minix/kernel/main.c

242 lines
8.9 KiB
C
Raw Normal View History

2005-04-21 16:53:53 +02:00
/* This file contains the main program of MINIX as well as its shutdown code.
* The routine main() initializes the system and starts the ball rolling by
* setting up the process table, interrupt vectors, and scheduling each task
* to run to initialize itself.
* The routine shutdown() does the opposite and brings down MINIX.
2005-04-21 16:53:53 +02:00
*
* The entries into this file are:
* main: MINIX main program
* prepare_shutdown: prepare to take MINIX down
*/
#include "kernel.h"
#include <signal.h>
#include <string.h>
2005-04-21 16:53:53 +02:00
#include <unistd.h>
#include <a.out.h>
#include <minix/callnr.h>
#include <minix/com.h>
'proc number' is process slot, 'endpoint' are generation-aware process instance numbers, encoded and decoded using macros in <minix/endpoint.h>. proc number -> endpoint migration . proc_nr in the interrupt hook is now an endpoint, proc_nr_e. . m_source for messages and notifies is now an endpoint, instead of proc number. . isokendpt() converts an endpoint to a process number, returns success (but fails if the process number is out of range, the process slot is not a living process, or the given endpoint number does not match the endpoint number in the process slot, indicating an old process). . okendpt() is the same as isokendpt(), but panic()s if the conversion fails. This is mainly used for decoding message.m_source endpoints, and other endpoint numbers in kernel data structures, which should always be correct. . if DEBUG_ENABLE_IPC_WARNINGS is enabled, isokendpt() and okendpt() get passed the __FILE__ and __LINE__ of the calling lines, and print messages about what is wrong with the endpoint number (out of range proc, empty proc, or inconsistent endpoint number), with the caller, making finding where the conversion failed easy without having to include code for every call to print where things went wrong. Sometimes this is harmless (wrong arg to a kernel call), sometimes it's a fatal internal inconsistency (bogus m_source). . some process table fields have been appended an _e to indicate it's become and endpoint. . process endpoint is stored in p_endpoint, without generation number. it turns out the kernel never needs the generation number, except when fork()ing, so it's decoded then. . kernel calls all take endpoints as arguments, not proc numbers. the one exception is sys_fork(), which needs to know in which slot to put the child.
2006-03-03 11:00:02 +01:00
#include <minix/endpoint.h>
2005-04-21 16:53:53 +02:00
#include "proc.h"
2005-05-02 16:30:04 +02:00
/* Prototype declarations for PRIVATE functions. */
FORWARD _PROTOTYPE( void announce, (void));
2005-04-21 16:53:53 +02:00
/*===========================================================================*
2005-09-11 18:44:06 +02:00
* main *
2005-04-21 16:53:53 +02:00
*===========================================================================*/
PUBLIC void main()
{
/* Start the ball rolling. */
struct boot_image *ip; /* boot image pointer */
register struct proc *rp; /* process pointer */
register struct priv *sp; /* privilege structure pointer */
2007-04-23 15:46:54 +02:00
register int i, j, s;
2005-04-21 16:53:53 +02:00
int hdrindex; /* index to array of a.out headers */
phys_clicks text_base;
vir_clicks text_clicks, data_clicks, st_clicks;
2005-04-21 16:53:53 +02:00
reg_t ktsb; /* kernel task stack base */
struct exec e_hdr; /* for a copy of an a.out header */
/* Clear the process table. Anounce each slot as empty and set up mappings
* for proc_addr() and proc_nr() macros. Do the same for the table with
* privilege structures for the system processes.
2005-04-21 16:53:53 +02:00
*/
for (rp = BEG_PROC_ADDR, i = -NR_TASKS; rp < END_PROC_ADDR; ++rp, ++i) {
rp->p_rts_flags = SLOT_FREE; /* initialize free slot */
#if DEBUG_SCHED_CHECK
rp->p_magic = PMAGIC;
#endif
2005-04-21 16:53:53 +02:00
rp->p_nr = i; /* proc number from ptr */
'proc number' is process slot, 'endpoint' are generation-aware process instance numbers, encoded and decoded using macros in <minix/endpoint.h>. proc number -> endpoint migration . proc_nr in the interrupt hook is now an endpoint, proc_nr_e. . m_source for messages and notifies is now an endpoint, instead of proc number. . isokendpt() converts an endpoint to a process number, returns success (but fails if the process number is out of range, the process slot is not a living process, or the given endpoint number does not match the endpoint number in the process slot, indicating an old process). . okendpt() is the same as isokendpt(), but panic()s if the conversion fails. This is mainly used for decoding message.m_source endpoints, and other endpoint numbers in kernel data structures, which should always be correct. . if DEBUG_ENABLE_IPC_WARNINGS is enabled, isokendpt() and okendpt() get passed the __FILE__ and __LINE__ of the calling lines, and print messages about what is wrong with the endpoint number (out of range proc, empty proc, or inconsistent endpoint number), with the caller, making finding where the conversion failed easy without having to include code for every call to print where things went wrong. Sometimes this is harmless (wrong arg to a kernel call), sometimes it's a fatal internal inconsistency (bogus m_source). . some process table fields have been appended an _e to indicate it's become and endpoint. . process endpoint is stored in p_endpoint, without generation number. it turns out the kernel never needs the generation number, except when fork()ing, so it's decoded then. . kernel calls all take endpoints as arguments, not proc numbers. the one exception is sys_fork(), which needs to know in which slot to put the child.
