/* This file contains essentially all of the process and message handling.

* Together with "mpx.s" it forms the lowest layer of the MINIX kernel.

* There is one entry point from the outside:

*

* sys_call: a system call, i.e., the kernel is trapped with an INT

*

* Changes:

* Aug 19, 2005 rewrote scheduling code (Jorrit N. Herder)

* Jul 25, 2005 rewrote system call handling (Jorrit N. Herder)

* May 26, 2005 rewrote message passing functions (Jorrit N. Herder)

* May 24, 2005 new notification system call (Jorrit N. Herder)

* Oct 28, 2004 nonblocking send and receive calls (Jorrit N. Herder)

*

* The code here is critical to make everything work and is important for the

* overall performance of the system. A large fraction of the code deals with

* list manipulation. To make this both easy to understand and fast to execute

* pointer pointers are used throughout the code. Pointer pointers prevent

* exceptions for the head or tail of a linked list.

*

* node_t *queue, *new_node; // assume these as global variables

* node_t **xpp = &queue; // get pointer pointer to head of queue

* while (*xpp != NULL) // find last pointer of the linked list

* xpp = &(*xpp)->next; // get pointer to next pointer

* *xpp = new_node; // now replace the end (the NULL pointer)

* new_node->next = NULL; // and mark the new end of the list

*

* For example, when adding a new node to the end of the list, one normally

* makes an exception for an empty list and looks up the end of the list for

* nonempty lists. As shown above, this is not required with pointer pointers.

*/

#include <stddef.h>

#include <signal.h>

#include <assert.h>

#include <string.h>

#include "vm.h"

#include "clock.h"

#include "spinlock.h"

#include "arch_proto.h"

#include <minix/syslib.h>

/* Scheduling and message passing functions */

static void idle(void);

/**

* Made public for use in clock.c (for user-space scheduling)

static int mini_send(struct proc *caller_ptr, endpoint_t dst_e, message

*m_ptr, int flags);

*/

static int mini_receive(struct proc *caller_ptr, endpoint_t src,

message *m_buff_usr, int flags);

static int mini_senda(struct proc *caller_ptr, asynmsg_t *table, size_t

size);

static int deadlock(int function, register struct proc *caller,

endpoint_t src_dst_e);

static int try_async(struct proc *caller_ptr);

static int try_one(endpoint_t receive_e, struct proc *src_ptr,

struct proc *dst_ptr);

static struct proc * pick_proc(void);

static void enqueue_head(struct proc *rp);

/* all idles share the same idle_priv structure */

static struct priv idle_priv;

static void set_idle_name(char * name, int n)

{

int i, c;

int p_z = 0;

if (n > 999)

n = 999;

name[0] = 'i';

name[1] = 'd';

name[2] = 'l';

name[3] = 'e';

for (i = 4, c = 100; c > 0; c /= 10) {

int digit;

digit = n / c;

n -= digit * c;

if (p_z || digit != 0 || c == 1) {

p_z = 1;

name[i++] = '0' + digit;

}

}

name[i] = '\0';

}

#define PICK_ANY 1

#define PICK_HIGHERONLY 2

#define BuildNotifyMessage(m_ptr, src, dst_ptr) \

memset((m_ptr), 0, sizeof(*(m_ptr))); \

(m_ptr)->m_type = NOTIFY_MESSAGE; \

(m_ptr)->m_notify.timestamp = get_monotonic(); \

switch (src) { \

case HARDWARE: \

(m_ptr)->m_notify.interrupts = \

priv(dst_ptr)->s_int_pending; \

priv(dst_ptr)->s_int_pending = 0; \

break; \

case SYSTEM: \

memcpy(&(m_ptr)->m_notify.sigset, \

&priv(dst_ptr)->s_sig_pending, \

sizeof(sigset_t)); \

sigemptyset(&priv(dst_ptr)->s_sig_pending); \

break; \

}

static message m_notify_buff = { 0, NOTIFY_MESSAGE };

void proc_init(void)

{

struct proc * rp;

struct priv *sp;

int i;

/* Clear the process table. Announce 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.

