12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394959697989910010110210310410510610710810911011111211311411511611711811912012112212312412512612712812913013113213313413513613713813914014114214314414514614714814915015115215315415515615715815916016116216316416516616716816917017117217317417517617717817918018118218318418518618718818919019119219319419519619719819920020120220320420520620720820921021121221321421521621721821922022122222322422522622722822923023123223323423523623723823924024124224324424524624724824925025125225325425525625725825926026126226326426526626726826927027127227327427527627727827928028128228328428528628728828929029129229329429529629729829930030130230330430530630730830931031131231331431531631731831932032132232332432532632732832933033133233333433533633733833934034134234334434534634734834935035135235335435535635735835936036136236336436536636736836937037137237337437537637737837938038138238338438538638738838939039139239339439539639739839940040140240340440540640740840941041141241341441541641741841942042142242342442542642742842943043143243343443543643743843944044144244344444544644744844945045145245345445545645745845946046146246346446546646746846947047147247347447547647747847948048148248348448548648748848949049149249349449549649749849950050150250350450550650750850951051151251351451551651751851952052152252352452552652752852953053153253353453553653753853954054154254354454554654754854955055155255355455555655755855956056156256356456556656756856957057157257357457557657757857958058158258358458558658758858959059159259359459559659759859960060160260360460560660760860961061161261361461561661761861962062162262362462562662762862963063163263363463563663763863964064164264364464564664764864965065165265365465565665765865966066166266366466566666766866967067167267367467567667767867968068168268368468568668768868969069169269369469569669769869970070170270370470570670770870971071171271371471571671771871972072172272372472572672772872973073173273373473573673773873974074174274374474574674774874975075175275375475575675775875976076176276376476576676776876977077177277377477577677777877978078178278378478578678778878979079179279379479579679779879980080180280380480580680780880981081181281381481581681781881982082182282382482582682782882983083183283383483583683783883984084184284384484584684784884985085185285385485585685785885986086186286386486586686786886987087187287387487587687787887988088188288388488588688788888989089189289389489589689789889990090190290390490590690790890991091191291391491591691791891992092192292392492592692792892993093193293393493593693793893994094194294394494594694794894995095195295395495595695795895996096196296396496596696796896997097197297397497597697797897998098198298398498598698798898999099199299399499599699799899910001001100210031004100510061007100810091010101110121013101410151016101710181019102010211022102310241025102610271028102910301031103210331034103510361037103810391040104110421043104410451046104710481049105010511052105310541055105610571058105910601061106210631064106510661067106810691070107110721073107410751076107710781079108010811082108310841085108610871088108910901091109210931094109510961097109810991100110111021103110411051106110711081109111011111112111311141115111611171118111911201121112211231124112511261127112811291130113111321133113411351136113711381139114011411142114311441145114611471148114911501151115211531154115511561157115811591160116111621163116411651166116711681169117011711172117311741175117611771178117911801181118211831184118511861187118811891190119111921193119411951196119711981199120012011202120312041205120612071208120912101211121212131214121512161217121812191220122112221223122412251226122712281229123012311232123312341235123612371238123912401241124212431244124512461247124812491250125112521253125412551256125712581259126012611262126312641265126612671268126912701271127212731274127512761277127812791280128112821283128412851286128712881289129012911292129312941295129612971298129913001301130213031304130513061307130813091310131113121313131413151316131713181319132013211322132313241325132613271328132913301331133213331334133513361337133813391340134113421343134413451346134713481349135013511352135313541355135613571358135913601361136213631364136513661367136813691370137113721373137413751376137713781379138013811382138313841385138613871388138913901391139213931394139513961397139813991400140114021403140414051406140714081409141014111412141314141415141614171418141914201421142214231424142514261427142814291430143114321433143414351436143714381439144014411442144314441445144614471448144914501451145214531454145514561457145814591460146114621463146414651466146714681469147014711472147314741475147614771478147914801481148214831484148514861487148814891490149114921493149414951496149714981499150015011502150315041505150615071508150915101511151215131514151515161517151815191520152115221523152415251526152715281529153015311532153315341535153615371538