void
BDirectMessageTarget::Acquire()
{
atomic_add(&fReferenceCount, 1);
}
status_t
load_image(char const* name, image_type type, const char* rpath,
image_t** _image)
{
int32 pheaderSize, sheaderSize;
char path[PATH_MAX];
ssize_t length;
char pheaderBuffer[4096];
int32 numRegions;
image_t* found;
image_t* image;
status_t status;
int fd;
struct Elf32_Ehdr eheader;
// Have we already loaded that image? Don't check for add-ons -- we always
// reload them.
if (type != B_ADD_ON_IMAGE) {
found = find_loaded_image_by_name(name, APP_OR_LIBRARY_TYPE);
if (found == NULL && type != B_APP_IMAGE && gProgramImage != NULL) {
// Special case for add-ons that link against the application
// executable, with the executable not having a soname set.
if (const char* lastSlash = strrchr(name, '/')) {
if (strcmp(gProgramImage->name, lastSlash + 1) == 0)
found = gProgramImage;
}
}
if (found) {
atomic_add(&found->ref_count, 1);
*_image = found;
KTRACE("rld: load_container(\"%s\", type: %d, rpath: \"%s\") "
"already loaded", name, type, rpath);
return B_OK;
}
}
KTRACE("rld: load_container(\"%s\", type: %d, rpath: \"%s\")", name, type,
rpath);
strlcpy(path, name, sizeof(path));
// find and open the file
fd = open_executable(path, type, rpath, get_program_path(),
sSearchPathSubDir);
if (fd < 0) {
FATAL("Cannot open file %s: %s\n", name, strerror(fd));
KTRACE("rld: load_container(\"%s\"): failed to open file", name);
return fd;
}
// normalize the image path
status = _kern_normalize_path(path, true, path);
if (status != B_OK)
goto err1;
// Test again if this image has been registered already - this time,
// we can check the full path, not just its name as noted.
// You could end up loading an image twice with symbolic links, else.
if (type != B_ADD_ON_IMAGE) {
found = find_loaded_image_by_name(path, APP_OR_LIBRARY_TYPE);
if (found) {
atomic_add(&found->ref_count, 1);
*_image = found;
_kern_close(fd);
KTRACE("rld: load_container(\"%s\"): already loaded after all",
name);
return B_OK;
}
}
length = _kern_read(fd, 0, &eheader, sizeof(eheader));
if (length != sizeof(eheader)) {
status = B_NOT_AN_EXECUTABLE;
FATAL("%s: Troubles reading ELF header\n", path);
goto err1;
}
status = parse_elf_header(&eheader, &pheaderSize, &sheaderSize);
if (status < B_OK) {
FATAL("%s: Incorrect ELF header\n", path);
goto err1;
}
// ToDo: what to do about this restriction??
if (pheaderSize > (int)sizeof(pheaderBuffer)) {
FATAL("%s: Cannot handle program headers bigger than %lu\n",
path, sizeof(pheaderBuffer));
status = B_UNSUPPORTED;
goto err1;
}
length = _kern_read(fd, eheader.e_phoff, pheaderBuffer, pheaderSize);
if (length != pheaderSize) {
FATAL("%s: Could not read program headers: %s\n", path,
strerror(length));
status = B_BAD_DATA;
goto err1;
}
numRegions = count_regions(path, pheaderBuffer, eheader.e_phnum,
eheader.e_phentsize);
if (numRegions <= 0) {
FATAL("%s: Troubles parsing Program headers, numRegions = %ld\n",
path, numRegions);
status = B_BAD_DATA;
goto err1;
}
image = create_image(name, path, numRegions);
if (image == NULL) {
FATAL("%s: Failed to allocate image_t object\n", path);
status = B_NO_MEMORY;
goto err1;
}
status = parse_program_headers(image, pheaderBuffer, eheader.e_phnum,
eheader.e_phentsize);
if (status < B_OK)
goto err2;
if (!assert_dynamic_loadable(image)) {
FATAL("%s: Dynamic segment must be loadable (implementation "
"restriction)\n", image->path);
status = B_UNSUPPORTED;
goto err2;
}
status = map_image(fd, path, image, type == B_APP_IMAGE);
if (status < B_OK) {
FATAL("%s: Could not map image: %s\n", image->path, strerror(status));
status = B_ERROR;
goto err2;
}
if (!parse_dynamic_segment(image)) {
FATAL("%s: Troubles handling dynamic section\n", image->path);
status = B_BAD_DATA;
goto err3;
}
if (eheader.e_entry != 0)
image->entry_point = eheader.e_entry + image->regions[0].delta;
analyze_image_haiku_version_and_abi(fd, image, eheader, sheaderSize,
pheaderBuffer, sizeof(pheaderBuffer));
// If this is the executable image, we init the search path
// subdir, if the compiler version doesn't match ours.
if (type == B_APP_IMAGE) {
#if __GNUC__ == 2
if ((image->abi & B_HAIKU_ABI_MAJOR) == B_HAIKU_ABI_GCC_4)
sSearchPathSubDir = "gcc4";
#elif __GNUC__ == 4
if ((image->abi & B_HAIKU_ABI_MAJOR) == B_HAIKU_ABI_GCC_2)
sSearchPathSubDir = "gcc2";
#endif
}
// init gcc version dependent image flags
// symbol resolution strategy
if (image->abi == B_HAIKU_ABI_GCC_2_ANCIENT)
image->find_undefined_symbol = find_undefined_symbol_beos;
// init version infos
status = init_image_version_infos(image);
image->type = type;
register_image(image, fd, path);
image_event(image, IMAGE_EVENT_LOADED);
_kern_close(fd);
enqueue_loaded_image(image);
*_image = image;
KTRACE("rld: load_container(\"%s\"): done: id: %ld (ABI: %#lx)", name,
image->id, image->abi);
return B_OK;
err3:
unmap_image(image);
err2:
delete_image_struct(image);
err1:
_kern_close(fd);
KTRACE("rld: load_container(\"%s\"): failed: %s", name,
strerror(status));
return status;
}
/*
* Check if this packet is complete.
* Returns NULL on failure by any reason, and pointer
* to current nexthdr field in reassembled frame.
*
* It is called with locked fq, and caller must check that
* queue is eligible for reassembly i.e. it is not COMPLETE,
* the last and the first frames arrived and all the bits are here.
*/
static struct sk_buff *
nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
{
struct sk_buff *fp, *op, *head = fq->fragments;
int payload_len;
fq_kill(fq);
BUG_TRAP(head != NULL);
BUG_TRAP(NFCT_FRAG6_CB(head)->offset == 0);
/* Unfragmented part is taken from the first segment. */
payload_len = (head->data - head->nh.raw) - sizeof(struct ipv6hdr) + fq->len - sizeof(struct frag_hdr);
if (payload_len > IPV6_MAXPLEN) {
DEBUGP("payload len is too large.\n");
goto out_oversize;
}
/* Head of list must not be cloned. */
if (skb_cloned(head) && pskb_expand_head(head, 0, 0, GFP_ATOMIC)) {
DEBUGP("skb is cloned but can't expand head");
goto out_oom;
}
/* If the first fragment is fragmented itself, we split
* it to two chunks: the first with data and paged part
* and the second, holding only fragments. */
if (skb_shinfo(head)->frag_list) {
struct sk_buff *clone;
int i, plen = 0;
if ((clone = alloc_skb(0, GFP_ATOMIC)) == NULL) {
DEBUGP("Can't alloc skb\n");
goto out_oom;
}
clone->next = head->next;
head->next = clone;
skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list;
skb_shinfo(head)->frag_list = NULL;
for (i=0; i<skb_shinfo(head)->nr_frags; i++)
plen += skb_shinfo(head)->frags[i].size;
clone->len = clone->data_len = head->data_len - plen;
head->data_len -= clone->len;
head->len -= clone->len;
clone->csum = 0;
clone->ip_summed = head->ip_summed;
NFCT_FRAG6_CB(clone)->orig = NULL;
atomic_add(clone->truesize, &nf_ct_frag6_mem);
}
/* We have to remove fragment header from datagram and to relocate
* header in order to calculate ICV correctly. */
head->nh.raw[fq->nhoffset] = head->h.raw[0];
memmove(head->head + sizeof(struct frag_hdr), head->head,
(head->data - head->head) - sizeof(struct frag_hdr));
head->mac.raw += sizeof(struct frag_hdr);
head->nh.raw += sizeof(struct frag_hdr);
skb_shinfo(head)->frag_list = head->next;
head->h.raw = head->data;
skb_push(head, head->data - head->nh.raw);
atomic_sub(head->truesize, &nf_ct_frag6_mem);
for (fp=head->next; fp; fp = fp->next) {
head->data_len += fp->len;
head->len += fp->len;
if (head->ip_summed != fp->ip_summed)
head->ip_summed = CHECKSUM_NONE;
else if (head->ip_summed == CHECKSUM_HW)
head->csum = csum_add(head->csum, fp->csum);
head->truesize += fp->truesize;
atomic_sub(fp->truesize, &nf_ct_frag6_mem);
}
head->next = NULL;
head->dev = dev;
skb_set_timestamp(head, &fq->stamp);
head->nh.ipv6h->payload_len = htons(payload_len);
/* Yes, and fold redundant checksum back. 8) */
if (head->ip_summed == CHECKSUM_HW)
head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum);
fq->fragments = NULL;
/* all original skbs are linked into the NFCT_FRAG6_CB(head).orig */
fp = skb_shinfo(head)->frag_list;
if (NFCT_FRAG6_CB(fp)->orig == NULL)
/* at above code, head skb is divided into two skbs. */
fp = fp->next;
op = NFCT_FRAG6_CB(head)->orig;
for (; fp; fp = fp->next) {
struct sk_buff *orig = NFCT_FRAG6_CB(fp)->orig;
op->next = orig;
op = orig;
NFCT_FRAG6_CB(fp)->orig = NULL;
}
return head;
out_oversize:
if (net_ratelimit())
printk(KERN_DEBUG "nf_ct_frag6_reasm: payload len = %d\n", payload_len);
goto out_fail;
out_oom:
if (net_ratelimit())
printk(KERN_DEBUG "nf_ct_frag6_reasm: no memory for reassembly\n");
out_fail:
return NULL;
}
asmlinkage long sys_msgsnd (int msqid, struct msgbuf *msgp, size_t msgsz, int msgflg)
{
struct msg_queue *msq;
struct msg_msg *msg;
long mtype;
int err;
if (msgsz > msg_ctlmax || (long) msgsz < 0 || msqid < 0)
return -EINVAL;
if (get_user(mtype, &msgp->mtype))
return -EFAULT;
if (mtype < 1)
return -EINVAL;
msg = load_msg(msgp->mtext, msgsz);
if(IS_ERR(msg))
return PTR_ERR(msg);
msg->m_type = mtype;
msg->m_ts = msgsz;
msq = msg_lock(msqid);
err=-EINVAL;
if(msq==NULL)
goto out_free;
retry:
err= -EIDRM;
if (msg_checkid(msq,msqid))
goto out_unlock_free;
err=-EACCES;
if (ipcperms(&msq->q_perm, S_IWUGO))
goto out_unlock_free;
if(msgsz + msq->q_cbytes > msq->q_qbytes ||
1 + msq->q_qnum > msq->q_qbytes) {
struct msg_sender s;
if(msgflg&IPC_NOWAIT) {
err=-EAGAIN;
goto out_unlock_free;
}
ss_add(msq, &s);
msg_unlock(msqid);
schedule();
current->state= TASK_RUNNING;
msq = msg_lock(msqid);
err = -EIDRM;
if(msq==NULL)
goto out_free;
ss_del(&s);
if (signal_pending(current)) {
err=-EINTR;
goto out_unlock_free;
}
goto retry;
}
if(!pipelined_send(msq,msg)) {
/* noone is waiting for this message, enqueue it */
list_add_tail(&msg->m_list,&msq->q_messages);
msq->q_cbytes += msgsz;
msq->q_qnum++;
atomic_add(msgsz,&msg_bytes);
atomic_inc(&msg_hdrs);
}
err = 0;
msg = NULL;
msq->q_lspid = current->pid;
msq->q_stime = CURRENT_TIME;
out_unlock_free:
msg_unlock(msqid);
out_free:
if(msg!=NULL)
free_msg(msg);
return err;
}
//BrowserWindow::BrowserWindow( BRect cFrame, std::vector<HistoryEntry*>* pcHistory, bool bLoadPos ) :
BrowserWindow::BrowserWindow( BRect cFrame, bool /*bLoadPos*/ )
:BWindow( cFrame, "ABrowse", B_TITLED_WINDOW, 0 )
,m_pcHTMLPart( NULL )
,m_pcStatusBar( NULL )
,m_pcURLView( NULL )
{
atomic_add( &s_nWndCount, 1 );
// BAutolock al( GlobalMutex::GetMutex() );
// GlobalMutex::Lock();
RootView *pcRoot = new RootView( Bounds() );
AddChild( pcRoot );
// AddChild( pcRoot );
// AddTimer( this, 1, 100000, false );
// GlobalMutex::Unlock();
//--------------------------------------------------------------------------
#if 0
LayoutView* pcFrame = new LayoutView( cWndBounds, "frame_view" );
VLayoutNode* pcRoot = new VLayoutNode( "root" );
FrameView* pcNavBarFrame = new FrameView( Rect(0,0,0,0), "nav_bar_frame", "" );
HLayoutNode* pcNavBar = new HLayoutNode( "nav_bar" );
m_pcURLView = new TextView( Rect(0,0,0,0), "url_view", ""/*g_pzDefaultURL*/ );
m_pcURLView->SetEventMask( os::TextView::EI_ENTER_PRESSED | os::TextView::EI_ESC_PRESSED );
m_pcURLView->SetMultiLine( false );
m_pcURLView->SetMessage( new Message( ID_URL_CHANGED ) );
m_pcToolBar = new ToolBar( Rect(0,0,0,0), "tool_bar" );
m_pcStatusBar = new StatusBar( Rect(), "status_bar" );
m_pcToolBar->AddButton( "hi16-action-back.png", os::Message( ID_PREV_URL ) );
m_pcToolBar->AddButton( "hi16-action-forward.png", os::Message( ID_NEXT_URL ) );
m_pcToolBar->AddButton( "hi16-action-gohome.png", os::Message( ID_GOHOME ) );
m_pcToolBar->AddButton( "hi16-action-reload.png", os::Message( ID_RELOAD ) );
m_pcToolBar->AddButton( "hi16-action-find.png", os::Message( ID_FIND ) );
m_pcToolBar->AddButton( "hi16-action-editcopy.png", os::Message( ID_COPY ) );
pcNavBar->AddChild( m_pcToolBar )->SetBorders( Rect( 4, 2, 10, 2 ) );
pcNavBar->AddChild( m_pcURLView );
pcNavBarFrame->SetRoot( pcNavBar );
pcRoot->AddChild( pcNavBarFrame );
QWidget* pcTopView = new TopView;
pcTopView->show();
pcTopView->SetResizeMask( CF_FOLLOW_ALL );
pcTopView->SetFrame( cWndBounds.Bounds() );
pcRoot->AddChild( pcTopView );
pcRoot->AddChild( m_pcStatusBar );
pcFrame->SetRoot( pcRoot );
m_pcHTMLPart = new KHTMLPart( pcTopView, "khtmlpart" ); // ### frame name
m_pcHTMLPart->browserExtension()->SetCallback( this );
AddChild( pcFrame );
KHTMLView* pcView = m_pcHTMLPart->view();
pcView->SetResizeMask( CF_FOLLOW_ALL );
pcView->SetFrame( pcTopView->GetBounds() );
pcView->MakeFocus();
pcView->SetMsgTarget( Messenger( this ) );
pcView->show();
m_pcURLView->SetTarget( this );
m_pcToolBar->SetTarget( Messenger( this ) );
m_pcToolBar->EnableButton( BI_BACK, false );
m_pcToolBar->EnableButton( BI_FORWARD, false );
m_pcToolBar->EnableButton( BI_COPY, false );
char* pzHome = getenv( "HOME" );
if ( bLoadPos && NULL != pzHome ) {
FILE* hFile;
char zPath[ 256 ];
strcpy( zPath, pzHome );
strcat( zPath, "/config/abrowse.cfg" );
hFile = fopen( zPath, "rb" );
if ( NULL != hFile ) {
Rect cNewFrame;
fread( &cNewFrame, sizeof( cNewFrame ), 1, hFile );
fclose( hFile );
SetFrame( cNewFrame );
}
}
if ( pcHistory != NULL ) {
for ( uint i = 0 ; i < pcHistory->size() ; ++i ) {
HistoryEntry* pcEntry = new HistoryEntry;
// *pcEntry = *(*pcHistory)[i];
pcEntry->m_cURL = (*pcHistory)[i]->m_cURL;
pcEntry->m_cState = (*pcHistory)[i]->m_cState;
m_cHistory.push_back( pcEntry );
}
m_nCurHistoryPos = m_cHistory.size() - 1;
m_pcToolBar->EnableButton( BI_BACK, true );
}
// AddTimer( this, 1, 100000, false );
GlobalMutex::PopLooper();
// GlobalMutex::Unlock();
#endif
}
/**
* The main polling loop
*
* This routine does the polling and despatches of IO events
* to the DCB's. It may be called either directly or as the entry point
* of a polling thread within the gateway.
