linux/net/atm/pppoatm.c
David Woodhouse 9d02daf754 pppoatm: Fix excessive queue bloat
We discovered that PPPoATM has an excessively deep transmit queue. A
queue the size of the default socket send buffer (wmem_default) is
maintained between the PPP generic core and the ATM device.

Fix it to queue a maximum of *two* packets. The one the ATM device is
currently working on, and one more for the ATM driver to process
immediately in its TX done interrupt handler. The PPP core is designed
to feed packets to the channel with minimal latency, so that really
ought to be enough to keep the ATM device busy.

While we're at it, fix the fact that we were triggering the wakeup
tasklet on *every* pppoatm_pop() call. The comment saying "this is
inefficient, but doing it right is too hard" turns out to be overly
pessimistic... I think :)

On machines like the Traverse Geos, with a slow Geode CPU and two
high-speed ADSL2+ interfaces, there were reports of extremely high CPU
usage which could partly be attributed to the extra wakeups.

(The wakeup handling could actually be made a whole lot easier if we
 stop checking sk->sk_sndbuf altogether. Given that we now only queue
 *two* packets ever, one wonders what the point is. As it is, you could
 already deadlock the thing by setting the sk_sndbuf to a value lower
 than the MTU of the device, and it'd just block for ever.)

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-13 13:03:45 -04:00

443 lines
14 KiB
C

/* net/atm/pppoatm.c - RFC2364 PPP over ATM/AAL5 */
/* Copyright 1999-2000 by Mitchell Blank Jr */
/* Based on clip.c; 1995-1999 by Werner Almesberger, EPFL LRC/ICA */
/* And on ppp_async.c; Copyright 1999 Paul Mackerras */
/* And help from Jens Axboe */
/*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*
* This driver provides the encapsulation and framing for sending
* and receiving PPP frames in ATM AAL5 PDUs.
*/
/*
* One shortcoming of this driver is that it does not comply with
* section 8 of RFC2364 - we are supposed to detect a change
* in encapsulation and immediately abort the connection (in order
* to avoid a black-hole being created if our peer loses state
* and changes encapsulation unilaterally. However, since the
* ppp_generic layer actually does the decapsulation, we need
* a way of notifying it when we _think_ there might be a problem)
* There's two cases:
* 1. LLC-encapsulation was missing when it was enabled. In
* this case, we should tell the upper layer "tear down
* this session if this skb looks ok to you"
* 2. LLC-encapsulation was present when it was disabled. Then
* we need to tell the upper layer "this packet may be
* ok, but if its in error tear down the session"
* These hooks are not yet available in ppp_generic
*/
#define pr_fmt(fmt) KBUILD_MODNAME ":%s: " fmt, __func__
#include <linux/module.h>
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/skbuff.h>
#include <linux/slab.h>
#include <linux/atm.h>
#include <linux/atmdev.h>
#include <linux/capability.h>
#include <linux/ppp_defs.h>
#include <linux/ppp-ioctl.h>
#include <linux/ppp_channel.h>
#include <linux/atmppp.h>
#include "common.h"
enum pppoatm_encaps {
e_autodetect = PPPOATM_ENCAPS_AUTODETECT,
e_vc = PPPOATM_ENCAPS_VC,
e_llc = PPPOATM_ENCAPS_LLC,
};
struct pppoatm_vcc {
struct atm_vcc *atmvcc; /* VCC descriptor */
void (*old_push)(struct atm_vcc *, struct sk_buff *);
void (*old_pop)(struct atm_vcc *, struct sk_buff *);
/* keep old push/pop for detaching */
enum pppoatm_encaps encaps;
atomic_t inflight;
unsigned long blocked;
int flags; /* SC_COMP_PROT - compress protocol */
struct ppp_channel chan; /* interface to generic ppp layer */
struct tasklet_struct wakeup_tasklet;
};
/*
