The QueueLib<T, MODE>
function allows the selection of either static or dynamic priority mode.
-
STATIC_MODE
: in static mode, the system employs a singly linked list. Elements within this list are arranged in descending order of priority (1 - 99). -
DYNAMIC_MODE
: dynamic mode involves the utilization of groups, each identified by a static integer ID assigned during group creation. These groups operate with dynamic priorities. Similarly to the static mode, the structure employed here is a singly linked list, where elements are arranged in a first-in-first-out (FIFO) manner. To manage the ordering of groups by priority, a separate singly linked list is used. When groups are created, they are sorted by decreasing priority (the first group created will be the first, the second will be the second and so on).
-
push(data, param)
: the function inserts an element into the correct position according to the priority (param). Data is the element we want to add. -
pull(data)
: the function retrieves and removes the first element from the list, that is the element with most priority.
-
push(idGroup)
: this function inserts a new group at the end of singly linked list. -
push(data, param)
: the function inserts an element at the end of the specified group (param). -
pull(data)
: it retrieves and removes the first element from the highest-priority group. -
pull(data, idGroup)
: this function performs the same task as the basic pull() function but is also able to target a specific group. Upon retrieval of an element from the specified group, that group becomes the highest-priority group and is moved to the head of the priority ordering list.
In both functionalities, in all functions, there is a paramater (idThread) which indicates the specific thread that is using the queue.
For implementation details and usage examples, refer to the accompanying code documentation.
Each method that modifies the queue (pull()
and push()
) are protected by a mutex (semaphore). In this way, only one thread at a time can make changes to the queue, avoiding conflicts.
Thread 1840230400 -> I'm inside
StaticLinkedList: {(1) data=[1]}
1840230400 -> I'm going to release the MUTEX
1840803840 -> I'm inside
StaticLinkedList: {(1) data=[2](2) data=[1]}
1840803840 -> I'm going to release the MUTEX
1841377280 -> I'm inside
StaticLinkedList: {(1) data=[5](2) data=[2](3) data=[1]}
1841377280 -> I'm going to release the MUTEX
1840230400 -> I'm inside
StaticLinkedList: {(1) data=[5](2) data=[3](3) data=[2](4) data=[1]}
1840230400 -> I'm going to release the MUTEX
1840803840 -> I'm inside
StaticLinkedList: {(1) data=[5](2) data=[4](3) data=[3](4) data=[2](5) data=[1]}
1840803840 -> I'm going to release the MUTEX
1841377280 -> I'm inside
StaticLinkedList: {(1) data=[6](2) data=[5](3) data=[4](4) data=[3](5) data=[2](6) data=[1]}
1841377280 -> I'm going to release the MUTEX
1840230400 -> I'm inside
StaticLinkedList: {(1) data=[5](2) data=[4](3) data=[3](4) data=[2](5) data=[1]}
1840230400 -> I'm going to release the MUTEX
6
1840803840 -> I'm inside
StaticLinkedList: {(1) data=[4](2) data=[3](3) data=[2](4) data=[1]}
1840803840 -> I'm going to release the MUTEX
5
1841377280 -> I'm inside
StaticLinkedList: {(1) data=[3](2) data=[2](3) data=[1]}
1841377280 -> I'm going to release the MUTEX
4
1840230400 -> I'm inside
StaticLinkedList: {(1) data=[2](2) data=[1]}
1840230400 -> I'm going to release the MUTEX
3
1840803840 -> I'm inside
StaticLinkedList: {(1) data=[1]}
1840803840 -> I'm going to release the MUTEX
2
1841377280 -> I'm inside
StaticLinkedList: {}
1841377280 -> I'm going to release the MUTEX
1
// init the queue in static mode, with int items and verbose = true
