minix/servers/pm/main.c

/* This file contains the main program of the process manager and some related
 * procedures.  When MINIX starts up, the kernel runs for a little while,
 * initializing itself and its tasks, and then it runs PM and FS.  Both PM
 * and FS initialize themselves as far as they can. PM asks the kernel for
 * all free memory and starts serving requests.
 *
 * The entry points into this file are:
 *   main:	starts PM running
 *   setreply:	set the reply to be sent to process making an PM system call
 */

#include "pm.h"
#include <minix/keymap.h>
#include <minix/callnr.h>
#include <minix/com.h>
#include <signal.h>
#include <stdlib.h>
#include <fcntl.h>
#include <sys/resource.h>
#include <string.h>
#include "mproc.h"
#include "param.h"

#include "../../kernel/const.h"
#include "../../kernel/config.h"
#include "../../kernel/type.h"
#include "../../kernel/proc.h"

FORWARD _PROTOTYPE( void get_work, (void)				);
FORWARD _PROTOTYPE( void pm_init, (void)				);
FORWARD _PROTOTYPE( int get_nice_value, (int queue)			);
FORWARD _PROTOTYPE( void get_mem_chunks, (struct memory *mem_chunks) 	);
FORWARD _PROTOTYPE( void patch_mem_chunks, (struct memory *mem_chunks, 
	struct mem_map *map_ptr) 	);
FORWARD _PROTOTYPE( void do_x86_vm, (struct memory mem_chunks[NR_MEMS])	);

#define click_to_round_k(n) \
	((unsigned) ((((unsigned long) (n) << CLICK_SHIFT) + 512) / 1024))

/*===========================================================================*
 *				main					     *
 *===========================================================================*/
PUBLIC int main()
{
/* Main routine of the process manager. */
  int result, s, proc_nr;
  struct mproc *rmp;
  sigset_t sigset;

  pm_init();			/* initialize process manager tables */

  /* This is PM's main loop-  get work and do it, forever and forever. */
  while (TRUE) {
	get_work();		/* wait for an PM system call */

	/* Check for system notifications first. Special cases. */
	if (call_nr == SYN_ALARM) {
		pm_expire_timers(m_in.NOTIFY_TIMESTAMP);
		result = SUSPEND;		/* don't reply */
	} else if (call_nr == SYS_SIG) {	/* signals pending */
		sigset = m_in.NOTIFY_ARG;
		if (sigismember(&sigset, SIGKSIG))  (void) ksig_pending();
		result = SUSPEND;		/* don't reply */
	}
	/* Else, if the system call number is valid, perform the call. */
	else if ((unsigned) call_nr >= NCALLS) {
		result = ENOSYS;
	} else {
		result = (*call_vec[call_nr])();
	}

	/* Send the results back to the user to indicate completion. */
	if (result != SUSPEND) setreply(who, result);

	swap_in();		/* maybe a process can be swapped in? */

	/* Send out all pending reply messages, including the answer to
	 * the call just made above.  The processes must not be swapped out.
	 */
	for (proc_nr=0, rmp=mproc; proc_nr < NR_PROCS; proc_nr++, rmp++) {
		/* In the meantime, the process may have been killed by a
		 * signal (e.g. if a lethal pending signal was unblocked)
		 * without the PM realizing it. If the slot is no longer in
		 * use or just a zombie, don't try to reply.
		 */
		if ((rmp->mp_flags & (REPLY | ONSWAP | IN_USE | ZOMBIE)) ==
		   (REPLY | IN_USE)) {
			if ((s=send(proc_nr, &rmp->mp_reply)) != OK) {
				panic(__FILE__,"PM can't reply to", proc_nr);
			}
			rmp->mp_flags &= ~REPLY;
		}
	}
  }
  return(OK);
}

/*===========================================================================*
 *				get_work				     *
 *===========================================================================*/
PRIVATE void get_work()
{
/* Wait for the next message and extract useful information from it. */
  if (receive(ANY, &m_in) != OK) panic(__FILE__,"PM receive error", NO_NUM);
  who = m_in.m_source;		/* who sent the message */
  call_nr = m_in.m_type;	/* system call number */