*
* The routine will loop as long as the variable "shutdown" is set to zero,
* setting this to a non-zero value will cause the polling loop to return.
*
* There are two options for the polling, a debug option that is only useful if
* you have a single thread. This blocks in epoll_wait until an event occurs.
*
* The non-debug option does an epoll with a time out. This allows the checking of
* shutdown value to be checked in all threads. The algorithm for polling in this
* mode is to do a poll with no-wait, if no events are detected then the poll is
* repeated with a time out. This allows for a quick check before making the call
* with timeout. The call with the timeout differs in that the Linux scheduler may
* deschedule a process if a timeout is included, but will not do this if a 0 timeout
* value is given. this improves performance when the gateway is under heavy load.
*
* In order to provide a fairer means of sharing the threads between the different
* DCB's the poll mechanism has been decoupled from the processing of the events.
* The events are now recieved via the epoll_wait call, a queue of DCB's that have
* events pending is maintained and as new events arrive the DCB is added to the end
* of this queue. If an eent arrives for a DCB alreayd in the queue, then the event
* bits are added to the DCB but the DCB mantains the same point in the queue unless
* the original events are already being processed. If they are being processed then
* the DCB is moved to the back of the queue, this means that a DCB that is receiving
* events at a high rate will not block the execution of events for other DCB's and
* should result in a fairer polling strategy.
*
* The introduction of the ability to inject "fake" write events into the event queue meant
* that there was a possibility to "starve" new events sicne the polling loop would
* consume the event queue before looking for new events. If the DCB that inject
* the fake event then injected another fake event as a result of the first it meant
* that new events did not get added to the queue. The strategy has been updated to
* not consume the entire event queue, but process one event before doing a non-blocking
* call to add any new events before processing any more events. A blocking call to
* collect events is only made if there are no pending events to be processed on the
* event queue.
*
* Also introduced a "timeout bias" mechanism. This mechansim control the length of
* of timeout passed to epoll_wait in blocking calls based on previous behaviour.
* The initial call will block for 10% of the define timeout peroid, this will be
* increased in increments of 10% until the full timeout value is used. If at any
* point there is an event to be processed then the value will be reduced to 10% again
* for the next blocking call.
*
* @param arg The thread ID passed as a void * to satisfy the threading package
*/
void
poll_waitevents(void *arg)
{
struct epoll_event events[MAX_EVENTS];
int i, nfds, timeout_bias = 1;
int thread_id = (int)arg;
DCB *zombies = NULL;
int poll_spins = 0;
/** Add this thread to the bitmask of running polling threads */
bitmask_set(&poll_mask, thread_id);
if (thread_data)
{
thread_data[thread_id].state = THREAD_IDLE;
}
/** Init mysql thread context for use with a mysql handle and a parser */
mysql_thread_init();
while (1)
{
if (pollStats.evq_pending == 0 && timeout_bias < 10)
{
timeout_bias++;
}
atomic_add(&n_waiting, 1);
#if BLOCKINGPOLL
nfds = epoll_wait(epoll_fd, events, MAX_EVENTS, -1);
atomic_add(&n_waiting, -1);
#else /* BLOCKINGPOLL */
#if MUTEX_EPOLL
simple_mutex_lock(&epoll_wait_mutex, TRUE);
#endif
if (thread_data)
{
thread_data[thread_id].state = THREAD_POLLING;
}
atomic_add(&pollStats.n_polls, 1);
if ((nfds = epoll_wait(epoll_fd, events, MAX_EVENTS, 0)) == -1)
{
atomic_add(&n_waiting, -1);
int eno = errno;
errno = 0;
LOGIF(LD, (skygw_log_write(
LOGFILE_DEBUG,
"%lu [poll_waitevents] epoll_wait returned "
"%d, errno %d",
pthread_self(),
nfds,
eno)));
atomic_add(&n_waiting, -1);
}
/*
* If there are no new descriptors from the non-blocking call
* and nothing to process on the event queue then for do a
* blocking call to epoll_wait.
*
* We calculate a timeout bias to alter the length of the blocking
* call based on the time since we last received an event to process
*/
else if (nfds == 0 && pollStats.evq_pending == 0 && poll_spins++ > number_poll_spins)
{
atomic_add(&pollStats.blockingpolls, 1);
nfds = epoll_wait(epoll_fd,
events,
MAX_EVENTS,
(max_poll_sleep * timeout_bias) / 10);
if (nfds == 0 && pollStats.evq_pending)
{
atomic_add(&pollStats.wake_evqpending, 1);
poll_spins = 0;
}
}
else
{
atomic_add(&n_waiting, -1);
}
if (n_waiting == 0)
atomic_add(&pollStats.n_nothreads, 1);
#if MUTEX_EPOLL
simple_mutex_unlock(&epoll_wait_mutex);
#endif
#endif /* BLOCKINGPOLL */
if (nfds > 0)
{
timeout_bias = 1;
if (poll_spins <= number_poll_spins + 1)
atomic_add(&pollStats.n_nbpollev, 1);
poll_spins = 0;
LOGIF(LD, (skygw_log_write(
LOGFILE_DEBUG,
"%lu [poll_waitevents] epoll_wait found %d fds",
pthread_self(),
nfds)));
atomic_add(&pollStats.n_pollev, 1);
if (thread_data)
{
thread_data[thread_id].n_fds = nfds;
thread_data[thread_id].cur_dcb = NULL;
thread_data[thread_id].event = 0;
thread_data[thread_id].state = THREAD_PROCESSING;
}
pollStats.n_fds[(nfds < MAXNFDS ? (nfds - 1) : MAXNFDS - 1)]++;
load_average = (load_average * load_samples + nfds)
/ (load_samples + 1);
atomic_add(&load_samples, 1);
atomic_add(&load_nfds, nfds);
/*
* Process every DCB that has a new event and add
* it to the poll queue.
* If the DCB is currently being processed then we
* or in the new eent bits to the pending event bits
* and leave it in the queue.
* If the DCB was not already in the queue then it was
* idle and is added to the queue to process after
* setting the event bits.
*/
for (i = 0; i < nfds; i++)
{
DCB *dcb = (DCB *)events[i].data.ptr;
__uint32_t ev = events[i].events;
spinlock_acquire(&pollqlock);
if (DCB_POLL_BUSY(dcb))
{
if (dcb->evq.pending_events == 0)
{
pollStats.evq_pending++;
dcb->evq.inserted = hkheartbeat;
}
dcb->evq.pending_events |= ev;
}
else
{
dcb->evq.pending_events = ev;
if (eventq)
{
dcb->evq.prev = eventq->evq.prev;
eventq->evq.prev->evq.next = dcb;
eventq->evq.prev = dcb;
dcb->evq.next = eventq;
}
else
{
eventq = dcb;
dcb->evq.prev = dcb;
dcb->evq.next = dcb;
}
pollStats.evq_length++;
pollStats.evq_pending++;
dcb->evq.inserted = hkheartbeat;
if (pollStats.evq_length > pollStats.evq_max)
{
pollStats.evq_max = pollStats.evq_length;
}
}
spinlock_release(&pollqlock);
}
}
/*
* Process of the queue of waiting requests
* This is done without checking the evq_pending count as a
* precautionary measure to avoid issues if the house keeping
* of the count goes wrong.
*/
if (process_pollq(thread_id))
timeout_bias = 1;
if (thread_data)
thread_data[thread_id].state = THREAD_ZPROCESSING;
zombies = dcb_process_zombies(thread_id);
if (thread_data)
thread_data[thread_id].state = THREAD_IDLE;
if (do_shutdown)
{
/*<
* Remove the thread from the bitmask of running
* polling threads.
*/
if (thread_data)
{
thread_data[thread_id].state = THREAD_STOPPED;
}
bitmask_clear(&poll_mask, thread_id);
/** Release mysql thread context */
mysql_thread_end();
return;
}
if (thread_data)
{
thread_data[thread_id].state = THREAD_IDLE;
}
} /*< while(1) */
}
static int host_cmd_motion_sense(struct host_cmd_handler_args *args)
{
const struct ec_params_motion_sense *in = args->params;
struct ec_response_motion_sense *out = args->response;
struct motion_sensor_t *sensor;
int i, ret = EC_RES_INVALID_PARAM, reported;
switch (in->cmd) {
case MOTIONSENSE_CMD_DUMP:
out->dump.module_flags =
(*(host_get_memmap(EC_MEMMAP_ACC_STATUS)) &
EC_MEMMAP_ACC_STATUS_PRESENCE_BIT) ?
MOTIONSENSE_MODULE_FLAG_ACTIVE : 0;
out->dump.sensor_count = motion_sensor_count;
args->response_size = sizeof(out->dump);
reported = MIN(motion_sensor_count, in->dump.max_sensor_count);
mutex_lock(&g_sensor_mutex);
for (i = 0; i < reported; i++) {
sensor = &motion_sensors[i];
out->dump.sensor[i].flags =
MOTIONSENSE_SENSOR_FLAG_PRESENT;
/* casting from int to s16 */
out->dump.sensor[i].data[X] = sensor->xyz[X];
out->dump.sensor[i].data[Y] = sensor->xyz[Y];
out->dump.sensor[i].data[Z] = sensor->xyz[Z];
}
mutex_unlock(&g_sensor_mutex);
args->response_size += reported *
sizeof(struct ec_response_motion_sensor_data);
break;
case MOTIONSENSE_CMD_DATA:
sensor = host_sensor_id_to_motion_sensor(
in->sensor_odr.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
out->data.flags = 0;
mutex_lock(&g_sensor_mutex);
out->data.data[X] = sensor->xyz[X];
out->data.data[Y] = sensor->xyz[Y];
out->data.data[Z] = sensor->xyz[Z];
mutex_unlock(&g_sensor_mutex);
args->response_size = sizeof(out->data);
break;
case MOTIONSENSE_CMD_INFO:
sensor = host_sensor_id_to_motion_sensor(
in->sensor_odr.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
out->info.type = sensor->type;
out->info.location = sensor->location;
out->info.chip = sensor->chip;
args->response_size = sizeof(out->info);
break;
case MOTIONSENSE_CMD_EC_RATE:
sensor = host_sensor_id_to_motion_sensor(
in->sensor_odr.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
/*
* Set new sensor sampling rate when AP is on, if the data arg
* has a value.
