International Business Machines Corporation

Lenovo (singapore) Pte. Ltd.

Shajith Chandran

Sivakumar Krishnasamy

Kavana Nandikur Bhat

Prateek Goel

Dilip Singh

Manish Kulshreshtha

Joshi Chandran

Vishal Ramachandra Mansur

Muthulakshmi Pearl Srinivasan

Sameer K Sinha

Anil Kumar Singh

Ismail Shamirulla Khan

Harikrlshna Manchineni

Ashok Kumar Jagadeeswaran

Disclosed aspects relate to address space management with respect to a Coherent Accelerator Processor Interface (CAPI) architecture. A shared CAPI component may be established to access a plurality of address spaces of a plurality of assets. The plurality of assets may include both a first asset which corresponds to a first address space of the plurality of address spaces and a second asset which corresponds to a second address space of the plurality of address spaces. A request may be detected to process a set of data and route a set of processed data from the first asset to the second asset. The set of data may be processed to form the set of processed data. The set of processed data may be routed from the first address space to the second address space to fulfill the request.

Address space management with respect to a coherent accelerator processor interface architecture

Mechanisms are provided, in a data processing system having a processor and a cache subsystem, for providing software controlled cache segmentation and cache segment utilization. The mechanisms segment a cache memory of the cache subsystem such that the cache memory comprises a plurality of cache segments. Each cache segment in the plurality of cache segments is associated with a different data property of data stored in the cache segment. The mechanisms configure software executing on the data processing system to direct cache accesses to one of the cache segments based on a corresponding data property of the cache accesses by the software. The mechanisms process, by the processor, a data access operation from software executing on the processor, based on an identifier of a cache segment in one of an effective address provided by the software or a page table entry corresponding to the effective address provided by the software.

Software controlled cache line replacement within a data property dependent cache segment of a cache using a cache segmentation enablement bit and cache segment selection bits

Embodiments of the present invention provide a system and method for a dynamic tracing framework for debugging in a virtualized environment. Embodiment of the present invention can include selecting a home node and a set of remote nodes to which apply a set of probes. Data collection is performed by a tracing agent of each node and the collection of data is shared across multiple software systems.

Dynamic tracing framework for debugging in virtualized environments

A computer-implemented method for lock-free use of a non-preemptive system resource by a preemptive thread, which may be interrupted. The method comprises registering a non-preemptive system resource and a first level reclaim handler for the non-preemptive system resource with the kernel of an operating system, registering a second level reclaim handler with the kernel, wherein the second level reclaim handler is included in an application program, and running the application program as a preemptive thread using the non-preemptive system resource. The first level reclaim handler is code that is a part of the implementation of the non-preemptive system resource in the kernel. The second level reclaim handler is code that is part of the application and is registered with the kernel before the application uses the non-preemptive system resource. The method enables a preemptive thread using a non-preemptive system resource to be preempted without crashing.

Lock free use of non-preemptive system resource

Techniques are disclosed for providing thread specific protection levels in a multithreaded processing environment. An associated method includes generating a group of threads in a process, one of the group of threads opening a thread entity, and that one of the group of threads specifying one or more levels of access to the thread entity for the other threads. In one embodiment, when a first of the threads attempts to perform a specified operation on the thread entity, the method of this invention determines whether that first thread is the one of the group of threads that opened the thread entity. When the first thread is not that one of the group of threads, the first thread is allowed to perform the specified operation if and only if such operation is permitted by the specified one or more levels of access.

Providing thread specific protection levels

An apparatus for monitoring changes to a block of data is disclosed. A first hardware watchpoint is set to monitor changes to the block of data at a current location of the block of data in memory and a second hardware watchpoint is set to monitor changes at a selected location in the memory where a reference to the block of data is located. Responsive to the second hardware watchpoint being triggered by a change at the selected location where the reference to the block of data is located, a new location of the block of data in the memory is identified based on the change that triggered the second hardware watchpoint. Subsequent to identifying the new location of the block of data, the first hardware watchpoint is reset to monitor changes to the block of data at the new location of the block of data.

Dynamic hardware watchpoint

A method and apparatus for monitoring changes to a block of data is disclosed. A computer sets a first hardware watchpoint to monitor changes to the block of data at a current location of the block of data in memory and a second hardware watchpoint to monitor changes at a selected location in the memory where a reference to the block of data is located. Responsive to the second hardware watchpoint being triggered by a change at the selected location where the reference to the block of data is located, the computer identifies a new location of the block of data in the memory based on the change that triggered the second hardware watchpoint. Subsequent to identifying the new location of the block of data, the computer then resets the first hardware watchpoint to monitor changes to the block of data at the new location of the block of data.

An approach to providing multiple concurrently executing debugging sessions for a currently executing operating system. The approach involves providing one first debugging session for debugging the currently executing operating system. The first debugging session has read access and write access to the data of the currently executing operating system. The approach also involves providing one or more second debugging sessions for the currently executing operating system. Each of the second debugging sessions has read-only access to the data of the currently executing operating system. The second debugging sessions run simultaneously with the first debugging session if the second debugging sessions are started while the first debugging session is active. As a result, multiple users can simultaneously debug the currently executing operating system. A lock may be used to ensure that only the first debugging session has write access to the data. The lock may be shared between the various debugging sessions for the operating system.

Virtual debugging sessions

The illustrative embodiments described herein provide a computer-implemented method, apparatus, and computer-usable program code for accessing a compatible library. The process receives a file name associated with a library reference. The library reference is included in an executable, and indicates a library used for executing the executable. The process receives environmental variable information in response to identifying an unavailable library in a secondary memory. The unavailable library is associated with the file name. The process accesses the compatible library from a shared library segment using the file name in response to receiving the environmental variable information. The compatible library is associated with the file name. The process utilizes the compatible library to execute the executable.

Growing community of inventors

Bengaluru, India

Shajith Chandran