*/

for (rp = BEG_PROC_ADDR, i = -NR_TASKS; rp < END_PROC_ADDR; ++rp, ++i) {

rp->p_rts_flags = RTS_SLOT_FREE;/* initialize free slot */

rp->p_magic = PMAGIC;

rp->p_nr = i; /* proc number from ptr */

rp->p_endpoint = _ENDPOINT(0, rp->p_nr); /* generation no. 0 */

rp->p_scheduler = NULL; /* no user space scheduler */

rp->p_priority = 0; /* no priority */

rp->p_quantum_size_ms = 0; /* no quantum size */

/* arch-specific initialization */

arch_proc_reset(rp);

}

for (sp = BEG_PRIV_ADDR, i = 0; sp < END_PRIV_ADDR; ++sp, ++i) {

sp->s_proc_nr = NONE; /* initialize as free */

sp->s_id = (sys_id_t) i; /* priv structure index */

ppriv_addr[i] = sp; /* priv ptr from number */

sp->s_sig_mgr = NONE; /* clear signal managers */

sp->s_bak_sig_mgr = NONE;

}

idle_priv.s_flags = IDL_F;

/* initialize IDLE structures for every CPU */

for (i = 0; i < CONFIG_MAX_CPUS; i++) {

struct proc * ip = get_cpu_var_ptr(i, idle_proc);

ip->p_endpoint = IDLE;

ip->p_priv = &idle_priv;

/* must not let idle ever get scheduled */

ip->p_rts_flags |= RTS_PROC_STOP;

set_idle_name(ip->p_name, i);

}

}

static void switch_address_space_idle(void)

{

#ifdef CONFIG_SMP

/*

* currently we bet that VM is always alive and its pages available so

* when the CPU wakes up the kernel is mapped and no surprises happen.

* This is only a problem if more than 1 cpus are available

*/

switch_address_space(proc_addr(VM_PROC_NR));

#endif

}

/*===========================================================================*

* idle *

*===========================================================================*/

static void idle(void)

{

struct proc * p;

/* This function is called whenever there is no work to do.

* Halt the CPU, and measure how many timestamp counter ticks are

* spent not doing anything. This allows test setups to measure

* the CPU utilization of certain workloads with high precision.

*/

p = get_cpulocal_var(proc_ptr) = get_cpulocal_var_ptr(idle_proc);

if (priv(p)->s_flags & BILLABLE)

get_cpulocal_var(bill_ptr) = p;

switch_address_space_idle();

#ifdef CONFIG_SMP

get_cpulocal_var(cpu_is_idle) = 1;

/* we don't need to keep time on APs as it is handled on the BSP */

if (cpuid != bsp_cpu_id)

stop_local_timer();

else

#endif

{

/*

* If the timer has expired while in kernel we must

* rearm it before we go to sleep

*/

restart_local_timer();

}

/* start accounting for the idle time */

context_stop(proc_addr(KERNEL));

#if !SPROFILE

halt_cpu();

#else

if (!sprofiling)

halt_cpu();

else {

volatile int * v;

v = get_cpulocal_var_ptr(idle_interrupted);

interrupts_enable();

while (!*v)

arch_pause();

interrupts_disable();

*v = 0;

}

#endif

/*

* end of accounting for the idle task does not happen here, the kernel

* is handling stuff for quite a while before it gets back here!

*/

}

/*===========================================================================*

* vm_suspend *

*===========================================================================*/

void vm_suspend(struct proc *caller, const struct proc *target,

const vir_bytes linaddr, const vir_bytes len, const int type,

const int writeflag)

{

/* This range is not OK for this process. Set parameters

* of the request and notify VM about the pending request.

*/

assert(!RTS_ISSET(caller, RTS_VMREQUEST));

assert(!RTS_ISSET(target, RTS_VMREQUEST));

RTS_SET(caller, RTS_VMREQUEST);

caller->p_vmrequest.req_type = VMPTYPE_CHECK;

caller->p_vmrequest.target = target->p_endpoint;

caller->p_vmrequest.params.check.start = linaddr;

caller->p_vmrequest.params.check.length = len;

caller->p_vmrequest.params.check.writeflag = writeflag;

caller->p_vmrequest.type = type;

