sanchayanmaity/minix
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
minix/servers/vm/main.c

495 lines
14 KiB
C
Raw Normal View History

Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
#define _POSIX_SOURCE 1
#define _MINIX 1
#define _SYSTEM 1
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
#include <minix/callnr.h>
#include <minix/com.h>
#include <minix/config.h>
#include <minix/const.h>
#include <minix/ds.h>
#include <minix/endpoint.h>
#include <minix/keymap.h>
#include <minix/minlib.h>
#include <minix/type.h>
#include <minix/ipc.h>
#include <minix/sysutil.h>
#include <minix/syslib.h>
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
#include <minix/const.h>
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
#include <minix/bitmap.h>
#include <minix/crtso.h>
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
#include <minix/rs.h>
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
#include <errno.h>
#include <string.h>
#include <env.h>
#include <stdio.h>
let vm use physically fragmented memory for allocations. map_copy_ph_block is replaced by map_clone_ph_block, which can replace a single physical block by multiple physical blocks. also, . merge map_mem.c with region.c, as they manipulate the same data structures . NOTRUNNABLE removed as sanity check . use direct functions for ALLOC_MEM and FREE_MEM again . add some checks to shared memory mapping code . fix for data structure integrity when using shared memory . fix sanity checks
2010-04-12 13:25:24 +02:00
#include <assert.h>
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
#include <memory.h>
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
#define _MAIN 1
#include "glo.h"
#include "proto.h"
#include "util.h"
#include "vm.h"
#include "sanitycheck.h"
basic sparepage optimisation some simplification of linear/virtual address translation (less hardcoding and more use of arch_*2* functions)
2009-05-12 13:38:29 +02:00
extern int missing_spares;
Move archtypes.h, fpu.h, and stackframe.h Move archtypes.h to include/ dir, since several servers require it. Move fpu.h and stackframe.h to arch-specific header directory. Make source files and makefiles aware of the new header locations.
2010-03-09 10:41:14 +01:00
#include <machine/archtypes.h>
Convert kernel over to bsdmake
2010-04-02 00:22:33 +02:00
#include "kernel/const.h"
#include "kernel/config.h"
#include "kernel/proc.h"
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
#include <signal.h>
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Table of calls and a macro to test for being in range. */
struct {
int (*vmc_func)(message *); /* Call handles message. */
char *vmc_name; /* Human-readable string. */
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
} vm_calls[NR_VM_CALLS];
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Macro to verify call range and map 'high' range to 'base' range
* (starting at 0) in one. Evaluates to zero-based call number if call
* number is valid, returns -1 otherwise.
*/
#define CALLNUMBER(c) (((c) >= VM_RQ_BASE && \
(c) < VM_RQ_BASE + ELEMENTS(vm_calls)) ? \
((c) - VM_RQ_BASE) : -1)
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
FORWARD _PROTOTYPE(int map_service, (struct rprocpub *rpub));
- enable remembering of device memory ranges set by PCI and told to kernel - makes VM ask the kernel if a certain process is allowed to map in a range of physical memory (VM rounds it to page boundaries afterwards - but it's impossible to map anything smaller otherwise so I assume this is safe, i.e. there won't be anything else in that page; certainly no regular memory) - VM permission check cleanup (no more hardcoded calls, less hardcoded logic, more readable main loop), a loose end left by GQ - remove do_copy warning, as the ipc server triggers this but it's no more harmful than the special cases already excluded explicitly (VFS, PM, etc).
2009-11-03 12:12:23 +01:00
FORWARD _PROTOTYPE(int vm_acl_ok, (endpoint_t caller, int call));
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
extern int unmap_ok;
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
/* SEF functions and variables. */
FORWARD _PROTOTYPE( void sef_local_startup, (void) );
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
FORWARD _PROTOTYPE( int sef_cb_init_fresh, (int type, sef_init_info_t *info) );
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
FORWARD _PROTOTYPE( void sef_cb_signal_handler, (int signo) );
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/*===========================================================================*
* main *
*===========================================================================*/
PUBLIC int main(void)
{
message msg;
Changed pagefault delivery to VM this patch changes the way pagefaults are delivered to VM. It adopts the same model as the out-of-quantum messages sent by kernel to a scheduler. - everytime a userspace pagefault occurs, kernel creates a message which is sent to VM on behalf of the faulting process - the process is blocked on delivery to VM in the standard IPC code instead of waiting in a spacial in-kernel queue (stack) and is not runnable until VM tell kernel that the pagefault is resolved and is free to clear the RTS_PAGEFAULT flag. - VM does not need call kernel and poll the pagefault information which saves many (1/2?) calls and kernel calls that return "no more data" - VM notification by kernel does not need to use signals - each entry in proc table is by 12 bytes smaller (~3k save)
2010-04-27 01:21:26 +02:00
int result, who_e, rcv_sts;
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
sigset_t sigset;
secondary cache feature in vm. A new call to vm lets processes yield a part of their memory to vm, together with an id, getting newly allocated memory in return. vm is allowed to forget about it if it runs out of memory. processes can ask for it back using the same id. (These two operations are normally combined in a single call.) It can be used as a as-big-as-memory-will-allow block cache for filesystems, which is how mfs now uses it.
