processes - computer scienceporter/courses/comp530/f18/slides/processes.pdf · • a process is the...

COMP530:OperatingSystems

Processes

DonPorter

PortionscourtesyEmmettWitchel

1


App

Whatisaprocess?

2-2

Hardware

Libraries

Kernel

User

Super-visor

App

Libraries

Daemon

Libraries

SystemCallTable(350—1200)

Intuitively,oneofthese


Whatisaprocess?• Aprocessisaprogramduringexecution.

– Program=staticfile(image)– Process=executingprogram=program+executionstate.

• Aprocessisthebasicunitofexecutioninanoperatingsystem– Eachprocesshasanumber,itsprocessidentifier(pid).

• Differentprocessesmayrundifferentinstancesofthesameprogram– E.g.,myjavac andyourjavac processboth runtheJavacompiler

• Ataminimum,processexecutionrequiresfollowingresources:– Memory tocontaintheprogramcodeanddata– AsetofCPUregisterstosupportexecution

3


• Wewriteaprogramine.g.,Java.• Acompilerturnsthatprogramintoaninstructionlist.• TheCPUinterpretstheinstructionlist(whichismoreagraphofbasic

blocks).

void X (int b) {

if(b == 1) {

…

int main() {

int a = 2;X(a);

}

Programtoprocess


void X (int b) {

if(b == 1) {

…

int main() {

int a = 2;X(a);

}

Whatyouwrote:

void X (int b) {

if(b == 1) {

…

int main() {

int a = 2;X(a);

} Code

main; a = 2X; b = 2

Heap

Stack

ProcessinmemoryWhatisinmemory:

Data


Wheretoprocessescomefrom?• WhenItype‘./a.out’,thebinaryruns,right?

– Reallyonlytrueforstaticbinaries(moreonthislater)

• Inrealityaloader setsuptheprogram– Usuallyauser-levelprogram– Canalsobein-kernel,orsplitbetweenboth

6


Wheretoprocessescomefrom?• Inordertorunaprogram,theloader:

– readsandinterpretstheexecutablefile– setsuptheprocess’smemorytocontainthecode&datafrom

executable– pushes“argc”,“argv” onthestack– setstheCPUregistersproperly&calls“_start()”

• Programstartsrunningat_start()_start(args) {

initialize_java();ret = main(args);exit(ret)

}

“process” isnowrunning;nolongerthinkof“program”

• Whenmain()returns,OScalls“exit()”whichdestroystheprocessandreturnsallresources

7WhatbookkeepingdoestheOSneedforprocesses?


• A process has code.– OS must track program counter (code location).

• A process has a stack.– OS must track stack pointer.

• OS stores state of processes’ computation in a process control block (PCB).– E.g., each process has an identifier (process identifier,

or PID)• Data (program instructions, stack & heap) resides

in memory, metadata is in PCB (which is a kernel data structure in memory)

Keepingtrackofaprocess


ContextSwitching• TheOSperiodicallyswitchesexecutionfromoneprocesstoanother

• Calleda contextswitch,becausetheOSsavesoneexecutioncontextandloadsanother


Whatcausescontextswitches?• WaitingforI/O(disk,network,etc.)

– MightaswellusetheCPUforsomethinguseful– Calledablockedstate

• Timerinterrupt(preemptivemultitasking)– Evenifaprocessisbusy,weneedtobefairtootherprograms

• Voluntaryyielding(cooperativemultitasking)• Afewothers

– Synchronization,IPC,etc.


Processlifecycle• Processesarealwayseither:

– Executing– Waitingtoexecute, or– Blockedwaitingforanevent tooccur

11

RunningReady

Blocked

Start Done


Operating System

“SystemSoftware”

User Process 1

User Program 2User Process 2

User Process n

...Process 1 Process 2OS I/O

Device

k: read()

k+1:

startIO()

endio{ interrupt

main{

main{

}

read{

}

}

schedule()

Memory

savestate schedule()

restorestate

savestate

Processcontexts


1. Ready2. Running3. Blocked4. Zombie5. Exited

WhenaprocessiswaitingforI/O,whatisitsstate?


CPUScheduling• Problemofchoosingwhichprocesstorunnext

– Andforhowlonguntilthenextprocessruns

• Whybother?– Improveperformance:amortizecontextswitchingcosts– Improveuserexperience:e.g.,lowlatencykeystrokes– Priorities:favor“important”workoverbackgroundwork– Fairness

14Wewillcovertechniqueslater


Whendoesschedulinghappen?• Whenaprocessblocks• WhenadeviceinterruptstheCPUtoindicateaneventoccurred(possiblyun-blockingaprocess)

• WhenaprocessyieldstheCPU

• Preemptivescheduling:SettingatimertointerrupttheCPUaftersometime– PlacesanupperboundonhowlongaCPU-boundprocesscanrunwithoutgivinganotherprocessaturn

• Non-preemptivescheduling:ProcessesmustexplicitlyyieldtheCPU

15


• OS uses PCBs to represent a process• Every resource is represented with a queue• OS puts PCB on an appropriate queue.

