/*
* Probe all of a vlserver's addresses to find out the best route and to
* query its capabilities.
*/
static bool afs_do_probe_vlserver(struct afs_net *net,
struct afs_vlserver *server,
struct key *key,
unsigned int server_index,
struct afs_error *_e)
{
struct afs_addr_cursor ac = {
.index = 0,
};
struct afs_call *call;
bool in_progress = false;
_enter("%s", server->name);
read_lock(&server->lock);
ac.alist = rcu_dereference_protected(server->addresses,
lockdep_is_held(&server->lock));
read_unlock(&server->lock);
atomic_set(&server->probe_outstanding, ac.alist->nr_addrs);
memset(&server->probe, 0, sizeof(server->probe));
server->probe.rtt = UINT_MAX;
for (ac.index = 0; ac.index < ac.alist->nr_addrs; ac.index++) {
call = afs_vl_get_capabilities(net, &ac, key, server,
server_index);
if (!IS_ERR(call)) {
afs_put_call(call);
in_progress = true;
} else {
afs_prioritise_error(_e, PTR_ERR(call), ac.abort_code);
}
}
if (!in_progress)
afs_vl_probe_done(server);
return in_progress;
}
/*
* Send off probes to all unprobed servers.
*/
int afs_send_vl_probes(struct afs_net *net, struct key *key,
struct afs_vlserver_list *vllist)
{
struct afs_vlserver *server;
struct afs_error e;
bool in_progress = false;
int i;
e.error = 0;
e.responded = false;
for (i = 0; i < vllist->nr_servers; i++) {
server = vllist->servers[i].server;
if (test_bit(AFS_VLSERVER_FL_PROBED, &server->flags))
continue;
if (!test_and_set_bit_lock(AFS_VLSERVER_FL_PROBING, &server->flags) &&
afs_do_probe_vlserver(net, server, key, i, &e))
in_progress = true;
}
return in_progress ? 0 : e.error;
}
/*
* Wait for the first as-yet untried server to respond.
*/
int afs_wait_for_vl_probes(struct afs_vlserver_list *vllist,
unsigned long untried)
{
struct wait_queue_entry *waits;
struct afs_vlserver *server;
unsigned int rtt = UINT_MAX;
bool have_responders = false;
int pref = -1, i;
_enter("%u,%lx", vllist->nr_servers, untried);
/* Only wait for servers that have a probe outstanding. */
for (i = 0; i < vllist->nr_servers; i++) {
if (test_bit(i, &untried)) {
server = vllist->servers[i].server;
if (!test_bit(AFS_VLSERVER_FL_PROBING, &server->flags))
__clear_bit(i, &untried);
if (server->probe.responded)
have_responders = true;
}
}
if (have_responders || !untried)
return 0;
waits = kmalloc(array_size(vllist->nr_servers, sizeof(*waits)), GFP_KERNEL);
if (!waits)
return -ENOMEM;
for (i = 0; i < vllist->nr_servers; i++) {
if (test_bit(i, &untried)) {
server = vllist->servers[i].server;
init_waitqueue_entry(&waits[i], current);
add_wait_queue(&server->probe_wq, &waits[i]);
}
}
for (;;) {
bool still_probing = false;
set_current_state(TASK_INTERRUPTIBLE);
for (i = 0; i < vllist->nr_servers; i++) {
if (test_bit(i, &untried)) {
server = vllist->servers[i].server;
if (server->probe.responded)
goto stop;
if (test_bit(AFS_VLSERVER_FL_PROBING, &server->flags))
still_probing = true;
}
}
if (!still_probing || signal_pending(current))
goto stop;
schedule();
}
stop:
set_current_state(TASK_RUNNING);
for (i = 0; i < vllist->nr_servers; i++) {
if (test_bit(i, &untried)) {
server = vllist->servers[i].server;
if (server->probe.responded &&
server->probe.rtt < rtt) {
pref = i;
rtt = server->probe.rtt;
}
remove_wait_queue(&server->probe_wq, &waits[i]);
}
}
kfree(waits);
if (pref == -1 && signal_pending(current))
return -ERESTARTSYS;
if (pref >= 0)
vllist->preferred = pref;
_leave(" = 0 [%u]", pref);
return 0;
}
static int try_to_freeze_tasks(bool user_only)
{
struct task_struct *g, *p;
unsigned long end_time;
unsigned int todo;
bool wq_busy = false;
struct timeval start, end;
u64 elapsed_msecs64;
unsigned int elapsed_msecs;
bool wakeup = false;
int sleep_usecs = USEC_PER_MSEC;
char suspend_abort[MAX_SUSPEND_ABORT_LEN];
do_gettimeofday(&start);
end_time = jiffies + msecs_to_jiffies(freeze_timeout_msecs);
if (!user_only)
freeze_workqueues_begin();
while (true) {
todo = 0;
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (p == current || !freeze_task(p))
continue;
if (!freezer_should_skip(p))
todo++;
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
if (!user_only) {
wq_busy = freeze_workqueues_busy();
todo += wq_busy;
}
if (!todo || time_after(jiffies, end_time))
break;
if (pm_wakeup_pending()) {
pm_get_active_wakeup_sources(suspend_abort,
MAX_SUSPEND_ABORT_LEN);
log_suspend_abort_reason(suspend_abort);
wakeup = true;
break;
}
/*
* We need to retry, but first give the freezing tasks some
* time to enter the refrigerator. Start with an initial
* 1 ms sleep followed by exponential backoff until 8 ms.
*/
usleep_range(sleep_usecs / 2, sleep_usecs);
if (sleep_usecs < 8 * USEC_PER_MSEC)
sleep_usecs *= 2;
}
do_gettimeofday(&end);
elapsed_msecs64 = timeval_to_ns(&end) - timeval_to_ns(&start);
do_div(elapsed_msecs64, NSEC_PER_MSEC);
elapsed_msecs = elapsed_msecs64;
if (wakeup) {
printk("\n");
printk(KERN_ERR "Freezing of tasks aborted after %d.%03d seconds",
elapsed_msecs / 1000, elapsed_msecs % 1000);
} else if (todo) {
printk("\n");
printk(KERN_ERR "Freezing of tasks failed after %d.%03d seconds"
" (%d tasks refusing to freeze, wq_busy=%d):\n",
elapsed_msecs / 1000, elapsed_msecs % 1000,
todo - wq_busy, wq_busy);
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (p != current && !freezer_should_skip(p)
&& freezing(p) && !frozen(p))
sched_show_task(p);
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
} else {
int do_getitimer(int which, struct itimerval *value)
{
struct task_struct *tsk = current;
cputime_t cinterval, cval;
switch (which) {
case ITIMER_REAL:
spin_lock_irq(&tsk->sighand->siglock);
value->it_value = itimer_get_remtime(&tsk->signal->real_timer);
value->it_interval =
ktime_to_timeval(tsk->signal->it_real_incr);
spin_unlock_irq(&tsk->sighand->siglock);
break;
case ITIMER_VIRTUAL:
read_lock(&tasklist_lock);
spin_lock_irq(&tsk->sighand->siglock);
cval = tsk->signal->it_virt_expires;
cinterval = tsk->signal->it_virt_incr;
if (!cputime_eq(cval, cputime_zero)) {
struct task_struct *t = tsk;
cputime_t utime = tsk->signal->utime;
do {
utime = cputime_add(utime, t->utime);
t = next_thread(t);
} while (t != tsk);
if (cputime_le(cval, utime)) { /* about to fire */
cval = jiffies_to_cputime(1);
} else {
cval = cputime_sub(cval, utime);
}
}
spin_unlock_irq(&tsk->sighand->siglock);
read_unlock(&tasklist_lock);
cputime_to_timeval(cval, &value->it_value);
cputime_to_timeval(cinterval, &value->it_interval);
break;
case ITIMER_PROF:
read_lock(&tasklist_lock);
spin_lock_irq(&tsk->sighand->siglock);
cval = tsk->signal->it_prof_expires;
cinterval = tsk->signal->it_prof_incr;
if (!cputime_eq(cval, cputime_zero)) {
struct task_struct *t = tsk;
cputime_t ptime = cputime_add(tsk->signal->utime,
tsk->signal->stime);
do {
ptime = cputime_add(ptime,
cputime_add(t->utime,
t->stime));
t = next_thread(t);
} while (t != tsk);
if (cputime_le(cval, ptime)) { /* about to fire */
cval = jiffies_to_cputime(1);
} else {
cval = cputime_sub(cval, ptime);
}
}
spin_unlock_irq(&tsk->sighand->siglock);
read_unlock(&tasklist_lock);
cputime_to_timeval(cval, &value->it_value);
cputime_to_timeval(cinterval, &value->it_interval);
break;
default:
return(-EINVAL);
}
return 0;
}
asmlinkage long
sys_timer_create(const clockid_t which_clock,
struct sigevent __user *timer_event_spec,
timer_t __user * created_timer_id)
{
int error = 0;
struct k_itimer *new_timer = NULL;
int new_timer_id;
struct task_struct *process = NULL;
unsigned long flags;
sigevent_t event;
int it_id_set = IT_ID_NOT_SET;
if (invalid_clockid(which_clock))
return -EINVAL;
new_timer = alloc_posix_timer();
if (unlikely(!new_timer))
return -EAGAIN;
spin_lock_init(&new_timer->it_lock);
retry:
if (unlikely(!idr_pre_get(&posix_timers_id, GFP_KERNEL))) {
error = -EAGAIN;
goto out;
}
spin_lock_irq(&idr_lock);
error = idr_get_new(&posix_timers_id, (void *) new_timer,
&new_timer_id);
spin_unlock_irq(&idr_lock);
if (error == -EAGAIN)
goto retry;
else if (error) {
/*
* Wierd looking, but we return EAGAIN if the IDR is
* full (proper POSIX return value for this)
*/
error = -EAGAIN;
goto out;
}
it_id_set = IT_ID_SET;
new_timer->it_id = (timer_t) new_timer_id;
new_timer->it_clock = which_clock;
new_timer->it_overrun = -1;
error = CLOCK_DISPATCH(which_clock, timer_create, (new_timer));
if (error)
goto out;
/*
* return the timer_id now. The next step is hard to
* back out if there is an error.
*/
if (copy_to_user(created_timer_id,
&new_timer_id, sizeof (new_timer_id))) {
error = -EFAULT;
goto out;
}
if (timer_event_spec) {
if (copy_from_user(&event, timer_event_spec, sizeof (event))) {
error = -EFAULT;
goto out;
}
new_timer->it_sigev_notify = event.sigev_notify;
new_timer->it_sigev_signo = event.sigev_signo;
new_timer->it_sigev_value = event.sigev_value;
read_lock(&tasklist_lock);
if ((process = good_sigevent(&event))) {
/*
* We may be setting up this process for another
* thread. It may be exiting. To catch this
* case the we check the PF_EXITING flag. If
* the flag is not set, the siglock will catch
* him before it is too late (in exit_itimers).
*
* The exec case is a bit more invloved but easy
* to code. If the process is in our thread
* group (and it must be or we would not allow
* it here) and is doing an exec, it will cause
* us to be killed. In this case it will wait
* for us to die which means we can finish this
* linkage with our last gasp. I.e. no code :)
*/
spin_lock_irqsave(&process->sighand->siglock, flags);
if (!(process->flags & PF_EXITING)) {
new_timer->it_process = process;
list_add(&new_timer->list,
&process->signal->posix_timers);
if (new_timer->it_sigev_notify == (SIGEV_SIGNAL|SIGEV_THREAD_ID))
get_task_struct(process);
spin_unlock_irqrestore(&process->sighand->siglock, flags);
} else {
spin_unlock_irqrestore(&process->sighand->siglock, flags);
process = NULL;
}
}
read_unlock(&tasklist_lock);
if (!process) {
error = -EINVAL;
goto out;
}
} else {
new_timer->it_sigev_notify = SIGEV_SIGNAL;
new_timer->it_sigev_signo = SIGALRM;
new_timer->it_sigev_value.sival_int = new_timer->it_id;
process = current->group_leader;
spin_lock_irqsave(&process->sighand->siglock, flags);
new_timer->it_process = process;
list_add(&new_timer->list, &process->signal->posix_timers);
spin_unlock_irqrestore(&process->sighand->siglock, flags);
}
/*
* In the case of the timer belonging to another task, after
* the task is unlocked, the timer is owned by the other task
* and may cease to exist at any time. Don't use or modify
* new_timer after the unlock call.
*/
out:
if (error)
release_posix_timer(new_timer, it_id_set);
return error;
}
static int ipgre_rcv(struct sk_buff *skb)
{
struct iphdr *iph;
u8 *h;
__be16 flags;
__sum16 csum = 0;
__be32 key = 0;
u32 seqno = 0;
struct ip_tunnel *tunnel;
int offset = 4;
__be16 gre_proto;
unsigned int len;
if (!pskb_may_pull(skb, 16))
goto drop_nolock;
iph = ip_hdr(skb);
h = skb->data;
flags = *(__be16*)h;
if (flags&(GRE_CSUM|GRE_KEY|GRE_ROUTING|GRE_SEQ|GRE_VERSION)) {
/* - Version must be 0.
- We do not support routing headers.
*/
if (flags&(GRE_VERSION|GRE_ROUTING))
goto drop_nolock;
if (flags&GRE_CSUM) {
switch (skb->ip_summed) {
case CHECKSUM_COMPLETE:
csum = csum_fold(skb->csum);
if (!csum)
break;
/* fall through */
case CHECKSUM_NONE:
skb->csum = 0;
csum = __skb_checksum_complete(skb);
skb->ip_summed = CHECKSUM_COMPLETE;
}
offset += 4;
}
if (flags&GRE_KEY) {
key = *(__be32*)(h + offset);
offset += 4;
}
if (flags&GRE_SEQ) {
seqno = ntohl(*(__be32*)(h + offset));
offset += 4;
}
}
gre_proto = *(__be16 *)(h + 2);
read_lock(&ipgre_lock);
if ((tunnel = ipgre_tunnel_lookup(skb->dev,
iph->saddr, iph->daddr, key,
gre_proto))) {
struct net_device_stats *stats = &tunnel->dev->stats;
secpath_reset(skb);
skb->protocol = gre_proto;
/* WCCP version 1 and 2 protocol decoding.