QueueLib<int, STATIC_MODE> queue(true);
queue.push(1, 1, tid);
queue.push(3, 3, tid);
queue.push(2, 2, tid);
int value = 0;
queue.pull(value, tid);
std::cout << value << std::endl;
In output:
StaticLinkedList: {(1) data=[1]}
StaticLinkedList: {(1) data=[3](2) data=[1]}
StaticLinkedList: {(1) data=[3](2) data=[2](3) data=[1]}
StaticLinkedList: {(1) data=[2](2) data=[1]}
3
StaticLinkedList: {(1) data=[1]}
2
StaticLinkedList: {}
1
// init the queue in dynamic mode, with int items and verbose = true
QueueLib<int, DYNAMIC_MODE> queue(true);
// creation group 1
queue.push(1, tid);
// adding values "11" and "12"
queue.push(11, 1, tid);
queue.push(12, 1, tid);
// creation group 2
queue.push(2, tid);
// adding values "21" and "22"
queue.push(21, 2, tid);
queue.push(22, 2, tid);
// creation group 3
queue.push(3, tid);
// adding values "31"
queue.push(31, 3, tid);
int value = 0;
queue.pull(value, tid);
std::cout << value << std::endl;
queue.pull(value, 3, tid);
std::cout << value << std::endl;
queue.pull(value, tid);
std::cout << value << std::endl;
queue.pull(value, 1, tid);
std::cout << value << std::endl;
queue.pull(value, tid);
std::cout << value << std::endl;
queue.pull(value, tid);
std::cout << value << std::endl;
In output:
GroupsLinkedList: {(1) data=[SinglyLinkedList: {(1) data=[11]}]}
GroupsLinkedList: {(1) data=[SinglyLinkedList: {(1) data=[11](2) data=[12]}]}
GroupsLinkedList: {(1) data=[SinglyLinkedList: {(1) data=[11](2) data=[12]}](2) data=[SinglyLinkedList: {(1) data=[21]}]}
GroupsLinkedList: {(1) data=[SinglyLinkedList: {(1) data=[11](2) data=[12]}](2) data=[SinglyLinkedList: {(1) data=[21](2) data=[22]}]}
GroupsLinkedList: {(1) data=[SinglyLinkedList: {(1) data=[11](2) data=[12]}](2) data=[SinglyLinkedList: {(1) data=[21](2) data=[22]}](3) data=[SinglyLinkedList: {(1) data=[31]}]}
GroupsLinkedList: {(1) data=[SinglyLinkedList: {(1) data=[12]}](2) data=[SinglyLinkedList: {(1) data=[21](2) data=[22]}](3) data=[SinglyLinkedList: {(1) data=[31]}]}
11
GroupsLinkedList: {(3) data=[SinglyLinkedList: {}](1) data=[SinglyLinkedList: {(1) data=[12]}](2) data=[SinglyLinkedList: {(1) data=[21](2) data=[22]}]}
31
GroupsLinkedList: {(3) data=[SinglyLinkedList: {}](1) data=[SinglyLinkedList: {}](2) data=[SinglyLinkedList: {(1) data=[21](2) data=[22]}]}
12
Group empty!
0
GroupsLinkedList: {(1) data=[SinglyLinkedList: {}](3) data=[SinglyLinkedList: {}](2) data=[SinglyLinkedList: {(1) data=[22]}]}
21
GroupsLinkedList: {(1) data=[SinglyLinkedList: {}](3) data=[SinglyLinkedList: {}](2) data=[SinglyLinkedList: {}]}
22
-
Main
[link]: this branch contains the files needed to implement this library within a project. -
Pthread
[link]: this branch contains all the files needed to run the library in thread safe mode. -
Bazel-test
[link]: this branch contains all the files needed to run the library’s automated tests. -
Static-mode
[link]: this branch was used to create the static linked list. -
Dynamic-mode
[link]: this branch was used to create the FIFO linked list and the groups linked list.
-
v1.0.0
: release for implementing this library in thread safe mode, without order. Obviously, when multiple threads are waiting to access the shared resource, they are waiting on the mutex semaphore. Once the semaphore POST is invoked, a random thread is awakened. -
v2.0.0
: Release for implementing this library in FIFO thread safe mode. When multiple threads are waiting to access the shared resource, they are waiting on a specific semaphore, one for each thread. A SinglyLinkedList to queue the semaphores that processes will suspend on in order of arrival. Once the shared resource is released, the thread at the head of the list will be woken up with token passing.