*/
if (in->ec_rate.data != EC_MOTION_SENSE_NO_VALUE) {
if (in->ec_rate.data == 0)
sensor->config[SENSOR_CONFIG_AP].ec_rate = 0;
else
sensor->config[SENSOR_CONFIG_AP].ec_rate =
MAX(in->ec_rate.data,
MIN_MOTION_SENSE_WAIT_TIME / MSEC);
/* Bound the new sampling rate. */
motion_sense_set_accel_interval();
}
out->ec_rate.ret = motion_sense_ec_rate(sensor) / MSEC;
args->response_size = sizeof(out->ec_rate);
break;
case MOTIONSENSE_CMD_SENSOR_ODR:
/* Verify sensor number is valid. */
sensor = host_sensor_id_to_motion_sensor(
in->sensor_odr.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
/* Set new data rate if the data arg has a value. */
if (in->sensor_odr.data != EC_MOTION_SENSE_NO_VALUE) {
sensor->config[SENSOR_CONFIG_AP].odr =
in->sensor_odr.data |
(in->sensor_odr.roundup ? ROUND_UP_FLAG : 0);
ret = motion_sense_set_data_rate(sensor);
if (ret != EC_SUCCESS)
return EC_RES_INVALID_PARAM;
/*
* To be sure timestamps are calculated properly,
* Send an event to have a timestamp inserted in the
* FIFO.
*/
task_set_event(TASK_ID_MOTIONSENSE,
TASK_EVENT_MOTION_ODR_CHANGE, 0);
/*
* If the sensor was suspended before, or now
* suspended, we have to recalculate the EC sampling
* rate
*/
motion_sense_set_accel_interval();
}
out->sensor_odr.ret = sensor->drv->get_data_rate(sensor);
args->response_size = sizeof(out->sensor_odr);
break;
case MOTIONSENSE_CMD_SENSOR_RANGE:
/* Verify sensor number is valid. */
sensor = host_sensor_id_to_motion_sensor(
in->sensor_range.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
/* Set new range if the data arg has a value. */
if (in->sensor_range.data != EC_MOTION_SENSE_NO_VALUE) {
if (sensor->drv->set_range(sensor,
in->sensor_range.data,
in->sensor_range.roundup)
!= EC_SUCCESS) {
return EC_RES_INVALID_PARAM;
}
}
out->sensor_range.ret = sensor->drv->get_range(sensor);
args->response_size = sizeof(out->sensor_range);
break;
case MOTIONSENSE_CMD_SENSOR_OFFSET:
/* Verify sensor number is valid. */
sensor = host_sensor_id_to_motion_sensor(
in->sensor_offset.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
/* Set new range if the data arg has a value. */
if (in->sensor_offset.flags & MOTION_SENSE_SET_OFFSET) {
ret = sensor->drv->set_offset(sensor,
in->sensor_offset.offset,
in->sensor_offset.temp);
if (ret != EC_SUCCESS)
return ret;
}
ret = sensor->drv->get_offset(sensor, out->sensor_offset.offset,
&out->sensor_offset.temp);
if (ret != EC_SUCCESS)
return ret;
args->response_size = sizeof(out->sensor_offset);
break;
case MOTIONSENSE_CMD_PERFORM_CALIB:
/* Verify sensor number is valid. */
sensor = host_sensor_id_to_motion_sensor(
in->sensor_offset.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
if (!sensor->drv->perform_calib)
return EC_RES_INVALID_COMMAND;
ret = sensor->drv->perform_calib(sensor);
if (ret != EC_SUCCESS)
return ret;
ret = sensor->drv->get_offset(sensor, out->sensor_offset.offset,
&out->sensor_offset.temp);
if (ret != EC_SUCCESS)
return ret;
args->response_size = sizeof(out->sensor_offset);
break;
#ifdef CONFIG_ACCEL_FIFO
case MOTIONSENSE_CMD_FIFO_FLUSH:
sensor = host_sensor_id_to_motion_sensor(
in->sensor_odr.sensor_num);
if (sensor == NULL)
return EC_RES_INVALID_PARAM;
atomic_add(&sensor->flush_pending, 1);
task_set_event(TASK_ID_MOTIONSENSE,
TASK_EVENT_MOTION_FLUSH_PENDING, 0);
/* passthrough */
case MOTIONSENSE_CMD_FIFO_INFO:
motion_sense_get_fifo_info(&out->fifo_info);
for (i = 0; i < motion_sensor_count; i++) {
out->fifo_info.lost[i] = motion_sensors[i].lost;
motion_sensors[i].lost = 0;
}
motion_sense_fifo_lost = 0;
args->response_size = sizeof(out->fifo_info) +
sizeof(uint16_t) * motion_sensor_count;
break;
case MOTIONSENSE_CMD_FIFO_READ:
mutex_lock(&g_sensor_mutex);
reported = MIN((args->response_max - sizeof(out->fifo_read)) /
motion_sense_fifo.unit_bytes,
MIN(queue_count(&motion_sense_fifo),
in->fifo_read.max_data_vector));
reported = queue_remove_units(&motion_sense_fifo,
out->fifo_read.data, reported);
mutex_unlock(&g_sensor_mutex);
out->fifo_read.number_data = reported;
args->response_size = sizeof(out->fifo_read) + reported *
motion_sense_fifo.unit_bytes;
break;
#else
case MOTIONSENSE_CMD_FIFO_INFO:
/* Only support the INFO command, to tell there is no FIFO. */
memset(&out->fifo_info, 0, sizeof(out->fifo_info));
args->response_size = sizeof(out->fifo_info);
break;
#endif
default:
/* Call other users of the motion task */
#ifdef CONFIG_LID_ANGLE
if (ret == EC_RES_INVALID_PARAM)
ret = host_cmd_motion_lid(args);
#endif
return ret;
}
return EC_RES_SUCCESS;
}
INLINE friend T operator++ (AtomicInternal& value) { return atomic_add(&value.data, 1) + 1; }
INLINE friend T operator-- (AtomicInternal& value, int) { return atomic_add(&value.data, -1); }
inline void ATOMIC_ADD(ATOMIC_T *v, int i)
{
atomic_add(i, v);
}
INLINE friend T operator+= (AtomicInternal& value, T input) { return atomic_add(&value.data, input) + input; }
/* Test various atomic.h macros. */
static int
do_test (void)
{
atomic_t mem;
int ret = 0;
#ifdef atomic_compare_and_exchange_val_acq
mem = 24;
if (atomic_compare_and_exchange_val_acq (&mem, 35, 24) != 24
|| mem != 35)
{
puts ("atomic_compare_and_exchange_val_acq test 1 failed");
ret = 1;
}
mem = 12;
if (atomic_compare_and_exchange_val_acq (&mem, 10, 15) != 12
|| mem != 12)
{
puts ("atomic_compare_and_exchange_val_acq test 2 failed");
ret = 1;
}
mem = -15;
if (atomic_compare_and_exchange_val_acq (&mem, -56, -15) != -15
|| mem != -56)
{
puts ("atomic_compare_and_exchange_val_acq test 3 failed");
ret = 1;
}
mem = -1;
if (atomic_compare_and_exchange_val_acq (&mem, 17, 0) != -1
|| mem != -1)
{
puts ("atomic_compare_and_exchange_val_acq test 4 failed");
ret = 1;
}
#endif
mem = 24;
if (atomic_compare_and_exchange_bool_acq (&mem, 35, 24)
|| mem != 35)
{
puts ("atomic_compare_and_exchange_bool_acq test 1 failed");
ret = 1;
}
mem = 12;
if (! atomic_compare_and_exchange_bool_acq (&mem, 10, 15)
|| mem != 12)
{
puts ("atomic_compare_and_exchange_bool_acq test 2 failed");
ret = 1;
}
mem = -15;
if (atomic_compare_and_exchange_bool_acq (&mem, -56, -15)
|| mem != -56)
{
puts ("atomic_compare_and_exchange_bool_acq test 3 failed");
ret = 1;
}
mem = -1;
if (! atomic_compare_and_exchange_bool_acq (&mem, 17, 0)
|| mem != -1)
{
puts ("atomic_compare_and_exchange_bool_acq test 4 failed");
ret = 1;
}
mem = 64;
if (atomic_exchange_acq (&mem, 31) != 64
|| mem != 31)
{
puts ("atomic_exchange_acq test failed");
ret = 1;
}
mem = 2;
if (atomic_exchange_and_add (&mem, 11) != 2
|| mem != 13)
{
puts ("atomic_exchange_and_add test failed");
ret = 1;
}
mem = -21;
atomic_add (&mem, 22);
if (mem != 1)
{
puts ("atomic_add test failed");
ret = 1;
}
mem = -1;
atomic_increment (&mem);
if (mem != 0)
{
puts ("atomic_increment test failed");
ret = 1;
}
mem = 2;
if (atomic_increment_val (&mem) != 3)
{
puts ("atomic_increment_val test failed");
ret = 1;
}
mem = 0;
if (atomic_increment_and_test (&mem)
|| mem != 1)
{
puts ("atomic_increment_and_test test 1 failed");
ret = 1;
}
mem = 35;
if (atomic_increment_and_test (&mem)
|| mem != 36)
{
puts ("atomic_increment_and_test test 2 failed");
ret = 1;
}
mem = -1;
if (! atomic_increment_and_test (&mem)
|| mem != 0)
{
puts ("atomic_increment_and_test test 3 failed");
ret = 1;
}
mem = 17;
atomic_decrement (&mem);
if (mem != 16)
{
puts ("atomic_decrement test failed");
ret = 1;
}
if (atomic_decrement_val (&mem) != 15)
{
puts ("atomic_decrement_val test failed");
ret = 1;
}
mem = 0;
if (atomic_decrement_and_test (&mem)
|| mem != -1)
{
puts ("atomic_decrement_and_test test 1 failed");
ret = 1;
}
mem = 15;
if (atomic_decrement_and_test (&mem)
|| mem != 14)
{
puts ("atomic_decrement_and_test test 2 failed");
ret = 1;
}
mem = 1;
if (! atomic_decrement_and_test (&mem)
|| mem != 0)
{
puts ("atomic_decrement_and_test test 3 failed");
ret = 1;
}
mem = 1;
if (atomic_decrement_if_positive (&mem) != 1
|| mem != 0)
{
puts ("atomic_decrement_if_positive test 1 failed");
ret = 1;
}
mem = 0;
if (atomic_decrement_if_positive (&mem) != 0
|| mem != 0)
{
puts ("atomic_decrement_if_positive test 2 failed");
ret = 1;
}
mem = -1;
if (atomic_decrement_if_positive (&mem) != -1
|| mem != -1)
{
puts ("atomic_decrement_if_positive test 3 failed");
ret = 1;
}
mem = -12;
if (! atomic_add_negative (&mem, 10)
|| mem != -2)
{
puts ("atomic_add_negative test 1 failed");
ret = 1;
}
mem = 0;
if (atomic_add_negative (&mem, 100)
|| mem != 100)
{
puts ("atomic_add_negative test 2 failed");
ret = 1;
}
mem = 15;
if (atomic_add_negative (&mem, -10)
|| mem != 5)
{
puts ("atomic_add_negative test 3 failed");
ret = 1;
}
mem = -12;
if (atomic_add_negative (&mem, 14)
|| mem != 2)
{
puts ("atomic_add_negative test 4 failed");
ret = 1;
}
mem = 0;
if (! atomic_add_negative (&mem, -1)
|| mem != -1)
{
puts ("atomic_add_negative test 5 failed");
ret = 1;
}
mem = -31;
if (atomic_add_negative (&mem, 31)
|| mem != 0)
{
puts ("atomic_add_negative test 6 failed");
ret = 1;
}
mem = -34;
if (atomic_add_zero (&mem, 31)
|| mem != -3)
{
puts ("atomic_add_zero test 1 failed");
ret = 1;
}
mem = -36;
if (! atomic_add_zero (&mem, 36)
|| mem != 0)
{
puts ("atomic_add_zero test 2 failed");
ret = 1;
}
mem = 113;
if (atomic_add_zero (&mem, -13)
|| mem != 100)
{
puts ("atomic_add_zero test 3 failed");
ret = 1;
}
mem = -18;
if (atomic_add_zero (&mem, 20)
|| mem != 2)
{
puts ("atomic_add_zero test 4 failed");
ret = 1;
}
mem = 10;
if (atomic_add_zero (&mem, -20)
|| mem != -10)
{
puts ("atomic_add_zero test 5 failed");
ret = 1;
}
mem = 10;
if (! atomic_add_zero (&mem, -10)
|| mem != 0)
{
puts ("atomic_add_zero test 6 failed");
ret = 1;
}
mem = 0;
atomic_bit_set (&mem, 1);
if (mem != 2)
{
puts ("atomic_bit_set test 1 failed");
ret = 1;
}
mem = 8;
atomic_bit_set (&mem, 3);
if (mem != 8)
{
puts ("atomic_bit_set test 2 failed");
ret = 1;
}
#ifdef TEST_ATOMIC64
mem = 16;
atomic_bit_set (&mem, 35);
if (mem != 0x800000010LL)
{
puts ("atomic_bit_set test 3 failed");
ret = 1;
}
#endif
mem = 0;
if (atomic_bit_test_set (&mem, 1)
|| mem != 2)
{
puts ("atomic_bit_test_set test 1 failed");
ret = 1;
}
mem = 8;
if (! atomic_bit_test_set (&mem, 3)
|| mem != 8)
{
puts ("atomic_bit_test_set test 2 failed");
ret = 1;
}
#ifdef TEST_ATOMIC64
mem = 16;
if (atomic_bit_test_set (&mem, 35)
|| mem != 0x800000010LL)
{
puts ("atomic_bit_test_set test 3 failed");
ret = 1;
}
mem = 0x100000000LL;
if (! atomic_bit_test_set (&mem, 32)
|| mem != 0x100000000LL)
{
puts ("atomic_bit_test_set test 4 failed");
ret = 1;