/* Connect caller on vmrequest wait queue. */

if(!(caller->p_vmrequest.nextrequestor = vmrequest))

if(OK != send_sig(VM_PROC_NR, SIGKMEM))

panic("send_sig failed");

vmrequest = caller;

}

/*===========================================================================*

* delivermsg *

*===========================================================================*/

static void delivermsg(struct proc *rp)

{

assert(!RTS_ISSET(rp, RTS_VMREQUEST));

assert(rp->p_misc_flags & MF_DELIVERMSG);

assert(rp->p_delivermsg.m_source != NONE);

if (copy_msg_to_user(&rp->p_delivermsg,

(message *) rp->p_delivermsg_vir)) {

if(rp->p_misc_flags & MF_MSGFAILED) {

/* 2nd consecutive failure means this won't succeed */

printf("WARNING wrong user pointer 0x%08lx from "

"process %s / %d\n",

rp->p_delivermsg_vir,

rp->p_name,

rp->p_endpoint);

cause_sig(rp->p_nr, SIGSEGV);

} else {

/* 1st failure means we have to ask VM to handle it */

vm_suspend(rp, rp, rp->p_delivermsg_vir,

sizeof(message), VMSTYPE_DELIVERMSG, 1);

rp->p_misc_flags |= MF_MSGFAILED;

}

} else {

/* Indicate message has been delivered; address is 'used'. */

rp->p_delivermsg.m_source = NONE;

rp->p_misc_flags &= ~(MF_DELIVERMSG|MF_MSGFAILED);

if(!(rp->p_misc_flags & MF_CONTEXT_SET)) {

rp->p_reg.retreg = OK;

}

}

}

/*===========================================================================*

* switch_to_user *

*===========================================================================*/

void switch_to_user(void)

{

/* This function is called an instant before proc_ptr is

* to be scheduled again.

*/

struct proc * p;

#ifdef CONFIG_SMP

int tlb_must_refresh = 0;

#endif

p = get_cpulocal_var(proc_ptr);

/*

* if the current process is still runnable check the misc flags and let

* it run unless it becomes not runnable in the meantime

*/

if (proc_is_runnable(p))

goto check_misc_flags;

/*

* if a process becomes not runnable while handling the misc flags, we

* need to pick a new one here and start from scratch. Also if the

* current process wasn't runnable, we pick a new one here

*/

not_runnable_pick_new:

if (proc_is_preempted(p)) {

p->p_rts_flags &= ~RTS_PREEMPTED;

if (proc_is_runnable(p)) {

if (p->p_cpu_time_left)

enqueue_head(p);

else

enqueue(p);

}

}

/*

* if we have no process to run, set IDLE as the current process for

* time accounting and put the cpu in an idle state. After the next

* timer interrupt the execution resumes here and we can pick another

* process. If there is still nothing runnable we "schedule" IDLE again

*/

while (!(p = pick_proc())) {

idle();

}

/* update the global variable */

get_cpulocal_var(proc_ptr) = p;

#ifdef CONFIG_SMP

if (p->p_misc_flags & MF_FLUSH_TLB && get_cpulocal_var(ptproc) == p)

tlb_must_refresh = 1;

#endif

switch_address_space(p);

check_misc_flags:

assert(p);

assert(proc_is_runnable(p));

while (p->p_misc_flags &

(MF_KCALL_RESUME | MF_DELIVERMSG |

MF_SC_DEFER | MF_SC_TRACE | MF_SC_ACTIVE)) {

assert(proc_is_runnable(p));

if (p->p_misc_flags & MF_KCALL_RESUME) {

kernel_call_resume(p);

}

else if (p->p_misc_flags & MF_DELIVERMSG) {

TRACE(VF_SCHEDULING, printf("delivering to %s / %d\n",

p->p_name, p->p_endpoint););

delivermsg(p);

}

else if (p->p_misc_flags & MF_SC_DEFER) {

/* Perform the system call that we deferred earlier. */

assert (!(p->p_misc_flags & MF_SC_ACTIVE));

arch_do_syscall(p);

/* If the process is stopped for signal delivery, and

* not blocked sending a message after the system call,

* inform PM.

