Mutexes are often used to prevent multiple threads from accessing critical resources at the same time. Sometimes, when locking mutexes, multiple threads hold each other's lock, and the program consequently deadlocks. There are four requirements for deadlock:
- mutual Mutual exclusion
- hold Hold and wait
- no No preemption
- circular Circular wait
Deadlock requires all four conditions, so to prevent deadlock, prevent any one of the four conditions. This guideline recommends locking the mutexes in a predefined order to prevent circular wait. This rule is a specific instance of CON35-C. Avoid deadlock by locking in predefined order using POSIX threads.
Noncompliant Code Example
The following This noncompliant code example has behavior that depends on the runtime environment and the platform's scheduler. However, with proper timing, the main() function will deadlock when running thr1 and thr2, where thr1 tries to lock ba2's mutex, while thr2 tries to lock on ba1's mutex in the deposit() function, and the program will not progress.
| Code Block | ||||
|---|---|---|---|---|
| ||||
typedef struct {
int balance;
pthread_mutex_t balance_mutex;
} bank_account;
typedef struct {
bank_account *from;
bank_account *to;
int amount;
} deposit_thr_args;
void create_bank_account(bank_account **ba, int initial_amount) {
int result;
bank_account *nba = malloc(sizeof(bank_account));
if (nba == NULL) {
/* Handle Errorerror */
}
nba->balance = initial_amount;
result = pthread_mutex_init(&nba->balance_mutex, NULL);
if (result) {
/* Handle Errorerror */
}
*ba = nba;
}
void *deposit(void *ptr) {
int result;
deposit_thr_args *args = (deposit_thr_args *)ptr;
if ((result = pthread_mutex_lock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
/* notNot enough balance to transfer */
if (args->from->balance < args->amount) {
if ((result = pthread_mutex_unlock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
return NULL;
}
if ((result = pthread_mutex_lock(&(args->to->balance_mutex))) != 0) {
/* Handle Errorerror */
}
args->from->balance -= args->amount;
args->to->balance += args->amount;
if ((result = pthread_mutex_unlock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
if ((result = pthread_mutex_unlock(&(args->to->balance_mutex))) != 0) {
/* Handle Errorerror */
}
free(ptr);
return NULL;
}
int main(void) {
pthread_t thr1, thr2;
int result;
bank_account *ba1;
bank_account *ba2;
create_bank_account(&ba1, 1000);
create_bank_account(&ba2, 1000);
deposit_thr_args *arg1 = malloc(sizeof(deposit_thr_args));
if (arg1 == NULL) {
/* Handle Errorerror */
}
deposit_thr_args *arg2 = malloc(sizeof(deposit_thr_args));
if (arg2 == NULL) {
/* Handle Errorerror */
}
arg1->from = ba1;
arg1->to = ba2;
arg1->amount = 100;
arg2->from = ba2;
arg2->to = ba1;
arg2->amount = 100;
/* performPerform the deposits */
if ((result = pthread_create(&thr1, NULL, deposit, (void *)arg1)) != 0) {
/* Handle Errorerror */
}
if ((result = pthread_create(&thr2, NULL, deposit, (void *)arg2)) != 0) {
/* Handle Errorerror */
}
pthread_exit(NULL);
return 0;
}
|
...
The solution to the deadlock problem is to use a predefined order for the locks in the deposit() function. In the following this compliant solution, each thread will lock an on the basis of the bank_account ID, defined in the struct initialization. This solution prevents the circular wait problem.:
| Code Block | ||||
|---|---|---|---|---|
| ||||
typedef struct {
int balance;
pthread_mutex_t balance_mutex;
unsigned int id; /* shouldShould never be changed after initialized */
} bank_account;
unsigned int global_id = 1;
void create_bank_account(bank_account **ba, int initial_amount) {
int result;
bank_account *nba = malloc(sizeof(bank_account));
if (nba == NULL) {
/* Handle Errorerror */
}
nba->balance = initial_amount;
result = pthread_mutex_init(&nba->balance_mutex, NULL);
if (result != 0) {
/* Handle Errorerror */
}
nba->id = global_id++;
*ba = nba;
}
void *deposit(void *ptr) {
deposit_thr_args *args = (deposit_thr_args *)ptr;
int result;
if (args->from->id == args->to->id)
return;
/* ensureEnsure proper ordering for locking */
if (args->from->id < args->to->id) {
if ((result = pthread_mutex_lock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
if ((result = pthread_mutex_lock(&(args->to->balance_mutex))) != 0) {
/* Handle Errorerror */
}
} else {
if ((result = pthread_mutex_lock(&(args->to->balance_mutex))) != 0) {
/* Handle Errorerror */
}
if ((result = pthread_mutex_lock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
}
/* notNot enough balance to transfer */
if (args->from->balance < args->amount) {
if ((result = pthread_mutex_unlock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
if ((result = pthread_mutex_unlock(&(args->to->balance_mutex))) != 0) {
/* Handle Errorerror */
}
return;
}
args->from->balance -= args->amount;
args->to->balance += args->amount;
if ((result = pthread_mutex_unlock(&(args->from->balance_mutex))) != 0) {
/* Handle Errorerror */
}
if ((result = pthread_mutex_unlock(&(args->to->balance_mutex))) != 0) {
/* Handle Errorerror */
}
free(ptr);
return;
}
|
Risk Assessment
Deadlock prevents multiple threads from progressing, thus halting the executing program. A denial-of-service attack is possible because the attacker can force deadlock situations. Deadlock is likely to occur in multithreaded programs that manage multiple shared resources.
Recommendation | Severity | Likelihood | Detectable |
|---|
Repairable | Priority | Level |
|---|---|---|
POS51-C | Low |
Probable |
No |
No |
P4
L3
Related Guidelines
...
P2 | L3 |
Automated Detection
| Tool | Version | Checker | Description | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Astrée |
| Supported: Astrée reports all potential deadlocks. | |||||||
| CodeSonar |
| CONCURRENCY.LOCK.ORDER | Conflicting lock order | ||||||
| Helix QAC |
| C1772, C1773 | |||||||
| Klocwork |
| CONC.DL | |||||||
| Parasoft C/C++test |
| CERT_C-POS51-a | Do not acquire locks in different order | ||||||
| Polyspace Bug Finder |
| CERT C: Rule POS51-C | Checks for deadlock (rule fully covered) |
Related Guidelines
Key here (explains table format and definitions)
Taxonomy | Taxonomy item | Relationship |
|---|---|---|
| CERT C | LCK07-J. Avoid deadlock by requesting and releasing locks in the same order |
MITRE CWE: CWE-764 Multiple locks of critical resources
...
| Prior to 2018-01-12: CERT: Unspecified Relationship |
CERT-CWE Mapping Notes
Key here for mapping notes
CWE-764 and POS51-C/POS35-C
Independent( CWE-764, POS51-C, POS35-C)
CWE-764 is about semaphores, or objects capable of being locked multiple times. Deadlock arises from multiple locks being acquired in a cyclic order, and generally does not arise from semaphores or recursive mutexes.
Bibliography
| [Barney 2010] | pthread_mutex tutorial |
| [Bryant 2003] | Chapter 13, "Concurrent Programming" |
...