2006-03-03 11:00:02 +01:00
rp->p_endpoint = _ENDPOINT(0, rp->p_nr); /* generation no. 0 */
2005-04-21 16:53:53 +02:00
(pproc_addr + NR_TASKS)[i] = rp; /* proc ptr from number */
}
for (sp = BEG_PRIV_ADDR, i = 0; sp < END_PRIV_ADDR; ++sp, ++i) {
sp->s_proc_nr = NONE; /* initialize as free */
sp->s_id = i; /* priv structure index */
ppriv_addr[i] = sp; /* priv ptr from number */
}
2005-04-21 16:53:53 +02:00
2005-10-02 21:02:05 +02:00
/* Set up proc table entries for processes in boot image. The stacks of the
2005-04-21 16:53:53 +02:00
* kernel tasks are initialized to an array in data space. The stacks
* of the servers have been added to the data segment by the monitor, so
* the stack pointer is set to the end of the data segment. All the
* processes are in low memory on the 8086. On the 386 only the kernel
* is in low memory, the rest is loaded in extended memory.
*/
/* Task stacks. */
ktsb = (reg_t) t_stack;
for (i=0; i < NR_BOOT_PROCS; ++i) {
int ci;
bitchunk_t fv;
ip = &image[i]; /* process' attributes */
rp = proc_addr(ip->proc_nr); /* get process pointer */
'proc number' is process slot, 'endpoint' are generation-aware process instance numbers, encoded and decoded using macros in <minix/endpoint.h>. proc number -> endpoint migration . proc_nr in the interrupt hook is now an endpoint, proc_nr_e. . m_source for messages and notifies is now an endpoint, instead of proc number. . isokendpt() converts an endpoint to a process number, returns success (but fails if the process number is out of range, the process slot is not a living process, or the given endpoint number does not match the endpoint number in the process slot, indicating an old process). . okendpt() is the same as isokendpt(), but panic()s if the conversion fails. This is mainly used for decoding message.m_source endpoints, and other endpoint numbers in kernel data structures, which should always be correct. . if DEBUG_ENABLE_IPC_WARNINGS is enabled, isokendpt() and okendpt() get passed the __FILE__ and __LINE__ of the calling lines, and print messages about what is wrong with the endpoint number (out of range proc, empty proc, or inconsistent endpoint number), with the caller, making finding where the conversion failed easy without having to include code for every call to print where things went wrong. Sometimes this is harmless (wrong arg to a kernel call), sometimes it's a fatal internal inconsistency (bogus m_source). . some process table fields have been appended an _e to indicate it's become and endpoint. . process endpoint is stored in p_endpoint, without generation number. it turns out the kernel never needs the generation number, except when fork()ing, so it's decoded then. . kernel calls all take endpoints as arguments, not proc numbers. the one exception is sys_fork(), which needs to know in which slot to put the child.
