Patent Application
20210096848
The use of cloud-hosted services and applications has exploded in the past decade and continues to grow at an exponential rate. Cloud-hosted services and applications are generally implemented in large data centers housing thousands of compute platforms such as servers, blade servers, server modules, micro-servers, etc. Oftentimes, the platforms are configured as virtualized execution used for hosting virtual machines and “containers” or the like in which software applications are run.
Each platform includes physical hardware, firmware (also referred to as BIOS—Basic Input-Output System), and software. The root of trust for the platform is the platform hardware and firmware, which although less susceptible to malicious actors that software still may pose a risk. For security and other reasons (e.g., performance), platform firmware may need to be updated.
Historically, the BIOS in personal computer (PC) platforms was a monolithic block of code that was installed in Read-Only Memory (ROM), wherein BIOS was updated by replacing the BIOS ROM chip. Subsequently, the BIOS was installed in EEPROM (Electrically Erasable Programmable Read-Only Memory) and could be replaced (in its entirely) via a firmware update. In approximately 1998, Intel® Corporation began development of a modular firmware architecture known as the Extensible Firmware Interface (EFI). In 2005, the Unified EFI forum was formed as an industry-wide organization to promote adoption and continue the development of the EFI Specification. Using the EFI 1.10 Specification as the starting point, this industry group released began releasing firmware specifications, renamed Unified EFI (UEFI). UEFI firmware dominates today's platform architectures. UEFI firmware has a modular architecture that includes a core block to which modules are added, wherein the core block is booted first and the booted code is used to load the modules during platform boot. Also, rather than EEPROMs, most of today's firmware is stored in flash memory (sometimes referred to as BIOS flash and referred to as persistent flash memory). More generally, platform firmware may be stored in a non-volatile storage device, which includes but is not limited to flash memory and EEPROMs.
Under some platform architectures, bootloaders may be used to load platform firmware. For example, bootloaders are used for mobile devices and some server platforms.
Traditionally, the uCode updates can be loaded into CPU (central processing unit, aka processor) either by boot time loading (if the uCode update corresponds to pre-boot uCode that is loaded prior to booting an operating system) or by operating system execution time loading (referred to as a hot-upgrade or run-time uCode update). Cloud environments are sensitive to system downtime caused by system firmware upgrades and the cloud service providers prefer the hot-upgrade method to load uCode patch through operating system utility in runtime if the uCode update does not have dependencies to the boot phase of platform hardware and operating system.
Cloud service providers provide a variety options for tenants, including virtualized environments under which the service provider provides and operating system and hypervisor or VMM (virtual machine monitor) on which tenant-managed virtual machines (VMs) are run and “bare” metal platforms under which the cloud service provider leases the hardware on which tenant VMs are run without a host operating system provided by the service provider. An example of a bare metal cloud environment 100 is shown in
Tenant environment 104 includes a host operating system 114 on which one or more applications 116 are run. For illustrative purposes, tenant environment is further depicted as including a uCode hot-upgrade utility 118 that would be used to support uCode hot-upgrades during operating run-time for a virtual environment that was not bare metal. Under bare metal environment 100, host operating system 114 is owned by a single tenant instead of the cloud service provider. This makes it much more complex for cloud service providers to facilitate uCode updates using hot-upgrade methods for tenant-owned operating system environments, such as illustrated for tenant environment 104. In particular, uCode hot-upgrade utility 118 cannot update uCode using an in-band method in bare metal cloud environment 100.
The foregoing aspects and many of the attendant advantages of this invention will become more readily appreciated as the same becomes better understood by reference to the following detailed description, when taken in conjunction with the accompanying drawings, wherein like reference numerals refer to like parts throughout the various views unless otherwise specified:
Embodiments of a microcode (uCode) hot-upgrade method for bare metal cloud deployment and associated apparatus configured to implement the method are described herein. In the following description, numerous specific details are set forth to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention can be practiced without one or more of the specific details, or with other methods, components, materials, etc. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of the invention.
Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
For clarity, individual components in the Figures herein may also be referred to by their labels in the Figures, rather than by a particular reference number. Additionally, reference numbers referring to a particular type of component (as opposed to a particular component) may be shown with a reference number followed by “(typ)” meaning “typical.” It will be understood that the configuration of these components will be typical of similar components that may exist but are not shown in the drawing Figures for simplicity and clarity or otherwise similar components that are not labeled with separate reference numbers. Conversely, “(typ)” is not to be construed as meaning the component, element, etc. is typically used for its disclosed function, implement, purpose, etc.
In accordance with aspects of the embodiments disclosed herein, a uCode hot-upgrade method is provided that applies the uCode patch to a one or more CPUs through an out-of-band controller, such as but not limited to a baseboard management controller (BMC). In one aspect, this innovation defines an interrupt service, such as a SMI (System Management Interrupt) service, to upgrade uCode patch to one or more CPUs in the bare-metal managed platform in runtime. The uCode patch is received by the BMC and buffered in memory onboard the BMC. The BMC then triggers an SMI handler to execute uCode update service to perform the uCode hot-upgrade.
This innovation enables cloud service providers to deploy uCode hot-patches to bare metal servers for persistent storage and live-patch without touching the tenant operating system environment. This approach significantly reduces the complexity and cost for cloud customers to deploy uCode patch in bare metal environment. The method is transparent for host operating system and the tenant's applications (meaning performed independent of the host OS) and helps cloud service providers to improve user experiences for tenants while deploying firmware patches for host bare metal systems without explicit interruption to tenant's applications.
An overview of the method deployed for a bare metal cloud environment 200 is shown in
As further shown in
In some embodiments, the uCode patch is encapsulated into a capsule format and a capsule firmware update interface is used to update uCode regions. The capsule format uCode image can be built by either an offline phase or online phase to support flexible integrity check method. Examples of capsule formats and packaging schemes are shown in
As illustrated in
An SMI (System Management Interrupt) hander is defined in BIOS to parse the uCode patch from the capsule image. Generally, the uCode capsule package 300 can be dynamically generated either by offline tools from the OEM/ODM/platform vendor or by online tools from the cloud customer configured to support flexible release and variable security check requirements. To support the integrity check for the update image, a signature may be added into AuthInfo field 306; this signature is created using platform credentials from either the OEM/ODM/platform vendor or from cloud customers. For such cases (employing signatures), in one embodiment the SMI handler is configured to check the integrity of the BIOS update image by validating this signature with an internal platform credential.
In some instances, a given uCode release will including multiple uCode patches that are configured to be implemented on the same CPU model while supporting different stepping levels. For example, stepping levels for a CPU may including A0, A1, A2, etc. In the cases where changes are more significant, the stepping levels may use a next letter, such as B0, then B1, B2, etc. The uCode patches for a CPU model for a given stepping level may differ from uCode patches for that CPU model for another stepping level, thus multiple uCode patches may be included in an update package.
Generally, an OEM/ODM (original equipment manufacturer or original device manufacturer) may create an uCode update package with multiple uCode patches to support multiple stepping levels for a CPU model or produce (e.g., blade server, server module, etc.) for which the uCode update package is targeted. For example, a platform vendor may manufacture and sell a given blade server for several years employing the same CPU model, while during that time the stepping versions of the CPU model will have changed. Thus, a uCode update for the blade server may require multiple patches.
In the case of a cloud operator, some uCode update packages may be more targeted to only be implemented on a CPU model with a specific stepping level. In this case, the uCode update package may only include a single uCode patch. In other cases, the cloud operator may generate uCode update packages with multiple uCode patches.
As shown in
Use of Out-of-Band Controller to Update uCode in CPU(s) During Runtime
Prior to performing the uCode hot-upgrade, BMC agent 510 validates the integrity of the uCode update package with internal platform credentials. For example, as discussed above and shown in
An SMI service in BIOS firmware is defined to implement a uCode hot-upgrade procedure for one or more of CPUs 208, as shown in flowchart 600 of
In a block 606, the BMC exposes the internal memory buffer as a PCIe Memory MMIO range. In one embodiment the BMC is connected to a host CPU via a PCIe interface (e.g., PCIe interface/link 508). The BMC is presented to the host CPU as a PCIe endpoint device. PCIe endpoint devices can expose internal memory resources as MMIO ranges using a standard PCIe mechanism, as is the case in block 606.