153915401541154215431544154515461547154815491550155115521553155415551556155715581559156015611562156315641565156615671568156915701571157215731574157515761577157815791580158115821583158415851586158715881589159015911592159315941595159615971598159916001601160216031604160516061607160816091610161116121613161416151616161716181619162016211622162316241625162616271628162916301631163216331634163516361637163816391640164116421643164416451646164716481649165016511652165316541655165616571658165916601661166216631664166516661667166816691670167116721673167416751676167716781679168016811682168316841685168616871688168916901691169216931694169516961697169816991700170117021703170417051706170717081709171017111712171317141715171617171718171917201721172217231724172517261727172817291730173117321733173417351736173717381739174017411742174317441745174617471748174917501751175217531754175517561757175817591760176117621763176417651766176717681769177017711772177317741775177617771778177917801781178217831784178517861787178817891790179117921793179417951796179717981799180018011802180318041805180618071808180918101811181218131814181518161817181818191820182118221823182418251826182718281829183018311832183318341835183618371838183918401841184218431844184518461847184818491850185118521853185418551856185718581859186018611862186318641865186618671868186918701871187218731874187518761877187818791880188118821883188418851886188718881889189018911892189318941895189618971898189919001901190219031904190519061907190819091910191119121913191419151916191719181919192019211922192319241925192619271928192919301931193219331934193519361937193819391940194119421943194419451946194719481949195019511952195319541955195619571958195919601961196219631964196519661967196819691970197119721973197419751976197719781979198019811982198319841985198619871988198919901991199219931994199519961997199819992000200120022003200420052006200720082009201020112012201320142015201620172018201920202021202220232024202520262027202820292030203120322033203420352036203720382039204020412042204320442045204620472048204920502051205220532054205520562057205820592060206120622063206420652066206720682069207020712072207320742075207620772078207920802081208220832084208520862087208820892090209120922093209420952096209720982099210021012102210321042105210621072108210921102111211221132114211521162117211821192120212121222123212421252126212721282129213021312132213321342135213621372138213921402141214221432144214521462147214821492150215121522153215421552156215721582159216021612162216321642165216621672168216921702171217221732174217521762177217821792180218121822183218421852186218721882189219021912192219321942195219621972198219922002201220222032204220522062207220822092210221122122213221422152216221722182219222022212222222322242225222622272228222922302231223222332234223522362237223822392240224122422243224422452246224722482249225022512252225322542255225622572258225922602261226222632264226522662267226822692270227122722273227422752276227722782279228022812282228322842285228622872288228922902291229222932294229522962297229822992300230123022303230423052306230723082309231023112312231323142315231623172318231923202321232223232324232523262327232823292330233123322333233423352336233723382339234023412342234323442345234623472348234923502351235223532354235523562357235823592360236123622363236423652366236723682369237023712372237323742375237623772378237923802381238223832384238523862387238823892390239123922393239423952396239723982399240024012402240324042405240624072408240924102411241224132414241524162417241824192420242124222423242424252426242724282429243024312432243324342435243624372438243924402441244224432444244524462447244824492450245124522453245424552456245724582459246024612462246324642465246624672468246924702471247224732474247524762477247824792480248124822483248424852486248724882489249024912492249324942495249624972498249925002501250225032504250525062507250825092510251125122513251425152516251725182519252025212522252325242525252625272528252925302531253225332534253525362537253825392540254125422543254425452546254725482549255025512552255325542555255625572558255925602561256225632564256525662567256825692570257125722573257425752576257725782579258025812582258325842585258625872588258925902591259225932594259525962597259825992600260126022603260426052606260726082609261026112612261326142615261626172618261926202621262226232624262526262627262826292630263126322633263426352636263726382639264026412642264326442645264626472648264926502651265226532654265526562657265826592660266126622663266426652666266726682669267026712672267326742675267626772678267926802681268226832684268526862687268826892690269126922693269426952696269726982699270027012702270327042705270627072708270927102711271227132714271527162717271827192720272127222723272427252726272727282729273027312732273327342735273627372738273927402741274227432744274527462747274827492750275127522753275427552756275727582759276027612762276327642765276627672768276927