2006-03-03 11:00:02 +01:00
ip->endpoint = rp->p_endpoint; /* ipc endpoint */
rp->p_max_priority = ip->priority; /* max scheduling priority */
rp->p_priority = ip->priority; /* current priority */
rp->p_quantum_size = ip->quantum; /* quantum size in ticks */
rp->p_ticks_left = ip->quantum; /* current credit */
strncpy(rp->p_name, ip->proc_name, P_NAME_LEN); /* set process name */
(void) get_priv(rp, (ip->flags & SYS_PROC)); /* assign structure */
priv(rp)->s_flags = ip->flags; /* process flags */
priv(rp)->s_trap_mask = ip->trap_mask; /* allowed traps */
IPC privileges fixes Kernel: o Remove s_ipc_sendrec, instead using s_ipc_to for all send primitives o Centralize s_ipc_to bit manipulation, - disallowing assignment of bits pointing to unused priv structs; - preventing send-to-self by not setting bit for own priv struct; - preserving send mask matrix symmetry in all cases o Add IPC send mask checks to SENDA, which were missing entirely somehow o Slightly improve IPC stats accounting for SENDA o Remove SYSTEM from user processes' send mask o Half-fix the dependency between boot image order and process numbers, - correcting the table order of the boot processes; - documenting the order requirement needed for proper send masks; - warning at boot time if the order is violated RS: o Add support in /etc/drivers.conf for servers that talk to user processes, - disallowing IPC to user processes if no "ipc" field is present - adding a special "USER" label to explicitly allow IPC to user processes o Always apply IPC masks when specified; remove -i flag from service(8) o Use kernel send mask symmetry to delay adding IPC permissions for labels that do not exist yet, adding them to that label's process upon creation o Add VM to ipc permissions list for rtl8139 and fxp in drivers.conf Left to future fixes: o Removal of the table order vs process numbers dependency altogether, possibly using per-process send list structures as used for SYSTEM calls o Proper assignment of send masks to boot processes; some of the assigned (~0) masks are much wider than necessary o Proper assignment of IPC send masks for many more servers in drivers.conf o Removal of the debugging warning about the now legitimate case where RS's add_forward_ipc cannot find the IPC destination's label yet
2009-07-02 18:25:31 +02:00
/* Warn about violations of the boot image table order consistency. */
if (priv_id(rp) != s_nr_to_id(ip->proc_nr))
kprintf("Warning: boot image table has wrong process order\n");
/* Initialize call mask bitmap from unordered set.
* A single SYS_ALL_CALLS is a special case - it
* means all calls are allowed.
*/
if(ip->nr_k_calls == 1 && ip->k_calls[0] == SYS_ALL_CALLS)
fv = ~0; /* fill call mask */
else
fv = 0; /* clear call mask */
for(ci = 0; ci < CALL_MASK_SIZE; ci++) /* fill or clear call mask */
priv(rp)->s_k_call_mask[ci] = fv;
if(!fv) /* not all full? enter calls bit by bit */
for(ci = 0; ci < ip->nr_k_calls; ci++)
SET_BIT(priv(rp)->s_k_call_mask,
ip->k_calls[ci]-KERNEL_CALL);
IPC privileges fixes Kernel: o Remove s_ipc_sendrec, instead using s_ipc_to for all send primitives o Centralize s_ipc_to bit manipulation, - disallowing assignment of bits pointing to unused priv structs; - preventing send-to-self by not setting bit for own priv struct; - preserving send mask matrix symmetry in all cases o Add IPC send mask checks to SENDA, which were missing entirely somehow o Slightly improve IPC stats accounting for SENDA o Remove SYSTEM from user processes' send mask o Half-fix the dependency between boot image order and process numbers, - correcting the table order of the boot processes; - documenting the order requirement needed for proper send masks; - warning at boot time if the order is violated RS: o Add support in /etc/drivers.conf for servers that talk to user processes, - disallowing IPC to user processes if no "ipc" field is present - adding a special "USER" label to explicitly allow IPC to user processes o Always apply IPC masks when specified; remove -i flag from service(8) o Use kernel send mask symmetry to delay adding IPC permissions for labels that do not exist yet, adding them to that label's process upon creation o Add VM to ipc permissions list for rtl8139 and fxp in drivers.conf Left to future fixes: o Removal of the table order vs process numbers dependency altogether, possibly using per-process send list structures as used for SYSTEM calls o Proper assignment of send masks to boot processes; some of the assigned (~0) masks are much wider than necessary o Proper assignment of IPC send masks for many more servers in drivers.conf o Removal of the debugging warning about the now legitimate case where RS's add_forward_ipc cannot find the IPC destination's label yet
2009-07-02 18:25:31 +02:00
for (j = 0; j < NR_SYS_PROCS && j < BITCHUNK_BITS; j++)
if (ip->ipc_to & (1 << j))
set_sendto_bit(rp, j); /* restrict targets */
2007-04-23 15:46:54 +02:00
if (iskerneln(proc_nr(rp))) { /* part of the kernel? */
if (ip->stksize > 0) { /* HARDWARE stack size is 0 */
rp->p_priv->s_stack_guard = (reg_t *) ktsb;
*rp->p_priv->s_stack_guard = STACK_GUARD;
2005-04-21 16:53:53 +02:00
}
ktsb += ip->stksize; /* point to high end of stack */
2005-04-21 16:53:53 +02:00
rp->p_reg.sp = ktsb; /* this task's initial stack ptr */
hdrindex = 0; /* all use the first a.out header */
} else {
hdrindex = 1 + i-NR_TASKS; /* servers, drivers, INIT */
2005-04-21 16:53:53 +02:00
}
/* Architecture-specific way to find out aout header of this
* boot process.