}
#endif
#ifdef catomic_compare_and_exchange_val_acq
mem = 24;
if (catomic_compare_and_exchange_val_acq (&mem, 35, 24) != 24
|| mem != 35)
{
puts ("catomic_compare_and_exchange_val_acq test 1 failed");
ret = 1;
}
mem = 12;
if (catomic_compare_and_exchange_val_acq (&mem, 10, 15) != 12
|| mem != 12)
{
puts ("catomic_compare_and_exchange_val_acq test 2 failed");
ret = 1;
}
mem = -15;
if (catomic_compare_and_exchange_val_acq (&mem, -56, -15) != -15
|| mem != -56)
{
puts ("catomic_compare_and_exchange_val_acq test 3 failed");
ret = 1;
}
mem = -1;
if (catomic_compare_and_exchange_val_acq (&mem, 17, 0) != -1
|| mem != -1)
{
puts ("catomic_compare_and_exchange_val_acq test 4 failed");
ret = 1;
}
#endif
mem = 24;
if (catomic_compare_and_exchange_bool_acq (&mem, 35, 24)
|| mem != 35)
{
puts ("catomic_compare_and_exchange_bool_acq test 1 failed");
ret = 1;
}
mem = 12;
if (! catomic_compare_and_exchange_bool_acq (&mem, 10, 15)
|| mem != 12)
{
puts ("catomic_compare_and_exchange_bool_acq test 2 failed");
ret = 1;
}
mem = -15;
if (catomic_compare_and_exchange_bool_acq (&mem, -56, -15)
|| mem != -56)
{
puts ("catomic_compare_and_exchange_bool_acq test 3 failed");
ret = 1;
}
mem = -1;
if (! catomic_compare_and_exchange_bool_acq (&mem, 17, 0)
|| mem != -1)
{
puts ("catomic_compare_and_exchange_bool_acq test 4 failed");
ret = 1;
}
mem = 2;
if (catomic_exchange_and_add (&mem, 11) != 2
|| mem != 13)
{
puts ("catomic_exchange_and_add test failed");
ret = 1;
}
mem = -21;
catomic_add (&mem, 22);
if (mem != 1)
{
puts ("catomic_add test failed");
ret = 1;
}
mem = -1;
catomic_increment (&mem);
if (mem != 0)
{
puts ("catomic_increment test failed");
ret = 1;
}
mem = 2;
if (catomic_increment_val (&mem) != 3)
{
puts ("catomic_increment_val test failed");
ret = 1;
}
mem = 17;
catomic_decrement (&mem);
if (mem != 16)
{
puts ("catomic_decrement test failed");
ret = 1;
}
if (catomic_decrement_val (&mem) != 15)
{
puts ("catomic_decrement_val test failed");
ret = 1;
}
return ret;
}
static int camera_v4l2_open(struct file *filep)
{
int rc = 0;
struct v4l2_event event;
struct msm_video_device *pvdev = video_drvdata(filep);
struct task_struct *qdaemon_task;
BUG_ON(!pvdev);
rc = camera_v4l2_fh_open(filep);
if (rc < 0) {
pr_err("%s : camera_v4l2_fh_open", __FUNCTION__);
goto fh_open_fail;
}
/* every stream has a vb2 queue */
rc = camera_v4l2_vb2_q_init(filep);
if (rc < 0) {
pr_err("%s : camera_v4l2_vb2_q_init", __FUNCTION__);
goto vb2_q_fail;
}
if (!atomic_read(&pvdev->opened)) {
/* create a new session when first opened */
rc = msm_create_session(pvdev->vdev->num, pvdev->vdev);
if (rc < 0) {
pr_err("%s : msm_create_session", __func__);
goto session_fail;
}
rc = msm_create_command_ack_q(pvdev->vdev->num, 0);
if (rc < 0) {
pr_err("%s : msm_create_command_ack_q", __FUNCTION__);
goto command_ack_q_fail;
}
camera_pack_event(filep, MSM_CAMERA_NEW_SESSION, 0, -1, &event);
rc = msm_post_event(&event, MSM_POST_EVT_TIMEOUT);
if (rc < 0) {
pr_err("%s, __dbg: post fail \n",__FUNCTION__);
pr_err("%s, pid: %d, tgid: %d\n",__FUNCTION__, qdaemon_pid, qdaemon_tgid);
qdaemon_task = pid_task(find_get_pid(qdaemon_tgid), PIDTYPE_PID);
if (qdaemon_task) {
if (!strncmp(qdaemon_task->comm, QDAEMON, strlen(QDAEMON))) {
pr_err("%s, kill daemon", __func__);
send_sig(SIGKILL, qdaemon_task, 0);
pr_err("%s, kill this", __func__);
send_sig(SIGKILL, current, 0);
} else
pr_err("%s, now (%s : %d)", __func__,
qdaemon_task->comm, task_pid_nr(qdaemon_task));
} else
pr_err("error!! can't look for daemon");
goto post_fail;
}
rc = camera_check_event_status(&event);
if (rc < 0) {
pr_err("%s : camera_check_event_status", __FUNCTION__);
goto post_fail;
}
} else {
rc = msm_create_command_ack_q(pvdev->vdev->num,
atomic_read(&pvdev->opened));
if (rc < 0) {
pr_err("%s : msm_create_command_ack_q", __FUNCTION__);
goto session_fail;
}
}
atomic_add(1, &pvdev->opened);
return rc;
post_fail:
msm_delete_command_ack_q(pvdev->vdev->num, 0);
command_ack_q_fail:
msm_destroy_session(pvdev->vdev->num);
session_fail:
camera_v4l2_vb2_q_release(filep);
vb2_q_fail:
camera_v4l2_fh_release(filep);
fh_open_fail:
return rc;
}
void
BDirectMessageTarget::Release()
{
if (atomic_add(&fReferenceCount, -1) == 1)
delete this;
}
static int send_via_shortcut(struct sk_buff *skb, struct mpoa_client *mpc)
{
in_cache_entry *entry;
struct iphdr *iph;
char *buff;
__be32 ipaddr = 0;
static struct {
struct llc_snap_hdr hdr;
__be32 tag;
} tagged_llc_snap_hdr = {
{0xaa, 0xaa, 0x03, {0x00, 0x00, 0x00}, {0x88, 0x4c}},
0
};
buff = skb->data + mpc->dev->hard_header_len;
iph = (struct iphdr *)buff;
ipaddr = iph->daddr;
ddprintk("(%s) ipaddr 0x%x\n",
mpc->dev->name, ipaddr);
entry = mpc->in_ops->get(ipaddr, mpc);
if (entry == NULL) {
entry = mpc->in_ops->add_entry(ipaddr, mpc);
if (entry != NULL)
mpc->in_ops->put(entry);
return 1;
}
/* threshold not exceeded or VCC not ready */
if (mpc->in_ops->cache_hit(entry, mpc) != OPEN) {
ddprintk("(%s) cache_hit: returns != OPEN\n",
mpc->dev->name);
mpc->in_ops->put(entry);
return 1;
}
ddprintk("(%s) using shortcut\n",
mpc->dev->name);
/* MPOA spec A.1.4, MPOA client must decrement IP ttl at least by one */
if (iph->ttl <= 1) {
ddprintk("(%s) IP ttl = %u, using LANE\n",
mpc->dev->name, iph->ttl);
mpc->in_ops->put(entry);
return 1;
}
iph->ttl--;
iph->check = 0;
iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
if (entry->ctrl_info.tag != 0) {
ddprintk("(%s) adding tag 0x%x\n",
mpc->dev->name, entry->ctrl_info.tag);
tagged_llc_snap_hdr.tag = entry->ctrl_info.tag;
skb_pull(skb, ETH_HLEN); /* get rid of Eth header */
skb_push(skb, sizeof(tagged_llc_snap_hdr));
/* add LLC/SNAP header */
skb_copy_to_linear_data(skb, &tagged_llc_snap_hdr,
sizeof(tagged_llc_snap_hdr));
} else {
skb_pull(skb, ETH_HLEN); /* get rid of Eth header */
skb_push(skb, sizeof(struct llc_snap_hdr));
/* add LLC/SNAP header + tag */
skb_copy_to_linear_data(skb, &llc_snap_mpoa_data,
sizeof(struct llc_snap_hdr));
}
atomic_add(skb->truesize, &sk_atm(entry->shortcut)->sk_wmem_alloc);
ATM_SKB(skb)->atm_options = entry->shortcut->atm_options;
entry->shortcut->send(entry->shortcut, skb);
entry->packets_fwded++;
mpc->in_ops->put(entry);
return 0;
}
/**
* Release a previously obtained readlock.
*
* Simply decrement the n_readers value for the hash table
*
* @param table The hash table to unlock
*/
static void
hashtable_read_unlock(HASHTABLE *table)
{
atomic_add(&table->n_readers, -1);
}
static int __ip6_append_data(struct sock *sk,
struct flowi6 *fl6,
struct sk_buff_head *queue,
struct inet_cork *cork,
struct inet6_cork *v6_cork,
struct page_frag *pfrag,
int getfrag(void *from, char *to, int offset,
int len, int odd, struct sk_buff *skb),
void *from, int length, int transhdrlen,
unsigned int flags, struct ipcm6_cookie *ipc6,
const struct sockcm_cookie *sockc)
{
struct sk_buff *skb, *skb_prev = NULL;
unsigned int maxfraglen, fragheaderlen, mtu, orig_mtu;
int exthdrlen = 0;
int dst_exthdrlen = 0;
int hh_len;
int copy;
int err;
int offset = 0;
__u8 tx_flags = 0;
u32 tskey = 0;
struct rt6_info *rt = (struct rt6_info *)cork->dst;
struct ipv6_txoptions *opt = v6_cork->opt;
int csummode = CHECKSUM_NONE;
unsigned int maxnonfragsize, headersize;
skb = skb_peek_tail(queue);
if (!skb) {
exthdrlen = opt ? opt->opt_flen : 0;
dst_exthdrlen = rt->dst.header_len - rt->rt6i_nfheader_len;
}
mtu = cork->fragsize;
orig_mtu = mtu;
hh_len = LL_RESERVED_SPACE(rt->dst.dev);
fragheaderlen = sizeof(struct ipv6hdr) + rt->rt6i_nfheader_len +
(opt ? opt->opt_nflen : 0);
maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen -
sizeof(struct frag_hdr);
headersize = sizeof(struct ipv6hdr) +
(opt ? opt->opt_flen + opt->opt_nflen : 0) +
(dst_allfrag(&rt->dst) ?
sizeof(struct frag_hdr) : 0) +
rt->rt6i_nfheader_len;
if (cork->length + length > mtu - headersize && ipc6->dontfrag &&
(sk->sk_protocol == IPPROTO_UDP ||
sk->sk_protocol == IPPROTO_RAW)) {
ipv6_local_rxpmtu(sk, fl6, mtu - headersize +
sizeof(struct ipv6hdr));
goto emsgsize;
}
if (ip6_sk_ignore_df(sk))
maxnonfragsize = sizeof(struct ipv6hdr) + IPV6_MAXPLEN;
else
maxnonfragsize = mtu;
if (cork->length + length > maxnonfragsize - headersize) {
emsgsize:
ipv6_local_error(sk, EMSGSIZE, fl6,
mtu - headersize +
sizeof(struct ipv6hdr));
return -EMSGSIZE;
}
/* CHECKSUM_PARTIAL only with no extension headers and when
* we are not going to fragment
*/
if (transhdrlen && sk->sk_protocol == IPPROTO_UDP &&
headersize == sizeof(struct ipv6hdr) &&
length < mtu - headersize &&
!(flags & MSG_MORE) &&
rt->dst.dev->features & (NETIF_F_IPV6_CSUM | NETIF_F_HW_CSUM))
csummode = CHECKSUM_PARTIAL;
if (sk->sk_type == SOCK_DGRAM || sk->sk_type == SOCK_RAW) {
sock_tx_timestamp(sk, sockc->tsflags, &tx_flags);
if (tx_flags & SKBTX_ANY_SW_TSTAMP &&
sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID)
tskey = sk->sk_tskey++;
}
/*
* Let's try using as much space as possible.
* Use MTU if total length of the message fits into the MTU.
* Otherwise, we need to reserve fragment header and
* fragment alignment (= 8-15 octects, in total).
*
* Note that we may need to "move" the data from the tail of
* of the buffer to the new fragment when we split
* the message.
*
* FIXME: It may be fragmented into multiple chunks
* at once if non-fragmentable extension headers
* are too large.
* --yoshfuji
*/
cork->length += length;
if (((length > mtu) ||
(skb && skb_is_gso(skb))) &&
(sk->sk_protocol == IPPROTO_UDP) &&
(rt->dst.dev->features & NETIF_F_UFO) && !rt->dst.header_len &&
(sk->sk_type == SOCK_DGRAM) && !udp_get_no_check6_tx(sk)) {
err = ip6_ufo_append_data(sk, queue, getfrag, from, length,
hh_len, fragheaderlen, exthdrlen,
transhdrlen, mtu, flags, fl6);
if (err)
goto error;
return 0;
}
if (!skb)
goto alloc_new_skb;
while (length > 0) {
/* Check if the remaining data fits into current packet. */
copy = (cork->length <= mtu && !(cork->flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - skb->len;
if (copy < length)
copy = maxfraglen - skb->len;
if (copy <= 0) {
char *data;
unsigned int datalen;
unsigned int fraglen;
unsigned int fraggap;
unsigned int alloclen;
alloc_new_skb:
/* There's no room in the current skb */
if (skb)
fraggap = skb->len - maxfraglen;
else
fraggap = 0;
/* update mtu and maxfraglen if necessary */
if (!skb || !skb_prev)
ip6_append_data_mtu(&mtu, &maxfraglen,
fragheaderlen, skb, rt,
orig_mtu);
skb_prev = skb;
/*
* If remaining data exceeds the mtu,
* we know we need more fragment(s).