*/

if ((p->p_misc_flags & MF_SIG_DELAY) &&

!RTS_ISSET(p, RTS_SENDING))

sig_delay_done(p);

}

else if (p->p_misc_flags & MF_SC_TRACE) {

/* Trigger a system call leave event if this was a

* system call. We must do this after processing the

* other flags above, both for tracing correctness and

* to be able to use 'break'.

*/

if (!(p->p_misc_flags & MF_SC_ACTIVE))

break;

p->p_misc_flags &=

~(MF_SC_TRACE | MF_SC_ACTIVE);

/* Signal the "leave system call" event.

* Block the process.

*/

cause_sig(proc_nr(p), SIGTRAP);

}

else if (p->p_misc_flags & MF_SC_ACTIVE) {

/* If MF_SC_ACTIVE was set, remove it now:

* we're leaving the system call.

*/

p->p_misc_flags &= ~MF_SC_ACTIVE;

break;

}

/*

* the selected process might not be runnable anymore. We have

* to checkit and schedule another one

*/

if (!proc_is_runnable(p))

goto not_runnable_pick_new;

}

/*

* check the quantum left before it runs again. We must do it only here

* as we are sure that a possible out-of-quantum message to the

* scheduler will not collide with the regular ipc

*/

if (!p->p_cpu_time_left)

proc_no_time(p);

/*

* After handling the misc flags the selected process might not be

* runnable anymore. We have to checkit and schedule another one

*/

if (!proc_is_runnable(p))

goto not_runnable_pick_new;

TRACE(VF_SCHEDULING, printf("cpu %d starting %s / %d "

"pc 0x%08x\n",

cpuid, p->p_name, p->p_endpoint, p->p_reg.pc););

#if DEBUG_TRACE

p->p_schedules++;

#endif

p = arch_finish_switch_to_user();

assert(p->p_cpu_time_left);

context_stop(proc_addr(KERNEL));

/* If the process isn't the owner of FPU, enable the FPU exception */

if (get_cpulocal_var(fpu_owner) != p)

enable_fpu_exception();

else

disable_fpu_exception();

/* If MF_CONTEXT_SET is set, don't clobber process state within

* the kernel. The next kernel entry is OK again though.

*/

p->p_misc_flags &= ~MF_CONTEXT_SET;

#if defined(__i386__)

assert(p->p_seg.p_cr3 != 0);

#elif defined(__arm__)

assert(p->p_seg.p_ttbr != 0);

#endif

#ifdef CONFIG_SMP

if (p->p_misc_flags & MF_FLUSH_TLB) {

if (tlb_must_refresh)

refresh_tlb();

p->p_misc_flags &= ~MF_FLUSH_TLB;

}

#endif

restart_local_timer();

/*

* restore_user_context() carries out the actual mode switch from kernel

* to userspace. This function does not return

*/

restore_user_context(p);

NOT_REACHABLE;

}

/*

* handler for all synchronous IPC calls

*/

static int do_sync_ipc(struct proc * caller_ptr, /* who made the call */

int call_nr, /* system call number and flags */

endpoint_t src_dst_e, /* src or dst of the call */

message *m_ptr) /* users pointer to a message */

{

int result; /* the system call's result */

int src_dst_p; /* Process slot number */

char *callname;

/* Check destination. RECEIVE is the only call that accepts ANY (in addition

* to a real endpoint). The other calls (SEND, SENDREC, and NOTIFY) require an

* endpoint to corresponds to a process. In addition, it is necessary to check

* whether a process is allowed to send to a given destination.

*/

assert(call_nr != SENDA);

/* Only allow non-negative call_nr values less than 32 */

if (call_nr < 0 || call_nr > IPCNO_HIGHEST || call_nr >= 32

|| !(callname = ipc_call_names[call_nr])) {

#if DEBUG_ENABLE_IPC_WARNINGS

printf("sys_call: trap %d not allowed, caller %d, src_dst %d\n",

call_nr, proc_nr(caller_ptr), src_dst_e);

#endif

return(ETRAPDENIED); /* trap denied by mask or kernel */

}

if (src_dst_e == ANY)

{

if (call_nr != RECEIVE)

{

#if 0

printf("sys_call: %s by %d with bad endpoint %d\n",

callname,

proc_nr(caller_ptr), src_dst_e);

#endif

return EINVAL;

}

src_dst_p = (int) src_dst_e;

}

else

{

/* Require a valid source and/or destination process. */

if(!isokendpt(src_dst_e, &src_dst_p)) {

#if 0

printf("sys_call: %s by %d with bad endpoint %d\n",

callname,

proc_nr(caller_ptr), src_dst_e);

#endif

return EDEADSRCDST;

}

/* If the call is to send to a process, i.e., for SEND, SENDNB,

* SENDREC or NOTIFY, verify that the caller is allowed to send to

* the given destination.

