Boot procedure

A) General Principles

The ALMOS-MKH boot procedure can be decomposed in two phases:

• The architecture dependent phase, implemented by an architecture specific boot_loader procedure.
• The architecture independent phase, implemented by a generic (architecture independent) kernel-init procedure.

As the generic (i.e. architecture independent) kernel initialization procedure is executed in parallel by all kernel instances in all clusters containing at least one core and one memory bank, the main task of the boot-loader is to load - in each cluster - a local copy of the ALMOS-MKH kernel code, and a description of the hardware architecture, contained in a local boot_info_t data-structure.

This fixed size boot_info_t structure is build by the boot-loader, and stored at the beginning of the local copy of the kdata segment. As it contains both general and cluster specific information, the content depends on the cluster:

• general hardware architecture features : number of clusters, topology, etc.
• available external (shared) peripherals : types and features.
• number of cores in cluster,
• available internal (private) peripherals in cluster : types and features.
• available physical memory in cluster.

This boot_info_t structure is defined in the boot_info.h file.

To build the various boot_info_t structures (one per cluster), the boot-loader uses the arch_info_t binary structure, that is described in section Hardware Platform Definition. This binary structure is contained in the arch_info.bin file, and must be stored in the file system root directory.

This method allows an intelligent boot_loader to check and - if required - reconfigure the hardware components, to guaranty that the generated boot_info_t structures contain only functionally tested hardware components.

We describe below the boot_loader for the TSAR architecture, the boot_loader for the I86 architecture, and the generic kernel initialization procedure.

B) Boot-loader for the TSAR architecture

The TSAR boot-loader uses an OS-independent pre-loader, stored in an addressable but non-volatile device, that load the TSAR boot-loader code from an external block-device to the cluster 0 physical memory. This preloader is specific for the TSAR architecture, but independent on the Operating System. It is used by ALMOS-MKH, but also by LINUX, NetBSD, ALMOS_MKH, or the GIET-VM.

The TSAR boot_loader allocates - in each cluster containing a physical memory bank - six fixed size memory zones, to store various binary files or data structures. The two first zones are permanently allocated: The PRE_LOADER zone is only defined in cluster 0, and contains the pre-loader code. The KERNEL_CODE zone containing the kcode and kdata sgments is directly used by the kernel when the boot_loader transfers control - in each cluster - to the kernel_init procedure. The BOOT_CODE, ARCH_INFO, KERNEL_ELF, and BOOT_STACK zones are temporary: they are only used - in each cluster - by the boot-loader code, and the corresponding physical memory can be freely re-allocated by the local kernel instance when it starts execution.

name	description	base address (physical)	size
PRE_LOADER	pre-loader code	PRELOADER_BASE (0)	PRELOADER_MAX_SIZE (16KB)
KERNEL_CODE	kernel code and data	KERNEL_CODE_BASE (16 KB)	KERNEL_CODE_MAX_SIZE (2 MB - 16 KB)
BOOT_CODE	boot-loader code and data	BOOT_CODE_BASE (2 MB)	BOOT_CODE_MAX_SIZE (1 MB)
ARCH_INFO	arch_info.bin file copy	ARCH_INFO_BASE (3 MB)	ARCH_INFO_MAX_SIZE (1 MB)
KERNEL_ELF	kernel.elf file copy	KERNEL_ELF_BASE (4 MB)	KERN_ELF_MAX_SIZE (2 MB)
BOOT_STACK	boot stacks (one per core)	BOOT_STACK_BASE (6 MB)	BOOT_STACK_MAX_SIZE (1MB)

The values given in this array are indicative. The actual values are defined by configuration parameters in the boot_config.h file. The two main constraint are the following:

• the kcode and kdata segments (in the KERNEL_CODE zone) must be entirely contained in one single big physical page (2 Mbytes), because it will be mapped as one single big page in all process virtual spaces.
• the BOOT_CODE zone (containing the boot loader instructions and data) must be entirely contained in the

next big physical page, because it will be mapped in the boot-loader page table to allow the cores to access locally the boot code as soon as it has been copied in the local cluster.