2010-05-05 13:35:04 +02:00
int caller_slot;
struct vmproc *vmp_caller;
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 0);
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
/* SEF local startup. */
sef_local_startup();
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 1);
let vm use physically fragmented memory for allocations. map_copy_ph_block is replaced by map_clone_ph_block, which can replace a single physical block by multiple physical blocks. also, . merge map_mem.c with region.c, as they manipulate the same data structures . NOTRUNNABLE removed as sanity check . use direct functions for ALLOC_MEM and FREE_MEM again . add some checks to shared memory mapping code . fix for data structure integrity when using shared memory . fix sanity checks
2010-04-12 13:25:24 +02:00
SANITYCHECK(SCL_TOP);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* This is VM's main loop. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 2);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
while (TRUE) {
int r, c;
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 3);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
SANITYCHECK(SCL_TOP);
basic sparepage optimisation some simplification of linear/virtual address translation (less hardcoding and more use of arch_*2* functions)
2009-05-12 13:38:29 +02:00
if(missing_spares > 0) {
pt_cycle(); /* pagetable code wants to be called */
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 4);
Changed pagefault delivery to VM this patch changes the way pagefaults are delivered to VM. It adopts the same model as the out-of-quantum messages sent by kernel to a scheduler. - everytime a userspace pagefault occurs, kernel creates a message which is sent to VM on behalf of the faulting process - the process is blocked on delivery to VM in the standard IPC code instead of waiting in a spacial in-kernel queue (stack) and is not runnable until VM tell kernel that the pagefault is resolved and is free to clear the RTS_PAGEFAULT flag. - VM does not need call kernel and poll the pagefault information which saves many (1/2?) calls and kernel calls that return "no more data" - VM notification by kernel does not need to use signals - each entry in proc table is by 12 bytes smaller (~3k save)
2010-04-27 01:21:26 +02:00
if ((r=sef_receive_status(ANY, &msg, &rcv_sts)) != OK)
panic("sef_receive_status() error: %d", r);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, msg.m_type, 5);
Changed pagefault delivery to VM this patch changes the way pagefaults are delivered to VM. It adopts the same model as the out-of-quantum messages sent by kernel to a scheduler. - everytime a userspace pagefault occurs, kernel creates a message which is sent to VM on behalf of the faulting process - the process is blocked on delivery to VM in the standard IPC code instead of waiting in a spacial in-kernel queue (stack) and is not runnable until VM tell kernel that the pagefault is resolved and is free to clear the RTS_PAGEFAULT flag. - VM does not need call kernel and poll the pagefault information which saves many (1/2?) calls and kernel calls that return "no more data" - VM notification by kernel does not need to use signals - each entry in proc table is by 12 bytes smaller (~3k save)
2010-04-27 01:21:26 +02:00
if (is_ipc_notify(rcv_sts)) {
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
/* Unexpected notify(). */
printf("VM: ignoring notify() from %d\n", msg.m_source);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
continue;
}
who_e = msg.m_source;
secondary cache feature in vm. A new call to vm lets processes yield a part of their memory to vm, together with an id, getting newly allocated memory in return. vm is allowed to forget about it if it runs out of memory. processes can ask for it back using the same id. (These two operations are normally combined in a single call.) It can be used as a as-big-as-memory-will-allow block cache for filesystems, which is how mfs now uses it.
2010-05-05 13:35:04 +02:00
if(vm_isokendpt(who_e, &caller_slot) != OK)
panic("invalid caller", who_e);
vmp_caller = &vmproc[caller_slot];
- enable remembering of device memory ranges set by PCI and told to kernel - makes VM ask the kernel if a certain process is allowed to map in a range of physical memory (VM rounds it to page boundaries afterwards - but it's impossible to map anything smaller otherwise so I assume this is safe, i.e. there won't be anything else in that page; certainly no regular memory) - VM permission check cleanup (no more hardcoded calls, less hardcoded logic, more readable main loop), a loose end left by GQ - remove do_copy warning, as the ipc server triggers this but it's no more harmful than the special cases already excluded explicitly (VFS, PM, etc).
