摘要:
Various technologies and techniques are disclosed for supporting parallel nested transactions in a transactional memory system. Multiple closed nested transactions are created for a single parent transaction, and the closed nested transactions are executed concurrently as parallel nested transactions. Various techniques are used to ensure effects of the parallel nested transactions are hidden from other transactions outside the parent transaction until the parent transaction commits. For example, versioned write locks are used with parallel nested transactions. When a transactional memory word changes from a write lock to a versioned write lock, an entry is made in a global versioned write lock map to store a pointer to a write log entry that the versioned write lock replaced. When the versioned write lock is encountered during transaction processing, the global versioned write lock map is consulted to translate the versioned write lock to the pointer to the write log entry.
摘要:
Various technologies and techniques are disclosed that improve implementation of concurrency control modes in a transactional memory system. A transactional memory word is provided for each piece of data. The transactional memory word includes a version number, a reader indicator, and an exclusive writer indicator. The transactional memory word is analyzed to determine if the particular concurrency control mode is proper. Using the transactional memory word to help with concurrency control allows multiple combinations of operations to be performed against the same memory location simultaneously and/or from different transactions. For example, a pessimistic read operation and an optimistic read operation can be performed against the same memory location.
摘要:
Various technologies and techniques are disclosed for implementing retrying transactions in a transactional memory system. The system allows a transaction to execute a retry operation. The system registers for waits on every read in a read set of the retrying transaction. The retrying transaction waits for notification that something in the read set has changed. A transaction knows if notification is required in one of two ways. If the transactional memory word contained a waiters bit during write lock acquisition, then during release the transactional memory word is looked up in an object waiters map, and waiting transactions are signaled. If a writing transaction finds a global count of waiting transactions to be greater than zero after releasing write locks, a transaction waiters map is used to determine which waiting transactions need to be signaled. In each case, the write lock is released using a normal store operation.
摘要:
Various technologies and techniques are disclosed that improve implementation of concurrency control modes in a transactional memory system. A transactional memory word is provided for each piece of data. The transactional memory word includes a version number, a reader indicator, and an exclusive writer indicator. The transactional memory word is analyzed to determine if the particular concurrency control mode is proper. Using the transactional memory word to help with concurrency control allows multiple combinations of operations to be performed against the same memory location simultaneously and/or from different transactions. For example, a pessimistic read operation and an optimistic read operation can be performed against the same memory location.
摘要:
Various technologies and techniques are disclosed that improve implementation of concurrency control modes in a transactional memory system. A transactional memory word is provided for each piece of data. The transactional memory word includes a version number, a reader indicator, and an exclusive writer indicator. The transactional memory word is analyzed to determine if the particular concurrency control mode is proper. Using the transactional memory word to help with concurrency control allows multiple combinations of operations to be performed against the same memory location simultaneously and/or from different transactions. For example, a pessimistic read operation and an optimistic read operation can be performed against the same memory location.
摘要:
Various technologies and techniques are disclosed for handling exceptions in sequential statements that are executed in parallel. A transactional memory system is provided with a contention manager. The contention manager is responsible for managing exceptions that occur within statements that were designed to be executed in an original sequential order, and that were transformed into ordered transactions for speculative execution in parallel. The contention manager ensures that any exceptions that are thrown from one or more speculatively executed blocks while the statements are being executed speculatively in parallel are handled in the original sequential order.
摘要:
Various technologies and techniques are disclosed that support buffered writes and enforced serialization order in a software transactional memory system. A buffered write process is provided that performs writes to shadow copies of objects and writes content back to the objects after validating a respective transaction during commit. When a write lock is first obtained for a particular transaction, a shadow copy is made of a particular object. Writes are performed to and reads from the shadow copy. After validating the particular transaction during commit, content is written from the shadow copy to the particular object. A transaction ordering process is provided that ensures that an order in which the transactions are committed matches an abstract serialization order of the transactions. Transactions are not allowed to commit until their ticket number matches a global number that tracks the next transaction that should commit.
摘要:
Various technologies and techniques facilitate stack read and write operations in a software transactional memory system. If the compiler determines that an address for a variable in a code segment is a stack location, the stack location is live on entry, and the address of the variable has not been taken and passed to another thread, the code is changed to ensure failure atomicity. One example includes modifying the code so a shadow copy is saved for local variables that are live on entry. If the same prior criteria are true except the stack location is not live on entry, the code is optimized by ensuring code for logging and software transactional memory operations are not included. If the compiler does not know the address is the stack location or that the address is not passed to another thread, the code is changed to ensure failure and concurrency atomicity.
摘要:
Various technologies and techniques are disclosed for detecting falsely doomed parent transactions of nested children in transactional memory systems. When rolling back nested transactions, a release count is tracked each time that a write lock is released due to rollback for a given nested transaction. For example, a write abort compensation map can be used to track the release count for each nested transaction. The number of times the nested transactions releases a write lock is recorded in their respective write abort compensation map. The release counts can be used during a validation of a parent transaction to determine if a failed optimistic read is really valid. If an aggregated release count for the nested children transactions accounts for the difference in version numbers exactly, then the optimistic read is valid.
摘要:
Various technologies and techniques are disclosed for implementing retrying transactions in a transactional memory system. The system allows a transaction to execute a retry operation. The system registers for waits on every read in a read set of the retrying transaction. The retrying transaction waits for notification that something in the read set has changed. A transaction knows if notification is required in one of two ways. If the transactional memory word contained a waiters bit during write lock acquisition, then during release the transactional memory word is looked up in an object waiters map, and waiting transactions are signaled. If a writing transaction finds a global count of waiting transactions to be greater than zero after releasing write locks, a transaction waiters map is used to determine which waiting transactions need to be signaled. In each case, the write lock is released using a normal store operation.