A core is identified by two indexes: cxy is the cluster identifier, an lid is the core local index in cluster. The CP0 register containing the core gid (global hardware identifier) has a fixed format: gid = (cxy << 2) + lid

All cores contribute to the boot procedure, but all cores are not simultaneously active:

• in the first phase - fully sequencial - only core[0][0] is running (core 0 in cluster 0).
• In the second phase - partially parallel - only core[cxy][0] is running in each cluster.
• in the last phase - fully parallel - all core[cxy][lid] are running.

We describe below the four phases of the TSAR boot-loader:

B1. Pre-loader phase

• In the TSAR_LETI architecture, the pre-loader code is stored in the first 16 kbytes of the physical address space in cluster 0.
• In the TSAR_IOB architecture, the preloader is stored in an external ROM, that is accessed throug the IO_bridge located in cluster 0.

At reset, the MMU is de-activated (for both data and instructions), and the extension address registers supporting direct access to remote memory banks (for data only) contain the 0 value. Therefore, all cores can only access the physical address space of cluster 0.

All cores execute the same pre-loader code, but the work done depends on the core identifier:

• The core[0][0] load in the BOOT_CODE zone of cluster 0, the boot-loader code stored on disk.
• All other cores do only one task before going to sleep (i.e. low-power state): each core activates its private WTI channel in the local ICU (Interrupt Controller Unit) to be later activated by an IPI (Inter Processor Interrupt).

B2. Boot-loader sequencial phase

In this phase, only core [0][0] is running, while all other cores are blocked in the preloaded, waiting to be activated by an IPI.

The first instructions of the boot-loader are defined in the boot_entry.S file. This assembly code is executed by all cores entering the boot-loader, but not at the same time.

Each core running this assembly code makes the 3 following actions:

• It initializes the core stack pointer depending on the lid value extracted from the gid, using the BOOT_STACK_BASE and BOOT_STACK_SIZE parameters defined in the boot_config.h file,
• It changes the value of the DATA address extension CP2 register, using the cxy value extracted from the gid, to force all cores to use the local stack segments.
• It jumps to the boot_loader() C function defined in the boot.c file, passing the two (cxy , lid) arguments.

In this sequencial phase, the core[0][0] executing this C function makes the following actions:

• The core[0][0] initializes 2 peripherals: The TTY terminal (channel 0) to display log messages, and the IOC peripheral to access the disk file system.
• The core[0][0] initializes the boot-loader FAT32, allowing the boot loader to access files stored in the FAT32 file system on disk.
• The core[0][0] load in the KERNEL_ELF zone the kernel.elf file from the disk file system..
• Then it copies in the KERNEL_CORE zone the kcode and kdata segments, using the addresses contained in the .elf file (identity mapping).
• The core[0][0] load in the ARCH_INFO zone the arch_info.bin file from the disk file system.
• Then it builds from this arch_info.t structure the specific boot_info_t structure for cluster 0, and stores it in the kdata segment.
• The core[0][0] send IPIs to activate all cores [cxy][0] in all other clusters.

B3. Boot-loader partially parallel phase

In this phase all core[cxy][0], other than the core[0][0] are running.

At this point, all DATA extension registers point already on the local cluster( to use the local stack).

The core[cxy][0] exécute the following tasks:

• To access the global data stored in cluster cxy, the core[cxy][0] copies the boot-loader code from BOOT_CODE zone in cluster 0 to BOOT_CORE zone in cluster cxy.
• To access the instructions stored in cluster cxy, the core[cxy][0] creates a minimal page table containing two big pages mapping respectively the local BOOT_CORE zone, and the local KERNEL_CODE zone, and activates the instruction MMU. [TO BE DONE]
• The core[cxy][0] copies the arch_info.bin structure from ARCH_INFO zone in cluster 0 to ARCH_INFO zone in cluster cxy.
• The core[cxy][0] copies the kcode and kdata segments from KERNEL_CODE zone in cluster 0 to KERNEL_CODE zone in cluster cxy.
• The core[cxy][0] builds from the arch_info.t the specific boot_info_t structure for cluster cxy, and stores it in the local kdata segment.
• All core[cxy][0], including core[0][0], synchronize using a global barrier.
• In each cluster cxy, the core[cxy][0] activates the other cores that are blocked in the pre-loader.