2009-11-03 12:12:23 +01:00
c = CALLNUMBER(msg.m_type);
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, c, 6);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
result = ENOSYS; /* Out of range or restricted calls return this. */
Changed pagefault delivery to VM this patch changes the way pagefaults are delivered to VM. It adopts the same model as the out-of-quantum messages sent by kernel to a scheduler. - everytime a userspace pagefault occurs, kernel creates a message which is sent to VM on behalf of the faulting process - the process is blocked on delivery to VM in the standard IPC code instead of waiting in a spacial in-kernel queue (stack) and is not runnable until VM tell kernel that the pagefault is resolved and is free to clear the RTS_PAGEFAULT flag. - VM does not need call kernel and poll the pagefault information which saves many (1/2?) calls and kernel calls that return "no more data" - VM notification by kernel does not need to use signals - each entry in proc table is by 12 bytes smaller (~3k save)
2010-04-27 01:21:26 +02:00
if (msg.m_type == VM_PAGEFAULT) {
if (!IPC_STATUS_FLAGS_TEST(rcv_sts, IPC_FLG_MSG_FROM_KERNEL)) {
printf("VM: process %d faked VM_PAGEFAULT "
"message!\n", msg.m_source);
}
do_pagefaults(&msg);
/*
* do not reply to this call, the caller is unblocked by
* a sys_vmctl() call in do_pagefaults if success. VM panics
* otherwise
*/
continue;
} else if(c < 0 || !vm_calls[c].vmc_func) {
secondary cache feature in vm. A new call to vm lets processes yield a part of their memory to vm, together with an id, getting newly allocated memory in return. vm is allowed to forget about it if it runs out of memory. processes can ask for it back using the same id. (These two operations are normally combined in a single call.) It can be used as a as-big-as-memory-will-allow block cache for filesystems, which is how mfs now uses it.
2010-05-05 13:35:04 +02:00
/* out of range or missing callnr */
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
} else {
Changed pagefault delivery to VM this patch changes the way pagefaults are delivered to VM. It adopts the same model as the out-of-quantum messages sent by kernel to a scheduler. - everytime a userspace pagefault occurs, kernel creates a message which is sent to VM on behalf of the faulting process - the process is blocked on delivery to VM in the standard IPC code instead of waiting in a spacial in-kernel queue (stack) and is not runnable until VM tell kernel that the pagefault is resolved and is free to clear the RTS_PAGEFAULT flag. - VM does not need call kernel and poll the pagefault information which saves many (1/2?) calls and kernel calls that return "no more data" - VM notification by kernel does not need to use signals - each entry in proc table is by 12 bytes smaller (~3k save)
2010-04-27 01:21:26 +02:00
if (vm_acl_ok(who_e, c) != OK) {
printf("VM: unauthorized %s by %d\n",
vm_calls[c].vmc_name, who_e);
} else {
SANITYCHECK(SCL_FUNCTIONS);
result = vm_calls[c].vmc_func(&msg);
SANITYCHECK(SCL_FUNCTIONS);
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
}
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, result, 7);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Send reply message, unless the return code is SUSPEND,
* which is a pseudo-result suppressing the reply message.
*/
if(result != SUSPEND) {
msg.m_type = result;
if((r=send(who_e, &msg)) != OK) {
printf("VM: couldn't send %d to %d (err %d)\n",
msg.m_type, who_e, r);
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("send() error");
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
}
}
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 8);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
}
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 9);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
return(OK);
}
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
/*===========================================================================*
* sef_local_startup *
*===========================================================================*/
PRIVATE void sef_local_startup()
{
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/* Register init callbacks. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 10, 10);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
sef_setcb_init_fresh(sef_cb_init_fresh);
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 11, 11);
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
sef_setcb_init_restart(sef_cb_init_fail);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
/* No live update support for now. */
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
/* Register signal callbacks. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 12, 12);
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
sef_setcb_signal_handler(sef_cb_signal_handler);
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
/* Let SEF perform startup. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 13, 13);
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
sef_startup();
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 14, 14);
Basic System Event Framework (SEF) with ping and live update. SYSLIB CHANGES: - SEF must be used by every system process and is thereby part of the system library. - The framework provides a receive() interface (sef_receive) for system processes to automatically catch known system even messages and process them. - SEF provides a default behavior for each type of system event, but allows system processes to register callbacks to override the default behavior. - Custom (local to the process) or predefined (provided by SEF) callback implementations can be registered to SEF. - SEF currently includes support for 2 types of system events: 1. SEF Ping. The event occurs every time RS sends a ping to figure out whether a system process is still alive. The default callback implementation provided by SEF is to notify RS back to let it know the process is alive and kicking. 2. SEF Live update. The event occurs every time RS sends a prepare to update message to let a system process know an update is available and to prepare for it. The live update support is very basic for now. SEF only deals with verifying if the prepare state can be supported by the process, dumping the state for debugging purposes, and providing an event-driven programming model to the process to react to state changes check-in when ready to update. - SEF should be extended in the future to integrate support for more types of system events. Ideally, all the cross-cutting concerns should be integrated into SEF to avoid duplicating code and ease extensibility. Examples include: * PM notify messages primarily used at shutdown. * SYSTEM notify messages primarily used for signals. * CLOCK notify messages used for system alarms. * Debug messages. IS could still be in charge of fkey handling but would forward the debug message to the target process (e.g. PM, if the user requested debug information about PM). SEF would then catch the message and do nothing unless the process has registered an appropriate callback to deal with the event. This simplifies the programming model to print debug information, avoids duplicating code, and reduces the effort to print debug information. SYSTEM PROCESSES CHANGES: - Every system process registers SEF callbacks it needs to override the default system behavior and calls sef_startup() right after being started. - sef_startup() does almost nothing now, but will be extended in the future to support callbacks of its own to let RS control and synchronize with every system process at initialization time. - Every system process calls sef_receive() now rather than receive() directly, to let SEF handle predefined system events. RS CHANGES: - RS supports a basic single-component live update protocol now, as follows: * When an update command is issued (via "service update *"), RS notifies the target system process to prepare for a specific update state. * If the process doesn't respond back in time, the update is aborted. * When the process responds back, RS kills it and marks it for refreshing. * The process is then automatically restarted as for a buggy process and can start running again. * Live update is currently prototyped as a controlled failure.