* We want to allow two packets in the queue. The one that's currently in
* flight, and *one* queued up ready for the ATM device to send immediately
* from its TX done IRQ. We want to be able to use atomic_inc_not_zero(), so
* inflight == -2 represents an empty queue, -1 one packet, and zero means
* there are two packets in the queue.
*/
#define NONE_INFLIGHT -2
#define BLOCKED 0
/*
* Header used for LLC Encapsulated PPP (4 bytes) followed by the LCP protocol
* ID (0xC021) used in autodetection
*/
static const unsigned char pppllc[6] = { 0xFE, 0xFE, 0x03, 0xCF, 0xC0, 0x21 };
#define LLC_LEN (4)
static inline struct pppoatm_vcc *atmvcc_to_pvcc(const struct atm_vcc *atmvcc)
{
return (struct pppoatm_vcc *) (atmvcc->user_back);
}
static inline struct pppoatm_vcc *chan_to_pvcc(const struct ppp_channel *chan)
{
return (struct pppoatm_vcc *) (chan->private);
}
/*
* We can't do this directly from our _pop handler, since the ppp code
* doesn't want to be called in interrupt context, so we do it from
* a tasklet
*/
static void pppoatm_wakeup_sender(unsigned long arg)
{
ppp_output_wakeup((struct ppp_channel *) arg);
}
/*
* This gets called every time the ATM card has finished sending our
* skb. The ->old_pop will take care up normal atm flow control,
* but we also need to wake up the device if we blocked it
*/
static void pppoatm_pop(struct atm_vcc *atmvcc, struct sk_buff *skb)
{
struct pppoatm_vcc *pvcc = atmvcc_to_pvcc(atmvcc);
pvcc->old_pop(atmvcc, skb);
atomic_dec(&pvcc->inflight);
/*
* We always used to run the wakeup tasklet unconditionally here, for
* fear of race conditions where we clear the BLOCKED flag just as we
* refuse another packet in pppoatm_send(). This was quite inefficient.
*
* In fact it's OK. The PPP core will only ever call pppoatm_send()
* while holding the channel->downl lock. And ppp_output_wakeup() as
* called by the tasklet will *also* grab that lock. So even if another
* CPU is in pppoatm_send() right now, the tasklet isn't going to race
* with it. The wakeup *will* happen after the other CPU is safely out
* of pppoatm_send() again.
*
* So if the CPU in pppoatm_send() has already set the BLOCKED bit and
* it about to return, that's fine. We trigger a wakeup which will
* happen later. And if the CPU in pppoatm_send() *hasn't* set the
* BLOCKED bit yet, that's fine too because of the double check in
* pppoatm_may_send() which is commented there.
*/
if (test_and_clear_bit(BLOCKED, &pvcc->blocked))
tasklet_schedule(&pvcc->wakeup_tasklet);
}
/*
* Unbind from PPP - currently we only do this when closing the socket,
* but we could put this into an ioctl if need be
*/
static void pppoatm_unassign_vcc(struct atm_vcc *atmvcc)
{
struct pppoatm_vcc *pvcc;
pvcc = atmvcc_to_pvcc(atmvcc);
atmvcc->push = pvcc->old_push;
atmvcc->pop = pvcc->old_pop;
tasklet_kill(&pvcc->wakeup_tasklet);
ppp_unregister_channel(&pvcc->chan);
atmvcc->user_back = NULL;
kfree(pvcc);
/* Gee, I hope we have the big kernel lock here... */
module_put(THIS_MODULE);
}
/* Called when an AAL5 PDU comes in */
static void pppoatm_push(struct atm_vcc *atmvcc, struct sk_buff *skb)
{
struct pppoatm_vcc *pvcc = atmvcc_to_pvcc(atmvcc);
pr_debug("\n");
if (skb == NULL) { /* VCC was closed */
pr_debug("removing ATMPPP VCC %p\n", pvcc);
pppoatm_unassign_vcc(atmvcc);
atmvcc->push(atmvcc, NULL); /* Pass along bad news */
return;
}
atm_return(atmvcc, skb->truesize);
switch (pvcc->encaps) {
case e_llc:
if (skb->len < LLC_LEN ||
memcmp(skb->data, pppllc, LLC_LEN))
goto error;
skb_pull(skb, LLC_LEN);
break;
case e_autodetect:
if (pvcc->chan.ppp == NULL) { /* Not bound yet! */
kfree_skb(skb);
return;
}
if (skb->len >= sizeof(pppllc) &&
!memcmp(skb->data, pppllc, sizeof(pppllc))) {
pvcc->encaps = e_llc;
skb_pull(skb, LLC_LEN);
