15 213 The course that gives CMU its Zip Synchronization November 19 2008 Topics lecture 24 ppt Synchronizing with semaphores Races and deadlocks Thread safety and reentrancy badcnt c An Improperly Synchronized Threaded Program shared volatile unsigned int cnt 0 define NITERS 100000000 int main pthread t tid1 tid2 Pthread create tid1 count Pthread create tid2 count NULL NULL NULL NULL Pthread join tid1 NULL Pthread join tid2 NULL if cnt unsigned NITERS 2 printf BOOM cnt d n cnt else printf OK cnt d n cnt 2 thread routine void count void arg int i for i 0 i NITERS i cnt return NULL linux badcnt BOOM cnt 198841183 linux badcnt BOOM cnt 198261801 linux badcnt BOOM cnt 198269672 cnt should be equal to 200 000 000 What went wrong 15 213 F 08 Assembly Code for Counter Loop C code for counter loop Corresponding asm code for i 0 i NITERS i cnt L9 movl 4 ebp eax cmpl 99999999 eax jle L12 jmp L10 Head Hi Load cnt Li Update cnt Ui Store cnt Si L12 movl cnt eax leal 1 eax edx movl edx cnt Load Update Store L11 movl 4 ebp eax leal 1 eax edx movl edx 4 ebp jmp L9 Tail Ti L10 3 15 213 F 08 Concurrent Execution Key idea In general any sequentially consistent interleaving is possible but some are incorrect 4 Ii denotes that thread i executes instruction I eaxi is the contents of eax in thread i s context i thread instri eax1 eax2 cnt 1 1 1 1 2 2 2 2 2 1 H1 L1 U1 S1 H2 L2 U2 S2 T2 T1 0 1 1 1 1 2 2 2 0 0 0 1 1 1 1 2 2 2 OK 15 213 F 08 Concurrent Execution cont Incorrect ordering two threads increment the counter but the result is 1 instead of 2 5 i thread instri eax1 eax2 cnt 1 1 1 2 2 1 1 2 2 2 H1 L1 U1 H2 L2 S1 T1 U2 S2 T2 0 1 1 1 0 1 1 1 0 0 0 0 0 1 1 1 1 1 Oops 15 213 F 08 Concurrent Execution cont How about this ordering i thread instri 1 1 2 2 2 2 1 1 1 2 H1 L1 H2 L2 U2 S2 U1 S1 T1 T2 eax1 eax2 cnt We can clarify our understanding of concurrent execution with the help of the progress graph 6 15 213 F 08 Progress Graphs A progress graph depicts the discrete execution state space of concurrent threads Thread 2 T2 L1 S2 Each axis corresponds to the sequential order of instructions in a thread S2 U2 Each point corresponds to a possible execution state Inst1 Inst2 L2 H2 H1 7 L1 U1 S1 T1 Thread 1 E g L1 S2 denotes state where thread 1 has completed L1 and thread 2 has completed S2 15 213 F 08 Trajectories in Progress Graphs Thread 2 A trajectory is a sequence of legal state transitions that describes one possible concurrent execution of the threads T2 S2 Example U2 H1 L1 U1 H2 L2 S1 T1 U2 S2 T2 L2 H2 H1 8 L1 U1 S1 T1 Thread 1 15 213 F 08 Critical Sections and Unsafe Regions Thread 2 L U and S form a critical section with respect to the shared variable cnt T2 Instructions in critical sections wrt to some shared variable should not be interleaved S2 critical section wrt cnt U2 Unsafe region Sets of states where such interleaving occurs form unsafe regions L2 H2 H1 L1 U1 S1 T1 Thread 1 critical section wrt cnt 9 15 213 F 08 Safe and Unsafe Trajectories Thread 2 T2 Safe trajectory S2 critical section wrt cnt Unsafe trajectory Unsafe region Def A trajectory is safe iff it doesn t touch any part of an unsafe region Claim A trajectory is correct wrt cnt iff it is safe U2 L2 H2 H1 L1 U1 S1 T1 Thread 1 critical section wrt cnt 10 15 213 F 08 Semaphores Question How can we guarantee a safe trajectory We must synchronize the threads so that they never enter an unsafe state Classic solution Dijkstra s P and V operations on semaphores semaphore non negative integer synchronization variable P s while s 0 wait s Dutch for Proberen test V s s Dutch for Verhogen increment OS guarantees that operations between brackets are executed indivisibly Only one P or V operation at a time can modify s When while loop in P terminates only that P can decrement s Semaphore invariant s 0 11 15 213 F 08 Locking with Semaphores Here is one way we could use P and V operations to synchronize the threads that update cnt Semaphore used like this referred to as a lock Semaphore s is initially 1 Thread routine void count void arg int i for i 0 i NITERS i P s cnt V s return NULL 12 15 213 F 08 Safe Sharing With Locks Thread 2 T2 V s 1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 Forbidden region 1 S2 U2 0 0 0 0 0 1 1 1 1 1 1 1 Unsafe region 1 1 0 1 1 1 1 0 0 0 0 1 1 1 1 0 0 0 0 1 1 L2 P s H2 1 Initially s 1 13 H1 P s L1 U1 1 1 1 1 S1 V s T1 Provide mutually exclusive access to shared variable by surrounding critical section with P and V operations on semaphore s initially set to 1 Semaphore invariant creates a forbidden region that encloses unsafe region and is never touched by any trajectory Thread 1 15 213 F 08 Wrappers on POSIX Semaphores Initialize semaphore sem to value pshared 0 if thread pshared 1 if process void Sem init sem t sem int pshared unsigned int value if sem init sem pshared value 0 unix error Sem init P operation on semaphore sem void P sem t sem if sem wait sem unix error P V operation on semaphore sem void V sem t sem if sem post sem unix error V 14 15 213 F 08 Sharing With POSIX Semaphores properly sync d counter program include csapp h define NITERS 10000000 thread routine void count void arg int i volatile unsigned int cnt sem t sem semaphore for i 0 i NITERS i P sem cnt V sem return NULL int main pthread t tid1 tid2 Sem init sem 0 1 sem 1 create 2 threads and wait if cnt unsigned NITERS 2 printf BOOM cnt d n cnt else printf OK cnt d n cnt exit 0 Warning It s really slow 15 15 213 F 08 One worry races A race occurs when the correctness of the program depends on one thread reaching point x before another thread reaches point y a threaded program with a race int main pthread t tid N int i for i 0 i N i …
View Full Document
Unlocking...