2009-12-21 15:12:21 +01:00
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/*===========================================================================*
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
* sef_cb_init_fresh *
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
*===========================================================================*/
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
PRIVATE int sef_cb_init_fresh(int type, sef_init_info_t *info)
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
{
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/* Initialize the vm server. */
Library call for cpu features; make kernel and vm use this to query cpu features (specifically: 4MB pages and TLB global bit). Only enable these features in CR4 if available. 4MB pages to be used in the near future.
2009-05-15 19:07:36 +02:00
int s, i;
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
int click, clicksforgotten = 0;
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
struct memory mem_chunks[NR_MEMS];
struct boot_image image[NR_BOOT_PROCS];
struct boot_image *ip;
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
struct rprocpub rprocpub[NR_BOOT_PROCS];
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
phys_bytes limit = 0;
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
#if SANITYCHECKS
incheck = nocheck = 0;
#endif
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 27);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
vm_paged = 1;
env_parse("vm_paged", "d", 0, &vm_paged, 0, 1);
#if SANITYCHECKS
env_parse("vm_sanitychecklevel", "d", 0, &vm_sanitychecklevel, 0, SCL_MAX);
#endif
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Get chunks of available memory. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 28);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
get_mem_chunks(mem_chunks);
/* Initialize VM's process table. Request a copy of the system
* image table that is defined at the kernel level to see which
* slots to fill in.
*/
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 28);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
if (OK != (s=sys_getimage(image)))
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("couldn't get image table: %d", s);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Set table to 0. This invalidates all slots (clear VMF_INUSE). */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 29);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
memset(vmproc, 0, sizeof(vmproc));
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 30);
Library call for cpu features; make kernel and vm use this to query cpu features (specifically: 4MB pages and TLB global bit). Only enable these features in CR4 if available. 4MB pages to be used in the near future.
2009-05-15 19:07:36 +02:00
for(i = 0; i < ELEMENTS(vmproc); i++) {
vmproc[i].vm_slot = i;
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Walk through boot-time system processes that are alive
* now and make valid slot entries for them.
*/
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 31);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
for (ip = &image[0]; ip < &image[NR_BOOT_PROCS]; ip++) {
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
phys_bytes proclimit;
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
struct vmproc *vmp;
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 32);
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if(ip->proc_nr >= _NR_PROCS) { panic("proc: %d", ip->proc_nr); }
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
if(ip->proc_nr < 0 && ip->proc_nr != SYSTEM) continue;
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
#define GETVMP(v, nr) \
if(nr >= 0) { \
vmp = &vmproc[ip->proc_nr]; \
} else if(nr == SYSTEM) { \
vmp = &vmproc[VMP_SYSTEM]; \
} else { \
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("init: crazy proc_nr: %d", nr); \
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Initialize normal process table slot or special SYSTEM
* table slot. Kernel memory is already reserved.
*/
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 33);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
GETVMP(vmp, ip->proc_nr);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* reset fields as if exited */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 34);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
clear_proc(vmp);
/* Get memory map for this process from the kernel. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 35);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
if ((s=get_mem_map(ip->proc_nr, vmp->vm_arch.vm_seg)) != OK)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("couldn't get process mem_map: %d", s);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Remove this memory from the free list. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 36);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
reserve_proc_mem(mem_chunks, vmp->vm_arch.vm_seg);
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
/* Set memory limit. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 37);
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
proclimit = CLICK2ABS(vmp->vm_arch.vm_seg[S].mem_phys +
vmp->vm_arch.vm_seg[S].mem_len) - 1;
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 38);
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
if(proclimit > limit)
limit = proclimit;
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
vmp->vm_flags = VMF_INUSE;
vmp->vm_endpoint = ip->endpoint;
vmp->vm_stacktop =
CLICK2ABS(vmp->vm_arch.vm_seg[S].mem_vir +
vmp->vm_arch.vm_seg[S].mem_len);
if (vmp->vm_arch.vm_seg[T].mem_len != 0)
vmp->vm_flags |= VMF_SEPARATE;
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 39);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
}
/* Architecture-dependent initialization. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 40);
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
pt_init(limit);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Initialize tables to all physical memory. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 41);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
mem_init(mem_chunks);
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
meminit_done = 1;
. map text (kernel's and processes') in readonly . map kernel in non-user . don't map in first pages of kernel code and data if possible these first pages could actually be freed but as the kernel isn't allowed to touch them either we can't reuse them until VM has totally taken over page table management and kernel doesn't rely on identity mapping any more.