  /* Process slot of caller. Misuse PM's own process slot if the kernel is
   * calling. This can happen in case of synchronous alarms (CLOCK) or or 
   * event like pending kernel signals (SYSTEM).
   */
  mp = &mproc[who < 0 ? PM_PROC_NR : who];
}

/*===========================================================================*
 *				setreply				     *
 *===========================================================================*/
PUBLIC void setreply(proc_nr, result)
int proc_nr;			/* process to reply to */
int result;			/* result of call (usually OK or error #) */
{
/* Fill in a reply message to be sent later to a user process.  System calls
 * may occasionally fill in other fields, this is only for the main return
 * value, and for setting the "must send reply" flag.
 */
  register struct mproc *rmp = &mproc[proc_nr];

  rmp->mp_reply.reply_res = result;
  rmp->mp_flags |= REPLY;	/* reply pending */

  if (rmp->mp_flags & ONSWAP)
	swap_inqueue(rmp);	/* must swap this process back in */
}

/*===========================================================================*
 *				pm_init					     *
 *===========================================================================*/
PRIVATE void pm_init()
{
/* Initialize the process manager. 
 * Memory use info is collected from the boot monitor, the kernel, and
 * all processes compiled into the system image. Initially this information
 * is put into an array mem_chunks. Elements of mem_chunks are struct memory,
 * and hold base, size pairs in units of clicks. This array is small, there
 * should be no more than 8 chunks. After the array of chunks has been built
 * the contents are used to initialize the hole list. Space for the hole list
 * is reserved as an array with twice as many elements as the maximum number
 * of processes allowed. It is managed as a linked list, and elements of the
 * array are struct hole, which, in addition to storage for a base and size in 
 * click units also contain space for a link, a pointer to another element.
*/
  int s;
  static struct boot_image image[NR_BOOT_PROCS];
  register struct boot_image *ip;
  static char core_sigs[] = { SIGQUIT, SIGILL, SIGTRAP, SIGABRT,
			SIGEMT, SIGFPE, SIGUSR1, SIGSEGV, SIGUSR2 };
  static char ign_sigs[] = { SIGCHLD, SIGWINCH, SIGCONT };
  static char mess_sigs[] = { SIGTERM, SIGHUP, SIGABRT, SIGQUIT };
  register struct mproc *rmp;
  register int i;
  register char *sig_ptr;
  phys_clicks total_clicks, minix_clicks, free_clicks;
  message mess;
  struct mem_map mem_map[NR_LOCAL_SEGS];
  struct memory mem_chunks[NR_MEMS];

  /* Initialize process table, including timers. */
  for (rmp=&mproc[0]; rmp<&mproc[NR_PROCS]; rmp++) {
	tmr_inittimer(&rmp->mp_timer);
  }

  /* Build the set of signals which cause core dumps, and the set of signals
   * that are by default ignored.
   */
  sigemptyset(&core_sset);
  for (sig_ptr = core_sigs; sig_ptr < core_sigs+sizeof(core_sigs); sig_ptr++)
	sigaddset(&core_sset, *sig_ptr);
  sigemptyset(&ign_sset);
  for (sig_ptr = ign_sigs; sig_ptr < ign_sigs+sizeof(ign_sigs); sig_ptr++)
	sigaddset(&ign_sset, *sig_ptr);

  /* Obtain a copy of the boot monitor parameters and the kernel info struct.  
   * Parse the list of free memory chunks. This list is what the boot monitor 
   * reported, but it must be corrected for the kernel and system processes.
   */
  if ((s=sys_getmonparams(monitor_params, sizeof(monitor_params))) != OK)
      panic(__FILE__,"get monitor params failed",s);
  get_mem_chunks(mem_chunks);
  if ((s=sys_getkinfo(&kinfo)) != OK)
      panic(__FILE__,"get kernel info failed",s);

