linux/drivers/net/ppp/ppp_synctty.c

791 lines
18 KiB
C
Raw Normal View History

/*
* PPP synchronous tty channel driver for Linux.
*
* This is a ppp channel driver that can be used with tty device drivers
* that are frame oriented, such as synchronous HDLC devices.
*
* Complete PPP frames without encoding/decoding are exchanged between
* the channel driver and the device driver.
*
* The async map IOCTL codes are implemented to keep the user mode
* applications happy if they call them. Synchronous PPP does not use
* the async maps.
*
* Copyright 1999 Paul Mackerras.
*
* Also touched by the grubby hands of Paul Fulghum paulkf@microgate.com
*
* 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 over sync serial lines. It relies on
* the generic PPP layer to give it frames to send and to process
* received frames. It implements the PPP line discipline.
*
* Part of the code in this driver was inspired by the old async-only
* PPP driver, written by Michael Callahan and Al Longyear, and
* subsequently hacked by Paul Mackerras.
*
* ==FILEVERSION 20040616==
*/
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>
#include <linux/tty.h>
#include <linux/netdevice.h>
#include <linux/poll.h>
#include <linux/ppp_defs.h>
#include <linux/ppp-ioctl.h>
#include <linux/ppp_channel.h>
#include <linux/spinlock.h>
#include <linux/completion.h>
#include <linux/init.h>
#include <linux/interrupt.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <asm/unaligned.h>
#include <asm/uaccess.h>
#define PPP_VERSION "2.4.2"
/* Structure for storing local state. */
struct syncppp {
struct tty_struct *tty;
unsigned int flags;
unsigned int rbits;
int mru;
spinlock_t xmit_lock;
spinlock_t recv_lock;
unsigned long xmit_flags;
u32 xaccm[8];
u32 raccm;
unsigned int bytes_sent;
unsigned int bytes_rcvd;
struct sk_buff *tpkt;
unsigned long last_xmit;
struct sk_buff_head rqueue;
struct tasklet_struct tsk;
atomic_t refcnt;
struct completion dead_cmp;
struct ppp_channel chan; /* interface to generic ppp layer */
};
/* Bit numbers in xmit_flags */
#define XMIT_WAKEUP 0
#define XMIT_FULL 1
/* Bits in rbits */
#define SC_RCV_BITS (SC_RCV_B7_1|SC_RCV_B7_0|SC_RCV_ODDP|SC_RCV_EVNP)
#define PPPSYNC_MAX_RQLEN 32 /* arbitrary */
/*
* Prototypes.
*/
static struct sk_buff* ppp_sync_txmunge(struct syncppp *ap, struct sk_buff *);
static int ppp_sync_send(struct ppp_channel *chan, struct sk_buff *skb);
static int ppp_sync_ioctl(struct ppp_channel *chan, unsigned int cmd,
unsigned long arg);
static void ppp_sync_process(unsigned long arg);
static int ppp_sync_push(struct syncppp *ap);
static void ppp_sync_flush_output(struct syncppp *ap);
static void ppp_sync_input(struct syncppp *ap, const unsigned char *buf,
char *flags, int count);
static const struct ppp_channel_ops sync_ops = {
.start_xmit = ppp_sync_send,
.ioctl = ppp_sync_ioctl,
};
/*
* Utility procedures to print a buffer in hex/ascii
*/
static void
ppp_print_hex (register __u8 * out, const __u8 * in, int count)
{
register __u8 next_ch;
2006-03-04 02:33:57 +00:00
static const char hex[] = "0123456789ABCDEF";
while (count-- > 0) {
next_ch = *in++;
*out++ = hex[(next_ch >> 4) & 0x0F];
*out++ = hex[next_ch & 0x0F];
++out;
}
}
static void
ppp_print_char (register __u8 * out, const __u8 * in, int count)
{
register __u8 next_ch;
while (count-- > 0) {
next_ch = *in++;
if (next_ch < 0x20 || next_ch > 0x7e)
*out++ = '.';
else {
*out++ = next_ch;
if (next_ch == '%') /* printk/syslogd has a bug !! */
*out++ = '%';
}
}
*out = '\0';
}
static void
ppp_print_buffer (const char *name, const __u8 *buf, int count)
{
__u8 line[44];
if (name != NULL)
printk(KERN_DEBUG "ppp_synctty: %s, count = %d\n", name, count);
while (count > 8) {
memset (line, 32, 44);
ppp_print_hex (line, buf, 8);
ppp_print_char (&line[8 * 3], buf, 8);
printk(KERN_DEBUG "%s\n", line);
count -= 8;
buf += 8;
}
if (count > 0) {
memset (line, 32, 44);
ppp_print_hex (line, buf, count);
ppp_print_char (&line[8 * 3], buf, count);
printk(KERN_DEBUG "%s\n", line);
}
}
/*
* Routines implementing the synchronous PPP line discipline.