2008-12-18 16:35:22 +01:00
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
/* Give these processes their own page table. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 42);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
for (ip = &image[0]; ip < &image[NR_BOOT_PROCS]; ip++) {
int s;
struct vmproc *vmp;
vir_bytes old_stacktop, old_stack;
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 43);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
if(ip->proc_nr < 0) continue;
GETVMP(vmp, ip->proc_nr);
- map in as much memory as is necessary in 4MB chunks to let boot processes run with segments - allow segment-only processes to fork() by copying them and giving them an identity page table
2009-12-07 13:10:44 +01:00
if(!(ip->flags & PROC_FULLVM))
continue;
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
old_stack =
vmp->vm_arch.vm_seg[S].mem_vir +
vmp->vm_arch.vm_seg[S].mem_len -
vmp->vm_arch.vm_seg[D].mem_len;
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 44);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
if(pt_new(&vmp->vm_pt) != OK)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("VM: no new pagetable");
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
#define BASICSTACK VM_PAGE_SIZE
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 77);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
old_stacktop = CLICK2ABS(vmp->vm_arch.vm_seg[S].mem_vir +
vmp->vm_arch.vm_seg[S].mem_len);
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, old_stacktop, 78);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
if(sys_vmctl(vmp->vm_endpoint, VMCTL_INCSP,
VM_STACKTOP - old_stacktop) != OK) {
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 79);
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("VM: vmctl for new stack failed");
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
}
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 45);
let vm use physically fragmented memory for allocations. map_copy_ph_block is replaced by map_clone_ph_block, which can replace a single physical block by multiple physical blocks. also, . merge map_mem.c with region.c, as they manipulate the same data structures . NOTRUNNABLE removed as sanity check . use direct functions for ALLOC_MEM and FREE_MEM again . add some checks to shared memory mapping code . fix for data structure integrity when using shared memory . fix sanity checks
2010-04-12 13:25:24 +02:00
free_mem(vmp->vm_arch.vm_seg[D].mem_phys +
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
vmp->vm_arch.vm_seg[D].mem_len,
old_stack);
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 46);
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
if(proc_new(vmp,
start all processes, including VM, in VM_PROCSTART in linear address space, to make space for kernel to be able to map in things below there.
2009-05-11 21:11:37 +02:00
VM_PROCSTART,
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
CLICK2ABS(vmp->vm_arch.vm_seg[T].mem_len),
CLICK2ABS(vmp->vm_arch.vm_seg[D].mem_len),
BASICSTACK,
CLICK2ABS(vmp->vm_arch.vm_seg[S].mem_vir +
vmp->vm_arch.vm_seg[S].mem_len -
vmp->vm_arch.vm_seg[D].mem_len) - BASICSTACK,
CLICK2ABS(vmp->vm_arch.vm_seg[T].mem_phys),
CLICK2ABS(vmp->vm_arch.vm_seg[D].mem_phys),
VM: fix kernel mappings for children of non-paged parents.
2010-03-18 18:17:31 +01:00
VM_STACKTOP, 0) != OK) {
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("failed proc_new for boot process");
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
}
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 47);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
}
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
/* Set up table of calls. */
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
#define CALLMAP(code, func) { int i; \
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
if((i=CALLNUMBER(code)) < 0) { panic(#code " invalid: %d", (code)); } \
if(i >= NR_VM_CALLS) { panic(#code " invalid: %d", (code)); } \
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
vm_calls[i].vmc_func = (func); \
vm_calls[i].vmc_name = #code; \
}
/* Set call table to 0. This invalidates all calls (clear
* vmc_func).