  /* Get the memory map of the kernel to see how much memory it uses. */
  if ((s=get_mem_map(SYSTASK, mem_map)) != OK)
  	panic(__FILE__,"couldn't get memory map of SYSTASK",s);
  minix_clicks = (mem_map[S].mem_phys+mem_map[S].mem_len)-mem_map[T].mem_phys;
  patch_mem_chunks(mem_chunks, mem_map);

  /* Initialize PM'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.
   */
  if (OK != (s=sys_getimage(image))) 
  	panic(__FILE__,"couldn't get image table: %d\n", s);
  procs_in_use = 0;				/* start populating table */
  printf("Building process table:");		/* show what's happening */
  for (ip = &image[0]; ip < &image[NR_BOOT_PROCS]; ip++) {		
  	if (ip->proc_nr >= 0) {			/* task have negative nrs */
  		procs_in_use += 1;		/* found user process */

		/* Set process details found in the image table. */
		rmp = &mproc[ip->proc_nr];	
  		strncpy(rmp->mp_name, ip->proc_name, PROC_NAME_LEN); 
		rmp->mp_parent = RS_PROC_NR;
		rmp->mp_nice = get_nice_value(ip->priority);
  		sigemptyset(&rmp->mp_sig2mess);
  		sigemptyset(&rmp->mp_ignore);	
  		sigemptyset(&rmp->mp_sigmask);
  		sigemptyset(&rmp->mp_catch);
		if (ip->proc_nr == INIT_PROC_NR) {	/* user process */
  			rmp->mp_procgrp = rmp->mp_pid = INIT_PID;
			rmp->mp_flags |= IN_USE; 
		}
		else {					/* system process */
  			rmp->mp_pid = get_free_pid();
			rmp->mp_flags |= IN_USE | DONT_SWAP | PRIV_PROC; 
#if DEAD_CODE
  			for (sig_ptr = mess_sigs; 
				sig_ptr < mess_sigs+sizeof(mess_sigs); 
				sig_ptr++)
			sigaddset(&rmp->mp_sig2mess, *sig_ptr);
#endif
		}

  		/* Get memory map for this process from the kernel. */
		if ((s=get_mem_map(ip->proc_nr, rmp->mp_seg)) != OK)
  			panic(__FILE__,"couldn't get process entry",s);
		if (rmp->mp_seg[T].mem_len != 0) rmp->mp_flags |= SEPARATE;
		minix_clicks += rmp->mp_seg[S].mem_phys + 
			rmp->mp_seg[S].mem_len - rmp->mp_seg[T].mem_phys;
  		patch_mem_chunks(mem_chunks, rmp->mp_seg);

		/* Tell FS about this system process. */
		mess.PR_PROC_NR = ip->proc_nr;
		mess.PR_PID = rmp->mp_pid;
  		if (OK != (s=send(FS_PROC_NR, &mess)))
			panic(__FILE__,"can't sync up with FS", s);
  		printf(" %s", ip->proc_name);	/* display process name */
  	}
  }
  printf(".\n");				/* last process done */

  /* Override some details. INIT, PM, FS and RS are somewhat special. */
  mproc[PM_PROC_NR].mp_pid = PM_PID;		/* PM has magic pid */
  mproc[RS_PROC_NR].mp_parent = INIT_PROC_NR;	/* INIT is root */
  sigfillset(&mproc[PM_PROC_NR].mp_ignore); 	/* guard against signals */
  sigfillset(&mproc[FS_PROC_NR].mp_sig2mess); 	/* forward signals */
  sigfillset(&mproc[TTY_PROC_NR].mp_sig2mess); 	/* forward signals */

  /* Tell FS that no more system processes follow and synchronize. */
  mess.PR_PROC_NR = NONE;
  if (sendrec(FS_PROC_NR, &mess) != OK || mess.m_type != OK)
	panic(__FILE__,"can't sync up with FS", NO_NUM);