702771277227732774277527762777277827792780278127822783278427852786278727882789279027912792279327942795279627972798279928002801280228032804280528062807280828092810281128122813281428152816281728182819282028212822282328242825282628272828282928302831283228332834283528362837283828392840284128422843284428452846284728482849285028512852285328542855285628572858285928602861286228632864286528662867286828692870287128722873287428752876287728782879288028812882288328842885288628872888288928902891289228932894289528962897289828992900290129022903290429052906290729082909291029112912291329142915291629172918291929202921292229232924292529262927292829292930293129322933293429352936293729382939294029412942294329442945294629472948294929502951295229532954295529562957295829592960296129622963296429652966296729682969297029712972297329742975297629772978297929802981298229832984298529862987298829892990299129922993299429952996299729982999300030013002300330043005300630073008300930103011301230133014301530163017301830193020302130223023302430253026302730283029303030313032303330343035303630373038303930403041304230433044304530463047304830493050305130523053305430553056305730583059306030613062306330643065306630673068306930703071307230733074307530763077307830793080308130823083308430853086308730883089309030913092309330943095309630973098309931003101310231033104310531063107310831093110311131123113311431153116311731183119312031213122312331243125312631273128312931303131313231333134313531363137313831393140314131423143314431453146314731483149315031513152315331543155315631573158315931603161316231633164316531663167316831693170317131723173317431753176317731783179318031813182318331843185318631873188318931903191319231933194319531963197319831993200320132023203320432053206320732083209321032113212321332143215321632173218321932203221322232233224322532263227322832293230323132323233323432353236323732383239324032413242324332443245324632473248324932503251325232533254325532563257325832593260326132623263326432653266326732683269327032713272327332743275327632773278327932803281328232833284328532863287328832893290329132923293329432953296329732983299330033013302330333043305330633073308330933103311331233133314331533163317331833193320332133223323332433253326332733283329333033313332333333343335333633373338333933403341334233433344334533463347334833493350335133523353335433553356335733583359336033613362336333643365336633673368336933703371337233733374337533763377337833793380338133823383338433853386338733883389339033913392339333943395339633973398339934003401340234033404340534063407340834093410341134123413341434153416341734183419342034213422342334243425342634273428342934303431343234333434343534363437343834393440344134423443344434453446344734483449345034513452345334543455345634573458345934603461346234633464346534663467346834693470347134723473347434753476347734783479348034813482348334843485348634873488348934903491349234933494349534963497349834993500350135023503350435053506350735083509351035113512351335143515351635173518351935203521352235233524352535263527352835293530353135323533353435353536353735383539354035413542354335443545354635473548354935503551355235533554355535563557355835593560356135623563356435653566356735683569357035713572357335743575357635773578357935803581358235833584358535863587358835893590359135923593359435953596359735983599360036013602360336043605360636073608360936103611361236133614361536163617361836193620362136223623362436253626362736283629363036313632363336343635363636373638363936403641364236433644364536463647364836493650365136523653365436553656365736583659366036613662366336643665366636673668366936703671367236733674367536763677367836793680368136823683368436853686368736883689369036913692369336943695369636973698369937003701370237033704370537063707370837093710371137123713371437153716371737183719372037213722372337243725372637273728372937303731373237333734373537363737373837393740374137423743374437453746374737483749375037513752375337543755375637573758375937603761376237633764376537663767376837693770377137723773377437753776377737783779378037813782378337843785378637873788378937903791379237933794379537963797379837993800380138023803380438053806380738083809381038113812381338143815381638173818381938203821382238233824382538263827382838293830383138323833383438353836383738383839384038413842384338443845384638473848384938503851385238533854385538563857385838593860386138623863386438653866386738683869387038713872387338743875387638773878387938803881388238833884388538863887388838893890389138923893389438953896389738983899390039013902390339043905390639073908390939103911391239133914391539163917391839193920392139223923392439253926392739283929393039313932393339343935393639373938393939403941394239433944394539463947394839493950395139523953395439553956395739583959396039613962396339643965396639673968396939703971397239733974397539763977397839793980398139823983398439853986398739883989399039913992399339943995399639973998399940004001