*/
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 48);
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
memset(vm_calls, 0, sizeof(vm_calls));
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/* Basic VM calls. */
CALLMAP(VM_MMAP, do_mmap);
CALLMAP(VM_MUNMAP, do_munmap);
CALLMAP(VM_MUNMAP_TEXT, do_munmap);
CALLMAP(VM_MAP_PHYS, do_map_phys);
CALLMAP(VM_UNMAP_PHYS, do_unmap_phys);
/* Calls from PM. */
CALLMAP(VM_EXIT, do_exit);
CALLMAP(VM_FORK, do_fork);
CALLMAP(VM_BRK, do_brk);
CALLMAP(VM_EXEC_NEWMEM, do_exec_newmem);
CALLMAP(VM_PUSH_SIG, do_push_sig);
CALLMAP(VM_WILLEXIT, do_willexit);
CALLMAP(VM_ADDDMA, do_adddma);
CALLMAP(VM_DELDMA, do_deldma);
CALLMAP(VM_GETDMA, do_getdma);
CALLMAP(VM_NOTIFY_SIG, do_notify_sig);
/* Calls from RS */
CALLMAP(VM_RS_SET_PRIV, do_rs_set_priv);
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
CALLMAP(VM_RS_UPDATE, do_rs_update);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/* Generic calls. */
CALLMAP(VM_REMAP, do_remap);
CALLMAP(VM_GETPHYS, do_get_phys);
CALLMAP(VM_SHM_UNMAP, do_shared_unmap);
CALLMAP(VM_GETREF, do_get_refcount);
VM information interface
2010-01-19 22:00:20 +01:00
CALLMAP(VM_INFO, do_info);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
CALLMAP(VM_QUERY_EXIT, do_query_exit);
secondary cache feature in vm. A new call to vm lets processes yield a part of their memory to vm, together with an id, getting newly allocated memory in return. vm is allowed to forget about it if it runs out of memory. processes can ask for it back using the same id. (These two operations are normally combined in a single call.) It can be used as a as-big-as-memory-will-allow block cache for filesystems, which is how mfs now uses it.
2010-05-05 13:35:04 +02:00
CALLMAP(VM_FORGETBLOCKS, do_forgetblocks);
CALLMAP(VM_FORGETBLOCK, do_forgetblock);
CALLMAP(VM_YIELDBLOCKGETBLOCK, do_yieldblockgetblock);
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
start all processes, including VM, in VM_PROCSTART in linear address space, to make space for kernel to be able to map in things below there.
2009-05-11 21:11:37 +02:00
/* Sanity checks */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 49);
start all processes, including VM, in VM_PROCSTART in linear address space, to make space for kernel to be able to map in things below there.
2009-05-11 21:11:37 +02:00
if(find_kernel_top() >= VM_PROCSTART)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("kernel loaded too high");
Basic VM and other minor improvements. Not complete, probably not fully debugged or optimized.
2008-11-19 13:26:10 +01:00
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
/* Initialize the structures for queryexit */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 50);
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
init_query_exit();
/* Unmap our own low pages. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 51);
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
unmap_ok = 1;
_minix_unmapzero();
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/* Map all the services in the boot image. */
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 52);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
if((s = sys_safecopyfrom(RS_PROC_NR, info->rproctab_gid, 0,
(vir_bytes) rprocpub, sizeof(rprocpub), S)) != OK) {
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("sys_safecopyfrom failed: %d", s);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
}
for(i=0;i < NR_BOOT_PROCS;i++) {
if(rprocpub[i].in_use) {
if((s = map_service(&rprocpub[i])) != OK) {
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("unable to map service: %d", s);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
}
}
}
More system processes, this was not enough for the release script to run on some configurations
2010-05-20 10:05:07 +02:00
/*XXX*/vmmcall(0x1234560c, 0, 53);
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
return(OK);
}
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
/*===========================================================================*
* sef_cb_signal_handler *
*===========================================================================*/
PRIVATE void sef_cb_signal_handler(int signo)
{
/* Check for known kernel signals, ignore anything else. */
switch(signo) {
/* There is a pending memory request from the kernel. */
case SIGKMEM:
do_memory();
break;
}
secondary cache feature in vm. A new call to vm lets processes yield a part of their memory to vm, together with an id, getting newly allocated memory in return. vm is allowed to forget about it if it runs out of memory. processes can ask for it back using the same id. (These two operations are normally combined in a single call.) It can be used as a as-big-as-memory-will-allow block cache for filesystems, which is how mfs now uses it.
2010-05-05 13:35:04 +02:00
/* It can happen that we get stuck receiving signals
* without sef_receive() returning. We could need more memory
* though.