* - Change protocol to IP
* - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
*/
if (flags == 0 && gre_proto == htons(ETH_P_WCCP)) {
skb->protocol = htons(ETH_P_IP);
if ((*(h + offset) & 0xF0) != 0x40)
offset += 4;
}
skb->mac_header = skb->network_header;
__pskb_pull(skb, offset);
skb_postpull_rcsum(skb, skb_transport_header(skb), offset);
skb->pkt_type = PACKET_HOST;
#ifdef CONFIG_NET_IPGRE_BROADCAST
if (ipv4_is_multicast(iph->daddr)) {
/* Looped back packet, drop it! */
if (skb_rtable(skb)->fl.iif == 0)
goto drop;
stats->multicast++;
skb->pkt_type = PACKET_BROADCAST;
}
#endif
if (((flags&GRE_CSUM) && csum) ||
(!(flags&GRE_CSUM) && tunnel->parms.i_flags&GRE_CSUM)) {
stats->rx_crc_errors++;
stats->rx_errors++;
goto drop;
}
if (tunnel->parms.i_flags&GRE_SEQ) {
if (!(flags&GRE_SEQ) ||
(tunnel->i_seqno && (s32)(seqno - tunnel->i_seqno) < 0)) {
stats->rx_fifo_errors++;
stats->rx_errors++;
goto drop;
}
tunnel->i_seqno = seqno + 1;
}
len = skb->len;
/* Warning: All skb pointers will be invalidated! */
if (tunnel->dev->type == ARPHRD_ETHER) {
if (!pskb_may_pull(skb, ETH_HLEN)) {
stats->rx_length_errors++;
stats->rx_errors++;
goto drop;
}
iph = ip_hdr(skb);
skb->protocol = eth_type_trans(skb, tunnel->dev);
skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
}
stats->rx_packets++;
stats->rx_bytes += len;
skb->dev = tunnel->dev;
skb_dst_drop(skb);
nf_reset(skb);
skb_reset_network_header(skb);
ipgre_ecn_decapsulate(iph, skb);
netif_rx(skb);
read_unlock(&ipgre_lock);
return(0);
}
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
drop:
read_unlock(&ipgre_lock);
drop_nolock:
kfree_skb(skb);
return(0);
}
void ipip_err(struct sk_buff *skb, u32 info)
{
#ifndef I_WISH_WORLD_WERE_PERFECT
/* It is not :-( All the routers (except for Linux) return only
8 bytes of packet payload. It means, that precise relaying of
ICMP in the real Internet is absolutely infeasible.
*/
struct iphdr *iph = (struct iphdr*)skb->data;
int type = skb->h.icmph->type;
int code = skb->h.icmph->code;
struct ip_tunnel *t;
switch (type) {
default:
case ICMP_PARAMETERPROB:
return;
case ICMP_DEST_UNREACH:
switch (code) {
case ICMP_SR_FAILED:
case ICMP_PORT_UNREACH:
/* Impossible event. */
return;
case ICMP_FRAG_NEEDED:
/* Soft state for pmtu is maintained by IP core. */
return;
default:
/* All others are translated to HOST_UNREACH.
rfc2003 contains "deep thoughts" about NET_UNREACH,
I believe they are just ether pollution. --ANK
*/
break;
}
break;
case ICMP_TIME_EXCEEDED:
if (code != ICMP_EXC_TTL)
return;
break;
}
read_lock(&ipip_lock);
t = ipip_tunnel_lookup(iph->daddr, iph->saddr);
if (t == NULL || t->parms.iph.daddr == 0)
goto out;
if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
goto out;
if (jiffies - t->err_time < IPTUNNEL_ERR_TIMEO)
t->err_count++;
else
t->err_count = 1;
t->err_time = jiffies;
out:
read_unlock(&ipip_lock);
return;
#else
struct iphdr *iph = (struct iphdr*)dp;
int hlen = iph->ihl<<2;
struct iphdr *eiph;
int type = skb->h.icmph->type;
int code = skb->h.icmph->code;
int rel_type = 0;
int rel_code = 0;
int rel_info = 0;
struct sk_buff *skb2;
struct rtable *rt;
if (len < hlen + sizeof(struct iphdr))
return;
eiph = (struct iphdr*)(dp + hlen);
switch (type) {
default:
return;
case ICMP_PARAMETERPROB:
if (skb->h.icmph->un.gateway < hlen)
return;
/* So... This guy found something strange INSIDE encapsulated
packet. Well, he is fool, but what can we do ?
*/
rel_type = ICMP_PARAMETERPROB;
rel_info = skb->h.icmph->un.gateway - hlen;
break;
case ICMP_DEST_UNREACH:
switch (code) {
case ICMP_SR_FAILED:
case ICMP_PORT_UNREACH:
/* Impossible event. */
return;
case ICMP_FRAG_NEEDED:
/* And it is the only really necesary thing :-) */
rel_info = ntohs(skb->h.icmph->un.frag.mtu);
if (rel_info < hlen+68)
return;
rel_info -= hlen;
/* BSD 4.2 MORE DOES NOT EXIST IN NATURE. */
if (rel_info > ntohs(eiph->tot_len))
return;
break;
default:
/* All others are translated to HOST_UNREACH.
rfc2003 contains "deep thoughts" about NET_UNREACH,
I believe, it is just ether pollution. --ANK
*/
rel_type = ICMP_DEST_UNREACH;
rel_code = ICMP_HOST_UNREACH;
break;
}
break;
case ICMP_TIME_EXCEEDED:
if (code != ICMP_EXC_TTL)
return;
break;
}
/* Prepare fake skb to feed it to icmp_send */
skb2 = skb_clone(skb, GFP_ATOMIC);
if (skb2 == NULL)
return;
dst_release(skb2->dst);
skb2->dst = NULL;
skb_pull(skb2, skb->data - (u8*)eiph);
skb2->nh.raw = skb2->data;
/* Try to guess incoming interface */
if (ip_route_output(&rt, eiph->saddr, 0, RT_TOS(eiph->tos), 0)) {
kfree_skb(skb2);
return;
}
skb2->dev = rt->u.dst.dev;
/* route "incoming" packet */
if (rt->rt_flags&RTCF_LOCAL) {
ip_rt_put(rt);
rt = NULL;
if (ip_route_output(&rt, eiph->daddr, eiph->saddr, eiph->tos, 0) ||
rt->u.dst.dev->type != ARPHRD_IPGRE) {
ip_rt_put(rt);
kfree_skb(skb2);
return;
}
} else {
ip_rt_put(rt);
if (ip_route_input(skb2, eiph->daddr, eiph->saddr, eiph->tos, skb2->dev) ||
skb2->dst->dev->type != ARPHRD_IPGRE) {
kfree_skb(skb2);
return;
}
}
/* change mtu on this route */
if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
if (rel_info > skb2->dst->pmtu) {
kfree_skb(skb2);
return;
}
skb2->dst->pmtu = rel_info;
rel_info = htonl(rel_info);
} else if (type == ICMP_TIME_EXCEEDED) {
struct ip_tunnel *t = (struct ip_tunnel*)skb2->dev->priv;
if (t->parms.iph.ttl) {
rel_type = ICMP_DEST_UNREACH;
rel_code = ICMP_HOST_UNREACH;
}
}
icmp_send(skb2, rel_type, rel_code, rel_info);
kfree_skb(skb2);
return;
#endif
}
static int try_to_freeze_tasks(bool user_only)
{
struct task_struct *g, *p;
unsigned long end_time;
unsigned int todo;
bool wq_busy = false;
struct timeval start, end;
u64 elapsed_msecs64;
unsigned int elapsed_msecs;
bool wakeup = false;
int sleep_usecs = USEC_PER_MSEC;
do_gettimeofday(&start);
end_time = jiffies + TIMEOUT;
if (!user_only)
freeze_workqueues_begin();
while (true) {
todo = 0;
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (p == current || !freeze_task(p))
continue;
/*
* Now that we've done set_freeze_flag, don't
* perturb a task in TASK_STOPPED or TASK_TRACED.
* It is "frozen enough". If the task does wake
* up, it will immediately call try_to_freeze.
*
* Because freeze_task() goes through p's scheduler lock, it's
* guaranteed that TASK_STOPPED/TRACED -> TASK_RUNNING
* transition can't race with task state testing here.
*/
if (!task_is_stopped_or_traced(p) &&
!freezer_should_skip(p))
todo++;
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
if (!user_only) {
wq_busy = freeze_workqueues_busy();
todo += wq_busy;
}
if (!todo || time_after(jiffies, end_time))
break;
if (pm_wakeup_pending()) {
wakeup = true;
break;
}
/*
* We need to retry, but first give the freezing tasks some
* time to enter the refrigerator. Start with an initial
* 1 ms sleep followed by exponential backoff until 8 ms.
*/
usleep_range(sleep_usecs / 2, sleep_usecs);
if (sleep_usecs < 8 * USEC_PER_MSEC)
sleep_usecs *= 2;
}
do_gettimeofday(&end);
elapsed_msecs64 = timeval_to_ns(&end) - timeval_to_ns(&start);
do_div(elapsed_msecs64, NSEC_PER_MSEC);
elapsed_msecs = elapsed_msecs64;
if (todo) {
/* This does not unfreeze processes that are already frozen
* (we have slightly ugly calling convention in that respect,
* and caller must call thaw_processes() if something fails),
* but it cleans up leftover PF_FREEZE requests.
*/
if(wakeup) {
printk("\n");
printk(KERN_ERR "Freezing of %s aborted\n",
user_only ? "user space " : "tasks ");
}
else {
printk("\n");
printk(KERN_ERR "Freezing of tasks %s after %d.%03d seconds "
"(%d tasks refusing to freeze, wq_busy=%d):\n",
wakeup ? "aborted" : "failed",
elapsed_msecs / 1000, elapsed_msecs % 1000,
todo - wq_busy, wq_busy);
}
if (!wakeup) {
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (p != current && !freezer_should_skip(p)
&& freezing(p) && !frozen(p) &&
elapsed_msecs > 1000)
sched_show_task(p);
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
}
} else {
static int ipip6_err(struct sk_buff *skb, u32 info)
{
/* All the routers (except for Linux) return only
8 bytes of packet payload. It means, that precise relaying of
ICMP in the real Internet is absolutely infeasible.
*/
struct iphdr *iph = (struct iphdr*)skb->data;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct ip_tunnel *t;
int err;
switch (type) {
default:
case ICMP_PARAMETERPROB:
return 0;
case ICMP_DEST_UNREACH:
switch (code) {
case ICMP_SR_FAILED:
case ICMP_PORT_UNREACH:
/* Impossible event. */
return 0;
case ICMP_FRAG_NEEDED:
/* Soft state for pmtu is maintained by IP core. */
return 0;
default:
/* All others are translated to HOST_UNREACH.
rfc2003 contains "deep thoughts" about NET_UNREACH,
I believe they are just ether pollution. --ANK
*/
break;
}
break;
case ICMP_TIME_EXCEEDED:
if (code != ICMP_EXC_TTL)
return 0;
break;
}
err = -ENOENT;
read_lock(&ipip6_lock);
t = ipip6_tunnel_lookup(dev_net(skb->dev), iph->daddr, iph->saddr);
if (t == NULL || t->parms.iph.daddr == 0)
goto out;
err = 0;
if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
goto out;
if (jiffies - t->err_time < IPTUNNEL_ERR_TIMEO)
t->err_count++;
else
t->err_count = 1;
t->err_time = jiffies;
out:
read_unlock(&ipip6_lock);
return err;
}
/*
* Weighted Least Connection scheduling
*/
static int
tcp_vs_wlc_schedule(struct tcp_vs_conn *conn, struct tcp_vs_service *svc)
{
register struct list_head *l, *e;
tcp_vs_dest_t *dest, *least;
TCP_VS_DBG(5, "tcp_vs_wlc_schedule(): Scheduling...\n");
/*
* We use the following formula to estimate the overhead:
* dest->conns / dest->weight
*
* Remember -- no floats in kernel mode!!!
* The comparison of h1*w2 > h2*w1 is equivalent to that of
* h1/w1 > h2/w2
* when each weight is larger than zero.
*
* The server with weight=0 is quiesced and will not receive any
* new connection.
*/
read_lock(&svc->lock);
l = &svc->destinations;
for (e = l->next; e != l; e = e->next) {
least = list_entry(e, tcp_vs_dest_t, n_list);
if (least->weight > 0) {
goto nextstage;
}
}
read_unlock(&svc->lock);
return -1;
/*
* Find the destination with the least load.