– Ready to run queue.– Blocked for IO queue (Queue per device).– Zombie queue.

• When CPU becomes available, choose from ready to run queue

• When an event occurs, remove waiting process from blocked queue, move to ready queue.

Schedulingprocesses


Consider a Web server:get network message (URL) from clientfetch URL data from diskcompose responsesend response

How well does this web server perform?With many incoming requests?

That access data all over the disk?

Whyusemultipleprocessesinoneapp?

AsingleprocesscannotoverlapCPUandI/O


Consider a Web serverget network message (URL) from clientcreate child process, send it URL

Childfetch URL data from diskcompose responsesend response

NowthechildcanblockonI/O,parentkeepsworkingDifferentchildrencanblockonreadingdifferentfiles

Howdoesserverknowifchildsucceededorfailed?

Whyusemultipleprocessesinoneapp?


• Aftertheprogramfinishesexecution,itcallsexit()• Thissystemcall:

– takesthe“result” of theprogramasanargument– closesallopen files,connections, etc.– deallocates memory– deallocates mostoftheOSstructuressupporting theprocess– checksifparentisalive:

v Ifso,itholdstheresultvalueuntilparentrequestsit;inthiscase,processdoesnotreallydie,but itentersthezombie/defunct state

v Ifnot,itdeallocates alldatastructures,theprocessisdead

• Processterminationistheultimategarbagecollection

Orderlytermination:exit()

Webserverex:Childusesexitcodeforsuccess/failure


• Childreturnsavaluetoparentviaexit()• Theparentreceivesthisvaluewithwait()

• Specifically,wait():– Blockstheparentuntil childfinishes(needawaitqueue)– Whenachildcallsexit(), theOSunblocks theparentandreturns thevalue

passedbyexit()asaresultof thewait()call(alongwiththepid ofthechild)– Iftherearenochildrenalive,wait()returnsimmediately

Thewait()systemcall


Zombies!!!

21

• Aparentcanwaitindefinitelytocallwait()• TheOStostoretheexitcodeforafinishedchilduntiltheparentcallswait()

• Hack:KeepPCBfordeadprocessesarounduntil:– Parentcallswait(),or– Parentexit()s(don’tneedtowait()ongrandkids)

• Andthatisazombie(donestate)– Willnotbescheduledagain


Wheredoprocessescomefrom?(redux)• Parent/childmodel• Anexistingprogramhastospawnanewone

– MostOSes haveaspecial‘init’programthatlaunchessystemservices,logondaemons,etc.

– Whenyoulogin(viaaterminalorssh),theloginprogramspawnsyourshell


Approach1:WindowsCreateProcess• InWindows,whenyoucreateanewprocess,youspecifytheprogram– Andcanoptionallyallowthechildtoinheritsomeresources(e.g.,anopenfilehandle)


Approach2:Unixfork/exec()• InUnix,aparentmakesacopy ofitselfusingfork()

– Childinheritseverything,runssameprogram– Onlydifferenceisthereturnvaluefromfork()

• Childgets0;parentgetschildpid

• Aseparateexec()systemcallloadsanewprogram– Likegettingabraintransplant

• Someprograms,likeourwebserverexample,fork()clones(withoutcallingexec()).– Commoncaseisprobablyfork+exec


• The exec() call allows a process to “load” a different program and start execution at main (actually _start).

• It allows a process to specify the number of arguments (argc) and the string argument array (argv).

• If the call is successful– it is the same process …– but it runs a different program !!

• Code, stack & heap is overwritten– Sometimes memory mapped files are preserved.

• Exec does not return!