*/
if(missing_spares > 0) {
pt_cycle(); /* pagetable code wants to be called */
}
New RS and new signal handling for system processes. UPDATING INFO: 20100317: /usr/src/etc/system.conf updated to ignore default kernel calls: copy it (or merge it) to /etc/system.conf. The hello driver (/dev/hello) added to the distribution: # cd /usr/src/commands/scripts && make clean install # cd /dev && MAKEDEV hello KERNEL CHANGES: - Generic signal handling support. The kernel no longer assumes PM as a signal manager for every process. The signal manager of a given process can now be specified in its privilege slot. When a signal has to be delivered, the kernel performs the lookup and forwards the signal to the appropriate signal manager. PM is the default signal manager for user processes, RS is the default signal manager for system processes. To enable ptrace()ing for system processes, it is sufficient to change the default signal manager to PM. This will temporarily disable crash recovery, though. - sys_exit() is now split into sys_exit() (i.e. exit() for system processes, which generates a self-termination signal), and sys_clear() (i.e. used by PM to ask the kernel to clear a process slot when a process exits). - Added a new kernel call (i.e. sys_update()) to swap two process slots and implement live update. PM CHANGES: - Posix signal handling is no longer allowed for system processes. System signals are split into two fixed categories: termination and non-termination signals. When a non-termination signaled is processed, PM transforms the signal into an IPC message and delivers the message to the system process. When a termination signal is processed, PM terminates the process. - PM no longer assumes itself as the signal manager for system processes. It now makes sure that every system signal goes through the kernel before being actually processes. The kernel will then dispatch the signal to the appropriate signal manager which may or may not be PM. SYSLIB CHANGES: - Simplified SEF init and LU callbacks. - Added additional predefined SEF callbacks to debug crash recovery and live update. - Fixed a temporary ack in the SEF init protocol. SEF init reply is now completely synchronous. - Added SEF signal event type to provide a uniform interface for system processes to deal with signals. A sef_cb_signal_handler() callback is available for system processes to handle every received signal. A sef_cb_signal_manager() callback is used by signal managers to process system signals on behalf of the kernel. - Fixed a few bugs with memory mapping and DS. VM CHANGES: - Page faults and memory requests coming from the kernel are now implemented using signals. - Added a new VM call to swap two process slots and implement live update. - The call is used by RS at update time and in turn invokes the kernel call sys_update(). RS CHANGES: - RS has been reworked with a better functional decomposition. - Better kernel call masks. com.h now defines the set of very basic kernel calls every system service is allowed to use. This makes system.conf simpler and easier to maintain. In addition, this guarantees a higher level of isolation for system libraries that use one or more kernel calls internally (e.g. printf). - RS is the default signal manager for system processes. By default, RS intercepts every signal delivered to every system process. This makes crash recovery possible before bringing PM and friends in the loop. - RS now supports fast rollback when something goes wrong while initializing the new version during a live update. - Live update is now implemented by keeping the two versions side-by-side and swapping the process slots when the old version is ready to update. - Crash recovery is now implemented by keeping the two versions side-by-side and cleaning up the old version only when the recovery process is complete. DS CHANGES: - Fixed a bug when the process doing ds_publish() or ds_delete() is not known by DS. - Fixed the completely broken support for strings. String publishing is now implemented in the system library and simply wraps publishing of memory ranges. Ideally, we should adopt a similar approach for other data types as well. - Test suite fixed. DRIVER CHANGES: - The hello driver has been added to the Minix distribution to demonstrate basic live update and crash recovery functionalities. - Other drivers have been adapted to conform the new SEF interface.
2010-03-17 02:15:29 +01:00
}
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/*===========================================================================*
* map_service *
*===========================================================================*/
PRIVATE int map_service(rpub)
struct rprocpub *rpub;
{
/* Map a new service by initializing its call mask. */
int r, proc_nr;
if ((r = vm_isokendpt(rpub->endpoint, &proc_nr)) != OK) {
return r;
}
/* Copy the call mask. */
memcpy(&vmproc[proc_nr].vm_call_mask, &rpub->vm_call_mask,
sizeof(vmproc[proc_nr].vm_call_mask));
return(OK);
- code shared with exec() letting boot-time processes have their own fully fledged virtual address space and freeing their pre-allocated heap+stack area (necessary to let memory driver map in arbitrary areas of memory for /dev/mem without sys_vm_map) - small optimization preallocating memory on exec - finished VR_DIRECT physical mapping code
2008-12-08 17:43:20 +01:00
}
- pages that points to page directory values of all processes, shared with the kernel, mapped into kernel address space; kernel is notified of its location. kernel segment size is increased to make it fit. - map in kernel and other processes that don't have their own page table using single 4MB (global) mapping. - new sanity check facility: objects that are allocated with the slab allocator are, when running with sanity checking on, marked readonly until they are explicitly unlocked using the USE() macro. - another sanity check facility: collect all uses of memory and see if they don't overlap with (a) eachother and (b) free memory - own munmap() and munmap_text() functions. - exec() recovers from out-of-memory conditions properly now; this solves some weird exec() behaviour - chew off memory from the same side of the chunk as where we start scanning, solving some memory fragmentation issues - use avl trees for freelist and phys_ranges in regions - implement most useful part of munmap() - remap() stuff is GQ's for shared memory
2009-09-21 16:49:49 +02:00
- enable remembering of device memory ranges set by PCI and told to kernel - makes VM ask the kernel if a certain process is allowed to map in a range of physical memory (VM rounds it to page boundaries afterwards - but it's impossible to map anything smaller otherwise so I assume this is safe, i.e. there won't be anything else in that page; certainly no regular memory) - VM permission check cleanup (no more hardcoded calls, less hardcoded logic, more readable main loop), a loose end left by GQ - remove do_copy warning, as the ipc server triggers this but it's no more harmful than the special cases already excluded explicitly (VFS, PM, etc).