*/
nextstage:
for (e = e->next; e != l; e = e->next) {
dest = list_entry(e, tcp_vs_dest_t, n_list);
if (atomic_read(&least->conns) * dest->weight
> atomic_read(&dest->conns) * least->weight) {
least = dest;
}
}
read_unlock(&svc->lock);
TCP_VS_DBG(5, "WLC: server %d.%d.%d.%d:%d "
"conns %d refcnt %d weight %d\n",
NIPQUAD(least->addr), ntohs(least->port),
atomic_read(&least->conns),
atomic_read(&least->refcnt), least->weight);
conn->dsock = tcp_vs_connect2dest(least);
if (!conn->dsock) {
TCP_VS_ERR_RL("The destination is not available\n");
return -1;
}
atomic_inc(&least->conns);
conn->dest = least;
return 0;
}
static int lowmem_shrink(struct shrinker *s, struct shrink_control *sc)
{
struct task_struct *tsk;
#ifdef ENHANCED_LMK_ROUTINE
struct task_struct *selected[LOWMEM_DEATHPENDING_DEPTH] = {NULL,};
#else
struct task_struct *selected = NULL;
#endif
int rem = 0;
int tasksize;
int i;
int min_score_adj = OOM_SCORE_ADJ_MAX + 1;
#ifdef ENHANCED_LMK_ROUTINE
int selected_tasksize[LOWMEM_DEATHPENDING_DEPTH] = {0,};
int selected_oom_score_adj[LOWMEM_DEATHPENDING_DEPTH] = {OOM_ADJUST_MAX,};
int all_selected_oom = 0;
int max_selected_oom_idx = 0;
#else
int selected_tasksize = 0;
int selected_oom_score_adj;
#endif
int array_size = ARRAY_SIZE(lowmem_adj);
int other_free = global_page_state(NR_FREE_PAGES);
int other_file = global_page_state(NR_FILE_PAGES) - global_page_state(NR_SHMEM);
if (lowmem_adj_size < array_size)
array_size = lowmem_adj_size;
if (lowmem_minfree_size < array_size)
array_size = lowmem_minfree_size;
for (i = 0; i < array_size; i++) {
if (other_free < lowmem_minfree[i] &&
other_file < lowmem_minfree[i]) {
min_score_adj = lowmem_adj[i];
break;
}
}
if (sc->nr_to_scan > 0)
lowmem_print(3, "lowmem_shrink %lu, %x, ofree %d %d, ma %d\n",
sc->nr_to_scan, sc->gfp_mask, other_free,
other_file, min_score_adj);
rem = global_page_state(NR_ACTIVE_ANON) +
global_page_state(NR_ACTIVE_FILE) +
global_page_state(NR_INACTIVE_ANON) +
global_page_state(NR_INACTIVE_FILE);
if (sc->nr_to_scan <= 0 || min_score_adj == OOM_SCORE_ADJ_MAX + 1) {
lowmem_print(5, "lowmem_shrink %lu, %x, return %d\n",
sc->nr_to_scan, sc->gfp_mask, rem);
return rem;
}
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++)
selected_oom_score_adj[i] = min_score_adj;
#else
selected_oom_score_adj = min_score_adj;
#endif
read_lock(&tasklist_lock);
for_each_process(tsk) {
struct task_struct *p;
int oom_score_adj;
#ifdef ENHANCED_LMK_ROUTINE
int is_exist_oom_task = 0;
#endif
if (tsk->flags & PF_KTHREAD)
continue;
p = find_lock_task_mm(tsk);
if (!p)
continue;
if (test_tsk_thread_flag(p, TIF_MEMDIE) &&
time_before_eq(jiffies, lowmem_deathpending_timeout)) {
task_unlock(p);
read_unlock(&tasklist_lock);
return 0;
}
oom_score_adj = p->signal->oom_score_adj;
if (oom_score_adj < min_score_adj) {
task_unlock(p);
continue;
}
tasksize = get_mm_rss(p->mm);
task_unlock(p);
if (tasksize <= 0)
continue;
#ifdef ENHANCED_LMK_ROUTINE
if (all_selected_oom < LOWMEM_DEATHPENDING_DEPTH) {
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (!selected[i]) {
is_exist_oom_task = 1;
max_selected_oom_idx = i;
break;
}
}
} else if (selected_oom_score_adj[max_selected_oom_idx] < oom_score_adj ||
(selected_oom_score_adj[max_selected_oom_idx] == oom_score_adj &&
selected_tasksize[max_selected_oom_idx] < tasksize)) {
is_exist_oom_task = 1;
}
if (is_exist_oom_task) {
selected[max_selected_oom_idx] = p;
selected_tasksize[max_selected_oom_idx] = tasksize;
selected_oom_score_adj[max_selected_oom_idx] = oom_score_adj;
if (all_selected_oom < LOWMEM_DEATHPENDING_DEPTH)
all_selected_oom++;
if (all_selected_oom == LOWMEM_DEATHPENDING_DEPTH) {
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (selected_oom_score_adj[i] < selected_oom_score_adj[max_selected_oom_idx])
max_selected_oom_idx = i;
else if (selected_oom_score_adj[i] == selected_oom_score_adj[max_selected_oom_idx] &&
selected_tasksize[i] < selected_tasksize[max_selected_oom_idx])
max_selected_oom_idx = i;
}
}
lowmem_print(2, "select %d (%s), adj %d, \
size %d, to kill\n",
p->pid, p->comm, oom_score_adj, tasksize);
}
#else
if (selected) {
if (oom_score_adj < selected_oom_score_adj)
continue;
if (oom_score_adj == selected_oom_score_adj &&
tasksize <= selected_tasksize)
continue;
}
selected = p;
selected_tasksize = tasksize;
selected_oom_score_adj = oom_score_adj;
lowmem_print(2, "select %d (%s), adj %d, size %d, to kill\n",
p->pid, p->comm, oom_score_adj, tasksize);
#endif
}
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (selected[i]) {
lowmem_print(1, "send sigkill to %d (%s), adj %d,\
size %d\n",
selected[i]->pid, selected[i]->comm,
selected_oom_score_adj[i],
selected_tasksize[i]);
lowmem_deathpending_timeout = jiffies + HZ;
send_sig(SIGKILL, selected[i], 0);
set_tsk_thread_flag(selected[i], TIF_MEMDIE);
rem -= selected_tasksize[i];
#ifdef LMK_COUNT_READ
lmk_count++;
#endif
}
}
#else
if (selected) {
lowmem_print(1, "send sigkill to %d (%s), adj %d, size %d\n",
selected->pid, selected->comm,
selected_oom_score_adj, selected_tasksize);
lowmem_deathpending_timeout = jiffies + HZ;
send_sig(SIGKILL, selected, 0);
set_tsk_thread_flag(selected, TIF_MEMDIE);
rem -= selected_tasksize;
#ifdef LMK_COUNT_READ
lmk_count++;
#endif
}
#endif
lowmem_print(4, "lowmem_shrink %lu, %x, return %d\n",
sc->nr_to_scan, sc->gfp_mask, rem);
read_unlock(&tasklist_lock);
return rem;
}
static int ipip6_tunnel_get_prl(struct ip_tunnel *t,
struct ip_tunnel_prl __user *a)
{
struct ip_tunnel_prl kprl, *kp;
struct ip_tunnel_prl_entry *prl;
unsigned int cmax, c = 0, ca, len;
int ret = 0;
if (copy_from_user(&kprl, a, sizeof(kprl)))
return -EFAULT;
cmax = kprl.datalen / sizeof(kprl);
if (cmax > 1 && kprl.addr != htonl(INADDR_ANY))
cmax = 1;
/* For simple GET or for root users,
* we try harder to allocate.
*/
kp = (cmax <= 1 || capable(CAP_NET_ADMIN)) ?
kcalloc(cmax, sizeof(*kp), GFP_KERNEL) :
NULL;
read_lock(&ipip6_lock);
ca = t->prl_count < cmax ? t->prl_count : cmax;
if (!kp) {
/* We don't try hard to allocate much memory for
* non-root users.
* For root users, retry allocating enough memory for
* the answer.
*/
kp = kcalloc(ca, sizeof(*kp), GFP_ATOMIC);
if (!kp) {
ret = -ENOMEM;
goto out;
}
}
c = 0;
for (prl = t->prl; prl; prl = prl->next) {
if (c > cmax)
break;
if (kprl.addr != htonl(INADDR_ANY) && prl->addr != kprl.addr)
continue;
kp[c].addr = prl->addr;
kp[c].flags = prl->flags;
c++;
if (kprl.addr != htonl(INADDR_ANY))
break;
}
out:
read_unlock(&ipip6_lock);
len = sizeof(*kp) * c;
ret = 0;
if ((len && copy_to_user(a + 1, kp, len)) || put_user(len, &a->datalen))
ret = -EFAULT;
kfree(kp);
return ret;
}
long sys_ptrace(long request, long pid, long addr, long data)
{
struct task_struct *child;
int i, ret;
lock_kernel();
ret = -EPERM;
if (request == PTRACE_TRACEME) {
/* are we already being traced? */
if (current->ptrace & PT_PTRACED)
goto out;
ret = security_ptrace(current->parent, current);
if (ret)
goto out;
/* set the ptrace bit in the process flags. */
current->ptrace |= PT_PTRACED;
ret = 0;
goto out;
}
ret = -ESRCH;
read_lock(&tasklist_lock);
child = find_task_by_pid(pid);
if (child)
get_task_struct(child);
read_unlock(&tasklist_lock);
if (!child)
goto out;
ret = -EPERM;
if (pid == 1) /* you may not mess with init */
goto out_tsk;
if (request == PTRACE_ATTACH) {
ret = ptrace_attach(child);
goto out_tsk;
}
#ifdef SUBACH_PTRACE_SPECIAL
SUBARCH_PTRACE_SPECIAL(child,request,addr,data);
#endif
ret = ptrace_check_attach(child, request == PTRACE_KILL);
if (ret < 0)
goto out_tsk;
switch (request) {
/* when I and D space are separate, these will need to be fixed. */
case PTRACE_PEEKTEXT: /* read word at location addr. */
case PTRACE_PEEKDATA: {
unsigned long tmp;
int copied;
ret = -EIO;
copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
if (copied != sizeof(tmp))
break;
ret = put_user(tmp, (unsigned long __user *) data);
break;
}
/* read the word at location addr in the USER area. */
case PTRACE_PEEKUSR:
ret = peek_user(child, addr, data);
break;
/* when I and D space are separate, this will have to be fixed. */
case PTRACE_POKETEXT: /* write the word at location addr. */
case PTRACE_POKEDATA:
ret = -EIO;
if (access_process_vm(child, addr, &data, sizeof(data),
1) != sizeof(data))
break;
ret = 0;
break;
case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
ret = poke_user(child, addr, data);
break;
case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */
case PTRACE_CONT: { /* restart after signal. */
ret = -EIO;
if (!valid_signal(data))
break;
set_singlestepping(child, 0);
if (request == PTRACE_SYSCALL) {
set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
}
else {
clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
}
child->exit_code = data;
wake_up_process(child);
ret = 0;
break;
}
/*
* make the child exit. Best I can do is send it a sigkill.
* perhaps it should be put in the status that it wants to
* exit.
*/
case PTRACE_KILL: {
ret = 0;
if (child->exit_state == EXIT_ZOMBIE) /* already dead */
break;
set_singlestepping(child, 0);
child->exit_code = SIGKILL;
wake_up_process(child);
break;
}
case PTRACE_SINGLESTEP: { /* set the trap flag. */
ret = -EIO;
if (!valid_signal(data))
break;
clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
set_singlestepping(child, 1);
child->exit_code = data;
/* give it a chance to run. */
wake_up_process(child);
ret = 0;
break;
}
case PTRACE_DETACH:
/* detach a process that was attached. */
ret = ptrace_detach(child, data);
break;
#ifdef PTRACE_GETREGS
case PTRACE_GETREGS: { /* Get all gp regs from the child. */
if (!access_ok(VERIFY_WRITE, (unsigned long *)data,
MAX_REG_OFFSET)) {
ret = -EIO;
break;
}
for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) {
__put_user(getreg(child, i),
(unsigned long __user *) data);
data += sizeof(long);
}
ret = 0;
break;
}
#endif
#ifdef PTRACE_SETREGS
case PTRACE_SETREGS: { /* Set all gp regs in the child. */
unsigned long tmp = 0;
if (!access_ok(VERIFY_READ, (unsigned *)data,
MAX_REG_OFFSET)) {
ret = -EIO;
break;
}
for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) {
__get_user(tmp, (unsigned long __user *) data);
putreg(child, i, tmp);
data += sizeof(long);
}
ret = 0;
break;
}
#endif
#ifdef PTRACE_GETFPREGS
case PTRACE_GETFPREGS: /* Get the child FPU state. */
ret = get_fpregs(data, child);
break;
#endif
#ifdef PTRACE_SETFPREGS
case PTRACE_SETFPREGS: /* Set the child FPU state. */
ret = set_fpregs(data, child);
break;
#endif
#ifdef PTRACE_GETFPXREGS
case PTRACE_GETFPXREGS: /* Get the child FPU state. */
ret = get_fpxregs(data, child);
break;
#endif
#ifdef PTRACE_SETFPXREGS
case PTRACE_SETFPXREGS: /* Set the child FPU state. */
ret = set_fpxregs(data, child);
break;
#endif
case PTRACE_FAULTINFO: {
/* Take the info from thread->arch->faultinfo,
* but transfer max. sizeof(struct ptrace_faultinfo).
* On i386, ptrace_faultinfo is smaller!