B4. Boot-loader fully parallel phase

In this phase all core[cxy][lid] are running.

Each core must initialise few registers, as described below, and jump to the kernel_entry address. This address is defined in the kernel.elf file, and registered in the kernel_entry global variable.

• argument : the kernel_init() function unique argument is a pointer on the boot_info_t structure, that is the first variable in the data segment.
• stack pointer : In each cluster an array of idle thread descriptors, indexed by the local core index, is defined in the kdatasegment, on top of the boot_info_t structure. For any thread, the thread descriptor contains the kernel stack, and this is used to initialize the stack pointer.
• base register : in each core, the cp0_ebase register, defines the kernel entry point in case of interrupt, exception, or syscall, and must be initialized.
• status register : in each core, the cp0_sr register defines the core state, and must be initialized (UM bit reset / IE bit reset / BEV bit reset ).

Moreover, all core[cxy][lid] with (lid != 0) must activate the instruction MMU to use both the local copy of the boot-loader code, and the local copy of the kernel.

• all core[cxy][lid] in cluster cxy synchronize on a local barrier.
• each core in cluster cxy initialize the following registers before jumping to kernel_init function:
  • status register : cp0_sr <= reset BEV bit
  • kernel_init() argument : a0 <= boot_info
  • stack pointer : sp <=
  • program counter : cp0_ebase <= kernel_entry

All cores other than core[cxy][0] This Chaque core CPi (lid non nul), réveillé par le CP0 local de son cluster, sort du code du preloader et exécute le boot-loader dans le cluster de boot puisque ses registres d'extension d'adresse ne sont pas encore mis à jour. Une fois sortis du preloader, ces cores décrémentent le compteur de la barrière de synchronisation et débloquent les CP0. Tous ces CP0 sauf un, se mettent tout de suite en attente jusqu'à ce que les CPi finissent leur exécution du boot-loader. Le seul CP0 qui n'arrive pas encore à cette barrière de synchronisation, celui du cluster(0,0), peut maintenant écraser le code du preloader en déplaçant l'image du noyau à l'adresse 0x0 de l'espace adressable physique du cluster(0,0), puisque tous les cores sont déjà sortis du preloader. Il rejoint ensuite les autres CP0 au dernier point de rendez-vous dans le boot-loader. Les CPi, quant à eux, exécutent, pour le moment, le code du boot-loader se trouvant dans le cluster de boot car leurs registres d'extension d'adresse ont toujours la valeur 0 par défaut. Chacun de ces CPi effectue les étapes suivantes:

• Il analyse le contenu de arch_info.bin (dans l'espace adressable physique du cluster de boot) en parcourant le tableau de descripteurs de core pour retrouver son identificateur de cluster cxy ainsi que son identificateur de core local dans son cluster lid. Notons que cette étape est exécutée parallèlement par tous les CPi, ce qui entraine une contention, encore plus forte que celle créée par les accès parallèles des CP0, au banc mémoire contenant ce tableau de descripteurs de core .
• Il peut maintenant, à partir de son cxy, mettre à jour les valeurs dans ses registres d'extension d'adresse de code et de données. Comme le CP0 du même cluster a déjà copié les informations nécessaires dans le banc mémoire local aux mêmes adresses que du cluster de boot, il peut toujours exécuter le code du boot-loader en local.
• Il alloue sa pile de boot en initialisant son pointeur de pile à l'adresse 0x600000 - 4K*lid dans l'espace adressable physique locale de son cluster (grâce à la nouvelle valeur dans le registre d'extension d'adresse de code).
• La structure boot_info_t du cluster étant déjà initialisée, chacun des CPi ne fait que vérifier les informations qui le concernent.
• Il arrive finalement au point de rendez-vous avec tous les CP0, décrémente le compteur de la barrière de synchronisation et se met en attente.
• Dès que le dernier core arrive à ce point et débloque les autres, tous les cores se branchent à la fonction kern_init().