2009-11-03 12:12:23 +01:00
/*===========================================================================*
* vm_acl_ok *
*===========================================================================*/
PRIVATE int vm_acl_ok(endpoint_t caller, int call)
{
int n, r;
if ((r = vm_isokendpt(caller, &n)) != OK)
panic() cleanup. this change - makes panic() variadic, doing full printf() formatting - no more NO_NUM, and no more separate printf() statements needed to print extra info (or something in hex) before panicing - unifies panic() - same panic() name and usage for everyone - vm, kernel and rest have different names/syntax currently in order to implement their own luxuries, but no longer - throws out the 1st argument, to make source less noisy. the panic() in syslib retrieves the server name from the kernel so it should be clear enough who is panicing; e.g. panic("sigaction failed: %d", errno); looks like: at_wini(73130): panic: sigaction failed: 0 syslib:panic.c: stacktrace: 0x74dc 0x2025 0x100a - throws out report() - printf() is more convenient and powerful - harmonizes/fixes the use of panic() - there were a few places that used printf-style formatting (didn't work) and newlines (messes up the formatting) in panic() - throws out a few per-server panic() functions - cleans up a tie-in of tty with panic() merging printf() and panic() statements to be done incrementally.
2010-03-05 16:05:11 +01:00
panic("VM: from strange source: %d", caller);
- enable remembering of device memory ranges set by PCI and told to kernel - makes VM ask the kernel if a certain process is allowed to map in a range of physical memory (VM rounds it to page boundaries afterwards - but it's impossible to map anything smaller otherwise so I assume this is safe, i.e. there won't be anything else in that page; certainly no regular memory) - VM permission check cleanup (no more hardcoded calls, less hardcoded logic, more readable main loop), a loose end left by GQ - remove do_copy warning, as the ipc server triggers this but it's no more harmful than the special cases already excluded explicitly (VFS, PM, etc).
2009-11-03 12:12:23 +01:00
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
/* See if the call is allowed. */
if (!GET_BIT(vmproc[n].vm_call_mask, call)) {
- enable remembering of device memory ranges set by PCI and told to kernel - makes VM ask the kernel if a certain process is allowed to map in a range of physical memory (VM rounds it to page boundaries afterwards - but it's impossible to map anything smaller otherwise so I assume this is safe, i.e. there won't be anything else in that page; certainly no regular memory) - VM permission check cleanup (no more hardcoded calls, less hardcoded logic, more readable main loop), a loose end left by GQ - remove do_copy warning, as the ipc server triggers this but it's no more harmful than the special cases already excluded explicitly (VFS, PM, etc).
2009-11-03 12:12:23 +01:00
return EPERM;
}
return OK;
}
Initialization protocol for system services. SYSLIB CHANGES: - SEF framework now supports a new SEF Init request type from RS. 3 different callbacks are available (init_fresh, init_lu, init_restart) to specify initialization code when a service starts fresh, starts after a live update, or restarts. SYSTEM SERVICE CHANGES: - Initialization code for system services is now enclosed in a callback SEF will automatically call at init time. The return code of the callback will tell RS whether the initialization completed successfully. - Each init callback can access information passed by RS to initialize. As of now, each system service has access to the public entries of RS's system process table to gather all the information required to initialize. This design eliminates many existing or potential races at boot time and provides a uniform initialization interface to system services. The same interface will be reused for the upcoming publish/subscribe model to handle dynamic registration / deregistration of system services. VM CHANGES: - Uniform privilege management for all system services. Every service uses the same call mask format. For boot services, VM copies the call mask from init data. For dynamic services, VM still receives the call mask via rs_set_priv call that will be soon replaced by the upcoming publish/subscribe model. RS CHANGES: - The system process table has been reorganized and split into private entries and public entries. Only the latter ones are exposed to system services. - VM call masks are now entirely configured in rs/table.c - RS has now its own slot in the system process table. Only kernel tasks and user processes not included in the boot image are now left out from the system process table. - RS implements the initialization protocol for system services. - For services in the boot image, RS blocks till initialization is complete and panics when failure is reported back. Services are initialized in their order of appearance in the boot image priv table and RS blocks to implements synchronous initialization for every system service having the flag SF_SYNCH_BOOT set. - For services started dynamically, the initialization protocol is implemented as though it were the first ping for the service. In this case, if the system service fails to report back (or reports failure), RS brings the service down rather than trying to restart it.
2010-01-08 02:20:42 +01:00
Reference in a new issue Copy permalink