*/
ret = copy_to_user((unsigned long __user *) data,
&child->thread.arch.faultinfo,
sizeof(struct ptrace_faultinfo));
if(ret)
break;
break;
}
#ifdef PTRACE_LDT
case PTRACE_LDT: {
struct ptrace_ldt ldt;
if(copy_from_user(&ldt, (unsigned long __user *) data,
sizeof(ldt))) {
ret = -EIO;
break;
}
/* This one is confusing, so just punt and return -EIO for
* now
*/
ret = -EIO;
break;
}
#endif
#ifdef CONFIG_PROC_MM
case PTRACE_SWITCH_MM: {
struct mm_struct *old = child->mm;
struct mm_struct *new = proc_mm_get_mm(data);
if(IS_ERR(new)) {
ret = PTR_ERR(new);
break;
}
atomic_inc(&new->mm_users);
child->mm = new;
child->active_mm = new;
mmput(old);
ret = 0;
break;
}
#endif
default:
ret = ptrace_request(child, request, addr, data);
break;
}
out_tsk:
put_task_struct(child);
out:
unlock_kernel();
return ret;
}
static int bfusb_send_frame(struct hci_dev *hdev, struct sk_buff *skb)
{
struct bfusb_data *data = hci_get_drvdata(hdev);
struct sk_buff *nskb;
unsigned char buf[3];
int sent = 0, size, count;
BT_DBG("hdev %p skb %p type %d len %d", hdev, skb, bt_cb(skb)->pkt_type, skb->len);
if (!test_bit(HCI_RUNNING, &hdev->flags))
return -EBUSY;
switch (bt_cb(skb)->pkt_type) {
case HCI_COMMAND_PKT:
hdev->stat.cmd_tx++;
break;
case HCI_ACLDATA_PKT:
hdev->stat.acl_tx++;
break;
case HCI_SCODATA_PKT:
hdev->stat.sco_tx++;
break;
};
/* Prepend skb with frame type */
memcpy(skb_push(skb, 1), &bt_cb(skb)->pkt_type, 1);
count = skb->len;
/* Max HCI frame size seems to be 1511 + 1 */
nskb = bt_skb_alloc(count + 32, GFP_ATOMIC);
if (!nskb) {
BT_ERR("Can't allocate memory for new packet");
return -ENOMEM;
}
nskb->dev = (void *) data;
while (count) {
size = min_t(uint, count, BFUSB_MAX_BLOCK_SIZE);
buf[0] = 0xc1 | ((sent == 0) ? 0x04 : 0) | ((count == size) ? 0x08 : 0);
buf[1] = 0x00;
buf[2] = (size == BFUSB_MAX_BLOCK_SIZE) ? 0 : size;
memcpy(skb_put(nskb, 3), buf, 3);
skb_copy_from_linear_data_offset(skb, sent, skb_put(nskb, size), size);
sent += size;
count -= size;
}
/* Don't send frame with multiple size of bulk max packet */
if ((nskb->len % data->bulk_pkt_size) == 0) {
buf[0] = 0xdd;
buf[1] = 0x00;
memcpy(skb_put(nskb, 2), buf, 2);
}
read_lock(&data->lock);
skb_queue_tail(&data->transmit_q, nskb);
bfusb_tx_wakeup(data);
read_unlock(&data->lock);
kfree_skb(skb);
return 0;
}
static void bfusb_rx_complete(struct urb *urb)
{
struct sk_buff *skb = (struct sk_buff *) urb->context;
struct bfusb_data *data = (struct bfusb_data *) skb->dev;
unsigned char *buf = urb->transfer_buffer;
int count = urb->actual_length;
int err, hdr, len;
BT_DBG("bfusb %p urb %p skb %p len %d", data, urb, skb, skb->len);
read_lock(&data->lock);
if (!test_bit(HCI_RUNNING, &data->hdev->flags))
goto unlock;
if (urb->status || !count)
goto resubmit;
data->hdev->stat.byte_rx += count;
skb_put(skb, count);
while (count) {
hdr = buf[0] | (buf[1] << 8);
if (hdr & 0x4000) {
len = 0;
count -= 2;
buf += 2;
} else {
len = (buf[2] == 0) ? 256 : buf[2];
count -= 3;
buf += 3;
}
if (count < len) {
BT_ERR("%s block extends over URB buffer ranges",
data->hdev->name);
}
if ((hdr & 0xe1) == 0xc1)
bfusb_recv_block(data, hdr, buf, len);
count -= len;
buf += len;
}
skb_unlink(skb, &data->pending_q);
kfree_skb(skb);
bfusb_rx_submit(data, urb);
read_unlock(&data->lock);
return;
resubmit:
urb->dev = data->udev;
err = usb_submit_urb(urb, GFP_ATOMIC);
if (err) {
BT_ERR("%s bulk resubmit failed urb %p err %d",
data->hdev->name, urb, err);
}
unlock:
read_unlock(&data->lock);
}
int timod_getmsg(unsigned int fd, char *ctl_buf, int ctl_maxlen, s32 *ctl_len,
char *data_buf, int data_maxlen, s32 *data_len, int *flags_p)
{
int error;
int oldflags;
struct file *filp;
struct inode *ino;
struct sol_socket_struct *sock;
struct T_unitdata_ind udi;
mm_segment_t old_fs = get_fs();
long args[6];
char *tmpbuf;
int tmplen;
int (*sys_socketcall)(int, unsigned long *) =
(int (*)(int, unsigned long *))SYS(socketcall);
int (*sys_recvfrom)(int, void *, size_t, unsigned, struct sockaddr *, int *);
SOLD("entry");
SOLDD(("%u %p %d %p %p %d %p %d\n", fd, ctl_buf, ctl_maxlen, ctl_len, data_buf, data_maxlen, data_len, *flags_p));
read_lock(¤t->files->file_lock);
filp = fcheck(fd);
read_unlock(¤t->files->file_lock);
if (!filp)
return -EBADF;
ino = filp->f_dentry->d_inode;
sock = (struct sol_socket_struct *)filp->private_data;
SOLDD(("%p %p\n", sock->pfirst, sock->pfirst ? sock->pfirst->next : NULL));
if ( ctl_maxlen > 0 && !sock->pfirst && ino->u.socket_i.type == SOCK_STREAM
&& sock->state == TS_IDLE) {
SOLD("calling LISTEN");
args[0] = fd;
args[1] = -1;
set_fs(KERNEL_DS);
sys_socketcall(SYS_LISTEN, args);
set_fs(old_fs);
SOLD("LISTEN done");
}
if (!(filp->f_flags & O_NONBLOCK)) {
poll_table wait_table, *wait;
poll_initwait(&wait_table);
wait = &wait_table;
for(;;) {
SOLD("loop");
set_current_state(TASK_INTERRUPTIBLE);
/* ! ( l<0 || ( l>=0 && ( ! pfirst || (flags == HIPRI && pri != HIPRI) ) ) ) */
/* ( ! l<0 && ! ( l>=0 && ( ! pfirst || (flags == HIPRI && pri != HIPRI) ) ) ) */
/* ( l>=0 && ( ! l>=0 || ! ( ! pfirst || (flags == HIPRI && pri != HIPRI) ) ) ) */
/* ( l>=0 && ( l<0 || ( pfirst && ! (flags == HIPRI && pri != HIPRI) ) ) ) */
/* ( l>=0 && ( l<0 || ( pfirst && (flags != HIPRI || pri == HIPRI) ) ) ) */
/* ( l>=0 && ( pfirst && (flags != HIPRI || pri == HIPRI) ) ) */
if (ctl_maxlen >= 0 && sock->pfirst && (*flags_p != MSG_HIPRI || sock->pfirst->pri == MSG_HIPRI))
break;
SOLD("cond 1 passed");
if (
#if 1
*flags_p != MSG_HIPRI &&
#endif
((filp->f_op->poll(filp, wait) & POLLIN) ||
(filp->f_op->poll(filp, NULL) & POLLIN) ||
signal_pending(current))
) {
break;
}
if( *flags_p == MSG_HIPRI ) {
SOLD("avoiding lockup");
break ;
}
if(wait_table.error) {
SOLD("wait-table error");
poll_freewait(&wait_table);
return wait_table.error;
}
SOLD("scheduling");
schedule();
}
SOLD("loop done");
current->state = TASK_RUNNING;
poll_freewait(&wait_table);
if (signal_pending(current)) {
SOLD("signal pending");
return -EINTR;
}
}
if (ctl_maxlen >= 0 && sock->pfirst) {
struct T_primsg *it = sock->pfirst;
int l = min_t(int, ctl_maxlen, it->length);
SCHECK_MAGIC((char*)((u64)(((char *)&it->type)+sock->offset+it->length+7)&~7),MKCTL_MAGIC);
SOLD("purting ctl data");
if(copy_to_user(ctl_buf,
(char*)&it->type + sock->offset, l))
return -EFAULT;
SOLD("pur it");
if(put_user(l, ctl_len))
return -EFAULT;
SOLD("set ctl_len");
*flags_p = it->pri;
it->length -= l;
if (it->length) {
SOLD("more ctl");
sock->offset += l;
return MORECTL;
} else {
SOLD("removing message");
sock->pfirst = it->next;
if (!sock->pfirst)
sock->plast = NULL;
SOLDD(("getmsg kfree %016lx->%016lx\n", it, sock->pfirst));
mykfree(it);
sock->offset = 0;
SOLD("ctl done");
return 0;
}
}
*flags_p = 0;
if (ctl_maxlen >= 0) {
SOLD("ACCEPT perhaps?");
if (ino->u.socket_i.type == SOCK_STREAM && sock->state == TS_IDLE) {
struct T_conn_ind ind;
char *buf = getpage();
int len = BUF_SIZE;
SOLD("trying ACCEPT");
if (put_user(ctl_maxlen - sizeof(ind), ctl_len))
return -EFAULT;
args[0] = fd;
args[1] = (long)buf;
args[2] = (long)&len;
oldflags = filp->f_flags;
filp->f_flags |= O_NONBLOCK;
SOLD("calling ACCEPT");
set_fs(KERNEL_DS);
error = sys_socketcall(SYS_ACCEPT, args);
set_fs(old_fs);
filp->f_flags = oldflags;
if (error < 0) {
SOLD("some error");
putpage(buf);
return error;
}
if (error) {
SOLD("connect");
putpage(buf);
if (sizeof(ind) > ctl_maxlen) {
SOLD("generating CONN_IND");
ind.PRIM_type = T_CONN_IND;
ind.SRC_length = len;
ind.SRC_offset = sizeof(ind);
ind.OPT_length = ind.OPT_offset = 0;
ind.SEQ_number = error;
if(copy_to_user(ctl_buf, &ind, sizeof(ind))||
put_user(sizeof(ind)+ind.SRC_length,ctl_len))
return -EFAULT;
SOLD("CONN_IND created");
}
if (data_maxlen >= 0)
put_user(0, data_len);
SOLD("CONN_IND done");
return 0;
}
if (len>ctl_maxlen) {
SOLD("data don't fit");
putpage(buf);
return -EFAULT; /* XXX - is this ok ? */
}
if(copy_to_user(ctl_buf,buf,len) || put_user(len,ctl_len)){
SOLD("can't copy data");
putpage(buf);
return -EFAULT;
}
SOLD("ACCEPT done");
putpage(buf);
}
}
SOLD("checking data req");
if (data_maxlen <= 0) {
if (data_maxlen == 0)
put_user(0, data_len);
if (ctl_maxlen >= 0)
put_user(0, ctl_len);
return -EAGAIN;
}
SOLD("wants data");
if (ctl_maxlen > sizeof(udi) && sock->state == TS_IDLE) {
SOLD("udi fits");
tmpbuf = ctl_buf + sizeof(udi);
tmplen = ctl_maxlen - sizeof(udi);
} else {
SOLD("udi does not fit");
tmpbuf = NULL;
tmplen = 0;
}
if (put_user(tmplen, ctl_len))
return -EFAULT;
SOLD("set ctl_len");
oldflags = filp->f_flags;
filp->f_flags |= O_NONBLOCK;
SOLD("calling recvfrom");
sys_recvfrom = (int (*)(int, void *, size_t, unsigned, struct sockaddr *, int *))SYS(recvfrom);
error = sys_recvfrom(fd, data_buf, data_maxlen, 0, (struct sockaddr*)tmpbuf, ctl_len);
filp->f_flags = oldflags;
if (error < 0)
return error;
SOLD("error >= 0" ) ;
if (error && ctl_maxlen > sizeof(udi) && sock->state == TS_IDLE) {
SOLD("generating udi");
udi.PRIM_type = T_UNITDATA_IND;
get_user(udi.SRC_length, ctl_len);
udi.SRC_offset = sizeof(udi);
udi.OPT_length = udi.OPT_offset = 0;
copy_to_user(ctl_buf, &udi, sizeof(udi));
put_user(sizeof(udi)+udi.SRC_length, ctl_len);
SOLD("udi done");
} else
put_user(0, ctl_len);
put_user(error, data_len);
SOLD("done");
return 0;
}
static int try_to_freeze_tasks(bool user_only)
{
struct task_struct *g, *p;
unsigned long end_time;
unsigned int todo;
bool wq_busy = false;
struct timeval start, end;
u64 elapsed_msecs64;
unsigned int elapsed_msecs;
bool wakeup = false;
int sleep_usecs = USEC_PER_MSEC;
do_gettimeofday(&start);
end_time = jiffies + TIMEOUT;
if (!user_only)
freeze_workqueues_begin();
while (true) {
todo = 0;
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (p == current || !freeze_task(p))
continue;
if (!task_is_stopped_or_traced(p) &&
!freezer_should_skip(p))
todo++;
}
while_each_thread(g, p);
read_unlock(&tasklist_lock);
if (!user_only) {
wq_busy = freeze_workqueues_busy();
todo += wq_busy;
}
if (!todo || time_after(jiffies, end_time))
break;
if (pm_wakeup_pending()) {
wakeup = true;
break;
}
/*
* We need to retry, but first give the freezing tasks some
* time to enter the regrigerator.