*/
datalen = length + fraggap;
if (datalen > (cork->length <= mtu && !(cork->flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - fragheaderlen)
datalen = maxfraglen - fragheaderlen - rt->dst.trailer_len;
if ((flags & MSG_MORE) &&
!(rt->dst.dev->features&NETIF_F_SG))
alloclen = mtu;
else
alloclen = datalen + fragheaderlen;
alloclen += dst_exthdrlen;
if (datalen != length + fraggap) {
/*
* this is not the last fragment, the trailer
* space is regarded as data space.
*/
datalen += rt->dst.trailer_len;
}
alloclen += rt->dst.trailer_len;
fraglen = datalen + fragheaderlen;
/*
* We just reserve space for fragment header.
* Note: this may be overallocation if the message
* (without MSG_MORE) fits into the MTU.
*/
alloclen += sizeof(struct frag_hdr);
if (transhdrlen) {
skb = sock_alloc_send_skb(sk,
alloclen + hh_len,
(flags & MSG_DONTWAIT), &err);
} else {
skb = NULL;
if (atomic_read(&sk->sk_wmem_alloc) <=
2 * sk->sk_sndbuf)
skb = sock_wmalloc(sk,
alloclen + hh_len, 1,
sk->sk_allocation);
if (unlikely(!skb))
err = -ENOBUFS;
}
if (!skb)
goto error;
/*
* Fill in the control structures
*/
skb->protocol = htons(ETH_P_IPV6);
skb->ip_summed = csummode;
skb->csum = 0;
/* reserve for fragmentation and ipsec header */
skb_reserve(skb, hh_len + sizeof(struct frag_hdr) +
dst_exthdrlen);
/* Only the initial fragment is time stamped */
skb_shinfo(skb)->tx_flags = tx_flags;
tx_flags = 0;
skb_shinfo(skb)->tskey = tskey;
tskey = 0;
/*
* Find where to start putting bytes
*/
data = skb_put(skb, fraglen);
skb_set_network_header(skb, exthdrlen);
data += fragheaderlen;
skb->transport_header = (skb->network_header +
fragheaderlen);
if (fraggap) {
skb->csum = skb_copy_and_csum_bits(
skb_prev, maxfraglen,
data + transhdrlen, fraggap, 0);
skb_prev->csum = csum_sub(skb_prev->csum,
skb->csum);
data += fraggap;
pskb_trim_unique(skb_prev, maxfraglen);
}
copy = datalen - transhdrlen - fraggap;
if (copy < 0) {
err = -EINVAL;
kfree_skb(skb);
goto error;
} else if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, fraggap, skb) < 0) {
err = -EFAULT;
kfree_skb(skb);
goto error;
}
offset += copy;
length -= datalen - fraggap;
transhdrlen = 0;
exthdrlen = 0;
dst_exthdrlen = 0;
/*
* Put the packet on the pending queue
*/
__skb_queue_tail(queue, skb);
continue;
}
if (copy > length)
copy = length;
if (!(rt->dst.dev->features&NETIF_F_SG)) {
unsigned int off;
off = skb->len;
if (getfrag(from, skb_put(skb, copy),
offset, copy, off, skb) < 0) {
__skb_trim(skb, off);
err = -EFAULT;
goto error;
}
} else {
int i = skb_shinfo(skb)->nr_frags;
err = -ENOMEM;
if (!sk_page_frag_refill(sk, pfrag))
goto error;
if (!skb_can_coalesce(skb, i, pfrag->page,
pfrag->offset)) {
err = -EMSGSIZE;
if (i == MAX_SKB_FRAGS)
goto error;
__skb_fill_page_desc(skb, i, pfrag->page,
pfrag->offset, 0);
skb_shinfo(skb)->nr_frags = ++i;
get_page(pfrag->page);
}
copy = min_t(int, copy, pfrag->size - pfrag->offset);
if (getfrag(from,
page_address(pfrag->page) + pfrag->offset,
offset, copy, skb->len, skb) < 0)
goto error_efault;
pfrag->offset += copy;
skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy);
skb->len += copy;
skb->data_len += copy;
skb->truesize += copy;
atomic_add(copy, &sk->sk_wmem_alloc);
}
offset += copy;
length -= copy;
}
return 0;
error_efault:
err = -EFAULT;
error:
cork->length -= length;
IP6_INC_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
return err;
}
/**
* Release the write lock on the hash table.
*
* @param table The hash table to unlock
*/
static void
hashtable_write_unlock(HASHTABLE *table)
{
atomic_add(&table->writelock, -1);
}
void zfs_do_throttle(zfs_throttle_t *zt, bool is_write, uint64_t size) {
int64_t _wait;
uint64_t _now;
uint64_t data;
struct semaphore *sem;
if (zt->is_enabled) {
if (is_write) {
if (zt->z_prop_write_bytes > 0) {
sem = zt->z_sem_real_write;
down(sem);
atomic_add(size, &(zt->z_real_write_bytes));
data = atomic_read(&(zt->z_real_write_bytes));
//dprintf("------- write bytes throttle 0: %llu\n", data);
if (data > zt->z_prop_write_bytes) {
if (zt->z_write_timestamp > 0) {
_now = now();
_wait = ((data * 1000000) / zt->z_prop_write_bytes)
- ((_now - zt->z_write_timestamp) / 1000);
//dprintf("------- write bytes throttle 1: %llu %llu %llu %llu %lld\n", data, zt->z_prop_write_bytes, _now, zt->z_write_timestamp, _wait);
if (_wait > 0) {
usleep_range(_wait, _wait);
}
}
atomic_set(&(zt->z_real_write_bytes), 0);
zt->z_write_timestamp = now();
}
up(sem);
} else {
if (zt->z_prop_write_iops > 0) {
sem = zt->z_sem_real_write;
down(sem);
atomic_add(1, &(zt->z_real_write_iops));
data = atomic_read(&(zt->z_real_write_iops));
//dprintf("------- write iops throttle 0: %llu\n", data);
if (data > zt->z_prop_write_iops) {
if (zt->z_write_timestamp > 0) {
_now = now();
_wait = ((data * 1000000) / zt->z_prop_write_iops)
- ((_now - zt->z_write_timestamp) / 1000);
//dprintf("------- write iops throttle 1: %llu %llu %llu %llu %lld\n", data, zt->z_prop_write_iops, _now, zt->z_write_timestamp, _wait);
if (_wait > 0) {
usleep_range(_wait, _wait);
}
}
atomic_set(&(zt->z_real_write_iops), 0);
zt->z_write_timestamp = now();
}
up(sem);
}
}
} else {
if (zt->z_prop_read_bytes > 0) {
sem = zt->z_sem_real_read;
down(sem);
atomic_add(size, &(zt->z_real_read_bytes));
data = atomic_read(&(zt->z_real_read_bytes));
//dprintf("------- read bytes throttle 0: %llu\n", data);
if (data > zt->z_prop_read_bytes) {
if (zt->z_read_timestamp > 0) {
_now = now();
_wait = ((data * 1000000) / zt->z_prop_read_bytes)
- ((_now - zt->z_read_timestamp) / 1000);
//dprintf("------- read bytes throttle 1: %llu %llu %llu %llu %lld\n", data, zt->z_prop_read_bytes, _now, zt->z_read_timestamp, _wait);
if (_wait > 0) {
usleep_range(_wait, _wait);
}
}
atomic_set(&(zt->z_real_read_bytes), 0);
zt->z_read_timestamp = now();
}
up(sem);
} else {
if (zt->z_prop_read_iops > 0) {
sem = zt->z_sem_real_read;
down(sem);
atomic_add(1, &(zt->z_real_read_iops));
data = atomic_read(&(zt->z_real_read_iops));
//dprintf("------- read iops throttle 0: %llu\n", data);
if (data > zt->z_prop_read_iops) {
if (zt->z_read_timestamp > 0) {
_now = now();
_wait = ((data * 1000000) / zt->z_prop_read_iops)
- ((_now - zt->z_read_timestamp) / 1000);
//dprintf("------- read iops throttle 1: %llu %llu %llu %llu %lld\n", data, zt->z_prop_read_iops, _now, zt->z_read_timestamp, _wait);
if (_wait > 0) {
usleep_range(_wait, _wait);
}
}
atomic_set(&(zt->z_real_read_iops), 0);
zt->z_read_timestamp = now();
}
up(sem);
}
}
}
}
// update counters
if (is_write) {
zt->z_io_serviced->write_bytes.value.ui64 += size;
zt->z_io_serviced->write_iops.value.ui64++;
} else {
zt->z_io_serviced->read_bytes.value.ui64 += size;
zt->z_io_serviced->read_iops.value.ui64++;
}
}
/*
* Decrements the reference-counter and
* tests whether it became zero.
*
* @return 1 reference-counter became zero
* @return 0 reference-counter didn't became zero
*/
int
ieee80211_node_dectestref(struct ieee80211_node* ni)
{
// atomic_add returns old value
return atomic_add((vint32*)&ni->ni_refcnt, -1) == 1;
}
long do_msgsnd(int msqid, long mtype, void __user *mtext,
size_t msgsz, int msgflg)
{
struct msg_queue *msq;
struct msg_msg *msg;
int err;
struct ipc_namespace *ns;
ns = current->nsproxy->ipc_ns;
if (msgsz > ns->msg_ctlmax || (long) msgsz < 0 || msqid < 0)
return -EINVAL;
if (mtype < 1)
return -EINVAL;
msg = load_msg(mtext, msgsz);
if (IS_ERR(msg))
return PTR_ERR(msg);
msg->m_type = mtype;
msg->m_ts = msgsz;
msq = msg_lock_check(ns, msqid);
if (IS_ERR(msq)) {
err = PTR_ERR(msq);
goto out_free;
}
for (;;) {
struct msg_sender s;
err = -EACCES;
if (ipcperms(ns, &msq->q_perm, S_IWUGO))
goto out_unlock_free;
err = security_msg_queue_msgsnd(msq, msg, msgflg);
if (err)
goto out_unlock_free;
if (msgsz + msq->q_cbytes <= msq->q_qbytes &&
1 + msq->q_qnum <= msq->q_qbytes) {
break;
}
/* queue full, wait: */
if (msgflg & IPC_NOWAIT) {
err = -EAGAIN;
goto out_unlock_free;
}
ss_add(msq, &s);
ipc_rcu_getref(msq);
msg_unlock(msq);
schedule();
ipc_lock_by_ptr(&msq->q_perm);
ipc_rcu_putref(msq);
if (msq->q_perm.deleted) {
err = -EIDRM;
goto out_unlock_free;
}
ss_del(&s);
if (signal_pending(current)) {
err = -ERESTARTNOHAND;
goto out_unlock_free;
}
}
msq->q_lspid = task_tgid_vnr(current);
msq->q_stime = get_seconds();
if (!pipelined_send(msq, msg)) {
/* no one is waiting for this message, enqueue it */
list_add_tail(&msg->m_list, &msq->q_messages);
msq->q_cbytes += msgsz;
msq->q_qnum++;
atomic_add(msgsz, &ns->msg_bytes);
atomic_inc(&ns->msg_hdrs);
}
err = 0;
msg = NULL;
out_unlock_free:
msg_unlock(msq);
out_free:
if (msg != NULL)
free_msg(msg);
return err;
}
static int clip_start_xmit(struct sk_buff *skb,struct net_device *dev)
{
struct clip_priv *clip_priv = PRIV(dev);
struct atmarp_entry *entry;
struct atm_vcc *vcc;
int old;
unsigned long flags;
DPRINTK("clip_start_xmit (skb %p)\n",skb);
if (!skb->dst) {
printk(KERN_ERR "clip_start_xmit: skb->dst == NULL\n");
dev_kfree_skb(skb);
clip_priv->stats.tx_dropped++;
return 0;
}
if (!skb->dst->neighbour) {
#if 0
skb->dst->neighbour = clip_find_neighbour(skb->dst,1);
if (!skb->dst->neighbour) {
dev_kfree_skb(skb); /* lost that one */
clip_priv->stats.tx_dropped++;
return 0;
}
#endif
printk(KERN_ERR "clip_start_xmit: NO NEIGHBOUR !\n");
dev_kfree_skb(skb);
clip_priv->stats.tx_dropped++;
return 0;
}
entry = NEIGH2ENTRY(skb->dst->neighbour);
if (!entry->vccs) {
if (time_after(jiffies, entry->expires)) {
/* should be resolved */
entry->expires = jiffies+ATMARP_RETRY_DELAY*HZ;
to_atmarpd(act_need,PRIV(dev)->number,entry->ip);
}
if (entry->neigh->arp_queue.qlen < ATMARP_MAX_UNRES_PACKETS)
skb_queue_tail(&entry->neigh->arp_queue,skb);
else {
dev_kfree_skb(skb);
clip_priv->stats.tx_dropped++;
}
return 0;
}
DPRINTK("neigh %p, vccs %p\n",entry,entry->vccs);
ATM_SKB(skb)->vcc = vcc = entry->vccs->vcc;
DPRINTK("using neighbour %p, vcc %p\n",skb->dst->neighbour,vcc);
if (entry->vccs->encap) {
void *here;
here = skb_push(skb,RFC1483LLC_LEN);
memcpy(here,llc_oui,sizeof(llc_oui));
((u16 *) here)[3] = skb->protocol;
}
atomic_add(skb->truesize, &sk_atm(vcc)->sk_wmem_alloc);
ATM_SKB(skb)->atm_options = vcc->atm_options;
entry->vccs->last_use = jiffies;
DPRINTK("atm_skb(%p)->vcc(%p)->dev(%p)\n",skb,vcc,vcc->dev);
old = xchg(&entry->vccs->xoff,1); /* assume XOFF ... */
if (old) {
printk(KERN_WARNING "clip_start_xmit: XOFF->XOFF transition\n");
return 0;
}
clip_priv->stats.tx_packets++;
clip_priv->stats.tx_bytes += skb->len;
(void) vcc->send(vcc,skb);
if (atm_may_send(vcc,0)) {
entry->vccs->xoff = 0;
return 0;
}
spin_lock_irqsave(&clip_priv->xoff_lock,flags);
netif_stop_queue(dev); /* XOFF -> throttle immediately */
barrier();
if (!entry->vccs->xoff)
netif_start_queue(dev);
/* Oh, we just raced with clip_pop. netif_start_queue should be
good enough, because nothing should really be asleep because
of the brief netif_stop_queue. If this isn't true or if it
changes, use netif_wake_queue instead. */
spin_unlock_irqrestore(&clip_priv->xoff_lock,flags);
return 0;
}
static int clip_start_xmit(struct sk_buff *skb, knet_netdev_t *dev)
{
struct clip_priv *clip_priv = PRIV(dev);
struct atmarp_entry *entry;
struct atm_vcc *vcc;
int old;
unsigned long flags;
DPRINTK("clip_start_xmit (skb %p)\n", skb);
if (!skb->dst)
{
printk(KERN_ERR "clip_start_xmit: skb->dst == NULL\n");
dev_kfree_skb(skb);
clip_priv->stats.tx_dropped++;
return 0;
}
if (!skb->dst->neighbour)
{
printk(KERN_ERR "clip_start_xmit: NO NEIGHBOUR!\n");
dev_kfree_skb(skb);
clip_priv->stats.tx_dropped++;
return 0;
}
entry = NEIGH2ENTRY(skb->dst->neighbour);
if (!entry->vccs)
{
if (time_after(jiffies, entry->expires))
{
/* should be resolved */
entry->expires = jiffies + ATMARP_RETRY_DELAY * HZ;
}
if (entry->neigh->arp_queue.qlen < ATMARP_MAX_UNRES_PACKETS)
skb_queue_tail(&entry->neigh->arp_queue, skb);
else
{
dev_kfree_skb(skb);
clip_priv->stats.tx_dropped++;
}
/* If a vcc was not resolved for a long time, it sends an InARP
* packet every 5 minutes. But if the other side connected now
* we do not want to wait.
