
	struct mlxsw_sp *mlxsw_sp = fib_work->mlxsw_sp;
	bool replace, append;
	int err;

	rtnl_lock();
	mlxsw_sp_span_respin(mlxsw_sp);

	switch (fib_work->event) {
	case FIB_EVENT_ENTRY_REPLACE: /* fall through */
	case FIB_EVENT_ENTRY_APPEND: /* fall through */
	case FIB_EVENT_ENTRY_ADD:
		replace = fib_work->event == FIB_EVENT_ENTRY_REPLACE;
		append = fib_work->event == FIB_EVENT_ENTRY_APPEND;
		err = mlxsw_sp_router_fib6_add(mlxsw_sp,
					       fib_work->fen6_info.rt, replace,
					       append);
		if (err)
			mlxsw_sp_router_fib_abort(mlxsw_sp);
		mlxsw_sp_rt6_release(fib_work->fen6_info.rt);
		break;
	case FIB_EVENT_ENTRY_DEL:
		mlxsw_sp_router_fib6_del(mlxsw_sp, fib_work->fen6_info.rt);
		mlxsw_sp_rt6_release(fib_work->fen6_info.rt);
		break;
	case FIB_EVENT_RULE_ADD:
		/* if we get here, a rule was added that we do not support.
		 * just do the fib_abort
		 */
		mlxsw_sp_router_fib_abort(mlxsw_sp);
		break;
	}
	rtnl_unlock();
	kfree(fib_work);
}

static void mlxsw_sp_router_fibmr_event_work(struct work_struct *work)
{
	struct mlxsw_sp_fib_event_work *fib_work =
		container_of(work, struct mlxsw_sp_fib_event_work, work);
	struct mlxsw_sp *mlxsw_sp = f