*/
usleep_range(sleep_usecs / 2, sleep_usecs);
if (sleep_usecs < 8 * USEC_PER_MSEC)
sleep_usecs *= 2;
}
do_gettimeofday(&end);
elapsed_msecs64 = timeval_to_ns(&end) - timeval_to_ns(&start);
do_div(elapsed_msecs64, NSEC_PER_MSEC);
elapsed_msecs = elapsed_msecs64;
if (todo) {
if(wakeup) {
printk("\n");
printk(KERN_ERR "Freezing of %s aborted\n",
user_only ? "user space " : "tasks ");
}
else {
printk("\n");
printk(KERN_ERR "Freezing of tasks %s after %d.%03d seconds "
"(%d tasks refusing to freeze, wq_busy=%d):\n",
wakeup ? "aborted" : "failed",
elapsed_msecs / 1000, elapsed_msecs % 1000,
todo - wq_busy, wq_busy);
}
if (!wakeup) {
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (p != current && !freezer_should_skip(p)
&& freezing(p) && !frozen(p) &&
elapsed_msecs > 1000)
sched_show_task(p);
}
while_each_thread(g, p);
read_unlock(&tasklist_lock);
}
} else {
asmlinkage int solaris_getmsg(unsigned int fd, u32 arg1, u32 arg2, u32 arg3)
{
struct file *filp;
struct inode *ino;
struct strbuf *ctlptr, *datptr;
struct strbuf ctl, dat;
int *flgptr;
int flags;
int error = -EBADF;
SOLD("entry");
lock_kernel();
if(fd >= NR_OPEN) goto out;
read_lock(¤t->files->file_lock);
filp = fcheck(fd);
read_unlock(¤t->files->file_lock);
if(!filp) goto out;
ino = filp->f_dentry->d_inode;
if (!ino) goto out;
if (!ino->i_sock)
goto out;
ctlptr = (struct strbuf *)A(arg1);
datptr = (struct strbuf *)A(arg2);
flgptr = (int *)A(arg3);
error = -EFAULT;
if (ctlptr) {
if (copy_from_user(&ctl,ctlptr,sizeof(struct strbuf)) ||
put_user(-1,&ctlptr->len))
goto out;
} else
ctl.maxlen = -1;
if (datptr) {
if (copy_from_user(&dat,datptr,sizeof(struct strbuf)) ||
put_user(-1,&datptr->len))
goto out;
} else
dat.maxlen = -1;
if (get_user(flags,flgptr))
goto out;
switch (flags) {
case 0:
case MSG_HIPRI:
case MSG_ANY:
case MSG_BAND:
break;
default:
error = -EINVAL;
goto out;
}
error = timod_getmsg(fd,(char*)A(ctl.buf),ctl.maxlen,&ctlptr->len,
(char*)A(dat.buf),dat.maxlen,&datptr->len,&flags);
if (!error && put_user(flags,flgptr))
error = -EFAULT;
out:
unlock_kernel();
SOLD("done");
return error;
}
static int
ip6_tnl_err(struct sk_buff *skb, __u8 ipproto, struct inet6_skb_parm *opt,
u8 *type, u8 *code, int *msg, __u32 *info, int offset)
{
struct ipv6hdr *ipv6h = (struct ipv6hdr *) skb->data;
struct ip6_tnl *t;
int rel_msg = 0;
u8 rel_type = ICMPV6_DEST_UNREACH;
u8 rel_code = ICMPV6_ADDR_UNREACH;
__u32 rel_info = 0;
__u16 len;
int err = -ENOENT;
/* If the packet doesn't contain the original IPv6 header we are
in trouble since we might need the source address for further
processing of the error. */
read_lock(&ip6_tnl_lock);
if ((t = ip6_tnl_lookup(dev_net(skb->dev), &ipv6h->daddr,
&ipv6h->saddr)) == NULL)
goto out;
if (t->parms.proto != ipproto && t->parms.proto != 0)
goto out;
err = 0;
switch (*type) {
__u32 teli;
struct ipv6_tlv_tnl_enc_lim *tel;
__u32 mtu;
case ICMPV6_DEST_UNREACH:
if (net_ratelimit())
printk(KERN_WARNING
"%s: Path to destination invalid "
"or inactive!\n", t->parms.name);
rel_msg = 1;
break;
case ICMPV6_TIME_EXCEED:
if ((*code) == ICMPV6_EXC_HOPLIMIT) {
if (net_ratelimit())
printk(KERN_WARNING
"%s: Too small hop limit or "
"routing loop in tunnel!\n",
t->parms.name);
rel_msg = 1;
}
break;
case ICMPV6_PARAMPROB:
teli = 0;
if ((*code) == ICMPV6_HDR_FIELD)
teli = parse_tlv_tnl_enc_lim(skb, skb->data);
if (teli && teli == *info - 2) {
tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli];
if (tel->encap_limit == 0) {
if (net_ratelimit())
printk(KERN_WARNING
"%s: Too small encapsulation "
"limit or routing loop in "
"tunnel!\n", t->parms.name);
rel_msg = 1;
}
} else if (net_ratelimit()) {
printk(KERN_WARNING
"%s: Recipient unable to parse tunneled "
"packet!\n ", t->parms.name);
}
break;
case ICMPV6_PKT_TOOBIG:
mtu = *info - offset;
if (mtu < IPV6_MIN_MTU)
mtu = IPV6_MIN_MTU;
t->dev->mtu = mtu;
if ((len = sizeof (*ipv6h) + ntohs(ipv6h->payload_len)) > mtu) {
rel_type = ICMPV6_PKT_TOOBIG;
rel_code = 0;
rel_info = mtu;
rel_msg = 1;
}
break;
}
*type = rel_type;
*code = rel_code;
*info = rel_info;
*msg = rel_msg;
out:
read_unlock(&ip6_tnl_lock);
return err;
}
void raw_seq_stop(struct seq_file *seq, void *v)
{
struct raw_iter_state *state = raw_seq_private(seq);
read_unlock(&state->h->lock);
}
static int should_defrag_range(struct inode *inode, u64 start, u64 len,
int thresh, u64 *last_len, u64 *skip,
u64 *defrag_end)
{
struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
struct extent_map *em = NULL;
struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
int ret = 1;
if (thresh == 0)
thresh = 256 * 1024;
/*
* make sure that once we start defragging and extent, we keep on
* defragging it
*/
if (start < *defrag_end)
return 1;
*skip = 0;
/*
* hopefully we have this extent in the tree already, try without
* the full extent lock
*/
read_lock(&em_tree->lock);
em = lookup_extent_mapping(em_tree, start, len);
read_unlock(&em_tree->lock);
if (!em) {
/* get the big lock and read metadata off disk */
lock_extent(io_tree, start, start + len - 1, GFP_NOFS);
em = btrfs_get_extent(inode, NULL, 0, start, len, 0);
unlock_extent(io_tree, start, start + len - 1, GFP_NOFS);
if (IS_ERR(em))
return 0;
}
/* this will cover holes, and inline extents */
if (em->block_start >= EXTENT_MAP_LAST_BYTE)
ret = 0;
/*
* we hit a real extent, if it is big don't bother defragging it again
*/
if ((*last_len == 0 || *last_len >= thresh) && em->len >= thresh)
ret = 0;
/*
* last_len ends up being a counter of how many bytes we've defragged.
* every time we choose not to defrag an extent, we reset *last_len
* so that the next tiny extent will force a defrag.
*
* The end result of this is that tiny extents before a single big
* extent will force at least part of that big extent to be defragged.
*/
if (ret) {
*last_len += len;
*defrag_end = extent_map_end(em);
} else {
*last_len = 0;
*skip = extent_map_end(em);
*defrag_end = 0;
}
free_extent_map(em);
return ret;
}
static void ipip6_err(struct sk_buff *skb, u32 info)
{
#ifndef I_WISH_WORLD_WERE_PERFECT
/* It is not :-( All the routers (except for Linux) return only
8 bytes of packet payload. It means, that precise relaying of
ICMP in the real Internet is absolutely infeasible.
*/
struct iphdr *iph = (struct iphdr*)skb->data;
int type = skb->h.icmph->type;
int code = skb->h.icmph->code;
struct ip_tunnel *t;
switch (type) {
default:
case ICMP_PARAMETERPROB:
return;
case ICMP_DEST_UNREACH:
switch (code) {
case ICMP_SR_FAILED:
case ICMP_PORT_UNREACH:
/* Impossible event. */
return;
case ICMP_FRAG_NEEDED:
/* Soft state for pmtu is maintained by IP core. */
return;
default:
/* All others are translated to HOST_UNREACH.
rfc2003 contains "deep thoughts" about NET_UNREACH,
I believe they are just ether pollution. --ANK
*/
break;
}
break;
case ICMP_TIME_EXCEEDED:
if (code != ICMP_EXC_TTL)
return;
break;
}
read_lock(&ipip6_lock);
t = ipip6_tunnel_lookup(iph->daddr, iph->saddr);
if (t == NULL || t->parms.iph.daddr == 0)
goto out;
if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
goto out;
if (jiffies - t->err_time < IPTUNNEL_ERR_TIMEO)
t->err_count++;
else
t->err_count = 1;
t->err_time = jiffies;
out:
read_unlock(&ipip6_lock);
return;
#else
struct iphdr *iph = (struct iphdr*)dp;
int hlen = iph->ihl<<2;
struct ipv6hdr *iph6;
int type = skb->h.icmph->type;
int code = skb->h.icmph->code;
int rel_type = 0;
int rel_code = 0;
int rel_info = 0;
struct sk_buff *skb2;
struct rt6_info *rt6i;
if (len < hlen + sizeof(struct ipv6hdr))
return;
iph6 = (struct ipv6hdr*)(dp + hlen);
switch (type) {
default:
return;
case ICMP_PARAMETERPROB:
if (skb->h.icmph->un.gateway < hlen)
return;
/* So... This guy found something strange INSIDE encapsulated
packet. Well, he is fool, but what can we do ?
*/
rel_type = ICMPV6_PARAMPROB;
rel_info = skb->h.icmph->un.gateway - hlen;
break;
case ICMP_DEST_UNREACH:
switch (code) {
case ICMP_SR_FAILED:
case ICMP_PORT_UNREACH:
/* Impossible event. */
return;
case ICMP_FRAG_NEEDED:
/* Too complicated case ... */
return;
default:
/* All others are translated to HOST_UNREACH.
rfc2003 contains "deep thoughts" about NET_UNREACH,
I believe, it is just ether pollution. --ANK
*/
rel_type = ICMPV6_DEST_UNREACH;
rel_code = ICMPV6_ADDR_UNREACH;
break;
}
break;
case ICMP_TIME_EXCEEDED:
if (code != ICMP_EXC_TTL)
return;
rel_type = ICMPV6_TIME_EXCEED;
rel_code = ICMPV6_EXC_HOPLIMIT;
break;
}
/* Prepare fake skb to feed it to icmpv6_send */
skb2 = skb_clone(skb, GFP_ATOMIC);
if (skb2 == NULL)
return;
dst_release(skb2->dst);
skb2->dst = NULL;
skb_pull(skb2, skb->data - (u8*)iph6);
skb2->nh.raw = skb2->data;
/* Try to guess incoming interface */
rt6i = rt6_lookup(&iph6->saddr, NULL, NULL, 0);
if (rt6i && rt6i->rt6i_dev) {
skb2->dev = rt6i->rt6i_dev;
rt6i = rt6_lookup(&iph6->daddr, &iph6->saddr, NULL, 0);
if (rt6i && rt6i->rt6i_dev && rt6i->rt6i_dev->type == ARPHRD_SIT) {
struct ip_tunnel *t = netdev_priv(rt6i->rt6i_dev);
if (rel_type == ICMPV6_TIME_EXCEED && t->parms.iph.ttl) {
rel_type = ICMPV6_DEST_UNREACH;
rel_code = ICMPV6_ADDR_UNREACH;
}
icmpv6_send(skb2, rel_type, rel_code, rel_info, skb2->dev);
}
}
kfree_skb(skb2);
return;
#endif
}
static int lowmem_shrink(struct shrinker *s, struct shrink_control *sc)
{
struct task_struct *tsk;
#ifdef ENHANCED_LMK_ROUTINE
struct task_struct *selected[LOWMEM_DEATHPENDING_DEPTH] = {NULL,};
#else
struct task_struct *selected = NULL;
#endif
#ifdef CONFIG_SEC_DEBUG_LMK_MEMINFO
#ifdef CONFIG_SEC_DEBUG_LMK_MEMINFO_VERBOSE
static DEFINE_RATELIMIT_STATE(lmk_rs, DEFAULT_RATELIMIT_INTERVAL, 0);
#else
static DEFINE_RATELIMIT_STATE(lmk_rs, 6*DEFAULT_RATELIMIT_INTERVAL, 0);
#endif
#endif
int rem = 0;
int tasksize;
int i;
int min_score_adj = OOM_SCORE_ADJ_MAX + 1;
#ifdef ENHANCED_LMK_ROUTINE
int selected_tasksize[LOWMEM_DEATHPENDING_DEPTH] = {0,};
int selected_oom_score_adj[LOWMEM_DEATHPENDING_DEPTH] = {OOM_ADJUST_MAX,};
int all_selected_oom = 0;
int max_selected_oom_idx = 0;
#else
int selected_tasksize = 0;
int selected_oom_score_adj;
#endif
#ifdef CONFIG_SAMP_HOTNESS
int selected_hotness_adj = 0;
#endif
int array_size = ARRAY_SIZE(lowmem_adj);
#if (!defined(CONFIG_MACH_JF) \
&& !defined(CONFIG_SEC_PRODUCT_8960)\
)
unsigned long nr_to_scan = sc->nr_to_scan;
#endif
struct reclaim_state *reclaim_state = current->reclaim_state;
#ifndef CONFIG_CMA
int other_free = global_page_state(NR_FREE_PAGES);
#else
int other_free = global_page_state(NR_FREE_PAGES) -
global_page_state(NR_FREE_CMA_PAGES);
#endif
int other_file = global_page_state(NR_FILE_PAGES) - global_page_state(NR_SHMEM);
#if defined(CONFIG_RUNTIME_COMPCACHE) || defined(CONFIG_ZSWAP)
other_file -= total_swapcache_pages;
#endif /* CONFIG_RUNTIME_COMPCACHE || CONFIG_ZSWAP */
if (lowmem_adj_size < array_size)
array_size = lowmem_adj_size;
if (lowmem_minfree_size < array_size)
array_size = lowmem_minfree_size;
for (i = 0; i < array_size; i++) {
if (other_free < lowmem_minfree[i] &&
other_file < lowmem_minfree[i]) {
min_score_adj = lowmem_adj[i];
break;
}
}
if (sc->nr_to_scan > 0)
lowmem_print(3, "lowmem_shrink %lu, %x, ofree %d %d, ma %d\n",
sc->nr_to_scan, sc->gfp_mask, other_free,
other_file, min_score_adj);
rem = global_page_state(NR_ACTIVE_ANON) +
global_page_state(NR_ACTIVE_FILE) +
global_page_state(NR_INACTIVE_ANON) +
global_page_state(NR_INACTIVE_FILE);
if (sc->nr_to_scan <= 0 || min_score_adj == OOM_SCORE_ADJ_MAX + 1) {
lowmem_print(5, "lowmem_shrink %lu, %x, return %d\n",
sc->nr_to_scan, sc->gfp_mask, rem);
return rem;
}
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++)