In a block 608, the BMC triggers an SMI on host CPU 802 (
In some embodiments under which multiple stepping version uCode patches are included in an update package, the SMI BIOS code may be configured to filter for only those uCode patches that have a stepping version that matches the stepping version of the processor. Under this approach, providers of uCode update packages can build and send out update package with multiple stepping versions without having to know the particular stepping version that is implemented by each processor for which uCode is to be updated.
Platform hardware 902 includes a processor 906 having a System on a Chip (SoC) architecture including a central processing unit (CPU) 908 with M processor cores 910, each coupled to a Level 1 and Level 2 (L1/L2) cache 912. Each of the processor cores and L1/L2 caches are connected to an interconnect 914 to which each of a memory interface 916 and a Last Level Cache (LLC) 918 is coupled, forming a coherent memory domain. Memory interface is used to access host memory 904 in which various software components are loaded and run via execution of associated software instructions on processor cores 910.
Processor 906 further includes an Input/Output (I/O) interconnect hierarchy, which includes one or more levels of interconnect circuitry and interfaces that are collectively depicted as I/O interconnect & interfaces 920 for simplicity. Various components and peripheral devices are coupled to processor 906 via respective interfaces (not all separately shown), including a network interface 922 and a BMC 924 including BMC firmware 926 and a BMC buffer 928. In the illustrated embodiment, BMC 924 is coupled to SoC 906 via a PCI interface 929. Network interface 922 is connected to a network 930. In some embodiments, BMC 924 is connected to a management network 931 that is separate from network 930. In other embodiments, BMC 924 either is connected to network 930 using a built-in or separate network interface (both not shown) or BMC 924 is configured to communicate with external entities coupled to network 930 via network interface 922.
Platform hardware 902 also includes a disk drive or solid-state disk (SSD) with controller 932 in which software components 934 are stored. Optionally, all or a portion of the software components used to implement the software aspects of embodiments herein may be loaded over a network 930 accessed by network interface 922.
During platform initialization, a current or new uCode image and various UEFI modules (not separately shown) are loaded into host memory 904 and booted, followed loading and initialization of various software components. The software components include a host operating system and a VMM 936 (that would sit above the host operating system but is not shown separately) used to host n virtual machines (VMs) VM 1, VM 2 . . . VM n, each including an operating system 938 on which one or more applications 940 are run. Platform architectures employing containers, such as Docker®-type containers, may be implemented in a similar manner. In addition, non-virtualized computing platforms that only run a single instance of an operating system (e.g., applications run directly on host operating system 936) may also be used.
As further illustrated in
For multi-socket bare metal cloud platforms, the platform architecture would be somewhat similar to that shown in
As used herein, “runtime” and “operating system runtime” refer to an operational phase of a platform following booting of a host operating system. Accordingly, when the host operating system is paused during an out-of-band hot-upgrade process, the host operating system is still in runtime through the process. Moreover, when the host operating system is paused it is unaware of any operations being performed by the CPUs on the bare metal platform and the out-of-band hot-upgrade process provided by the embodiments herein are transparent to the host operating system.
As used herein, “out-of-band” means a communication channel that does not employ communication facilities provided by an operating system, such as a network software stack. As a result, out-of-band communications are implemented separate and apart from communications in the tenant environment involving the host operating system.
In addition to applying uCode patches to CPUs, the teaching and principles disclosed herein may be applied to Other Processing Units (collectively termed XPUs) including one or more of Graphic Processor Units (GPUs) or General Purpose GPUs (GP-GPUs), Tensor Processing Unit (TPU) Data Processor Units (DPUs), Artificial Intelligence (AI) processors or AI inference units and/or other accelerators, FPGAs and/or other programmable logic (used for compute purposes), etc. While some of the diagrams herein show the use of CPUs, this is merely exemplary and non-limiting. Generally, any type of XPU may be used in place of a CPU in the illustrated embodiments. Moreover, as used in the following claims, CPUs and all forms of XPUs comprise processing units.
Although some embodiments have been described in reference to particular implementations, other implementations are possible according to some embodiments. Additionally, the arrangement and/or order of elements or other features illustrated in the drawings and/or described herein need not be arranged in the particular way illustrated and described. Many other arrangements are possible according to some embodiments.
