Aspects of the disclosure are directed to hardware interconnects and corresponding devices and systems for non-coherently accessing data in shared memory devices. Data produced and consumed by devices implementing the hardware interconnect can read and write directly to a memory device shared by multiple devices, and limit coherent memory transactions to relatively smaller flags and descriptors used to facilitate data transmission as described herein. Devices can communicate less data on input/output channels, and more data on memory and cache channels that are more efficient for data transmission. Aspects of the disclosure are directed to devices configured to process data that is read from the shared memory device. Devices, such as hardware accelerators, can receive data indicating addresses for different data buffers with data for processing, and non-coherently read or write the contents of the data buffers on a memory device shared between the accelerators and a host device.

In described examples, a processor system includes a processor core generating memory transactions, a lower level cache memory with a lower memory controller, and a higher level cache memory with a higher memory controller having a memory pipeline. The higher memory controller is connected to the lower memory controller by a bypass path that skips the memory pipeline. The higher memory controller: determines whether a memory transaction is a bypass write, which is a memory write request indicated not to result in a corresponding write being directed to the higher level cache memory; if the memory transaction is determined a bypass write, determines whether a memory transaction that prevents passing is in the memory pipeline; and if no transaction that prevents passing is determined to be in the memory pipeline, sends the memory transaction to the lower memory controller using the bypass path.

In described examples, a processor system includes a processor core generating memory transactions, a lower level cache memory with a lower memory controller, and a higher level cache memory with a higher memory controller having a memory pipeline. The higher memory controller is connected to the lower memory controller by a bypass path that skips the memory pipeline. The higher memory controller: determines whether a memory transaction is a bypass write, which is a memory write request indicated not to result in a corresponding write being directed to the higher level cache memory; if the memory transaction is determined a bypass write, determines whether a memory transaction that prevents passing is in the memory pipeline; and if no transaction that prevents passing is determined to be in the memory pipeline, sends the memory transaction to the lower memory controller using the bypass path.

An information processing apparatus including: a first management data storing region that stores a plurality of first links being provided one for each of multiple calculating cores and representing an order of migration of pages of a page group allocated to the calculating core among a plurality of the pages; a second management data storing region that stores a second link being provided for an operating system and managing a plurality of pages selected in accordance with the order of migration among the page group of the plurality of first links as a group of candidate pages to be migrated to the second memory; and a migration processor that migrates data of a page selected from the group of the second link from the first memory to the second memory. With this configuration, occurrence of a spinlock is reduced, so that the load on processor is reduced.

Cache-based trace logging using tags in an upper cache level. A processor influxes a cache line into a first cache level from an upper second cache level. Influxing the cache line into the first cache level includes, based on the first cache level being a recording cache, the processor reading a tag that is (i) stored in the second cache level and (ii) associated with the cache line. Based on reading the tag, the processor determines whether a first value of the cache line within the second cache level has been previously captured by a trace. The processor performs one of (i) when the first value is determined to have been previously logged, following a logged value logic path when influxing the cache line; or (ii) when the first value is determined to have not been previously logged, following a non-logged value logic path when influxing the cache line.

Apparatus and method for managing pipeline depth of a data processing device. For example, one embodiment of an apparatus comprises: an interface to receive a plurality of work requests from a plurality of clients; and a plurality of engines to perform the plurality of work requests; wherein the work requests are to be dispatched to the plurality of engines from a plurality of work queues, the work queues to store a work descriptor per work request, each work descriptor to include information needed to perform a corresponding work request, wherein the plurality of work queues include a first work queue to store work descriptors associated with first latency characteristics and a second work queue to store work descriptors associated with second latency characteristics; engine configuration circuitry to configure a first engine to have a first pipeline depth based on the first latency characteristics and to configure a second engine to have a second pipeline depth based on the second latency characteristics.

A method for managing memory within a computing system. The method includes one or more computer processors identifying a range of physical memory addresses that store a first data. The method further includes determining whether a second data is stored within the range of physical memory addresses that stores the first data. The method further includes responding to determining that the second data is stored within the range of physical memory addresses that store the first data, by determining whether a process accessing the second data is identified as associated with a side-channel attack. The method further includes responding to determining that the process accessing the second data is associated with the side-channel attack, by initiating a response associated with the process accessing the second data.

A method for identifying, in a system including two or more computing devices that are able to communicate with each other, with each computing device having with a cache and connected to a corresponding memory, a computing device accessing one of the memories, includes monitoring memory access to any of the memories; monitoring cache coherency commands between computing devices; and identifying the computing device accessing one of the memories by using information related to the memory access and cache coherency commands.

A method for identifying, in a system including two or more computing devices that are able to communicate with each other, with each computing device having with a cache and connected to a corresponding memory, a computing device accessing one of the memories, includes monitoring memory access to any of the memories; monitoring cache coherency commands between computing devices; and identifying the computing device accessing one of the memories by using information related to the memory access and cache coherency commands.

Embodiments of the present invention provide a memory resource optimization method and apparatus, relate to the computer field, solve a problem that existing multi-level memory resources affect each other, and optimize an existing single partitioning mechanism. A specific solution is: obtaining performance data of each program in a working set by using a page coloring technology, obtaining a category of each program in light of a memory access frequency, selecting, according to the category of each program, a page coloring-based partitioning policy corresponding to the working set, and writing the page coloring-based partitioning policy to an operating system kernel, to complete corresponding page coloring-based partitioning processing. The present invention is used to eliminate or reduce mutual interference of processes or threads on a memory resource in light of a feature of the working set, thereby improving overall performance of a computer.