*/
all_clip_vccs_start_resolving();
return 0;
}
DPRINTK("neigh %p, vccs %p\n", entry, entry->vccs);
ATM_SKB(skb)->vcc = vcc = entry->vccs->vcc;
DPRINTK("using neighbour %p, vcc %p\n", skb->dst->neighbour, vcc);
if (entry->vccs->encap)
{
void *here;
here = skb_push(skb, RFC1483LLC_LEN);
memcpy(here, llc_oui, sizeof(llc_oui));
((u16 *) here)[3] = skb->protocol;
}
atomic_add(skb->truesize, &vcc->tx_inuse);
ATM_SKB(skb)->iovcnt = 0;
ATM_SKB(skb)->atm_options = vcc->atm_options;
entry->vccs->last_use = jiffies;
DPRINTK("atm_skb(%p)->vcc(%p)->dev(%p)\n", skb, vcc,vcc->dev);
old = xchg(&entry->vccs->xoff, 1); /* assume XOFF ... */
if (old)
{
printk(KERN_WARNING "clip_start_xmit: XOFF->XOFF transition\n");
return 0;
}
clip_priv->stats.tx_packets++;
clip_priv->stats.tx_bytes += skb->len;
vcc->dev->ops->send(vcc, skb);
if (atm_may_send(vcc, 0))
{
entry->vccs->xoff = 0;
return 0;
}
spin_lock_irqsave(&clip_priv->xoff_lock, flags);
knet_netdev_stop_queue(dev); /* XOFF -> throttle immediately */
barrier();
if (!entry->vccs->xoff)
knet_netdev_start_queue(dev);
/* Oh, we just raced with clip_pop. netif_start_queue should be
good enough, because nothing should really be asleep because
of the brief netif_stop_queue. If this isn't true or if it
changes, use netif_wake_queue instead. */
spin_unlock_irqrestore(&clip_priv->xoff_lock, flags);
return 0;
}
/* Add new segment to existing queue. */
static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
{
struct sk_buff *prev, *next;
struct net_device *dev;
int flags, offset;
int ihl, end;
int err = -ENOENT;
u8 ecn;
if (qp->q.last_in & INET_FRAG_COMPLETE)
goto err;
if (!(IPCB(skb)->flags & IPSKB_FRAG_COMPLETE) &&
unlikely(ip_frag_too_far(qp)) &&
unlikely(err = ip_frag_reinit(qp))) {
ipq_kill(qp);
goto err;
}
ecn = ip4_frag_ecn(ip_hdr(skb)->tos);
offset = ntohs(ip_hdr(skb)->frag_off);
flags = offset & ~IP_OFFSET;
offset &= IP_OFFSET;
offset <<= 3; /* offset is in 8-byte chunks */
ihl = ip_hdrlen(skb);
/* Determine the position of this fragment. */
end = offset + skb->len - ihl;
err = -EINVAL;
/* Is this the final fragment? */
if ((flags & IP_MF) == 0) {
/* If we already have some bits beyond end
* or have different end, the segment is corrrupted.
*/
if (end < qp->q.len ||
((qp->q.last_in & INET_FRAG_LAST_IN) && end != qp->q.len))
goto err;
qp->q.last_in |= INET_FRAG_LAST_IN;
qp->q.len = end;
} else {
if (end&7) {
end &= ~7;
if (skb->ip_summed != CHECKSUM_UNNECESSARY)
skb->ip_summed = CHECKSUM_NONE;
}
if (end > qp->q.len) {
/* Some bits beyond end -> corruption. */
if (qp->q.last_in & INET_FRAG_LAST_IN)
goto err;
qp->q.len = end;
}
}
if (end == offset)
goto err;
err = -ENOMEM;
if (pskb_pull(skb, ihl) == NULL)
goto err;
err = pskb_trim_rcsum(skb, end - offset);
if (err)
goto err;
/* Find out which fragments are in front and at the back of us
* in the chain of fragments so far. We must know where to put
* this fragment, right?
*/
prev = qp->q.fragments_tail;
if (!prev || FRAG_CB(prev)->offset < offset) {
next = NULL;
goto found;
}
prev = NULL;
for (next = qp->q.fragments; next != NULL; next = next->next) {
if (FRAG_CB(next)->offset >= offset)
break; /* bingo! */
prev = next;
}
found:
/* We found where to put this one. Check for overlap with
* preceding fragment, and, if needed, align things so that
* any overlaps are eliminated.
*/
if (prev) {
int i = (FRAG_CB(prev)->offset + prev->len) - offset;
if (i > 0) {
offset += i;
err = -EINVAL;
if (end <= offset)
goto err;
err = -ENOMEM;
if (!pskb_pull(skb, i))
goto err;
if (skb->ip_summed != CHECKSUM_UNNECESSARY)
skb->ip_summed = CHECKSUM_NONE;
}
}
err = -ENOMEM;
while (next && FRAG_CB(next)->offset < end) {
int i = end - FRAG_CB(next)->offset; /* overlap is 'i' bytes */
if (i < next->len) {
/* Eat head of the next overlapped fragment
* and leave the loop. The next ones cannot overlap.
*/
if (!pskb_pull(next, i))
goto err;
FRAG_CB(next)->offset += i;
qp->q.meat -= i;
if (next->ip_summed != CHECKSUM_UNNECESSARY)
next->ip_summed = CHECKSUM_NONE;
break;
} else {
struct sk_buff *free_it = next;
/* Old fragment is completely overridden with
* new one drop it.
*/
next = next->next;
if (prev)
prev->next = next;
else
qp->q.fragments = next;
qp->q.meat -= free_it->len;
frag_kfree_skb(qp->q.net, free_it);
}
}
FRAG_CB(skb)->offset = offset;
/* Insert this fragment in the chain of fragments. */
skb->next = next;
if (!next)
qp->q.fragments_tail = skb;
if (prev)
prev->next = skb;
else
qp->q.fragments = skb;
dev = skb->dev;
if (dev) {
qp->iif = dev->ifindex;
skb->dev = NULL;
}
qp->q.stamp = skb->tstamp;
qp->q.meat += skb->len;
qp->ecn |= ecn;
atomic_add(skb->truesize, &qp->q.net->mem);
if (offset == 0)
qp->q.last_in |= INET_FRAG_FIRST_IN;
if (qp->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) &&
qp->q.meat == qp->q.len)
return ip_frag_reasm(qp, prev, dev);
write_lock(&ip4_frags.lock);
list_move_tail(&qp->q.lru_list, &qp->q.net->lru_list);
write_unlock(&ip4_frags.lock);
return -EINPROGRESS;
err:
kfree_skb(skb);
return err;
}
int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
int offset, int len, int odd, struct sk_buff *skb),
void *from, int length, int transhdrlen,
int hlimit, int tclass, struct ipv6_txoptions *opt, struct flowi6 *fl6,
struct rt6_info *rt, unsigned int flags, int dontfrag)
{
struct inet_sock *inet = inet_sk(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct inet_cork *cork;
struct sk_buff *skb, *skb_prev = NULL;
unsigned int maxfraglen, fragheaderlen, mtu, orig_mtu;
int exthdrlen;
int dst_exthdrlen;
int hh_len;
int copy;
int err;
int offset = 0;
__u8 tx_flags = 0;
if (flags&MSG_PROBE)
return 0;
cork = &inet->cork.base;
if (skb_queue_empty(&sk->sk_write_queue)) {
/*
* setup for corking
*/
if (opt) {
if (WARN_ON(np->cork.opt))
return -EINVAL;
np->cork.opt = kzalloc(opt->tot_len, sk->sk_allocation);
if (unlikely(np->cork.opt == NULL))
return -ENOBUFS;
np->cork.opt->tot_len = opt->tot_len;
np->cork.opt->opt_flen = opt->opt_flen;
np->cork.opt->opt_nflen = opt->opt_nflen;
np->cork.opt->dst0opt = ip6_opt_dup(opt->dst0opt,
sk->sk_allocation);
if (opt->dst0opt && !np->cork.opt->dst0opt)
return -ENOBUFS;
np->cork.opt->dst1opt = ip6_opt_dup(opt->dst1opt,
sk->sk_allocation);
if (opt->dst1opt && !np->cork.opt->dst1opt)
return -ENOBUFS;
np->cork.opt->hopopt = ip6_opt_dup(opt->hopopt,
sk->sk_allocation);
if (opt->hopopt && !np->cork.opt->hopopt)
return -ENOBUFS;
np->cork.opt->srcrt = ip6_rthdr_dup(opt->srcrt,
sk->sk_allocation);
if (opt->srcrt && !np->cork.opt->srcrt)
return -ENOBUFS;
/* need source address above miyazawa*/
}
dst_hold(&rt->dst);
cork->dst = &rt->dst;
inet->cork.fl.u.ip6 = *fl6;
np->cork.hop_limit = hlimit;
np->cork.tclass = tclass;
if (rt->dst.flags & DST_XFRM_TUNNEL)
mtu = np->pmtudisc >= IPV6_PMTUDISC_PROBE ?
rt->dst.dev->mtu : dst_mtu(&rt->dst);
else
mtu = np->pmtudisc >= IPV6_PMTUDISC_PROBE ?
rt->dst.dev->mtu : dst_mtu(rt->dst.path);
if (np->frag_size < mtu) {
if (np->frag_size)
mtu = np->frag_size;
}
cork->fragsize = mtu;
if (dst_allfrag(rt->dst.path))
cork->flags |= IPCORK_ALLFRAG;
cork->length = 0;
exthdrlen = (opt ? opt->opt_flen : 0);
length += exthdrlen;
transhdrlen += exthdrlen;
dst_exthdrlen = rt->dst.header_len - rt->rt6i_nfheader_len;
} else {
rt = (struct rt6_info *)cork->dst;
fl6 = &inet->cork.fl.u.ip6;
opt = np->cork.opt;
transhdrlen = 0;
exthdrlen = 0;
dst_exthdrlen = 0;
mtu = cork->fragsize;
}
orig_mtu = mtu;
hh_len = LL_RESERVED_SPACE(rt->dst.dev);
fragheaderlen = sizeof(struct ipv6hdr) + rt->rt6i_nfheader_len +
(opt ? opt->opt_nflen : 0);
maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen -
sizeof(struct frag_hdr);
if (mtu <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN) {
unsigned int maxnonfragsize, headersize;
headersize = sizeof(struct ipv6hdr) +
(opt ? opt->tot_len : 0) +
(dst_allfrag(&rt->dst) ?
sizeof(struct frag_hdr) : 0) +
rt->rt6i_nfheader_len;
if (ip6_sk_local_df(sk))
maxnonfragsize = sizeof(struct ipv6hdr) + IPV6_MAXPLEN;
else
maxnonfragsize = mtu;
/* dontfrag active */
if ((cork->length + length > mtu - headersize) && dontfrag &&
(sk->sk_protocol == IPPROTO_UDP ||
sk->sk_protocol == IPPROTO_RAW)) {
ipv6_local_rxpmtu(sk, fl6, mtu - headersize +
sizeof(struct ipv6hdr));
goto emsgsize;
}
if (cork->length + length > maxnonfragsize - headersize) {
emsgsize:
ipv6_local_error(sk, EMSGSIZE, fl6,
mtu - headersize +
sizeof(struct ipv6hdr));
return -EMSGSIZE;
}
}
/* For UDP, check if TX timestamp is enabled */
if (sk->sk_type == SOCK_DGRAM)
sock_tx_timestamp(sk, &tx_flags);
/*
* Let's try using as much space as possible.