400240034004400540064007400840094010401140124013401440154016401740184019402040214022402340244025402640274028402940304031403240334034403540364037403840394040404140424043404440454046404740484049405040514052405340544055405640574058405940604061406240634064406540664067406840694070407140724073407440754076407740784079408040814082408340844085408640874088408940904091409240934094409540964097409840994100410141024103410441054106410741084109411041114112411341144115411641174118411941204121412241234124412541264127412841294130413141324133413441354136413741384139414041414142414341444145414641474148414941504151415241534154415541564157415841594160416141624163416441654166416741684169417041714172417341744175417641774178417941804181418241834184418541864187418841894190419141924193419441954196419741984199420042014202420342044205420642074208420942104211421242134214421542164217421842194220422142224223422442254226422742284229423042314232423342344235423642374238423942404241424242434244424542464247424842494250425142524253425442554256425742584259426042614262426342644265426642674268426942704271427242734274427542764277427842794280428142824283428442854286428742884289429042914292429342944295429642974298429943004301430243034304430543064307430843094310431143124313431443154316431743184319432043214322432343244325432643274328432943304331433243334334433543364337433843394340434143424343434443454346434743484349435043514352435343544355435643574358435943604361436243634364436543664367436843694370437143724373437443754376437743784379438043814382438343844385438643874388438943904391439243934394439543964397439843994400440144024403440444054406440744084409441044114412441344144415441644174418441944204421442244234424442544264427442844294430443144324433443444354436443744384439444044414442444344444445444644474448444944504451445244534454445544564457445844594460446144624463446444654466446744684469447044714472447344744475447644774478447944804481448244834484448544864487448844894490449144924493449444954496449744984499450045014502450345044505450645074508450945104511451245134514451545164517451845194520452145224523452445254526452745284529453045314532453345344535453645374538453945404541454245434544454545464547454845494550455145524553455445554556455745584559456045614562456345644565456645674568456945704571457245734574457545764577457845794580458145824583458445854586458745884589459045914592459345944595459645974598459946004601460246034604460546064607460846094610461146124613461446154616461746184619462046214622462346244625462646274628462946304631463246334634463546364637463846394640464146424643464446454646464746484649465046514652465346544655465646574658465946604661466246634664466546664667466846694670467146724673467446754676467746784679468046814682468346844685468646874688468946904691469246934694469546964697469846994700470147024703470447054706470747084709471047114712471347144715471647174718471947204721472247234724472547264727472847294730473147324733473447354736473747384739474047414742474347444745474647474748474947504751475247534754475547564757475847594760476147624763476447654766476747684769477047714772477347744775477647774778477947804781478247834784478547864787478847894790479147924793479447954796479747984799480048014802480348044805480648074808480948104811481248134814481548164817481848194820482148224823482448254826482748284829483048314832483348344835483648374838483948404841484248434844484548464847484848494850485148524853485448554856485748584859486048614862486348644865486648674868486948704871487248734874487548764877487848794880488148824883488448854886488748884889489048914892489348944895489648974898489949004901490249034904490549064907490849094910491149124913491449154916491749184919492049214922492349244925492649274928492949304931493249334934493549364937493849394940494149424943494449454946494749484949495049514952495349544955495649574958495949604961496249634964496549664967496849694970497149724973497449754976497749784979498049814982498349844985498649874988498949904991499249934994499549964997499849995000500150025003500450055006500750085009501050115012501350145015501650175018501950205021502250235024502550265027502850295030503150325033503450355036503750385039504050415042504350445045504650475048504950505051505250535054505550565057505850595060506150625063506450655066506750685069507050715072507350745075507650775078507950805081508250835084508550865087508850895090509150925093509450955096509750985099510051015102510351045105510651075108510951105111511251135114511551165117511851195120512151225123512451255126512751285129513051315132513351345135513651375138513951405141514251435144514551465147514851495150515151525153515451555156515751585159516051615162516351645165516651675168516951705171517251735174517551765177517851795180518151825183518451855186518751885189519051915192519351945195519651975198519952005201520252035204520552065207520852095210521152125213521452155216521752185219522052215222522352245225522652275228522952305231523