*/

if (call_nr != RECEIVE)

{

if (!may_send_to(caller_ptr, src_dst_p)) {

#if DEBUG_ENABLE_IPC_WARNINGS

printf(

"sys_call: ipc mask denied %s from %d to %d\n",

callname,

caller_ptr->p_endpoint, src_dst_e);

#endif

return(ECALLDENIED); /* call denied by ipc mask */

}

}

}

/* Check if the process has privileges for the requested call. Calls to the

* kernel may only be SENDREC, because tasks always reply and may not block

* if the caller doesn't do receive().

*/

if (!(priv(caller_ptr)->s_trap_mask & (1 << call_nr))) {

#if DEBUG_ENABLE_IPC_WARNINGS

printf("sys_call: %s not allowed, caller %d, src_dst %d\n",

callname, proc_nr(caller_ptr), src_dst_p);

#endif

return(ETRAPDENIED); /* trap denied by mask or kernel */

}

if (call_nr != SENDREC && call_nr != RECEIVE && iskerneln(src_dst_p)) {

#if DEBUG_ENABLE_IPC_WARNINGS

printf("sys_call: trap %s not allowed, caller %d, src_dst %d\n",

callname, proc_nr(caller_ptr), src_dst_e);

#endif

return(ETRAPDENIED); /* trap denied by mask or kernel */

}

switch(call_nr) {

case SENDREC:

/* A flag is set so that notifications cannot interrupt SENDREC. */

caller_ptr->p_misc_flags |= MF_REPLY_PEND;

/* fall through */

case SEND:

result = mini_send(caller_ptr, src_dst_e, m_ptr, 0);

if (call_nr == SEND || result != OK)

break; /* done, or SEND failed */

/* fall through for SENDREC */

case RECEIVE:

if (call_nr == RECEIVE) {

caller_ptr->p_misc_flags &= ~MF_REPLY_PEND;

IPC_STATUS_CLEAR(caller_ptr); /* clear IPC status code */

}

result = mini_receive(caller_ptr, src_dst_e, m_ptr, 0);

break;

case NOTIFY:

result = mini_notify(caller_ptr, src_dst_e);

break;

case SENDNB:

result = mini_send(caller_ptr, src_dst_e, m_ptr, NON_BLOCKING);

break;

default:

result = EBADCALL; /* illegal system call */

}

/* Now, return the result of the system call to the caller. */

return(result);

}

int do_ipc(reg_t r1, reg_t r2, reg_t r3)

{

struct proc *const caller_ptr = get_cpulocal_var(proc_ptr); /* get pointer to caller */

int call_nr = (int) r1;

assert(!RTS_ISSET(caller_ptr, RTS_SLOT_FREE));

/* bill kernel time to this process. */

kbill_ipc = caller_ptr;

/* If this process is subject to system call tracing, handle that first. */

if (caller_ptr->p_misc_flags & (MF_SC_TRACE | MF_SC_DEFER)) {

/* Are we tracing this process, and is it the first sys_call entry? */

if ((caller_ptr->p_misc_flags & (MF_SC_TRACE | MF_SC_DEFER)) ==

MF_SC_TRACE) {

/* We must notify the tracer before processing the actual

* system call. If we don't, the tracer could not obtain the

* input message. Postpone the entire system call.