* Use MTU if total length of the message fits into the MTU.
* Otherwise, we need to reserve fragment header and
* fragment alignment (= 8-15 octects, in total).
*
* Note that we may need to "move" the data from the tail of
* of the buffer to the new fragment when we split
* the message.
*
* FIXME: It may be fragmented into multiple chunks
* at once if non-fragmentable extension headers
* are too large.
* --yoshfuji
*/
skb = skb_peek_tail(&sk->sk_write_queue);
cork->length += length;
if (((length > mtu) ||
(skb && skb_is_gso(skb))) &&
(sk->sk_protocol == IPPROTO_UDP) &&
(rt->dst.dev->features & NETIF_F_UFO)) {
err = ip6_ufo_append_data(sk, getfrag, from, length,
hh_len, fragheaderlen,
transhdrlen, mtu, flags, rt);
if (err)
goto error;
return 0;
}
if (!skb)
goto alloc_new_skb;
while (length > 0) {
/* Check if the remaining data fits into current packet. */
copy = (cork->length <= mtu && !(cork->flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - skb->len;
if (copy < length)
copy = maxfraglen - skb->len;
if (copy <= 0) {
char *data;
unsigned int datalen;
unsigned int fraglen;
unsigned int fraggap;
unsigned int alloclen;
alloc_new_skb:
/* There's no room in the current skb */
if (skb)
fraggap = skb->len - maxfraglen;
else
fraggap = 0;
/* update mtu and maxfraglen if necessary */
if (skb == NULL || skb_prev == NULL)
ip6_append_data_mtu(&mtu, &maxfraglen,
fragheaderlen, skb, rt,
orig_mtu);
skb_prev = skb;
/*
* If remaining data exceeds the mtu,
* we know we need more fragment(s).
*/
datalen = length + fraggap;
if (datalen > (cork->length <= mtu && !(cork->flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - fragheaderlen)
datalen = maxfraglen - fragheaderlen - rt->dst.trailer_len;
if ((flags & MSG_MORE) &&
!(rt->dst.dev->features&NETIF_F_SG))
alloclen = mtu;
else
alloclen = datalen + fragheaderlen;
alloclen += dst_exthdrlen;
if (datalen != length + fraggap) {
/*
* this is not the last fragment, the trailer
* space is regarded as data space.
*/
datalen += rt->dst.trailer_len;
}
alloclen += rt->dst.trailer_len;
fraglen = datalen + fragheaderlen;
/*
* We just reserve space for fragment header.
* Note: this may be overallocation if the message
* (without MSG_MORE) fits into the MTU.
*/
alloclen += sizeof(struct frag_hdr);
if (transhdrlen) {
skb = sock_alloc_send_skb(sk,
alloclen + hh_len,
(flags & MSG_DONTWAIT), &err);
} else {
skb = NULL;
if (atomic_read(&sk->sk_wmem_alloc) <=
2 * sk->sk_sndbuf)
skb = sock_wmalloc(sk,
alloclen + hh_len, 1,
sk->sk_allocation);
if (unlikely(skb == NULL))
err = -ENOBUFS;
else {
/* Only the initial fragment
* is time stamped.
*/
tx_flags = 0;
}
}
if (skb == NULL)
goto error;
/*
* Fill in the control structures
*/
skb->protocol = htons(ETH_P_IPV6);
skb->ip_summed = CHECKSUM_NONE;
skb->csum = 0;
/* reserve for fragmentation and ipsec header */
skb_reserve(skb, hh_len + sizeof(struct frag_hdr) +
dst_exthdrlen);
if (sk->sk_type == SOCK_DGRAM)
skb_shinfo(skb)->tx_flags = tx_flags;
/*
* Find where to start putting bytes
*/
data = skb_put(skb, fraglen);
skb_set_network_header(skb, exthdrlen);
data += fragheaderlen;
skb->transport_header = (skb->network_header +
fragheaderlen);
if (fraggap) {
skb->csum = skb_copy_and_csum_bits(
skb_prev, maxfraglen,
data + transhdrlen, fraggap, 0);
skb_prev->csum = csum_sub(skb_prev->csum,
skb->csum);
data += fraggap;
pskb_trim_unique(skb_prev, maxfraglen);
}
copy = datalen - transhdrlen - fraggap;
if (copy < 0) {
err = -EINVAL;
kfree_skb(skb);
goto error;
} else if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, fraggap, skb) < 0) {
err = -EFAULT;
kfree_skb(skb);
goto error;
}
offset += copy;
length -= datalen - fraggap;
transhdrlen = 0;
exthdrlen = 0;
dst_exthdrlen = 0;
/*
* Put the packet on the pending queue
*/
__skb_queue_tail(&sk->sk_write_queue, skb);
continue;
}
if (copy > length)
copy = length;
if (!(rt->dst.dev->features&NETIF_F_SG)) {
unsigned int off;
off = skb->len;
if (getfrag(from, skb_put(skb, copy),
offset, copy, off, skb) < 0) {
__skb_trim(skb, off);
err = -EFAULT;
goto error;
}
} else {
int i = skb_shinfo(skb)->nr_frags;
struct page_frag *pfrag = sk_page_frag(sk);
err = -ENOMEM;
if (!sk_page_frag_refill(sk, pfrag))
goto error;
if (!skb_can_coalesce(skb, i, pfrag->page,
pfrag->offset)) {
err = -EMSGSIZE;
if (i == MAX_SKB_FRAGS)
goto error;
__skb_fill_page_desc(skb, i, pfrag->page,
pfrag->offset, 0);
skb_shinfo(skb)->nr_frags = ++i;
get_page(pfrag->page);
}
copy = min_t(int, copy, pfrag->size - pfrag->offset);
if (getfrag(from,
page_address(pfrag->page) + pfrag->offset,
offset, copy, skb->len, skb) < 0)
goto error_efault;
pfrag->offset += copy;
skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy);
skb->len += copy;
skb->data_len += copy;
skb->truesize += copy;
atomic_add(copy, &sk->sk_wmem_alloc);
}
offset += copy;
length -= copy;
}
return 0;
error_efault:
err = -EFAULT;
error:
cork->length -= length;
IP6_INC_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
return err;
}
static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
struct net_device *dev)
{
struct net *net = container_of(qp->q.net, struct net, ipv4.frags);
struct iphdr *iph;
struct sk_buff *fp, *head = qp->q.fragments;
int len;
int ihlen;
int err;
u8 ecn;
ipq_kill(qp);
ecn = ip4_frag_ecn_table[qp->ecn];
if (unlikely(ecn == 0xff)) {
err = -EINVAL;
goto out_fail;
}
/* Make the one we just received the head. */
if (prev) {
head = prev->next;
fp = skb_clone(head, GFP_ATOMIC);
if (!fp)
goto out_nomem;
fp->next = head->next;
if (!fp->next)
qp->q.fragments_tail = fp;
prev->next = fp;
skb_morph(head, qp->q.fragments);
head->next = qp->q.fragments->next;
kfree_skb(qp->q.fragments);
qp->q.fragments = head;
}
WARN_ON(head == NULL);
WARN_ON(FRAG_CB(head)->offset != 0);
/* Allocate a new buffer for the datagram. */
ihlen = ip_hdrlen(head);
len = ihlen + qp->q.len;
err = -E2BIG;
if (len > 65535)
goto out_oversize;
/* Head of list must not be cloned. */
if (skb_cloned(head) && pskb_expand_head(head, 0, 0, GFP_ATOMIC))
goto out_nomem;
/* If the first fragment is fragmented itself, we split
* it to two chunks: the first with data and paged part
* and the second, holding only fragments. */
if (skb_has_frag_list(head)) {
struct sk_buff *clone;
int i, plen = 0;
if ((clone = alloc_skb(0, GFP_ATOMIC)) == NULL)
goto out_nomem;
clone->next = head->next;
head->next = clone;
skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list;
skb_frag_list_init(head);
for (i=0; i<skb_shinfo(head)->nr_frags; i++)
plen += skb_shinfo(head)->frags[i].size;
clone->len = clone->data_len = head->data_len - plen;
head->data_len -= clone->len;
head->len -= clone->len;
clone->csum = 0;
clone->ip_summed = head->ip_summed;
atomic_add(clone->truesize, &qp->q.net->mem);
}
skb_shinfo(head)->frag_list = head->next;
skb_push(head, head->data - skb_network_header(head));
for (fp=head->next; fp; fp = fp->next) {
head->data_len += fp->len;
head->len += fp->len;
if (head->ip_summed != fp->ip_summed)
head->ip_summed = CHECKSUM_NONE;
else if (head->ip_summed == CHECKSUM_COMPLETE)
head->csum = csum_add(head->csum, fp->csum);
head->truesize += fp->truesize;
}
atomic_sub(head->truesize, &qp->q.net->mem);
head->next = NULL;
head->dev = dev;
head->tstamp = qp->q.stamp;
iph = ip_hdr(head);
iph->frag_off = 0;
iph->tot_len = htons(len);
iph->tos |= ecn;
IP_INC_STATS_BH(net, IPSTATS_MIB_REASMOKS);
qp->q.fragments = NULL;
qp->q.fragments_tail = NULL;
return 0;
out_nomem:
LIMIT_NETDEBUG(KERN_ERR "IP: queue_glue: no memory for gluing "
"queue %p\n", qp);
err = -ENOMEM;
goto out_fail;
out_oversize:
if (net_ratelimit())
printk(KERN_INFO "Oversized IP packet from %pI4.\n",
&qp->saddr);
out_fail:
IP_INC_STATS_BH(net, IPSTATS_MIB_REASMFAILS);
return err;
}
static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb,
struct frag_hdr *fhdr, int nhoff)
{
struct sk_buff *prev, *next;
int offset, end;
if (fq->last_in & COMPLETE) {
DEBUGP("Allready completed\n");
goto err;
}
offset = ntohs(fhdr->frag_off) & ~0x7;
end = offset + (ntohs(skb->nh.ipv6h->payload_len) -
((u8 *) (fhdr + 1) - (u8 *) (skb->nh.ipv6h + 1)));
if ((unsigned int)end > IPV6_MAXPLEN) {
DEBUGP("offset is too large.\n");
return -1;
}
if (skb->ip_summed == CHECKSUM_HW)
skb->csum = csum_sub(skb->csum,
csum_partial(skb->nh.raw,
(u8*)(fhdr + 1) - skb->nh.raw,
0));
/* Is this the final fragment? */
if (!(fhdr->frag_off & htons(IP6_MF))) {
/* If we already have some bits beyond end
* or have different end, the segment is corrupted.
*/
if (end < fq->len ||
((fq->last_in & LAST_IN) && end != fq->len)) {
DEBUGP("already received last fragment\n");
goto err;
}
fq->last_in |= LAST_IN;
fq->len = end;
} else {
/* Check if the fragment is rounded to 8 bytes.
* Required by the RFC.
*/
if (end & 0x7) {
/* RFC2460 says always send parameter problem in
* this case. -DaveM
*/
DEBUGP("the end of this fragment is not rounded to 8 bytes.\n");
return -1;
}
if (end > fq->len) {
/* Some bits beyond end -> corruption. */
if (fq->last_in & LAST_IN) {
DEBUGP("last packet already reached.\n");
goto err;
}
fq->len = end;
}
}
if (end == offset)
goto err;
/* Point into the IP datagram 'data' part. */
if (!pskb_pull(skb, (u8 *) (fhdr + 1) - skb->data)) {
DEBUGP("queue: message is too short.\n");
goto err;
}
if (end-offset < skb->len) {
if (pskb_trim(skb, end - offset)) {
DEBUGP("Can't trim\n");
goto err;
}
if (skb->ip_summed != CHECKSUM_UNNECESSARY)
skb->ip_summed = CHECKSUM_NONE;
}
/* Find out which fragments are in front and at the back of us
* in the chain of fragments so far. We must know where to put
* this fragment, right?
*/
prev = NULL;
for (next = fq->fragments; next != NULL; next = next->next) {
if (NFCT_FRAG6_CB(next)->offset >= offset)
break; /* bingo! */
prev = next;
}
/* We found where to put this one. Check for overlap with
* preceding fragment, and, if needed, align things so that
* any overlaps are eliminated.
*/
if (prev) {
int i = (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset;
if (i > 0) {
offset += i;
if (end <= offset) {
DEBUGP("overlap\n");
goto err;
}
if (!pskb_pull(skb, i)) {
DEBUGP("Can't pull\n");
goto err;
}
if (skb->ip_summed != CHECKSUM_UNNECESSARY)
skb->ip_summed = CHECKSUM_NONE;
}
}
/* Look for overlap with succeeding segments.
* If we can merge fragments, do it.
*/
while (next && NFCT_FRAG6_CB(next)->offset < end) {
/* overlap is 'i' bytes */
int i = end - NFCT_FRAG6_CB(next)->offset;
if (i < next->len) {
/* Eat head of the next overlapped fragment
* and leave the loop. The next ones cannot overlap.
*/
DEBUGP("Eat head of the overlapped parts.: %d", i);
if (!pskb_pull(next, i))
goto err;
/* next fragment */
NFCT_FRAG6_CB(next)->offset += i;
fq->meat -= i;
if (next->ip_summed != CHECKSUM_UNNECESSARY)
next->ip_summed = CHECKSUM_NONE;
break;
} else {
struct sk_buff *free_it = next;
/* Old fragmnet is completely overridden with
* new one drop it.