252335234523552365237523852395240524152425243524452455246524752485249525052515252525352545255525652575258525952605261526252635264526552665267526852695270527152725273527452755276527752785279528052815282528352845285528652875288528952905291529252935294529552965297529852995300530153025303530453055306530753085309531053115312531353145315531653175318531953205321532253235324532553265327532853295330533153325333533453355336533753385339534053415342534353445345534653475348534953505351535253535354535553565357535853595360536153625363536453655366536753685369537053715372537353745375537653775378537953805381538253835384538553865387538853895390539153925393539453955396539753985399540054015402540354045405540654075408540954105411541254135414541554165417541854195420542154225423542454255426542754285429543054315432543354345435543654375438543954405441544254435444544554465447544854495450545154525453545454555456545754585459546054615462546354645465546654675468546954705471547254735474547554765477547854795480548154825483548454855486548754885489549054915492549354945495549654975498549955005501550255035504550555065507550855095510551155125513551455155516551755185519552055215522552355245525552655275528552955305531553255335534553555365537553855395540554155425543554455455546554755485549555055515552555355545555555655575558555955605561556255635564556555665567556855695570557155725573557455755576557755785579558055815582558355845585558655875588558955905591559255935594559555965597559855995600560156025603560456055606560756085609561056115612561356145615561656175618561956205621562256235624562556265627562856295630563156325633563456355636563756385639564056415642564356445645564656475648564956505651565256535654565556565657565856595660566156625663566456655666566756685669567056715672567356745675567656775678567956805681568256835684568556865687568856895690569156925693569456955696569756985699570057015702570357045705570657075708570957105711571257135714571557165717571857195720572157225723572457255726572757285729573057315732573357345735573657375738573957405741574257435744574557465747574857495750575157525753575457555756575757585759576057615762576357645765576657675768576957705771577257735774577557765777577857795780578157825783578457855786578757885789579057915792579357945795579657975798579958005801580258035804580558065807580858095810581158125813581458155816581758185819582058215822582358245825582658275828582958305831583258335834583558365837583858395840584158425843584458455846584758485849585058515852585358545855585658575858585958605861586258635864586558665867586858695870587158725873587458755876587758785879588058815882588358845885588658875888588958905891589258935894589558965897589858995900590159025903590459055906590759085909591059115912591359145915591659175918591959205921592259235924592559265927592859295930593159325933593459355936593759385939594059415942594359445945594659475948594959505951595259535954595559565957595859595960596159625963596459655966596759685969597059715972597359745975597659775978597959805981598259835984598559865987598859895990599159925993599459955996599759985999600060016002600360046005600660076008600960106011601260136014601560166017601860196020602160226023602460256026602760286029603060316032603360346035603660376038603960406041604260436044604560466047604860496050605160526053605460556056605760586059606060616062606360646065606660676068606960706071607260736074607560766077607860796080608160826083608460856086608760886089609060916092609360946095609660976098609961006101610261036104610561066107610861096110611161126113611461156116611761186119612061216122612361246125612661276128612961306131613261336134613561366137613861396140614161426143614461456146614761486149615061516152615361546155615661576158615961606161616261636164616561666167616861696170617161726173617461756176617761786179618061816182618361846185618661876188618961906191619261936194619561966197619861996200620162026203620462056206620762086209621062116212621362146215621662176218621962206221622262236224622562266227622862296230623162326233623462356236623762386239624062416242624362446245624662476248624962506251625262536254625562566257625862596260626162626263626462656266626762686269627062716272627362746275627662776278627962806281628262836284628562866287628862896290629162926293629462956296629762986299630063016302630363046305630663076308630963106311631263136314631563166317631863196320632163226323632463256326632763286329633063316332633363346335633663376338633963406341634263436344634563466347634863496350635163526353635463556356635763586359636063616362636363646365636663676368636963706371637263736374637563766377637863796380638163826383638463856386638763886389639063916392639363946395639663976398639964006401640264036404640564066407640864096410641164126413641464156416641764186419642064216422642364246425642664276428642964306431643264336434643564366437643864396440644164426443644464456446644764486449645064516452645364546455645664576458645964606461646264