*/

caller_ptr->p_misc_flags &= ~MF_SC_TRACE;

assert(!(caller_ptr->p_misc_flags & MF_SC_DEFER));

caller_ptr->p_misc_flags |= MF_SC_DEFER;

caller_ptr->p_defer.r1 = r1;

caller_ptr->p_defer.r2 = r2;

caller_ptr->p_defer.r3 = r3;

/* Signal the "enter system call" event. Block the process. */

cause_sig(proc_nr(caller_ptr), SIGTRAP);

/* Preserve the return register's value. */

return caller_ptr->p_reg.retreg;

}

/* If the MF_SC_DEFER flag is set, the syscall is now being resumed. */

caller_ptr->p_misc_flags &= ~MF_SC_DEFER;

assert (!(caller_ptr->p_misc_flags & MF_SC_ACTIVE));

/* Set a flag to allow reliable tracing of leaving the system call. */

caller_ptr->p_misc_flags |= MF_SC_ACTIVE;

}

if(caller_ptr->p_misc_flags & MF_DELIVERMSG) {

panic("sys_call: MF_DELIVERMSG on for %s / %d\n",

caller_ptr->p_name, caller_ptr->p_endpoint);

}

/* Now check if the call is known and try to perform the request. The only

* system calls that exist in MINIX are sending and receiving messages.

* - SENDREC: combines SEND and RECEIVE in a single system call

* - SEND: sender blocks until its message has been delivered

* - RECEIVE: receiver blocks until an acceptable message has arrived

* - NOTIFY: asynchronous call; deliver notification or mark pending

* - SENDA: list of asynchronous send requests

*/

switch(call_nr) {

case SENDREC:

case SEND:

case RECEIVE:

case NOTIFY:

case SENDNB:

{

/* Process accounting for scheduling */

caller_ptr->p_accounting.ipc_sync++;

return do_sync_ipc(caller_ptr, call_nr, (endpoint_t) r2,

(message *) r3);

}

case SENDA:

{

/*

* Get and check the size of the argument in bytes as it is a

* table

*/

size_t msg_size = (size_t) r2;

/* Process accounting for scheduling */

caller_ptr->p_accounting.ipc_async++;

/* Limit size to something reasonable. An arbitrary choice is 16

* times the number of process table entries.

*/

if (msg_size > 16*(NR_TASKS + NR_PROCS))

return EDOM;

return mini_senda(caller_ptr, (asynmsg_t *) r3, msg_size);

}

case MINIX_KERNINFO:

{

/* It might not be initialized yet. */

if(!minix_kerninfo_user) {

return EBADCALL;

}

arch_set_secondary_ipc_return(caller_ptr, minix_kerninfo_user);

return OK;

}

default:

return EBADCALL; /* illegal system call */

}

}

/*===========================================================================*

* deadlock *

*===========================================================================*/

static int deadlock(

int function, /* trap number */

register struct proc *cp, /* pointer to caller */

endpoint_t src_dst_e /* src or dst process */

)

{

/* Check for deadlock. This can happen if 'caller_ptr' and 'src_dst' have

* a cyclic dependency of blocking send and receive calls. The only cyclic

* dependency that is not fatal is if the caller and target directly SEND(REC)

* and RECEIVE to each other. If a deadlock is found, the group size is

* returned. Otherwise zero is returned.

*/

register struct proc *xp; /* process pointer */

int group_size = 1; /* start with only caller */

#if DEBUG_ENABLE_IPC_WARNINGS

static struct proc *processes[NR_PROCS + NR_TASKS];

processes[0] = cp;

#endif

while (src_dst_e != ANY) { /* check while process nr */

int src_dst_slot;

okendpt(src_dst_e, &src_dst_slot);

xp = proc_addr(src_dst_slot); /* follow chain of processes */

assert(proc_ptr_ok(xp));

assert(!RTS_ISSET(xp, RTS_SLOT_FREE));

#if DEBUG_ENABLE_IPC_WARNINGS

processes[group_size] = xp;

#endif

group_size ++; /* extra process in group */

/* Check whether the last process in the chain has a dependency. If it

* has not, the cycle cannot be closed and we are done.

*/

if((src_dst_e = P_BLOCKEDON(xp)) == NONE)

return 0;

/* Now check if there is a cyclic dependency. For group sizes of two,

* a combination of SEND(REC) and RECEIVE is not fatal. Larger groups

* or other combinations indicate a deadlock.