break;
}
if (skb->len >= (sizeof(pppllc) - LLC_LEN) &&
!memcmp(skb->data, &pppllc[LLC_LEN],
sizeof(pppllc) - LLC_LEN)) {
pvcc->encaps = e_vc;
pvcc->chan.mtu += LLC_LEN;
break;
}
pr_debug("Couldn't autodetect yet (skb: %02X %02X %02X %02X %02X %02X)\n",
skb->data[0], skb->data[1], skb->data[2],
skb->data[3], skb->data[4], skb->data[5]);
goto error;
case e_vc:
break;
}
ppp_input(&pvcc->chan, skb);
return;
error:
kfree_skb(skb);
ppp_input_error(&pvcc->chan, 0);
}
static inline int pppoatm_may_send(struct pppoatm_vcc *pvcc, int size)
{
/*
* It's not clear that we need to bother with using atm_may_send()
* to check we don't exceed sk->sk_sndbuf. If userspace sets a
* value of sk_sndbuf which is lower than the MTU, we're going to
* block for ever. But the code always did that before we introduced
* the packet count limit, so...
*/
if (atm_may_send(pvcc->atmvcc, size) &&
atomic_inc_not_zero_hint(&pvcc->inflight, NONE_INFLIGHT))
return 1;
/*
* We use test_and_set_bit() rather than set_bit() here because
* we need to ensure there's a memory barrier after it. The bit
* *must* be set before we do the atomic_inc() on pvcc->inflight.
* There's no smp_mb__after_set_bit(), so it's this or abuse
* smp_mb__after_clear_bit().
*/
test_and_set_bit(BLOCKED, &pvcc->blocked);
/*
* We may have raced with pppoatm_pop(). If it ran for the
* last packet in the queue, *just* before we set the BLOCKED
* bit, then it might never run again and the channel could
* remain permanently blocked. Cope with that race by checking
* *again*. If it did run in that window, we'll have space on
* the queue now and can return success. It's harmless to leave
* the BLOCKED flag set, since it's only used as a trigger to
* run the wakeup tasklet. Another wakeup will never hurt.
* If pppoatm_pop() is running but hasn't got as far as making
* space on the queue yet, then it hasn't checked the BLOCKED
* flag yet either, so we're safe in that case too. It'll issue
* an "immediate" wakeup... where "immediate" actually involves
* taking the PPP channel's ->downl lock, which is held by the
* code path that calls pppoatm_send(), and is thus going to
* wait for us to finish.
*/
if (atm_may_send(pvcc->atmvcc, size) &&
atomic_inc_not_zero(&pvcc->inflight))
return 1;
return 0;
}
/*
* Called by the ppp_generic.c to send a packet - returns true if packet
* was accepted. If we return false, then it's our job to call
* ppp_output_wakeup(chan) when we're feeling more up to it.
* Note that in the ENOMEM case (as opposed to the !atm_may_send case)
* we should really drop the packet, but the generic layer doesn't
* support this yet. We just return 'DROP_PACKET' which we actually define
* as success, just to be clear what we're really doing.
*/
#define DROP_PACKET 1
static int pppoatm_send(struct ppp_channel *chan, struct sk_buff *skb)
{
struct pppoatm_vcc *pvcc = chan_to_pvcc(chan);
ATM_SKB(skb)->vcc = pvcc->atmvcc;
pr_debug("(skb=0x%p, vcc=0x%p)\n", skb, pvcc->atmvcc);
if (skb->data[0] == '\0' && (pvcc->flags & SC_COMP_PROT))
(void) skb_pull(skb, 1);
switch (pvcc->encaps) { /* LLC encapsulation needed */
case e_llc:
if (skb_headroom(skb) < LLC_LEN) {
struct sk_buff *n;
n = skb_realloc_headroom(skb, LLC_LEN);
if (n != NULL &&
!pppoatm_may_send(pvcc, n->truesize)) {
kfree_skb(n);
goto nospace;
}
kfree_skb(skb);
skb = n;
if (skb == NULL)
return DROP_PACKET;
} else if (!pppoatm_may_send(pvcc, skb->truesize))
goto nospace;
memcpy(skb_push(skb, LLC_LEN), pppllc, LLC_LEN);
break;
case e_vc:
if (!pppoatm_may_send(pvcc, skb->truesize))
goto nospace;
break;
case e_autodetect:
pr_debug("Trying to send without setting encaps!\n");
kfree_skb(skb);
return 1;
}
atomic_add(skb->truesize, &sk_atm(ATM_SKB(skb)->vcc)->sk_wmem_alloc);