*/
next = next->next;
if (prev)
prev->next = next;
else
fq->fragments = next;
fq->meat -= free_it->len;
frag_kfree_skb(free_it, NULL);
}
}
NFCT_FRAG6_CB(skb)->offset = offset;
/* Insert this fragment in the chain of fragments. */
skb->next = next;
if (prev)
prev->next = skb;
else
fq->fragments = skb;
skb->dev = NULL;
skb_get_timestamp(skb, &fq->stamp);
fq->meat += skb->len;
atomic_add(skb->truesize, &nf_ct_frag6_mem);
/* The first fragment.
* nhoffset is obtained from the first fragment, of course.
*/
if (offset == 0) {
fq->nhoffset = nhoff;
fq->last_in |= FIRST_IN;
}
write_lock(&nf_ct_frag6_lock);
list_move_tail(&fq->lru_list, &nf_ct_frag6_lru_list);
write_unlock(&nf_ct_frag6_lock);
return 0;
err:
return -1;
}
/* replenish the buffers for a pool. note that we don't need to
* skb_reserve these since they are used for incoming...
*/
static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter,
struct ibmveth_buff_pool *pool)
{
u32 i;
u32 count = pool->size - atomic_read(&pool->available);
u32 buffers_added = 0;
struct sk_buff *skb;
unsigned int free_index, index;
u64 correlator;
unsigned long lpar_rc;
dma_addr_t dma_addr;
mb();
for (i = 0; i < count; ++i) {
union ibmveth_buf_desc desc;
skb = netdev_alloc_skb(adapter->netdev, pool->buff_size);
if (!skb) {
netdev_dbg(adapter->netdev,
"replenish: unable to allocate skb\n");
adapter->replenish_no_mem++;
break;
}
free_index = pool->consumer_index;
pool->consumer_index++;
if (pool->consumer_index >= pool->size)
pool->consumer_index = 0;
index = pool->free_map[free_index];
BUG_ON(index == IBM_VETH_INVALID_MAP);
BUG_ON(pool->skbuff[index] != NULL);
dma_addr = dma_map_single(&adapter->vdev->dev, skb->data,
pool->buff_size, DMA_FROM_DEVICE);
if (dma_mapping_error(&adapter->vdev->dev, dma_addr))
goto failure;
pool->free_map[free_index] = IBM_VETH_INVALID_MAP;
pool->dma_addr[index] = dma_addr;
pool->skbuff[index] = skb;
correlator = ((u64)pool->index << 32) | index;
*(u64 *)skb->data = correlator;
desc.fields.flags_len = IBMVETH_BUF_VALID | pool->buff_size;
desc.fields.address = dma_addr;
if (rx_flush) {
unsigned int len = min(pool->buff_size,
adapter->netdev->mtu +
IBMVETH_BUFF_OH);
ibmveth_flush_buffer(skb->data, len);
}
lpar_rc = h_add_logical_lan_buffer(adapter->vdev->unit_address,
desc.desc);
if (lpar_rc != H_SUCCESS) {
goto failure;
} else {
buffers_added++;
adapter->replenish_add_buff_success++;
}
}
mb();
atomic_add(buffers_added, &(pool->available));
return;
failure:
pool->free_map[free_index] = index;
pool->skbuff[index] = NULL;
if (pool->consumer_index == 0)
pool->consumer_index = pool->size - 1;
else
pool->consumer_index--;
if (!dma_mapping_error(&adapter->vdev->dev, dma_addr))
dma_unmap_single(&adapter->vdev->dev,
pool->dma_addr[index], pool->buff_size,
DMA_FROM_DEVICE);
dev_kfree_skb_any(skb);
adapter->replenish_add_buff_failure++;
mb();
atomic_add(buffers_added, &(pool->available));
}
int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
int offset, int len, int odd, struct sk_buff *skb),
void *from, int length, int transhdrlen,
int hlimit, int tclass, struct ipv6_txoptions *opt, struct flowi *fl,
struct rt6_info *rt, unsigned int flags)
{
struct inet_sock *inet = inet_sk(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct sk_buff *skb;
unsigned int maxfraglen, fragheaderlen;
int exthdrlen;
int hh_len;
int mtu;
int copy;
int err;
int offset = 0;
int csummode = CHECKSUM_NONE;
if (flags&MSG_PROBE)
return 0;
if (skb_queue_empty(&sk->sk_write_queue)) {
/*
* setup for corking
*/
if (opt) {
if (WARN_ON(np->cork.opt))
return -EINVAL;
np->cork.opt = kmalloc(opt->tot_len, sk->sk_allocation);
if (unlikely(np->cork.opt == NULL))
return -ENOBUFS;
np->cork.opt->tot_len = opt->tot_len;
np->cork.opt->opt_flen = opt->opt_flen;
np->cork.opt->opt_nflen = opt->opt_nflen;
np->cork.opt->dst0opt = ip6_opt_dup(opt->dst0opt,
sk->sk_allocation);
if (opt->dst0opt && !np->cork.opt->dst0opt)
return -ENOBUFS;
np->cork.opt->dst1opt = ip6_opt_dup(opt->dst1opt,
sk->sk_allocation);
if (opt->dst1opt && !np->cork.opt->dst1opt)
return -ENOBUFS;
np->cork.opt->hopopt = ip6_opt_dup(opt->hopopt,
sk->sk_allocation);
if (opt->hopopt && !np->cork.opt->hopopt)
return -ENOBUFS;
np->cork.opt->srcrt = ip6_rthdr_dup(opt->srcrt,
sk->sk_allocation);
if (opt->srcrt && !np->cork.opt->srcrt)
return -ENOBUFS;
/* need source address above miyazawa*/
}
dst_hold(&rt->u.dst);
inet->cork.dst = &rt->u.dst;
inet->cork.fl = *fl;
np->cork.hop_limit = hlimit;
np->cork.tclass = tclass;
mtu = np->pmtudisc == IPV6_PMTUDISC_PROBE ?
rt->u.dst.dev->mtu : dst_mtu(rt->u.dst.path);
if (np->frag_size < mtu) {
if (np->frag_size)
mtu = np->frag_size;
}
inet->cork.fragsize = mtu;
if (dst_allfrag(rt->u.dst.path))
inet->cork.flags |= IPCORK_ALLFRAG;
inet->cork.length = 0;
sk->sk_sndmsg_page = NULL;
sk->sk_sndmsg_off = 0;
exthdrlen = rt->u.dst.header_len + (opt ? opt->opt_flen : 0) -
rt->rt6i_nfheader_len;
length += exthdrlen;
transhdrlen += exthdrlen;
} else {
rt = (struct rt6_info *)inet->cork.dst;
fl = &inet->cork.fl;
opt = np->cork.opt;
transhdrlen = 0;
exthdrlen = 0;
mtu = inet->cork.fragsize;
}
hh_len = LL_RESERVED_SPACE(rt->u.dst.dev);
fragheaderlen = sizeof(struct ipv6hdr) + rt->rt6i_nfheader_len +
(opt ? opt->opt_nflen : 0);
maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen - sizeof(struct frag_hdr);
if (mtu <= sizeof(struct ipv6hdr) + IPV6_MAXPLEN) {
if (inet->cork.length + length > sizeof(struct ipv6hdr) + IPV6_MAXPLEN - fragheaderlen) {
ipv6_local_error(sk, EMSGSIZE, fl, mtu-exthdrlen);
return -EMSGSIZE;
}
}
/*
* Let's try using as much space as possible.
* Use MTU if total length of the message fits into the MTU.
* Otherwise, we need to reserve fragment header and
* fragment alignment (= 8-15 octects, in total).
*
* Note that we may need to "move" the data from the tail of
* of the buffer to the new fragment when we split
* the message.
*
* FIXME: It may be fragmented into multiple chunks
* at once if non-fragmentable extension headers
* are too large.
* --yoshfuji
*/
inet->cork.length += length;
if (((length > mtu) && (sk->sk_protocol == IPPROTO_UDP)) &&
(rt->u.dst.dev->features & NETIF_F_UFO)) {
err = ip6_ufo_append_data(sk, getfrag, from, length, hh_len,
fragheaderlen, transhdrlen, mtu,
flags, rt);
if (err)
goto error;
return 0;
}
if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL)
goto alloc_new_skb;
while (length > 0) {
/* Check if the remaining data fits into current packet. */
copy = (inet->cork.length <= mtu && !(inet->cork.flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - skb->len;
if (copy < length)
copy = maxfraglen - skb->len;
if (copy <= 0) {
char *data;
unsigned int datalen;
unsigned int fraglen;
unsigned int fraggap;
unsigned int alloclen;
struct sk_buff *skb_prev;
alloc_new_skb:
skb_prev = skb;
/* There's no room in the current skb */
if (skb_prev)
fraggap = skb_prev->len - maxfraglen;
else
fraggap = 0;
/*
* If remaining data exceeds the mtu,
* we know we need more fragment(s).
*/
datalen = length + fraggap;
if (datalen > (inet->cork.length <= mtu && !(inet->cork.flags & IPCORK_ALLFRAG) ? mtu : maxfraglen) - fragheaderlen)
datalen = maxfraglen - fragheaderlen;
fraglen = datalen + fragheaderlen;
if ((flags & MSG_MORE) &&
!(rt->u.dst.dev->features&NETIF_F_SG))
alloclen = mtu;
else
alloclen = datalen + fragheaderlen;
/*
* The last fragment gets additional space at tail.
* Note: we overallocate on fragments with MSG_MODE
* because we have no idea if we're the last one.
*/
if (datalen == length + fraggap)
alloclen += rt->u.dst.trailer_len;
/*
* We just reserve space for fragment header.
* Note: this may be overallocation if the message
* (without MSG_MORE) fits into the MTU.
*/
alloclen += sizeof(struct frag_hdr);
if (transhdrlen) {
skb = sock_alloc_send_skb(sk,
alloclen + hh_len,
(flags & MSG_DONTWAIT), &err);
} else {
skb = NULL;
if (atomic_read(&sk->sk_wmem_alloc) <=
2 * sk->sk_sndbuf)
skb = sock_wmalloc(sk,
alloclen + hh_len, 1,
sk->sk_allocation);
if (unlikely(skb == NULL))
err = -ENOBUFS;
}
if (skb == NULL)
goto error;
/*
* Fill in the control structures
*/
skb->ip_summed = csummode;
skb->csum = 0;
/* reserve for fragmentation */
skb_reserve(skb, hh_len+sizeof(struct frag_hdr));
/*
* Find where to start putting bytes
*/
data = skb_put(skb, fraglen);
skb_set_network_header(skb, exthdrlen);
data += fragheaderlen;
skb->transport_header = (skb->network_header +
fragheaderlen);
if (fraggap) {
skb->csum = skb_copy_and_csum_bits(
skb_prev, maxfraglen,
data + transhdrlen, fraggap, 0);
skb_prev->csum = csum_sub(skb_prev->csum,
skb->csum);
data += fraggap;
pskb_trim_unique(skb_prev, maxfraglen);
}
copy = datalen - transhdrlen - fraggap;
if (copy < 0) {
err = -EINVAL;
kfree_skb(skb);
goto error;
} else if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, fraggap, skb) < 0) {
err = -EFAULT;
kfree_skb(skb);
goto error;
}
offset += copy;
length -= datalen - fraggap;
transhdrlen = 0;
exthdrlen = 0;
csummode = CHECKSUM_NONE;
/*
* Put the packet on the pending queue
*/
__skb_queue_tail(&sk->sk_write_queue, skb);
continue;
}
if (copy > length)
copy = length;
if (!(rt->u.dst.dev->features&NETIF_F_SG)) {
unsigned int off;
off = skb->len;
if (getfrag(from, skb_put(skb, copy),
offset, copy, off, skb) < 0) {
__skb_trim(skb, off);
err = -EFAULT;
goto error;
}
} else {
int i = skb_shinfo(skb)->nr_frags;
skb_frag_t *frag = &skb_shinfo(skb)->frags[i-1];
struct page *page = sk->sk_sndmsg_page;
int off = sk->sk_sndmsg_off;
unsigned int left;
if (page && (left = PAGE_SIZE - off) > 0) {
if (copy >= left)
copy = left;
if (page != frag->page) {
if (i == MAX_SKB_FRAGS) {
err = -EMSGSIZE;
goto error;
}
get_page(page);
skb_fill_page_desc(skb, i, page, sk->sk_sndmsg_off, 0);
frag = &skb_shinfo(skb)->frags[i];
}
} else if(i < MAX_SKB_FRAGS) {
if (copy > PAGE_SIZE)
copy = PAGE_SIZE;
page = alloc_pages(sk->sk_allocation, 0);
if (page == NULL) {
err = -ENOMEM;
goto error;
}
sk->sk_sndmsg_page = page;
sk->sk_sndmsg_off = 0;
skb_fill_page_desc(skb, i, page, 0, 0);
frag = &skb_shinfo(skb)->frags[i];
} else {
err = -EMSGSIZE;
goto error;
}
if (getfrag(from, page_address(frag->page)+frag->page_offset+frag->size, offset, copy, skb->len, skb) < 0) {
err = -EFAULT;
goto error;
}
sk->sk_sndmsg_off += copy;
frag->size += copy;
skb->len += copy;
skb->data_len += copy;
skb->truesize += copy;
atomic_add(copy, &sk->sk_wmem_alloc);
}
offset += copy;
length -= copy;
}
return 0;
error:
inet->cork.length -= length;
IP6_INC_STATS(sock_net(sk), rt->rt6i_idev, IPSTATS_MIB_OUTDISCARDS);
return err;
}
static inline int unlock_ben(struct benaphore *ben)
{
if (atomic_add(&ben->atom, -1) > 1)
return release_sem(ben->sem);
return B_OK;
}