selected_oom_score_adj[i] = min_score_adj;
#else
selected_oom_score_adj = min_score_adj;
#endif
read_lock(&tasklist_lock);
for_each_process(tsk) {
struct task_struct *p;
int oom_score_adj;
#ifdef ENHANCED_LMK_ROUTINE
int is_exist_oom_task = 0;
#endif
#ifdef CONFIG_SAMP_HOTNESS
int hotness_adj = 0;
#endif
if (tsk->flags & PF_KTHREAD)
continue;
p = find_lock_task_mm(tsk);
if (!p)
continue;
if (test_tsk_thread_flag(p, TIF_MEMDIE) &&
time_before_eq(jiffies, lowmem_deathpending_timeout)) {
task_unlock(p);
read_unlock(&tasklist_lock);
return 0;
}
oom_score_adj = p->signal->oom_score_adj;
if (oom_score_adj < min_score_adj) {
task_unlock(p);
continue;
}
tasksize = get_mm_rss(p->mm);
#if defined(CONFIG_ZSWAP)
if (atomic_read(&zswap_stored_pages)) {
lowmem_print(3, "shown tasksize : %d\n", tasksize);
tasksize += atomic_read(&zswap_pool_pages) * get_mm_counter(p->mm, MM_SWAPENTS)
/ atomic_read(&zswap_stored_pages);
lowmem_print(3, "real tasksize : %d\n", tasksize);
}
#endif
#ifdef CONFIG_SAMP_HOTNESS
hotness_adj = p->signal->hotness_adj;
#endif
task_unlock(p);
if (tasksize <= 0)
continue;
#ifdef ENHANCED_LMK_ROUTINE
if (all_selected_oom < LOWMEM_DEATHPENDING_DEPTH) {
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (!selected[i]) {
is_exist_oom_task = 1;
max_selected_oom_idx = i;
break;
}
}
} else if (selected_oom_score_adj[max_selected_oom_idx] < oom_score_adj ||
(selected_oom_score_adj[max_selected_oom_idx] == oom_score_adj &&
selected_tasksize[max_selected_oom_idx] < tasksize)) {
is_exist_oom_task = 1;
}
if (is_exist_oom_task) {
selected[max_selected_oom_idx] = p;
selected_tasksize[max_selected_oom_idx] = tasksize;
selected_oom_score_adj[max_selected_oom_idx] = oom_score_adj;
if (all_selected_oom < LOWMEM_DEATHPENDING_DEPTH)
all_selected_oom++;
if (all_selected_oom == LOWMEM_DEATHPENDING_DEPTH) {
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (selected_oom_score_adj[i] < selected_oom_score_adj[max_selected_oom_idx])
max_selected_oom_idx = i;
else if (selected_oom_score_adj[i] == selected_oom_score_adj[max_selected_oom_idx] &&
selected_tasksize[i] < selected_tasksize[max_selected_oom_idx])
max_selected_oom_idx = i;
}
}
lowmem_print(2, "select %d (%s), adj %d, \
size %d, to kill\n",
p->pid, p->comm, oom_score_adj, tasksize);
}
#else
if (selected) {
#ifdef CONFIG_SAMP_HOTNESS
if (min_score_adj <= lowmem_adj[4]) {
#endif
if (oom_score_adj < selected_oom_score_adj)
continue;
if (oom_score_adj == selected_oom_score_adj &&
tasksize <= selected_tasksize)
continue;
#ifdef CONFIG_SAMP_HOTNESS
} else {
if (hotness_adj > selected_hotness_adj)
continue;
if (hotness_adj == selected_hotness_adj && tasksize <= selected_tasksize)
continue;
}
#endif
}
selected = p;
selected_tasksize = tasksize;
selected_oom_score_adj = oom_score_adj;
#ifdef CONFIG_SAMP_HOTNESS
selected_hotness_adj = hotness_adj;
#endif
lowmem_print(2, "select %d (%s), adj %d, size %d, to kill\n",
p->pid, p->comm, oom_score_adj, tasksize);
#endif
}
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (selected[i]) {
#ifdef CONFIG_SAMP_HOTNESS
lowmem_print(1, "send sigkill to %d (%s), adj %d,\
size %d, free memory = %d, reclaimable memory = %d ,hotness %d\n",
selected[i]->pid, selected[i]->comm,
selected_oom_score_adj[i],
selected_tasksize[i],
other_free, other_file,
selected_hotness_adj);
#else
lowmem_print(1, "send sigkill to %d (%s), adj %d,\
size %d, free memory = %d, reclaimable memory = %d\n",
selected[i]->pid, selected[i]->comm,
selected_oom_score_adj[i],
selected_tasksize[i],
other_free, other_file);
#endif
lowmem_deathpending_timeout = jiffies + HZ;
send_sig(SIGKILL, selected[i], 0);
set_tsk_thread_flag(selected[i], TIF_MEMDIE);
rem -= selected_tasksize[i];
if(reclaim_state)
reclaim_state->reclaimed_slab += selected_tasksize[i];
#ifdef LMK_COUNT_READ
lmk_count++;
#endif
}
}
#else
if (selected) {
#ifdef CONFIG_SAMP_HOTNESS
lowmem_print(1, "send sigkill to %d (%s), adj %d, size %d ,hotness %d\n",
selected->pid, selected->comm,
selected_oom_score_adj, selected_tasksize,selected_hotness_adj);
#else
lowmem_print(1, "send sigkill to %d (%s), adj %d, size %d\n",
selected->pid, selected->comm,
selected_oom_score_adj, selected_tasksize);
#endif
lowmem_deathpending_timeout = jiffies + HZ;
send_sig(SIGKILL, selected, 0);
set_tsk_thread_flag(selected, TIF_MEMDIE);
rem -= selected_tasksize;
if(reclaim_state)
reclaim_state->reclaimed_slab = selected_tasksize;
#ifdef LMK_COUNT_READ
lmk_count++;
#endif
}
#endif
#ifdef CONFIG_SEC_DEBUG_LMK_MEMINFO
if (__ratelimit(&lmk_rs)) {
lowmem_print(1, "lowmem_shrink %lu, %x, ofree %d %d, ma %d\n",
nr_to_scan, sc->gfp_mask, other_free,
other_file, min_score_adj);
#ifdef CONFIG_SEC_DEBUG_LMK_MEMINFO_VERBOSE
show_mem(SHOW_MEM_FILTER_NODES);
dump_tasks_info();
#endif
}
#endif
lowmem_print(4, "lowmem_shrink %lu, %x, return %d\n",
sc->nr_to_scan, sc->gfp_mask, rem);
read_unlock(&tasklist_lock);
return rem;
}
static void ipgre_err(struct sk_buff *skb, u32 info)
{
/* All the routers (except for Linux) return only
8 bytes of packet payload. It means, that precise relaying of
ICMP in the real Internet is absolutely infeasible.
Moreover, Cisco "wise men" put GRE key to the third word
in GRE header. It makes impossible maintaining even soft state for keyed
GRE tunnels with enabled checksum. Tell them "thank you".
Well, I wonder, rfc1812 was written by Cisco employee,
what the hell these idiots break standrads established
by themself???
*/
struct iphdr *iph = (struct iphdr *)skb->data;
__be16 *p = (__be16*)(skb->data+(iph->ihl<<2));
int grehlen = (iph->ihl<<2) + 4;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct ip_tunnel *t;
__be16 flags;
flags = p[0];
if (flags&(GRE_CSUM|GRE_KEY|GRE_SEQ|GRE_ROUTING|GRE_VERSION)) {
if (flags&(GRE_VERSION|GRE_ROUTING))
return;
if (flags&GRE_KEY) {
grehlen += 4;
if (flags&GRE_CSUM)
grehlen += 4;
}
}
/* If only 8 bytes returned, keyed message will be dropped here */
if (skb_headlen(skb) < grehlen)
return;
switch (type) {
default:
case ICMP_PARAMETERPROB:
return;
case ICMP_DEST_UNREACH:
switch (code) {
case ICMP_SR_FAILED:
case ICMP_PORT_UNREACH:
/* Impossible event. */
return;
case ICMP_FRAG_NEEDED:
/* Soft state for pmtu is maintained by IP core. */
return;
default:
/* All others are translated to HOST_UNREACH.
rfc2003 contains "deep thoughts" about NET_UNREACH,
I believe they are just ether pollution. --ANK
*/
break;
}
break;
case ICMP_TIME_EXCEEDED:
if (code != ICMP_EXC_TTL)
return;
break;
}
read_lock(&ipgre_lock);
t = ipgre_tunnel_lookup(skb->dev, iph->daddr, iph->saddr,
flags & GRE_KEY ?
*(((__be32 *)p) + (grehlen / 4) - 1) : 0,
p[1]);
if (t == NULL || t->parms.iph.daddr == 0 ||
ipv4_is_multicast(t->parms.iph.daddr))
goto out;
if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
goto out;
if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
t->err_count++;
else
t->err_count = 1;
t->err_time = jiffies;
out:
read_unlock(&ipgre_lock);
return;
}
static int android_oom_handler(struct notifier_block *nb,
unsigned long val, void *data)
{
struct task_struct *tsk;
#ifdef MULTIPLE_OOM_KILLER
struct task_struct *selected[OOM_DEPTH] = {NULL,};
#else
struct task_struct *selected = NULL;
#endif
int rem = 0;
int tasksize;
int i;
int min_score_adj = OOM_SCORE_ADJ_MAX + 1;
#ifdef MULTIPLE_OOM_KILLER
int selected_tasksize[OOM_DEPTH] = {0,};
int selected_oom_score_adj[OOM_DEPTH] = {OOM_ADJUST_MAX,};
int all_selected_oom = 0;
int max_selected_oom_idx = 0;
#else
int selected_tasksize = 0;
int selected_oom_score_adj;
#endif
static DEFINE_RATELIMIT_STATE(oom_rs, DEFAULT_RATELIMIT_INTERVAL/5, 1);
unsigned long *freed = data;
/* show status */
pr_warning("%s invoked Android-oom-killer: "
"oom_adj=%d, oom_score_adj=%d\n",
current->comm, current->signal->oom_adj,
current->signal->oom_score_adj);
dump_stack();
show_mem(SHOW_MEM_FILTER_NODES);
if (__ratelimit(&oom_rs))
dump_tasks_info();
min_score_adj = 0;
#ifdef MULTIPLE_OOM_KILLER
for (i = 0; i < OOM_DEPTH; i++)
selected_oom_score_adj[i] = min_score_adj;
#else
selected_oom_score_adj = min_score_adj;
#endif
read_lock(&tasklist_lock);
for_each_process(tsk) {
struct task_struct *p;
int oom_score_adj;
#ifdef MULTIPLE_OOM_KILLER
int is_exist_oom_task = 0;
#endif
if (tsk->flags & PF_KTHREAD)
continue;
p = find_lock_task_mm(tsk);
if (!p)
continue;
oom_score_adj = p->signal->oom_score_adj;
if (oom_score_adj < min_score_adj) {
task_unlock(p);
continue;
}
tasksize = get_mm_rss(p->mm);
task_unlock(p);
if (tasksize <= 0)
continue;
lowmem_print(2, "oom: ------ %d (%s), adj %d, size %d\n",
p->pid, p->comm, oom_score_adj, tasksize);
#ifdef MULTIPLE_OOM_KILLER
if (all_selected_oom < OOM_DEPTH) {
for (i = 0; i < OOM_DEPTH; i++) {
if (!selected[i]) {
is_exist_oom_task = 1;
max_selected_oom_idx = i;
break;
}
}
} else if (selected_oom_score_adj[max_selected_oom_idx] < oom_score_adj ||
(selected_oom_score_adj[max_selected_oom_idx] == oom_score_adj &&
selected_tasksize[max_selected_oom_idx] < tasksize)) {
is_exist_oom_task = 1;
}
if (is_exist_oom_task) {
selected[max_selected_oom_idx] = p;
selected_tasksize[max_selected_oom_idx] = tasksize;
selected_oom_score_adj[max_selected_oom_idx] = oom_score_adj;
if (all_selected_oom < OOM_DEPTH)
all_selected_oom++;
if (all_selected_oom == OOM_DEPTH) {
for (i = 0; i < OOM_DEPTH; i++) {
if (selected_oom_score_adj[i] < selected_oom_score_adj[max_selected_oom_idx])
max_selected_oom_idx = i;
else if (selected_oom_score_adj[i] == selected_oom_score_adj[max_selected_oom_idx] &&
selected_tasksize[i] < selected_tasksize[max_selected_oom_idx])
max_selected_oom_idx = i;
}
}
lowmem_print(2, "oom: max_selected_oom_idx(%d) select %d (%s), adj %d, \
size %d, to kill\n",
max_selected_oom_idx, p->pid, p->comm, oom_score_adj, tasksize);
}
#else
if (selected) {
if (oom_score_adj < selected_oom_score_adj)
continue;
if (oom_score_adj == selected_oom_score_adj &&
tasksize <= selected_tasksize)
continue;
}
selected = p;
selected_tasksize = tasksize;
selected_oom_score_adj = oom_score_adj;
lowmem_print(2, "oom: select %d (%s), adj %d, size %d, to kill\n",
p->pid, p->comm, oom_score_adj, tasksize);
#endif
}
#ifdef MULTIPLE_OOM_KILLER
for (i = 0; i < OOM_DEPTH; i++) {
if (selected[i]) {
lowmem_print(1, "oom: send sigkill to %d (%s), adj %d,\
size %d\n",
selected[i]->pid, selected[i]->comm,
selected_oom_score_adj[i],
selected_tasksize[i]);
send_sig(SIGKILL, selected[i], 0);
rem -= selected_tasksize[i];
*freed += (unsigned long)selected_tasksize[i];
#ifdef OOM_COUNT_READ
oom_count++;
#endif
}
}
#else
if (selected) {
lowmem_print(1, "oom: send sigkill to %d (%s), adj %d, size %d\n",
selected->pid, selected->comm,
selected_oom_score_adj, selected_tasksize);
send_sig(SIGKILL, selected, 0);
set_tsk_thread_flag(selected, TIF_MEMDIE);
rem -= selected_tasksize;
*freed += (unsigned long)selected_tasksize;
#ifdef OOM_COUNT_READ
oom_count++;
#endif
}
#endif
read_unlock(&tasklist_lock);
lowmem_print(2, "oom: get memory %lu", *freed);
return rem;
}
static int lowmem_shrink(struct shrinker *s, int nr_to_scan, gfp_t gfp_mask)
{
struct task_struct *p;
struct task_struct *selected = NULL;
int rem = 0;
int tasksize;
int i;
int min_adj = OOM_ADJUST_MAX + 1;
int selected_tasksize = 0;
int selected_oom_adj;
int array_size = ARRAY_SIZE(lowmem_adj);
int other_free = global_page_state(NR_FREE_PAGES);
int other_file = global_page_state(NR_FILE_PAGES) -
global_page_state(NR_SHMEM);
/*
* If we already have a death outstanding, then
* bail out right away; indicating to vmscan
* that we have nothing further to offer on
* this pass.