636464646564666467646864696470647164726473647464756476647764786479648064816482648364846485648664876488648964906491649264936494649564966497649864996500650165026503650465056506650765086509651065116512651365146515651665176518651965206521652265236524652565266527652865296530653165326533653465356536653765386539654065416542654365446545654665476548654965506551655265536554655565566557655865596560656165626563656465656566656765686569657065716572657365746575657665776578657965806581658265836584658565866587658865896590659165926593659465956596659765986599660066016602660366046605660666076608660966106611661266136614661566166617661866196620662166226623662466256626662766286629663066316632663366346635663666376638663966406641664266436644664566466647664866496650665166526653665466556656665766586659666066616662666366646665666666676668666966706671667266736674667566766677667866796680668166826683668466856686668766886689669066916692669366946695669666976698669967006701670267036704670567066707670867096710671167126713671467156716671767186719672067216722672367246725672667276728672967306731673267336734673567366737673867396740674167426743674467456746674767486749675067516752675367546755675667576758675967606761676267636764676567666767676867696770677167726773677467756776677767786779678067816782678367846785678667876788678967906791679267936794679567966797679867996800680168026803680468056806680768086809681068116812681368146815681668176818681968206821682268236824682568266827682868296830683168326833683468356836683768386839684068416842684368446845684668476848684968506851685268536854685568566857685868596860686168626863686468656866686768686869687068716872687368746875687668776878687968806881688268836884688568866887688868896890689168926893689468956896689768986899690069016902690369046905690669076908690969106911691269136914691569166917691869196920692169226923692469256926692769286929693069316932693369346935693669376938693969406941694269436944694569466947694869496950695169526953695469556956695769586959696069616962696369646965696669676968696969706971697269736974697569766977697869796980698169826983698469856986698769886989699069916992699369946995699669976998699970007001700270037004700570067007700870097010701170127013701470157016701770187019702070217022702370247025702670277028702970307031703270337034703570367037703870397040704170427043704470457046704770487049705070517052705370547055705670577058705970607061706270637064706570667067706870697070707170727073707470757076707770787079708070817082708370847085708670877088708970907091709270937094709570967097709870997100710171027103710471057106710771087109711071117112711371147115711671177118711971207121712271237124712571267127712871297130713171327133713471357136713771387139714071417142714371447145714671477148714971507151715271537154715571567157715871597160716171627163716471657166716771687169717071717172717371747175717671777178717971807181718271837184718571867187718871897190719171927193719471957196719771987199720072017202720372047205720672077208720972107211721272137214721572167217721872197220722172227223722472257226722772287229723072317232723372347235723672377238723972407241724272437244724572467247724872497250725172527253725472557256725772587259726072617262726372647265726672677268726972707271727272737274727572767277727872797280728172827283728472857286728772887289729072917292729372947295729672977298729973007301730273037304730573067307730873097310731173127313731473157316731773187319732073217322732373247325732673277328732973307331733273337334733573367337733873397340734173427343734473457346734773487349735073517352735373547355735673577358735973607361736273637364736573667367736873697370737173727373737473757376737773787379738073817382738373847385738673877388738973907391739273937394739573967397739873997400740174027403740474057406740774087409741074117412741374147415741674177418741974207421742274237424742574267427742874297430743174327433743474357436743774387439744074417442744374447445744674477448744974507451745274537454745574567457745874597460746174627463746474657466746774687469747074717472747374747475747674777478747974807481748274837484748574867487748874897490749174927493749474957496749774987499750075017502750375047505750675077508750975107511751275137514751575167517751875197520752175227523752475257526752775287529753075317532753375347535753675377538753975407541754275437544754575467547754875497550755175527553755475557556755775587559756075617562756375647565756675677568756975707571757275737574757575767577757875797580758175827583758475857586758775887589759075917592759375947595759675977598759976007601760276037604760576067607760876097610761176127613761476157616761776187619762076217622762376247625762676277628762976307631763276337634763576367637763876397640764176427643764476457646764776487649765076517652765376547655765676577658765976607661// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0
/* Copyright (c) 2016-2018 Mellanox Technologies. All rights reserved */