/ At this point, the boot-loader completed its job:

• The kernel code kcode and kdata segments are loaded - in all clusters - in the first offset physical pages.
• The hardware architecture described by the arch_info.binfile has been analyzed, and copied - in each cluster - in the boot_info_t structure, stored in the kdata segment.
• Each local kernel instance can use all the physical memory that is not used to store the kernel kcode and kdata segments themselves.

C) Boot-loader for the I86 architecture

TODO

D) Generic kernel initialization procedure

The kernel_init( boot_info_t * info ) function is the kernel entry point when the boot_loader transfer control to the kernel. The info argument is a pointer on the fixed size boot_info_t structure, that is stored in the data kernel segment.

All cores execute this procedure in parallel, but some tasks are only executed by the CP0 core. This procedure uses two synchronisation barriers, defined as global variables in the data segment:

• the global_barrier variable is used to synchronize all CP0 cores in all clusters containing a kernel instance.
• the local_barrier variable is used to synchronize all cores in a given cluster.

The kernel initialization procedure execute sequentially the following steps:

D1) Core and cluster identification

Each core has an unique hardware identifier, called gid, that is hard-wired in a read-only register. From the kernel point of view a core is identified by a composite index (cxy,lid), where cxy is the cluster identifier, and lid is a local (continuous) index in the cluster. The association between the gid hardware index and the (cxy,lid) composite index is defined in the boot_info_t structure. In this first step, each core makes an associative search in the boot_info_t structure to obtain the (cxy,lid) indexes from the gid index. Then the CP0 initialize the global variable local_cxy defining the local cluster identifier.

D2) TXT0 device initialization

The core[io_cxy][0] (i.e. CP0 in I/O cluster) initializes the chdev descriptor associated to the kernel text terminal TXT0. This terminal is used by any kernel instance running on any core to display log or debug messages. This terminal is configured in non-descheduling mode : the calling thread call directly the relevant TXT driver, without using a server thread.

A first synchonization barrier is used to avoid other cores to use the TXT0 terminal before initialization completion.

D3) Cluster manager initialization

In each cluster, the CP0 makes the cluster manager initialization, namely the cores descriptors array, and the physical memory allocators. Then it initializes the local process_zero, containing al kernel threads in a given cluster.

A second synchonization barrier is used to avoid other cores to access cluster manager before initialization completion.

D4) Internal & external devices initialization

In each cluster, the CP0 makes the devices initialization. For multi-channels devices, there is one channel device (called chdev_t) per channel. For internal (replicated) devices, the khedive descriptors are allocated in the local cluster. For external (shared) devices, the chdev descriptors are regularly distributed on all clusters. These external chdev are indexed by a global index, and the host cluster is computed from this index by a modulo.

The internal devices descriptors are created first( ICU, then MMC, then DMA ), because the ICU device is used by all other devices. Then the WTI mailboxes used for IPIs (Inter Processor Interrupt) are allocated in local ICU : one WTI mailbox per core. Then each external chdev descriptor is created by the CP0 in the cluster where it must be created.

A third synchonization barrier is used to avoid other cores to access devices before initialization completion.

D5) Idle thread initialization

In this step, each core creates and initializes its private idle thread descriptor.

D6) File system initialization

The CP0 in I/O cluster) initializes the file system.

A fourth synchonization barrier is used to avoid other cores to access file system before initialization completion.

D7) Scheduler activation

Finally, each core enables its private timer IRQ to activate its private scheduler, and jump to the idle thread code.= Boot procedure =