*
*/
if (lowmem_deathpending &&
time_before_eq(jiffies, lowmem_deathpending_timeout))
return 0;
if (lowmem_adj_size < array_size)
array_size = lowmem_adj_size;
if (lowmem_minfree_size < array_size)
array_size = lowmem_minfree_size;
for (i = 0; i < array_size; i++) {
if (other_free < lowmem_minfree[i] &&
other_file < lowmem_minfree[i]) {
min_adj = lowmem_adj[i];
break;
}
}
if (min_adj == OOM_ADJUST_MAX + 1)
return 0;
if (nr_to_scan > 0)
lowmem_print(3, "lowmem_shrink %d, %x, ofree %d %d, ma %d\n",
nr_to_scan, gfp_mask, other_free, other_file,
min_adj);
rem = global_page_state(NR_ACTIVE_ANON) +
global_page_state(NR_ACTIVE_FILE) +
global_page_state(NR_INACTIVE_ANON) +
global_page_state(NR_INACTIVE_FILE);
if (nr_to_scan <= 0) {
lowmem_print(5, "lowmem_shrink %d, %x, return %d\n",
nr_to_scan, gfp_mask, rem);
return rem;
}
selected_oom_adj = min_adj;
read_lock(&tasklist_lock);
for_each_process(p) {
struct mm_struct *mm;
struct signal_struct *sig;
int oom_adj;
task_lock(p);
mm = p->mm;
sig = p->signal;
if (!mm || !sig) {
task_unlock(p);
continue;
}
oom_adj = sig->oom_adj;
if (oom_adj < min_adj) {
task_unlock(p);
continue;
}
tasksize = get_mm_rss(mm);
task_unlock(p);
if (tasksize <= 0)
continue;
if (selected) {
if (oom_adj < selected_oom_adj)
continue;
if (oom_adj == selected_oom_adj &&
tasksize <= selected_tasksize)
continue;
}
selected = p;
selected_tasksize = tasksize;
selected_oom_adj = oom_adj;
lowmem_print(2, "select %d (%s), adj %d, size %d, to kill\n",
p->pid, p->comm, oom_adj, tasksize);
}
if (selected) {
lowmem_print(1, "send sigkill to %d (%s), adj %d, size %d\n",
selected->pid, selected->comm,
selected_oom_adj, selected_tasksize);
lowmem_deathpending = selected;
lowmem_deathpending_timeout = jiffies + HZ;
force_sig(SIGKILL, selected);
rem -= selected_tasksize;
} else
rem = -1;
lowmem_print(4, "lowmem_shrink %d, %x, return %d\n",
nr_to_scan, gfp_mask, rem);
read_unlock(&tasklist_lock);
return rem;
}
void __iomem * __arm_ioremap_pfn_caller(unsigned long pfn,
unsigned long offset, size_t size, unsigned int mtype, void *caller)
{
const struct mem_type *type;
int err;
unsigned long addr;
struct vm_struct * area;
#ifndef CONFIG_ARM_LPAE
/*
* High mappings must be supersection aligned
*/
if (pfn >= 0x100000 && (__pfn_to_phys(pfn) & ~SUPERSECTION_MASK))
return NULL;
#endif
type = get_mem_type(mtype);
if (!type)
return NULL;
/*
* Page align the mapping size, taking account of any offset.
*/
size = PAGE_ALIGN(offset + size);
/*
* Try to reuse one of the static mapping whenever possible.
*/
read_lock(&vmlist_lock);
for (area = vmlist; area; area = area->next) {
if (!size || (sizeof(phys_addr_t) == 4 && pfn >= 0x100000))
break;
if (!(area->flags & VM_ARM_STATIC_MAPPING))
continue;
if ((area->flags & VM_ARM_MTYPE_MASK) != VM_ARM_MTYPE(mtype))
continue;
if (__phys_to_pfn(area->phys_addr) > pfn ||
__pfn_to_phys(pfn) + offset + size-1 >
area->phys_addr + area->size-1)
continue;
/* we can drop the lock here as we know *area is static */
read_unlock(&vmlist_lock);
addr = (unsigned long)area->addr;
addr += __pfn_to_phys(pfn) - area->phys_addr;
return (void __iomem *) (offset + addr);
}
read_unlock(&vmlist_lock);
/*
* Don't allow RAM to be mapped - this causes problems with ARMv6+
*/
if (WARN_ON(pfn_valid(pfn)))
return NULL;
area = get_vm_area_caller(size, VM_IOREMAP, caller);
if (!area)
return NULL;
addr = (unsigned long)area->addr;
#if !defined(CONFIG_SMP) && !defined(CONFIG_ARM_LPAE)
if (DOMAIN_IO == 0 &&
(((cpu_architecture() >= CPU_ARCH_ARMv6) && (get_cr() & CR_XP)) ||
cpu_is_xsc3()) && pfn >= 0x100000 &&
!((__pfn_to_phys(pfn) | size | addr) & ~SUPERSECTION_MASK)) {
area->flags |= VM_ARM_SECTION_MAPPING;
err = remap_area_supersections(addr, pfn, size, type);
} else if (!((__pfn_to_phys(pfn) | size | addr) & ~PMD_MASK)) {
area->flags |= VM_ARM_SECTION_MAPPING;
err = remap_area_sections(addr, pfn, size, type);
} else
#endif
err = ioremap_page_range(addr, addr + size, __pfn_to_phys(pfn),
__pgprot(type->prot_pte));
if (err) {
vunmap((void *)addr);
return NULL;
}
flush_cache_vmap(addr, addr + size);
return (void __iomem *) (offset + addr);
}
int do_setitimer(int which, struct itimerval *value, struct itimerval *ovalue)
{
struct task_struct *tsk = current;
struct hrtimer *timer;
ktime_t expires;
cputime_t cval, cinterval, nval, ninterval;
/*
* Validate the timevals in value.
*
* Note: Although the spec requires that invalid values shall
* return -EINVAL, we just fixup the value and print a limited
* number of warnings in order not to break users of this
* historical misfeature.
*
* Scheduled for replacement in March 2007
*/
check_itimerval(value);
MARK(kernel_timer_set_itimer, "%d %ld %ld %ld %ld",
which,
value->it_interval.tv_sec,
value->it_interval.tv_usec,
value->it_value.tv_sec,
value->it_value.tv_usec);
switch (which) {
case ITIMER_REAL:
again:
spin_lock_irq(&tsk->sighand->siglock);
timer = &tsk->signal->real_timer;
if (ovalue) {
ovalue->it_value = itimer_get_remtime(timer);
ovalue->it_interval
= ktime_to_timeval(tsk->signal->it_real_incr);
}
/* We are sharing ->siglock with it_real_fn() */
if (hrtimer_try_to_cancel(timer) < 0) {
spin_unlock_irq(&tsk->sighand->siglock);
goto again;
}
tsk->signal->it_real_incr =
timeval_to_ktime(value->it_interval);
expires = timeval_to_ktime(value->it_value);
if (expires.tv64 != 0)
hrtimer_start(timer, expires, HRTIMER_REL);
spin_unlock_irq(&tsk->sighand->siglock);
break;
case ITIMER_VIRTUAL:
nval = timeval_to_cputime(&value->it_value);
ninterval = timeval_to_cputime(&value->it_interval);
read_lock(&tasklist_lock);
spin_lock_irq(&tsk->sighand->siglock);
cval = tsk->signal->it_virt_expires;
cinterval = tsk->signal->it_virt_incr;
if (!cputime_eq(cval, cputime_zero) ||
!cputime_eq(nval, cputime_zero)) {
if (cputime_gt(nval, cputime_zero))
nval = cputime_add(nval,
jiffies_to_cputime(1));
set_process_cpu_timer(tsk, CPUCLOCK_VIRT,
&nval, &cval);
}
tsk->signal->it_virt_expires = nval;
tsk->signal->it_virt_incr = ninterval;
spin_unlock_irq(&tsk->sighand->siglock);
read_unlock(&tasklist_lock);
if (ovalue) {
cputime_to_timeval(cval, &ovalue->it_value);
cputime_to_timeval(cinterval, &ovalue->it_interval);
}
break;
case ITIMER_PROF:
nval = timeval_to_cputime(&value->it_value);
ninterval = timeval_to_cputime(&value->it_interval);
read_lock(&tasklist_lock);
spin_lock_irq(&tsk->sighand->siglock);
cval = tsk->signal->it_prof_expires;
cinterval = tsk->signal->it_prof_incr;
if (!cputime_eq(cval, cputime_zero) ||
!cputime_eq(nval, cputime_zero)) {
if (cputime_gt(nval, cputime_zero))
nval = cputime_add(nval,
jiffies_to_cputime(1));
set_process_cpu_timer(tsk, CPUCLOCK_PROF,
&nval, &cval);
}
tsk->signal->it_prof_expires = nval;
tsk->signal->it_prof_incr = ninterval;
spin_unlock_irq(&tsk->sighand->siglock);
read_unlock(&tasklist_lock);
if (ovalue) {
cputime_to_timeval(cval, &ovalue->it_value);
cputime_to_timeval(cinterval, &ovalue->it_interval);
}
break;
default:
return -EINVAL;
}
return 0;
}
int timod_putmsg(unsigned int fd, char *ctl_buf, int ctl_len,
char *data_buf, int data_len, int flags)
{
int ret, error, terror;
char *buf;
struct file *filp;
struct inode *ino;
struct sol_socket_struct *sock;
mm_segment_t old_fs = get_fs();
long args[6];
int (*sys_socketcall)(int, unsigned long *) =
(int (*)(int, unsigned long *))SYS(socketcall);
int (*sys_sendto)(int, void *, size_t, unsigned, struct sockaddr *, int) =
(int (*)(int, void *, size_t, unsigned, struct sockaddr *, int))SYS(sendto);
read_lock(¤t->files->file_lock);
filp = fcheck(fd);
read_unlock(¤t->files->file_lock);
if (!filp)
return -EBADF;
ino = filp->f_dentry->d_inode;
sock = (struct sol_socket_struct *)filp->private_data;
SOLD("entry");
if (get_user(ret, (int *)A(ctl_buf)))
return -EFAULT;
switch (ret) {
case T_BIND_REQ:
{
struct T_bind_req req;
SOLDD(("bind %016lx(%016lx)\n", sock, filp));
SOLD("T_BIND_REQ");
if (sock->state != TS_UNBND) {
timod_error(fd, T_BIND_REQ, TOUTSTATE, 0);
return 0;
}
SOLD("state ok");
if (copy_from_user(&req, ctl_buf, sizeof(req))) {
timod_error(fd, T_BIND_REQ, TSYSERR, EFAULT);
return 0;
}
SOLD("got ctl req");
if (req.ADDR_offset && req.ADDR_length) {
if (req.ADDR_length > BUF_SIZE) {
timod_error(fd, T_BIND_REQ, TSYSERR, EFAULT);
return 0;
}
SOLD("req size ok");
buf = getpage();
if (copy_from_user(buf, ctl_buf + req.ADDR_offset, req.ADDR_length)) {
timod_error(fd, T_BIND_REQ, TSYSERR, EFAULT);
putpage(buf);
return 0;
}
SOLD("got ctl data");
args[0] = fd;
args[1] = (long)buf;
args[2] = req.ADDR_length;
SOLD("calling BIND");
set_fs(KERNEL_DS);
error = sys_socketcall(SYS_BIND, args);
set_fs(old_fs);
putpage(buf);
SOLD("BIND returned");
} else
error = 0;
if (!error) {
struct T_primsg *it;
if (req.CONIND_number) {
args[0] = fd;
args[1] = req.CONIND_number;
SOLD("calling LISTEN");
set_fs(KERNEL_DS);
error = sys_socketcall(SYS_LISTEN, args);
set_fs(old_fs);
SOLD("LISTEN done");
}
it = timod_mkctl(sizeof(struct T_bind_ack)+sizeof(struct sockaddr));
if (it) {
struct T_bind_ack *ack;
ack = (struct T_bind_ack *)&it->type;
ack->PRIM_type = T_BIND_ACK;
ack->ADDR_offset = sizeof(*ack);
ack->ADDR_length = sizeof(struct sockaddr);
ack->CONIND_number = req.CONIND_number;
args[0] = fd;
args[1] = (long)(ack+sizeof(*ack));
args[2] = (long)&ack->ADDR_length;
set_fs(KERNEL_DS);
sys_socketcall(SYS_GETSOCKNAME,args);
set_fs(old_fs);
sock->state = TS_IDLE;
timod_ok(fd, T_BIND_REQ);
timod_queue_end(fd, it);
SOLD("BIND done");
return 0;
}
}
SOLD("some error");
switch (error) {
case -EINVAL:
terror = TOUTSTATE;
error = 0;
break;
case -EACCES:
terror = TACCES;
error = 0;
break;
case -EADDRNOTAVAIL:
case -EADDRINUSE:
terror = TNOADDR;
error = 0;
break;
default:
terror = TSYSERR;
break;
}
timod_error(fd, T_BIND_REQ, terror, -error);
SOLD("BIND done");
return 0;
}
case T_CONN_REQ:
{
struct T_conn_req req;
unsigned short oldflags;
struct T_primsg *it;
SOLD("T_CONN_REQ");
if (sock->state != TS_UNBND && sock->state != TS_IDLE) {
timod_error(fd, T_CONN_REQ, TOUTSTATE, 0);
return 0;
}
SOLD("state ok");
if (copy_from_user(&req, ctl_buf, sizeof(req))) {