#if ENABLE_BOOTDEV
  /* Possibly we must correct the memory chunks for the boot device. */
  if (kinfo.bootdev_size > 0) {
      mem_map[T].mem_phys = kinfo.bootdev_base >> CLICK_SHIFT;
      mem_map[T].mem_len = 0;
      mem_map[D].mem_len = (kinfo.bootdev_size+CLICK_SIZE-1) >> CLICK_SHIFT;
      patch_mem_chunks(mem_chunks, mem_map);
  }
#endif /* ENABLE_BOOTDEV */

  /* Withhold some memory from x86 VM */
  do_x86_vm(mem_chunks);

  /* Initialize tables to all physical memory and print memory information. */
  printf("Physical memory:");
  mem_init(mem_chunks, &free_clicks);
  total_clicks = minix_clicks + free_clicks;
  printf(" total %u KB,", click_to_round_k(total_clicks));
  printf(" system %u KB,", click_to_round_k(minix_clicks));
  printf(" free %u KB.\n", click_to_round_k(free_clicks));
}

/*===========================================================================*
 *				get_nice_value				     *
 *===========================================================================*/
PRIVATE int get_nice_value(queue)
int queue;				/* store mem chunks here */
{
/* Processes in the boot image have a priority assigned. The PM doesn't know
 * about priorities, but uses 'nice' values instead. The priority is between 
 * MIN_USER_Q and MAX_USER_Q. We have to scale between PRIO_MIN and PRIO_MAX.
 */ 
  int nice_val = (queue - USER_Q) * (PRIO_MAX-PRIO_MIN+1) / 
      (MIN_USER_Q-MAX_USER_Q+1);
  if (nice_val > PRIO_MAX) nice_val = PRIO_MAX;	/* shouldn't happen */
  if (nice_val < PRIO_MIN) nice_val = PRIO_MIN;	/* shouldn't happen */
  return nice_val;
}

#if _WORD_SIZE == 2
/* In real mode only 1M can be addressed, and in 16-bit protected we can go
 * no further than we can count in clicks.  (The 286 is further limited by
 * its 24 bit address bus, but we can assume in that case that no more than
 * 16M memory is reported by the BIOS.)
 */
#define MAX_REAL	0x00100000L
#define MAX_16BIT	(0xFFF0L << CLICK_SHIFT)
#endif

/*===========================================================================*
 *				get_mem_chunks				     *
 *===========================================================================*/
PRIVATE void get_mem_chunks(mem_chunks)
struct memory *mem_chunks;			/* store mem chunks here */
{
/* Initialize the free memory list from the 'memory' boot variable.  Translate
 * the byte offsets and sizes in this list to clicks, properly truncated. Also
 * make sure that we don't exceed the maximum address space of the 286 or the
 * 8086, i.e. when running in 16-bit protected mode or real mode.
 */
  long base, size, limit;
  char *s, *end;			/* use to parse boot variable */ 
  int i, done = 0;
  struct memory *memp;
#if _WORD_SIZE == 2
  unsigned long max_address;
  struct machine machine;
  if (OK != (i=sys_getmachine(&machine)))
	panic(__FILE__, "sys_getmachine failed", i);
#endif

  /* Initialize everything to zero. */
  for (i = 0; i < NR_MEMS; i++) {
	memp = &mem_chunks[i];		/* next mem chunk is stored here */
	memp->base = memp->size = 0;
  }
  
  /* The available memory is determined by MINIX' boot loader as a list of 
   * (base:size)-pairs in boothead.s. The 'memory' boot variable is set in
   * in boot.s.  The format is "b0:s0,b1:s1,b2:s2", where b0:s0 is low mem,
   * b1:s1 is mem between 1M and 16M, b2:s2 is mem above 16M. Pairs b1:s1 
   * and b2:s2 are combined if the memory is adjacent. 
   */
  s = find_param("memory");		/* get memory boot variable */
  for (i = 0; i < NR_MEMS && !done; i++) {
	memp = &mem_chunks[i];		/* next mem chunk is stored here */
	base = size = 0;		/* initialize next base:size pair */
	if (*s != 0) {			/* get fresh data, unless at end */	

	    /* Read fresh base and expect colon as next char. */ 
	    base = strtoul(s, &end, 0x10);		/* get number */
	    if (end != s && *end == ':') s = ++end;	/* skip ':' */ 
	    else *s=0;			/* terminate, should not happen */