2005-04-21 16:53:53 +02:00
*/
arch_get_aout_headers(hdrindex, &e_hdr);
2005-04-21 16:53:53 +02:00
/* Convert addresses to clicks and build process memory map */
text_base = e_hdr.a_syms >> CLICK_SHIFT;
text_clicks = (e_hdr.a_text + CLICK_SIZE-1) >> CLICK_SHIFT;
data_clicks = (e_hdr.a_data+e_hdr.a_bss + CLICK_SIZE-1) >> CLICK_SHIFT;
st_clicks= (e_hdr.a_total + CLICK_SIZE-1) >> CLICK_SHIFT;
if (!(e_hdr.a_flags & A_SEP))
{
data_clicks= (e_hdr.a_text+e_hdr.a_data+e_hdr.a_bss +
CLICK_SIZE-1) >> CLICK_SHIFT;
text_clicks = 0; /* common I&D */
}
2005-04-21 16:53:53 +02:00
rp->p_memmap[T].mem_phys = text_base;
rp->p_memmap[T].mem_len = text_clicks;
rp->p_memmap[D].mem_phys = text_base + text_clicks;
rp->p_memmap[D].mem_len = data_clicks;
rp->p_memmap[S].mem_phys = text_base + text_clicks + st_clicks;
rp->p_memmap[S].mem_vir = st_clicks;
rp->p_memmap[S].mem_len = 0;
2005-04-21 16:53:53 +02:00
/* Set initial register values. The processor status word for tasks
* is different from that of other processes because tasks can
* access I/O; this is not allowed to less-privileged processes
*/
rp->p_reg.pc = (reg_t) ip->initial_pc;
rp->p_reg.psw = (iskernelp(rp)) ? INIT_TASK_PSW : INIT_PSW;
2005-04-21 16:53:53 +02:00
/* Initialize the server stack pointer. Take it down one word
* to give crtso.s something to use as "argc".
*/
if (isusern(proc_nr(rp))) { /* user-space process? */
2005-04-21 16:53:53 +02:00
rp->p_reg.sp = (rp->p_memmap[S].mem_vir +
rp->p_memmap[S].mem_len) << CLICK_SHIFT;
rp->p_reg.sp -= sizeof(reg_t);
}
/* If this process has its own page table, VM will set the
* PT up and manage it. VM will signal the kernel when it has
* done this; until then, don't let it run.
*/
if(priv(rp)->s_flags & PROC_FULLVM)
RTS_SET(rp, VMINHIBIT);
2005-04-21 16:53:53 +02:00
/* Set ready. The HARDWARE task is never ready. */
if (rp->p_nr == HARDWARE) RTS_SET(rp, NO_PRIORITY);
RTS_UNSET(rp, SLOT_FREE); /* remove SLOT_FREE and schedule */
2005-04-21 16:53:53 +02:00
alloc_segments(rp);
}
#if SPROFILE
sprofiling = 0; /* we're not profiling until instructed to */
#endif /* SPROFILE */
cprof_procs_no = 0; /* init nr of hash table slots used */
vm_running = 0;
krandom.random_sources = RANDOM_SOURCES;
krandom.random_elements = RANDOM_ELEMENTS;
/* MINIX is now ready. All boot image processes are on the ready queue.
* Return to the assembly code to start running the current process.
2005-04-21 16:53:53 +02:00
*/
bill_ptr = proc_addr(IDLE); /* it has to point somewhere */
announce(); /* print MINIX startup banner */
2005-04-21 16:53:53 +02:00
restart();
}
2005-09-11 18:44:06 +02:00
/*===========================================================================*
* announce *
*===========================================================================*/
2005-04-21 16:53:53 +02:00
PRIVATE void announce(void)
{
/* Display the MINIX startup banner. */
2005-10-06 12:21:24 +02:00
kprintf("\nMINIX %s.%s. "
#ifdef _SVN_REVISION
"(" _SVN_REVISION ")\n"
#endif
"Copyright 2009, Vrije Universiteit, Amsterdam, The Netherlands\n",
OS_RELEASE, OS_VERSION);
kprintf("MINIX is open source software, see http://www.minix3.org\n");
2005-04-21 16:53:53 +02:00
}
2005-09-11 18:44:06 +02:00
/*===========================================================================*
* prepare_shutdown *
*===========================================================================*/
PUBLIC void prepare_shutdown(how)
int how;
2005-04-21 16:53:53 +02:00
{
/* This function prepares to shutdown MINIX. */
static timer_t shutdown_timer;
register struct proc *rp;
message m;
2005-04-21 16:53:53 +02:00
/* Continue after 1 second, to give processes a chance to get scheduled to
* do shutdown work. Set a watchog timer to call shutdown(). The timer
* argument passes the shutdown status.