timod_error(fd, T_CONN_REQ, TSYSERR, EFAULT);
return 0;
}
SOLD("got ctl req");
if (ctl_len > BUF_SIZE) {
timod_error(fd, T_CONN_REQ, TSYSERR, EFAULT);
return 0;
}
SOLD("req size ok");
buf = getpage();
if (copy_from_user(buf, ctl_buf, ctl_len)) {
timod_error(fd, T_CONN_REQ, TSYSERR, EFAULT);
putpage(buf);
return 0;
}
#ifdef DEBUG_SOLARIS
{
char * ptr = buf;
int len = ctl_len;
printk("returned data (%d bytes): ",len);
while( len-- ) {
if (!(len & 7))
printk(" ");
printk("%02x",(unsigned char)*ptr++);
}
printk("\n");
}
#endif
SOLD("got ctl data");
args[0] = fd;
args[1] = (long)buf+req.DEST_offset;
args[2] = req.DEST_length;
oldflags = filp->f_flags;
filp->f_flags &= ~O_NONBLOCK;
SOLD("calling CONNECT");
set_fs(KERNEL_DS);
error = sys_socketcall(SYS_CONNECT, args);
set_fs(old_fs);
filp->f_flags = oldflags;
SOLD("CONNECT done");
if (!error) {
struct T_conn_con *con;
SOLD("no error");
it = timod_mkctl(ctl_len);
if (!it) {
putpage(buf);
return -ENOMEM;
}
con = (struct T_conn_con *)&it->type;
#ifdef DEBUG_SOLARIS
{
char * ptr = buf;
int len = ctl_len;
printk("returned data (%d bytes): ",len);
while( len-- ) {
if (!(len & 7))
printk(" ");
printk("%02x",(unsigned char)*ptr++);
}
printk("\n");
}
#endif
memcpy(con, buf, ctl_len);
SOLD("copied ctl_buf");
con->PRIM_type = T_CONN_CON;
sock->state = TS_DATA_XFER;
} else {
struct T_discon_ind *dis;
SOLD("some error");
it = timod_mkctl(sizeof(*dis));
if (!it) {
putpage(buf);
return -ENOMEM;
}
SOLD("got primsg");
dis = (struct T_discon_ind *)&it->type;
dis->PRIM_type = T_DISCON_IND;
dis->DISCON_reason = -error; /* FIXME: convert this as in iABI_errors() */
dis->SEQ_number = 0;
}
putpage(buf);
timod_ok(fd, T_CONN_REQ);
it->pri = 0;
timod_queue_end(fd, it);
SOLD("CONNECT done");
return 0;
}
case T_OPTMGMT_REQ:
{
struct T_optmgmt_req req;
SOLD("OPTMGMT_REQ");
if (copy_from_user(&req, ctl_buf, sizeof(req)))
return -EFAULT;
SOLD("got req");
return timod_optmgmt(fd, req.MGMT_flags,
req.OPT_offset > 0 ? ctl_buf + req.OPT_offset : NULL,
req.OPT_length, 1);
}
case T_UNITDATA_REQ:
{
struct T_unitdata_req req;
int err;
SOLD("T_UNITDATA_REQ");
if (sock->state != TS_IDLE && sock->state != TS_DATA_XFER) {
timod_error(fd, T_CONN_REQ, TOUTSTATE, 0);
return 0;
}
SOLD("state ok");
if (copy_from_user(&req, ctl_buf, sizeof(req))) {
timod_error(fd, T_CONN_REQ, TSYSERR, EFAULT);
return 0;
}
SOLD("got ctl req");
#ifdef DEBUG_SOLARIS
{
char * ptr = ctl_buf+req.DEST_offset;
int len = req.DEST_length;
printk("socket address (%d bytes): ",len);
while( len-- ) {
char c;
if (get_user(c,ptr))
printk("??");
else
printk("%02x",(unsigned char)c);
ptr++;
}
printk("\n");
}
#endif
err = sys_sendto(fd, data_buf, data_len, 0, req.DEST_length > 0 ? (struct sockaddr*)(ctl_buf+req.DEST_offset) : NULL, req.DEST_length);
if (err == data_len)
return 0;
if(err >= 0) {
printk("timod: sendto failed to send all the data\n");
return 0;
}
timod_error(fd, T_CONN_REQ, TSYSERR, -err);
return 0;
}
default:
printk(KERN_INFO "timod_putmsg: unsupported command %u.\n", ret);
break;
}
return -EINVAL;
}
static int lowmem_shrink(struct shrinker *s, struct shrink_control *sc)
{
struct task_struct *p;
#ifdef ENHANCED_LMK_ROUTINE
struct task_struct *selected[LOWMEM_DEATHPENDING_DEPTH] = {NULL,};
#else
struct task_struct *selected = NULL;
#endif
int rem = 0;
int tasksize;
int i;
int min_adj = OOM_ADJUST_MAX + 1;
#ifdef ENHANCED_LMK_ROUTINE
int selected_tasksize[LOWMEM_DEATHPENDING_DEPTH] = {0,};
int selected_oom_adj[LOWMEM_DEATHPENDING_DEPTH] = {OOM_ADJUST_MAX,};
int all_selected_oom = 0;
#else
int selected_tasksize = 0;
int selected_oom_adj;
#endif
int array_size = ARRAY_SIZE(lowmem_adj);
int other_free = global_page_state(NR_FREE_PAGES);
int other_file = global_page_state(NR_FILE_PAGES) -
global_page_state(NR_SHMEM);
/*
* If we already have a death outstanding, then
* bail out right away; indicating to vmscan
* that we have nothing further to offer on
* this pass.
*
*/
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (lowmem_deathpending[i] &&
time_before_eq(jiffies, lowmem_deathpending_timeout))
return 0;
}
#else
if (lowmem_deathpending &&
time_before_eq(jiffies, lowmem_deathpending_timeout))
return 0;
#endif
if (lowmem_adj_size < array_size)
array_size = lowmem_adj_size;
if (lowmem_minfree_size < array_size)
array_size = lowmem_minfree_size;
for (i = 0; i < array_size; i++) {
if (other_free < lowmem_minfree[i] &&
other_file < lowmem_minfree[i]) {
min_adj = lowmem_adj[i];
break;
}
}
if (sc->nr_to_scan > 0)
lowmem_print(3, "lowmem_shrink %lu, %x, ofree %d %d, ma %d\n",
sc->nr_to_scan, sc->gfp_mask, other_free, other_file,
min_adj);
rem = global_page_state(NR_ACTIVE_ANON) +
global_page_state(NR_ACTIVE_FILE) +
global_page_state(NR_INACTIVE_ANON) +
global_page_state(NR_INACTIVE_FILE);
if (sc->nr_to_scan <= 0 || min_adj == OOM_ADJUST_MAX + 1) {
lowmem_print(5, "lowmem_shrink %lu, %x, return %d\n",
sc->nr_to_scan, sc->gfp_mask, rem);
return rem;
}
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++)
selected_oom_adj[i] = min_adj;
#else
selected_oom_adj = min_adj;
#endif
read_lock(&tasklist_lock);
for_each_process(p) {
struct mm_struct *mm;
struct signal_struct *sig;
int oom_adj;
task_lock(p);
mm = p->mm;
sig = p->signal;
if (!mm || !sig) {
task_unlock(p);
continue;
}
oom_adj = sig->oom_adj;
if (oom_adj < min_adj) {
task_unlock(p);
continue;
}
tasksize = get_mm_rss(mm);
task_unlock(p);
if (tasksize <= 0)
continue;
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (all_selected_oom >= LOWMEM_DEATHPENDING_DEPTH) {
if (oom_adj < selected_oom_adj[i])
continue;
if (oom_adj == selected_oom_adj[i] &&
tasksize <= selected_tasksize[i])
continue;
} else if (selected[i])
continue;
selected[i] = p;
selected_tasksize[i] = tasksize;
selected_oom_adj[i] = oom_adj;
if (all_selected_oom < LOWMEM_DEATHPENDING_DEPTH)
all_selected_oom++;
lowmem_print(2, "select %d (%s), adj %d, size %d, to kill\n",
p->pid, p->comm, oom_adj, tasksize);
break;
}
#else
if (selected) {
if (oom_adj < selected_oom_adj)
continue;
if (oom_adj == selected_oom_adj &&
tasksize <= selected_tasksize)
continue;
}
selected = p;
selected_tasksize = tasksize;
selected_oom_adj = oom_adj;
lowmem_print(2, "select %d (%s), adj %d, size %d, to kill\n",
p->pid, p->comm, oom_adj, tasksize);
#endif
}
#ifdef ENHANCED_LMK_ROUTINE
for (i = 0; i < LOWMEM_DEATHPENDING_DEPTH; i++) {
if (selected[i]) {
lowmem_print(1, "send sigkill to %d (%s), adj %d, size %d\n",
selected[i]->pid, selected[i]->comm,
selected_oom_adj[i], selected_tasksize[i]);
lowmem_deathpending[i] = selected[i];
lowmem_deathpending_timeout = jiffies + HZ;
force_sig(SIGKILL, selected[i]);
rem -= selected_tasksize[i];
}
}
#else
if (selected) {
lowmem_print(1, "send sigkill to %d (%s), adj %d, size %d\n",
selected->pid, selected->comm,
selected_oom_adj, selected_tasksize);
lowmem_deathpending = selected;
lowmem_deathpending_timeout = jiffies + HZ;
force_sig(SIGKILL, selected);
rem -= selected_tasksize;
}
#endif
lowmem_print(4, "lowmem_shrink %lu, %x, return %d\n",
sc->nr_to_scan, sc->gfp_mask, rem);
read_unlock(&tasklist_lock);
return rem;
}
static int copy_files(unsigned long clone_flags, struct task_struct * tsk)
{
struct files_struct *oldf, *newf;
struct file **old_fds, **new_fds;
int open_files, nfds, size, i, error = 0;
/*
* A background process may not have any files ...
*/
oldf = current->files;
if (!oldf)
goto out;
if (clone_flags & CLONE_FILES) {
atomic_inc(&oldf->count);
goto out;
}
/*
* Note: we may be using current for both targets (See exec.c)
* This works because we cache current->files (old) as oldf. Don't
* break this.
*/
tsk->files = NULL;
error = -ENOMEM;
newf = kmem_cache_alloc(files_cachep, SLAB_KERNEL);
if (!newf)
goto out;
atomic_set(&newf->count, 1);
newf->file_lock = RW_LOCK_UNLOCKED;
newf->next_fd = 0;
newf->max_fds = NR_OPEN_DEFAULT;
newf->max_fdset = __FD_SETSIZE;
newf->close_on_exec = &newf->close_on_exec_init;
newf->open_fds = &newf->open_fds_init;
newf->fd = &newf->fd_array[0];
/* We don't yet have the oldf readlock, but even if the old
fdset gets grown now, we'll only copy up to "size" fds */
size = oldf->max_fdset;
if (size > __FD_SETSIZE) {
newf->max_fdset = 0;
write_lock(&newf->file_lock);
error = expand_fdset(newf, size-1);
write_unlock(&newf->file_lock);
if (error)
goto out_release;
}
read_lock(&oldf->file_lock);
open_files = count_open_files(oldf, size);
/*
* Check whether we need to allocate a larger fd array.
* Note: we're not a clone task, so the open count won't
* change.
*/
nfds = NR_OPEN_DEFAULT;
if (open_files > nfds) {
read_unlock(&oldf->file_lock);
newf->max_fds = 0;
write_lock(&newf->file_lock);
error = expand_fd_array(newf, open_files-1);
write_unlock(&newf->file_lock);
if (error)
goto out_release;
nfds = newf->max_fds;
read_lock(&oldf->file_lock);
}
old_fds = oldf->fd;
new_fds = newf->fd;
memcpy(newf->open_fds->fds_bits, oldf->open_fds->fds_bits, open_files/8);
memcpy(newf->close_on_exec->fds_bits, oldf->close_on_exec->fds_bits, open_files/8);
for (i = open_files; i != 0; i--) {
struct file *f = *old_fds++;
if (f) {
get_file(f);
} else {
/*
* The fd may be claimed in the fd bitmap but not yet
* instantiated in the files array if a sibling thread
* is partway through open(). So make sure that this
* fd is available to the new process.
*/
FD_CLR(open_files - i, newf->open_fds);
}
*new_fds++ = f;
}
read_unlock(&oldf->file_lock);
/* compute the remainder to be cleared */
size = (newf->max_fds - open_files) * sizeof(struct file *);
/* This is long word aligned thus could use a optimized version */
memset(new_fds, 0, size);
if (newf->max_fdset > open_files) {
int left = (newf->max_fdset-open_files)/8;
int start = open_files / (8 * sizeof(unsigned long));
memset(&newf->open_fds->fds_bits[start], 0, left);
memset(&newf->close_on_exec->fds_bits[start], 0, left);
}
tsk->files = newf;
error = 0;
out:
return error;
out_release:
free_fdset (newf->close_on_exec, newf->max_fdset);
free_fdset (newf->open_fds, newf->max_fdset);
kmem_cache_free(files_cachep, newf);
goto out;
}