*/

if (src_dst_e == cp->p_endpoint) { /* possible deadlock */

if (group_size == 2) { /* caller and src_dst */

/* The function number is magically converted to flags. */

if ((xp->p_rts_flags ^ (function << 2)) & RTS_SENDING) {

return(0); /* not a deadlock */

}

}

#if DEBUG_ENABLE_IPC_WARNINGS

{

int i;

printf("deadlock between these processes:\n");

for(i = 0; i < group_size; i++) {

printf(" %10s ", processes[i]->p_name);

}

printf("\n\n");

for(i = 0; i < group_size; i++) {

print_proc(processes[i]);

proc_stacktrace(processes[i]);

}

}

#endif

return(group_size); /* deadlock found */

}

}

return(0); /* not a deadlock */

}

/*===========================================================================*

* has_pending *

*===========================================================================*/

static int has_pending(sys_map_t *map, int src_p, int asynm)

{

/* Check to see if there is a pending message from the desired source

* available.

*/

int src_id;

sys_id_t id = NULL_PRIV_ID;

#ifdef CONFIG_SMP

struct proc * p;

#endif

/* Either check a specific bit in the mask map, or find the first bit set in

* it (if any), depending on whether the receive was called on a specific

* source endpoint.

*/

if (src_p != ANY) {

src_id = nr_to_id(src_p);

if (get_sys_bit(*map, src_id)) {

#ifdef CONFIG_SMP

p = proc_addr(id_to_nr(src_id));

if (asynm && RTS_ISSET(p, RTS_VMINHIBIT))

p->p_misc_flags |= MF_SENDA_VM_MISS;

else

#endif

id = src_id;

}

} else {

/* Find a source with a pending message */

for (src_id = 0; src_id < NR_SYS_PROCS; src_id += BITCHUNK_BITS) {

if (get_sys_bits(*map, src_id) != 0) {

#ifdef CONFIG_SMP

while (src_id < NR_SYS_PROCS) {

while (!get_sys_bit(*map, src_id)) {

if (src_id == NR_SYS_PROCS)

goto quit_search;

src_id++;

}

p = proc_addr(id_to_nr(src_id));

/*

* We must not let kernel fiddle with pages of a

* process which are currently being changed by

* VM. It is dangerous! So do not report such a

* process as having pending async messages.

* Skip it.

*/

if (asynm && RTS_ISSET(p, RTS_VMINHIBIT)) {

p->p_misc_flags |= MF_SENDA_VM_MISS;

src_id++;

} else

goto quit_search;

}

#else

while (!get_sys_bit(*map, src_id)) src_id++;

goto quit_search;

#endif

}

}

quit_search:

if (src_id < NR_SYS_PROCS) /* Found one */

id = src_id;

}

return(id);

}

/*===========================================================================*

* has_pending_notify *

*===========================================================================*/

int has_pending_notify(struct proc * caller, int src_p)

{

sys_map_t * map = &priv(caller)->s_notify_pending;

return has_pending(map, src_p, 0);

}

/*===========================================================================*

* has_pending_asend *

*===========================================================================*/

int has_pending_asend(struct proc * caller, int src_p)

{

sys_map_t * map = &priv(caller)->s_asyn_pending;

return has_pending(map, src_p, 1);

}

/*===========================================================================*

* unset_notify_pending *

*===========================================================================*/

void unset_notify_pending(struct proc * caller, int src_p)

{

sys_map_t * map = &priv(caller)->s_notify_pending;

unset_sys_bit(*map, src_p);

}

/*===========================================================================*

* mini_send *

*===========================================================================*/

int mini_send(

register struct proc *caller_ptr, /* who is trying to send a message? */

endpoint_t dst_e, /* to whom is message being sent? */

message *m_ptr, /* pointer to message buffer */

const int flags

)

{

/* Send a message from 'caller_ptr' to 'dst'. If 'dst' is blocked waiting

* for this message, copy the message to it and unblock 'dst'. If 'dst' is

* not waiting at all, or is waiting for another source, queue 'caller_ptr'.

*/

register struct proc *dst_ptr;

register struct proc **xpp;

int dst_p;

dst_p = _ENDPOINT_P(dst_e);

dst_ptr = proc_addr(dst_p);

if (RTS_ISSET(dst_ptr, RTS_NO_ENDPOINT))

{

return EDEADSRCDST;

}

/* Check if 'dst' is blocked waiting for this message. The destination's

* RTS_SENDING flag may be set when its SENDREC call blocked while sending.