	    /* Read fresh size and expect comma or assume end. */ 
	    size = strtoul(s, &end, 0x10);		/* get number */
	    if (end != s && *end == ',') s = ++end;	/* skip ',' */
	    else done = 1;
	}
	limit = base + size;	
#if _WORD_SIZE == 2
	max_address = machine.protected ? MAX_16BIT : MAX_REAL;
	if (limit > max_address) limit = max_address;
#endif
	base = (base + CLICK_SIZE-1) & ~(long)(CLICK_SIZE-1);
	limit &= ~(long)(CLICK_SIZE-1);
	if (limit <= base) continue;
	memp->base = base >> CLICK_SHIFT;
	memp->size = (limit - base) >> CLICK_SHIFT;
  }
}

/*===========================================================================*
 *				patch_mem_chunks			     *
 *===========================================================================*/
PRIVATE void patch_mem_chunks(mem_chunks, map_ptr)
struct memory *mem_chunks;			/* store mem chunks here */
struct mem_map *map_ptr;			/* memory to remove */
{
/* Remove server memory from the free memory list. The boot monitor
 * promises to put processes at the start of memory chunks. The 
 * tasks all use same base address, so only the first task changes
 * the memory lists. The servers and init have their own memory
 * spaces and their memory will be removed from the list. 
 */
  struct memory *memp;
  for (memp = mem_chunks; memp < &mem_chunks[NR_MEMS]; memp++) {
	if (memp->base == map_ptr[T].mem_phys) {
		memp->base += map_ptr[T].mem_len + map_ptr[D].mem_len;
		memp->size -= map_ptr[T].mem_len + map_ptr[D].mem_len;
	}
  }
}

#define PAGE_SIZE	4096
#define PAGE_TABLE_COVER (1024*PAGE_SIZE)
/*=========================================================================*
 *				do_x86_vm				   *
 *=========================================================================*/
PRIVATE void do_x86_vm(mem_chunks)
struct memory mem_chunks[NR_MEMS];
{
	phys_bytes high, bytes;
	phys_clicks clicks, base_click;
	unsigned pages;
	int i, r;

	/* Compute the highest memory location */
	high= 0;
	for (i= 0; i<NR_MEMS; i++)
	{
		if (mem_chunks[i].size == 0)
			continue;
		if (mem_chunks[i].base + mem_chunks[i].size > high)
			high= mem_chunks[i].base + mem_chunks[i].size;
	}

	high <<= CLICK_SHIFT;
#if VERBOSE_VM
	printf("do_x86_vm: found high 0x%x\n", high);
#endif

	/* The number of pages we need is one for the page directory, enough
	 * page tables to cover the memory, and one page for alignement.
	 */
	pages= 1 + (high + PAGE_TABLE_COVER-1)/PAGE_TABLE_COVER + 1;
	bytes= pages*PAGE_SIZE;
	clicks= (bytes + CLICK_SIZE-1) >> CLICK_SHIFT;

#if VERBOSE_VM
	printf("do_x86_vm: need %d pages\n", pages);
	printf("do_x86_vm: need %d bytes\n", bytes);
	printf("do_x86_vm: need %d clicks\n", clicks);
#endif

	for (i= 0; i<NR_MEMS; i++)
	{
		if (mem_chunks[i].size <= clicks)
			continue;
		break;
	}
	if (i >= NR_MEMS)
		panic("PM", "not enough memory for VM page tables?", NO_NUM);
	base_click= mem_chunks[i].base;
	mem_chunks[i].base += clicks;
	mem_chunks[i].size -= clicks;

#if VERBOSE_VM
	printf("do_x86_vm: using 0x%x clicks @ 0x%x\n", clicks, base_click);
#endif
	r= sys_vm_setbuf(base_click << CLICK_SHIFT, clicks << CLICK_SHIFT,
		high);
	if (r != 0)
		printf("do_x86_vm: sys_vm_setbuf failed: %d\n", r);
}