#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 <linux/if_macvlan.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 <net/switchdev.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,
					 struct netlink_ext_ack *extack);
	void (*fdb_del)(struct mlxsw_sp_rif *rif, const char *mac);
};

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);
}

#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,
	MLXSW_SP_FIB_ENTRY_TYPE_NVE_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;
}

int mlxsw_sp_router_tb_id_vr_id(struct mlxsw_sp *mlxsw_sp, u32 tb_id,
				u16 *vr_id)
{
	struct mlxsw_sp_vr *vr;

	vr = mlxsw_sp_vr_find(mlxsw_sp, tb_id);
	if (!vr)
		return -ESRCH;
	*vr_id = vr->id;

	return 0;
}

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, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
				  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, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
			   1, 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);
}

static struct mlxsw_sp_fib_entry *
mlxsw_sp_router_ip2me_fib_entry_find(struct mlxsw_sp *mlxsw_sp, u32 tb_id,
				     enum mlxsw_sp_l3proto proto,
				     const union mlxsw_sp_l3addr *addr,
				     enum mlxsw_sp_fib_entry_type type)
{
	struct mlxsw_sp_fib_entry *fib_entry;
	struct mlxsw_sp_fib_node *fib_node;
	unsigned char addr_prefix_len;
	struct mlxsw_sp_fib *fib;
	struct mlxsw_sp_vr *vr;
	const void *addrp;
	size_t addr_len;
	u32 addr4;

	vr = mlxsw_sp_vr_find(mlxsw_sp, tb_id);
	if (!vr)
		return NULL;
	fib = mlxsw_sp_vr_fib(vr, proto);

	switch (proto) {
	case MLXSW_SP_L3_PROTO_IPV4:
		addr4 = be32_to_cpu(addr->addr4);
		addrp = &addr4;
		addr_len = 4;
		addr_prefix_len = 32;
		break;
	case MLXSW_SP_L3_PROTO_IPV6: /* fall through */
	default:
		WARN_ON(1);
		return NULL;
	}

	fib_node = mlxsw_sp_fib_node_lookup(fib, addrp, addr_len,
					    addr_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 != type)
		return NULL;

	return fib_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;
}

int mlxsw_sp_router_nve_promote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id,
				      enum mlxsw_sp_l3proto ul_proto,
				      const union mlxsw_sp_l3addr *ul_sip,
				      u32 tunnel_index)
{
	enum mlxsw_sp_fib_entry_type type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP;
	struct mlxsw_sp_fib_entry *fib_entry;
	int err;