*/

if (WILLRECEIVE(caller_ptr->p_endpoint, dst_ptr, (vir_bytes)m_ptr, NULL)) {

int call;

/* Destination is indeed waiting for this message. */

assert(!(dst_ptr->p_misc_flags & MF_DELIVERMSG));

if (!(flags & FROM_KERNEL)) {

if(copy_msg_from_user(m_ptr, &dst_ptr->p_delivermsg))

return EFAULT;

} else {

dst_ptr->p_delivermsg = *m_ptr;

IPC_STATUS_ADD_FLAGS(dst_ptr, IPC_FLG_MSG_FROM_KERNEL);

}

dst_ptr->p_delivermsg.m_source = caller_ptr->p_endpoint;

dst_ptr->p_misc_flags |= MF_DELIVERMSG;

call = (caller_ptr->p_misc_flags & MF_REPLY_PEND ? SENDREC

: (flags & NON_BLOCKING ? SENDNB : SEND));

IPC_STATUS_ADD_CALL(dst_ptr, call);

if (dst_ptr->p_misc_flags & MF_REPLY_PEND)

dst_ptr->p_misc_flags &= ~MF_REPLY_PEND;

RTS_UNSET(dst_ptr, RTS_RECEIVING);

#if DEBUG_IPC_HOOK

hook_ipc_msgsend(&dst_ptr->p_delivermsg, caller_ptr, dst_ptr);

hook_ipc_msgrecv(&dst_ptr->p_delivermsg, caller_ptr, dst_ptr);

#endif

} else {

if(flags & NON_BLOCKING) {

return(ENOTREADY);

}

/* Check for a possible deadlock before actually blocking. */

if (deadlock(SEND, caller_ptr, dst_e)) {

return(ELOCKED);

}

/* Destination is not waiting. Block and dequeue caller. */

if (!(flags & FROM_KERNEL)) {

if(copy_msg_from_user(m_ptr, &caller_ptr->p_sendmsg))

return EFAULT;

} else {

caller_ptr->p_sendmsg = *m_ptr;

/*

* we need to remember that this message is from kernel so we

* can set the delivery status flags when the message is

* actually delivered

*/

caller_ptr->p_misc_flags |= MF_SENDING_FROM_KERNEL;

}

RTS_SET(caller_ptr, RTS_SENDING);

caller_ptr->p_sendto_e = dst_e;

/* Process is now blocked. Put in on the destination's queue. */

assert(caller_ptr->p_q_link == NULL);

xpp = &dst_ptr->p_caller_q; /* find end of list */

while (*xpp) xpp = &(*xpp)->p_q_link;

*xpp = caller_ptr; /* add caller to end */

#if DEBUG_IPC_HOOK

hook_ipc_msgsend(&caller_ptr->p_sendmsg, caller_ptr, dst_ptr);

#endif

}

return(OK);

}

/*===========================================================================*

* mini_receive *

*===========================================================================*/

static int mini_receive(struct proc * caller_ptr,

endpoint_t src_e, /* which message source is wanted */

message * m_buff_usr, /* pointer to message buffer */

const int flags)

{

/* A process or task wants to get a message. If a message is already queued,

* acquire it and deblock the sender. If no message from the desired source

* is available block the caller.

*/

register struct proc **xpp;

int r, src_id, found, src_proc_nr, src_p;

endpoint_t sender_e;

assert(!(caller_ptr->p_misc_flags & MF_DELIVERMSG));

/* This is where we want our message. */

caller_ptr->p_delivermsg_vir = (vir_bytes) m_buff_usr;

if(src_e == ANY) src_p = ANY;

else

{

okendpt(src_e, &src_p);

if (RTS_ISSET(proc_addr(src_p), RTS_NO_ENDPOINT))

{

return EDEADSRCDST;

}

}

/* Check to see if a message from desired source is already available. The

* caller's RTS_SENDING flag may be set if SENDREC couldn't send. If it is

* set, the process should be blocked.

*/

if (!RTS_ISSET(caller_ptr, RTS_SENDING)) {