2005-04-21 16:53:53 +02:00
*/
kprintf("MINIX will now be shut down ...\n");
tmr_arg(&shutdown_timer)->ta_int = how;
set_timer(&shutdown_timer, get_uptime() + system_hz, minix_shutdown);
2005-04-21 16:53:53 +02:00
}
Split of architecture-dependent and -independent functions for i386, mainly in the kernel and headers. This split based on work by Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture port. . kernel does not program the interrupt controller directly, do any other architecture-dependent operations, or contain assembly any more, but uses architecture-dependent functions in arch/$(ARCH)/. . architecture-dependent constants and types defined in arch/$(ARCH)/include. . <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now, architecture-independent functions. . int86, sdevio, readbios, and iopenable are now i386-specific kernel calls and live in arch/i386/do_* now. . i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have gone, and 'machine.protected' is gone (and always taken to be 1 in i386). If 86 support is to return, it should be a new architecture. . prototypes for the architecture-dependent functions defined in kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h . /etc/make.conf included in makefiles and shell scripts that need to know the building architecture; it defines ARCH=<arch>, currently only i386. . some basic per-architecture build support outside of the kernel (lib) . in clock.c, only dequeue a process if it was ready . fixes for new include files files deleted: . mpx/klib.s - only for choosing between mpx/klib86 and -386 . klib86.s - only for 86 i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/: . mpx386.s (entry point) . klib386.s . sconst.h . exception.c . protect.c . protect.h . i8269.c
2006-12-22 16:22:27 +01:00
2005-09-11 18:44:06 +02:00
/*===========================================================================*
* shutdown *
*===========================================================================*/
PUBLIC void minix_shutdown(tp)
timer_t *tp;
2005-04-21 16:53:53 +02:00
{
/* This function is called from prepare_shutdown or stop_sequence to bring
* down MINIX. How to shutdown is in the argument: RBT_HALT (return to the
* monitor), RBT_MONITOR (execute given code), RBT_RESET (hard reset).
2005-04-21 16:53:53 +02:00
*/
Split of architecture-dependent and -independent functions for i386, mainly in the kernel and headers. This split based on work by Ingmar Alting <iaalting@cs.vu.nl> done for his Minix PowerPC architecture port. . kernel does not program the interrupt controller directly, do any other architecture-dependent operations, or contain assembly any more, but uses architecture-dependent functions in arch/$(ARCH)/. . architecture-dependent constants and types defined in arch/$(ARCH)/include. . <ibm/portio.h> moved to <minix/portio.h>, as they have become, for now, architecture-independent functions. . int86, sdevio, readbios, and iopenable are now i386-specific kernel calls and live in arch/i386/do_* now. . i386 arch now supports even less 86 code; e.g. mpx86.s and klib86.s have gone, and 'machine.protected' is gone (and always taken to be 1 in i386). If 86 support is to return, it should be a new architecture. . prototypes for the architecture-dependent functions defined in kernel/arch/$(ARCH)/*.c but used in kernel/ are in kernel/proto.h . /etc/make.conf included in makefiles and shell scripts that need to know the building architecture; it defines ARCH=<arch>, currently only i386. . some basic per-architecture build support outside of the kernel (lib) . in clock.c, only dequeue a process if it was ready . fixes for new include files files deleted: . mpx/klib.s - only for choosing between mpx/klib86 and -386 . klib86.s - only for 86 i386-specific files files moved (or arch-dependent stuff moved) to arch/i386/: . mpx386.s (entry point) . klib386.s . sconst.h . exception.c . protect.c . protect.h . i8269.c
2006-12-22 16:22:27 +01:00
intr_init(INTS_ORIG);
2005-04-21 16:53:53 +02:00
clock_stop();
arch_shutdown(tp ? tmr_arg(tp)->ta_int : RBT_PANIC);
2005-04-21 16:53:53 +02:00
}