	/* It is valid to create a tunnel with a local IP and only later
	 * assign this IP address to a local interface
	 */
	fib_entry = mlxsw_sp_router_ip2me_fib_entry_find(mlxsw_sp, ul_tb_id,
							 ul_proto, ul_sip,
							 type);
	if (!fib_entry)
		return 0;

	fib_entry->decap.tunnel_index = tunnel_index;
	fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP;

	err = mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
	if (err)
		goto err_fib_entry_update;

	return 0;

err_fib_entry_update:
	fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP;
	mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
	return err;
}

void mlxsw_sp_router_nve_demote_decap(struct mlxsw_sp *mlxsw_sp, u32 ul_tb_id,
				      enum mlxsw_sp_l3proto ul_proto,
				      const union mlxsw_sp_l3addr *ul_sip)
{
	enum mlxsw_sp_fib_entry_type type = MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP;
	struct mlxsw_sp_fib_entry *fib_entry;

	fib_entry = mlxsw_sp_router_ip2me_fib_entry_find(mlxsw_sp, ul_tb_id,
							 ul_proto, ul_sip,
							 type);
	if (!fib_entry)
		return;

	fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_TRAP;
	mlxsw_sp_fib_entry_update(mlxsw_sp, fib_entry);
}

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_init(struct mlxsw_sp *mlxsw_sp);

static void mlxsw_sp_router_update_priority_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_router_init(mlxsw_sp);
	kfree(net_work);
}

static int mlxsw_sp_router_schedule_work(struct net *net,
					 struct notifier_block *nb,
					 void (*cb)(struct work_struct *))
{
	struct mlxsw_sp_netevent_work *net_work;
	struct mlxsw_sp_router *router;

	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, cb);
	net_work->mlxsw_sp = router->mlxsw_sp;
	mlxsw_core_schedule_work(&net_work->work);
	return NOTIFY_DONE;
}

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 *mlxsw_sp;
	unsigned long interval;
	struct neigh_parms *p;
	struct neighbour *n;

	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:
		return mlxsw_sp_router_schedule_work(ptr, nb,
				mlxsw_sp_router_mp_hash_event_work);

	case NETEVENT_IPV4_FWD_UPDATE_PRIORITY_UPDATE:
		return mlxsw_sp_router_schedule_work(ptr, nb,
				mlxsw_sp_router_update_priority_work);
	}

	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_count_query(mlxsw_sp,
					      MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
					      *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, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
				  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, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
			   old_ecmp_size, 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, MLXSW_SP_KVDL_ENTRY_TYPE_ADJ,
				   nh_grp->ecmp_size, 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:
	case MLXSW_SP_FIB_ENTRY_TYPE_NVE_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 ||
	    fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_NVE_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_nve_decap(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_tun_pack(ralue_pl,
					   fib_entry->decap.tunnel_index);
	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(ralue), ralue_pl);
}

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);
	case MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP:
		return mlxsw_sp_fib_entry_op_nve_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) };
	u32 tb_id = mlxsw_sp_fix_tb_id(fen_info->tb_id);
	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);
		}
		if (mlxsw_sp_nve_ipv4_route_is_decap(mlxsw_sp, tb_id,
						     dip.addr4)) {
			u32 t_index;

			t_index = mlxsw_sp_nve_decap_tunnel_index_get(mlxsw_sp);
			fib_entry->decap.tunnel_index = t_index;
			fib_entry->type = MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP;
			return 0;
		}
		/* 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 bool mlxsw_sp_fib6_rt_can_mp(const struct fib6_info *rt)
{
	/* RTF_CACHE routes are ignored */
	return (rt->fib6_flags & (RTF_GATEWAY | RTF_ADDRCONF)) == RTF_GATEWAY;
}

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 replace)
{
	struct mlxsw_sp_fib6_entry *fib6_entry;

	if (!mlxsw_sp_fib6_rt_can_mp(nrt) || replace)
		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 &&
		    mlxsw_sp_fib6_rt_can_mp(rt))
			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, *fallback = NULL;

	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) {
			if (mlxsw_sp_fib6_rt_can_mp(rt) ==
			    mlxsw_sp_fib6_rt_can_mp(nrt))
				return fib6_entry;
			if (mlxsw_sp_fib6_rt_can_mp(nrt))
				fallback = fallback ?: fib6_entry;
		}
		if (rt->fib6_metric > nrt->fib6_metric)
			return fallback ?: fib6_entry;
	}

	return fallback;
}

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)
{
	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, replace);
	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;
	}

	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_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 <