In each system shown in a figure, the elements in some cases may each have a same reference number or a different reference number to suggest that the elements represented could be different and/or similar. However, an element may be flexible enough to have different implementations and work with some or all of the systems shown or described herein. The various elements shown in the figures may be the same or different. Which one is referred to as a first element and which is called a second element is arbitrary.
In the description and claims, the terms “coupled” and “connected,” along with their derivatives, may be used. It should be understood that these terms are not intended as synonyms for each other. Rather, in particular embodiments, “connected” may be used to indicate that two or more elements are in direct physical or electrical contact with each other. “Coupled” may mean that two or more elements are in direct physical or electrical contact. However, “coupled” may also mean that two or more elements are not in direct contact with each other, but yet still co-operate or interact with each other. Additionally, “communicatively coupled” means that two or more elements that may or may not be in direct contact with each other, are enabled to communicate with each other. For example, if component A is connected to component B, which in turn is connected to component C, component A may be communicatively coupled to component C using component B as an intermediary component.
An embodiment is an implementation or example of the inventions. Reference in the specification to “an embodiment,” “one embodiment,” “some embodiments,” or “other embodiments” means that a particular feature, structure, or characteristic described in connection with the embodiments is included in at least some embodiments, but not necessarily all embodiments, of the inventions. The various appearances “an embodiment,” “one embodiment,” or “some embodiments” are not necessarily all referring to the same embodiments.
Not all components, features, structures, characteristics, etc. described and illustrated herein need be included in a particular embodiment or embodiments. If the specification states a component, feature, structure, or characteristic “may”, “might”, “can” or “could” be included, for example, that particular component, feature, structure, or characteristic is not required to be included. If the specification or claim refers to “a” or “an” element, that does not mean there is only one of the element. If the specification or claims refer to “an additional” element, that does not preclude there being more than one of the additional element.
Italicized letters, such as ‘M’, ‘n’, etc. in the foregoing detailed description are used to depict an integer number, and the use of a particular letter is not limited to particular embodiments. Moreover, the same letter may be used in separate claims to represent separate integer numbers, or different letters may be used. In addition, use of a particular letter in the detailed description may or may not match the letter used in a claim that pertains to the same subject matter in the detailed description.
As discussed above, various aspects of the embodiments herein may be facilitated by corresponding software and/or firmware components and applications. Thus, embodiments of this invention may be used as or to support a software program, software modules, and/or firmware (BIOS), executed upon some form of processor, processing core or embedded logic, a virtual machine running on a processor or core or otherwise implemented or realized upon or within a non-transitory computer-readable or machine-readable storage medium. A non-transitory computer-readable or machine-readable storage medium includes any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer). For example, a non-transitory computer-readable or machine-readable storage medium includes any mechanism that provides (i.e., stores and/or transmits) information in a form accessible by a computer or computing machine (e.g., computing device, electronic system, etc.), such as recordable/non-recordable media (e.g., read only memory (ROM), random access memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, etc.). The content may be directly executable (“object” or “executable” form), source code, or difference code (“delta” or “patch” code). A non-transitory computer-readable or machine-readable storage medium may also include a storage or database from which content can be downloaded. The non-transitory computer-readable or machine-readable storage medium may also include a device or product having content stored thereon at a time of sale or delivery. Thus, delivering a device with stored content, or offering content for download over a communication medium may be understood as providing an article of manufacture comprising a non-transitory computer-readable or machine-readable storage medium with such content described herein.
As used herein, a list of items joined by the term “at least one of” can mean any combination of the listed terms. For example, the phrase “at least one of A, B or C” can mean A; B; C; A and B; A and C; B and C; or A, B and C.
The above description of illustrated embodiments of the invention, including what is described in the Abstract, is not intended to be exhaustive or to limit the invention to the precise forms disclosed. While specific embodiments of, and examples for, the invention are described herein for illustrative purposes, various equivalent modifications are possible within the scope of the invention, as those skilled in the relevant art will recognize.
These modifications can be made to the invention in light of the above detailed description. The terms used in the following claims should not be construed to limit the invention to the specific embodiments disclosed in the specification and the drawings. Rather, the scope of the invention is to be determined entirely by the following claims, which are to be construed in accordance with established doctrines of claim interpretation.