ATM_SKB(skb)->atm_options = ATM_SKB(skb)->vcc->atm_options;
pr_debug("atm_skb(%p)->vcc(%p)->dev(%p)\n",
skb, ATM_SKB(skb)->vcc, ATM_SKB(skb)->vcc->dev);
return ATM_SKB(skb)->vcc->send(ATM_SKB(skb)->vcc, skb)
? DROP_PACKET : 1;
nospace:
/*
* We don't have space to send this SKB now, but we might have
* already applied SC_COMP_PROT compression, so may need to undo
*/
if ((pvcc->flags & SC_COMP_PROT) && skb_headroom(skb) > 0 &&
skb->data[-1] == '\0')
(void) skb_push(skb, 1);
return 0;
}
/* This handles ioctls sent to the /dev/ppp interface */
static int pppoatm_devppp_ioctl(struct ppp_channel *chan, unsigned int cmd,
unsigned long arg)
{
switch (cmd) {
case PPPIOCGFLAGS:
return put_user(chan_to_pvcc(chan)->flags, (int __user *) arg)
? -EFAULT : 0;
case PPPIOCSFLAGS:
return get_user(chan_to_pvcc(chan)->flags, (int __user *) arg)
? -EFAULT : 0;
}
return -ENOTTY;
}
static const struct ppp_channel_ops pppoatm_ops = {
.start_xmit = pppoatm_send,
.ioctl = pppoatm_devppp_ioctl,
};
static int pppoatm_assign_vcc(struct atm_vcc *atmvcc, void __user *arg)
{
struct atm_backend_ppp be;
struct pppoatm_vcc *pvcc;
int err;
/*
* Each PPPoATM instance has its own tasklet - this is just a
* prototypical one used to initialize them
*/
static const DECLARE_TASKLET(tasklet_proto, pppoatm_wakeup_sender, 0);
if (copy_from_user(&be, arg, sizeof be))
return -EFAULT;
if (be.encaps != PPPOATM_ENCAPS_AUTODETECT &&
be.encaps != PPPOATM_ENCAPS_VC && be.encaps != PPPOATM_ENCAPS_LLC)
return -EINVAL;
pvcc = kzalloc(sizeof(*pvcc), GFP_KERNEL);
if (pvcc == NULL)
return -ENOMEM;
pvcc->atmvcc = atmvcc;
/* Maximum is zero, so that we can use atomic_inc_not_zero() */
atomic_set(&pvcc->inflight, NONE_INFLIGHT);
pvcc->old_push = atmvcc->push;
pvcc->old_pop = atmvcc->pop;
pvcc->encaps = (enum pppoatm_encaps) be.encaps;
pvcc->chan.private = pvcc;
pvcc->chan.ops = &pppoatm_ops;
pvcc->chan.mtu = atmvcc->qos.txtp.max_sdu - PPP_HDRLEN -
(be.encaps == e_vc ? 0 : LLC_LEN);
pvcc->wakeup_tasklet = tasklet_proto;
pvcc->wakeup_tasklet.data = (unsigned long) &pvcc->chan;
err = ppp_register_channel(&pvcc->chan);
if (err != 0) {
kfree(pvcc);
return err;
}
atmvcc->user_back = pvcc;
atmvcc->push = pppoatm_push;
atmvcc->pop = pppoatm_pop;
__module_get(THIS_MODULE);
/* re-process everything received between connection setup and
backend setup */
vcc_process_recv_queue(atmvcc);
return 0;
}
/*
* This handles ioctls actually performed on our vcc - we must return
* -ENOIOCTLCMD for any unrecognized ioctl
*/
static int pppoatm_ioctl(struct socket *sock, unsigned int cmd,
unsigned long arg)
{
struct atm_vcc *atmvcc = ATM_SD(sock);
void __user *argp = (void __user *)arg;
if (cmd != ATM_SETBACKEND && atmvcc->push != pppoatm_push)
return -ENOIOCTLCMD;
switch (cmd) {
case ATM_SETBACKEND: {
atm_backend_t b;
if (get_user(b, (atm_backend_t __user *) argp))
return -EFAULT;
if (b != ATM_BACKEND_PPP)
return -ENOIOCTLCMD;
if (!capable(CAP_NET_ADMIN))
return -EPERM;
return pppoatm_assign_vcc(atmvcc, argp);
}
case PPPIOCGCHAN:
return put_user(ppp_channel_index(&atmvcc_to_pvcc(atmvcc)->
chan), (int __user *) argp) ? -EFAULT : 0;
case PPPIOCGUNIT:
return put_user(ppp_unit_number(&atmvcc_to_pvcc(atmvcc)->
chan), (int __user *) argp) ? -EFAULT : 0;
}
return -ENOIOCTLCMD;
}
static struct atm_ioctl pppoatm_ioctl_ops = {
.owner = THIS_MODULE,
.ioctl = pppoatm_ioctl,
};
static int __init pppoatm_init(void)
{
register_atm_ioctl(&pppoatm_ioctl_ops);
return 0;
}
static void __exit pppoatm_exit(void)
{
deregister_atm_ioctl(&pppoatm_ioctl_ops);
}
module_init(pppoatm_init);
module_exit(pppoatm_exit);
MODULE_AUTHOR("Mitchell Blank Jr <mitch@sfgoth.com>");
MODULE_DESCRIPTION("RFC2364 PPP over ATM/AAL5");
MODULE_LICENSE("GPL");