Programloading:exec()


In the parent process:main()…int rv =fork(); // create a childif(0 == rv) { // child continues here

exec_status = exec(“calc”, argc, argv0, argv1, …);printf(“Something is horribly wrong\n”); exit(exec_status);

} else { // parent continues hereprintf(“Shall I be mother?”);…child_status = wait(rv);

}

Exec should not return

fork()+exec()example


pid = 127open files = “.history”last_cpu = 0


int rv = fork();if(rv == 0) {close(“.history”);exec(“/bin/calc”);} else {wait(rv);


Process Control Blocks (PCBs)

OSUSER


int rvc_main(){irvq = 7;do_init();ln = get_input();exec_in(ln);

pid = 128open files = last_cpu = 0


Ashellforksandexecsacalculator




int shell_main() {int a = 2;… Code

main; a = 2

Heap

Stack

0xFC0933CA

int shell_main() {int a = 2;… Code

main; a = 2

Heap

Stack

0xFC0933CA

int calc_main() {int q = 7;… Code

Heap

Stack

0x43178050

pid = 128open files =last_cpu = 0

OSUSER

Process Control Blocks (PCBs)

Ashellforksandthenexecsacalculator


Whyseparatefork&exec?• Keyissue:Inheritance offiledescriptors,environment,etc.– Or,makingtheshellwork

• Rememberhowtheshellcandoredirection?– ./warmup <testinput.txt– Filehandle0(stdin)isopenedtoreadtestinput.txt

• Theparent(shell)openstestinput.txt beforefork()– Thechild(warmup)inheritsthisopenfilehandle

• Evenafterexec()

29


• Decoupling fork and exec lets you do anything to the child’s process environment without adding it to the CreateProcess API.

int rv = fork(); // create a childIf(0 == rv) { // child continues here

// Do anything (unmap memory, close net connections…)exec(“program”, argc, argv0, argv1, …);

}fork() creates a child process that inherits:Ø identical copy of all parent’s variables & memoryØ identical copy of all parent’s CPU registers (except one)

Parent and child execute at the same point after fork() returns:Ø by convention, for the child, fork() returns 0Ø by convention, for the parent, fork() returns the pid of the child

Theconvenienceofseparatefork/exec


TheCreateProcess alternative• Windowsdoesallowyoutocreateaprocessthatisinitiallysuspended– Youcanalsochangememoryandhandlesofanotherprocess

– Andthenunblockit

• Somewhatisomorphic– Butabitcumbersome– Andpronetosecurityissues(loadingthreadsandlibrariesinanotherapp!)

31


• Simple implementation of fork():– allocate memory for the child process– copy parent’s memory and CPU registers to child’s– Expensive !!

• In 99% of the time, we call exec() after calling fork()– the memory copying during fork() operation is useless– the child process will likely close the open files & connections– overhead is therefore high

Atwhatcost,fork()?

Anyideastoimprovethis?


Protool:vfork• Ifyouknowyouaregoingtocallexec()almostimmediately:

– CreateanewPCB,stack,registerstate– Butnotanewcopyofthefullmemory

• YoucanchangeOSstateandcallexecsafely• Youcannot:

– Returnfromthefunctionthatcalledfork()– Touchtheheap– Probablyotherstuff

• Whydoesitimproveperformance?Avoidscopies

• Unfortunateexampleofimplementationinfluenceoninterface– CurrentLinux&BSD4.4haveitforbackwardscompatibility

33


Copy-on-writefork(preview)• Idea:writeprotecteverythinginmemoryafterafork()– Detectandcopyonlywhatyoutouch,untiltheexec()– Afterexec(),removewriteprotectionfromchildmemory

• Commoncase:execquickly– Someoverheadtosettingcopy-on-write,butcheaperthancopyingeverything

• Uncommoncase:forkneverexecs– Eventuallycopyeverything

• Wewillseemoreaboutthislater…

34


OSmustincludecallstoenablespecialcontrolofaprocess:

• Prioritymanipulation:– nice(),whichspecifiesbaseprocesspriority (initialpriority)– InUNIX,processprioritydecaysastheprocessconsumesCPU

• Debuggingsupport:– ptrace(),allowsaprocesstobeputundercontrolofanotherprocess– Theotherprocesscansetbreakpoints, examineregisters,etc.

• Alarmsandtime:– Sleepputsaprocessonatimerqueuewaitingforsomenumberofseconds,

supporting analarmfunctionality

Processcontrol


while(! EOF) {read inputhandle regular expressionsint rv = fork(); // create a childif(rv == 0) { // child continues here

exec(“program”, argc, argv0, argv1, …);}else { // parent continues here…}

Translates<CTRL-C>tothekill()systemcallwithSIGKILL

Translates<CTRL-Z>tothekill()systemcallwithSIGSTOP

Allowsinput-output redirections,pipes,andalotofotherstuffthatwewillseelater

Tyingitalltogether:TheUnixshell


Summary• Understandwhataprocessis• Thehigh-levelideaofcontextswitchingandprocessstates

• Howaprocessiscreated• ProsandconsofdifferentcreationAPIs

– Intuitionofcopy-on-writeforkandvfork

processes - computer scienceporter/courses/comp530/f18/slides/processes.pdf · • a process is the...

Documents