*/
/*
* We have a potential race on dereferencing tty->disc_data,
* because the tty layer provides no locking at all - thus one
* cpu could be running ppp_synctty_receive while another
* calls ppp_synctty_close, which zeroes tty->disc_data and
* frees the memory that ppp_synctty_receive is using. The best
* way to fix this is to use a rwlock in the tty struct, but for now
* we use a single global rwlock for all ttys in ppp line discipline.
*
* FIXME: Fixed in tty_io nowadays.
*/
static DEFINE_RWLOCK(disc_data_lock);
static struct syncppp *sp_get(struct tty_struct *tty)
{
struct syncppp *ap;
read_lock(&disc_data_lock);
ap = tty->disc_data;
if (ap != NULL)
atomic_inc(&ap->refcnt);
read_unlock(&disc_data_lock);
return ap;
}
static void sp_put(struct syncppp *ap)
{
if (atomic_dec_and_test(&ap->refcnt))
complete(&ap->dead_cmp);
}
/*
* Called when a tty is put into sync-PPP line discipline.
*/
static int
ppp_sync_open(struct tty_struct *tty)
{
struct syncppp *ap;
int err;
ppp: ppp_mp_explode() redesign I found the PPP subsystem to not work properly when connecting channels with different speeds to the same bundle. Problem Description: As the "ppp_mp_explode" function fragments the sk_buff buffer evenly among the PPP channels that are connected to a certain PPP unit to make up a bundle, if we are transmitting using an upper layer protocol that requires an Ack before sending the next packet (like TCP/IP for example), we will have a bandwidth bottleneck on the slowest channel of the bundle. Let's clarify by an example. Let's consider a scenario where we have two PPP links making up a bundle: a slow link (10KB/sec) and a fast link (1000KB/sec) working at the best (full bandwidth). On the top we have a TCP/IP stack sending a 1000 Bytes sk_buff buffer down to the PPP subsystem. The "ppp_mp_explode" function will divide the buffer in two fragments of 500B each (we are neglecting all the headers, crc, flags etc?.). Before the TCP/IP stack sends out the next buffer, it will have to wait for the ACK response from the remote peer, so it will have to wait for both fragments to have been sent over the two PPP links, received by the remote peer and reconstructed. The resulting behaviour is that, rather than having a bundle working @1010KB/sec (the sum of the channels bandwidths), we'll have a bundle working @20KB/sec (the double of the slowest channels bandwidth). Problem Solution: The problem has been solved by redesigning the "ppp_mp_explode" function in such a way to make it split the sk_buff buffer according to the speeds of the underlying PPP channels (the speeds of the serial interfaces respectively attached to the PPP channels). Referring to the above example, the redesigned "ppp_mp_explode" function will now divide the 1000 Bytes buffer into two fragments whose sizes are set according to the speeds of the channels where they are going to be sent on (e.g . 10 Byets on 10KB/sec channel and 990 Bytes on 1000KB/sec channel). The reworked function grants the same performances of the original one in optimal working conditions (i.e. a bundle made up of PPP links all working at the same speed), while greatly improving performances on the bundles made up of channels working at different speeds. Signed-off-by: Gabriele Paoloni <gabriele.paoloni@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 23:09:12 +00:00
int speed;
if (tty->ops->write == NULL)
return -EOPNOTSUPP;
2007-07-19 08:49:03 +00:00
ap = kzalloc(sizeof(*ap), GFP_KERNEL);
err = -ENOMEM;
if (!ap)
goto out;
/* initialize the syncppp structure */
ap->tty = tty;
ap->mru = PPP_MRU;
spin_lock_init(&ap->xmit_lock);
spin_lock_init(&ap->recv_lock);
ap->xaccm[0] = ~0U;
ap->xaccm[3] = 0x60000000U;
ap->raccm = ~0U;
skb_queue_head_init(&ap->rqueue);
tasklet_init(&ap->tsk, ppp_sync_process, (unsigned long) ap);
atomic_set(&ap->refcnt, 1);
init_completion(&ap->dead_cmp);
ap->chan.private = ap;
ap->chan.ops = &sync_ops;
ap->chan.mtu = PPP_MRU;
ap->chan.hdrlen = 2; /* for A/C bytes */
ppp: ppp_mp_explode() redesign I found the PPP subsystem to not work properly when connecting channels with different speeds to the same bundle. Problem Description: As the "ppp_mp_explode" function fragments the sk_buff buffer evenly among the PPP channels that are connected to a certain PPP unit to make up a bundle, if we are transmitting using an upper layer protocol that requires an Ack before sending the next packet (like TCP/IP for example), we will have a bandwidth bottleneck on the slowest channel of the bundle. Let's clarify by an example. Let's consider a scenario where we have two PPP links making up a bundle: a slow link (10KB/sec) and a fast link (1000KB/sec) working at the best (full bandwidth). On the top we have a TCP/IP stack sending a 1000 Bytes sk_buff buffer down to the PPP subsystem. The "ppp_mp_explode" function will divide the buffer in two fragments of 500B each (we are neglecting all the headers, crc, flags etc?.). Before the TCP/IP stack sends out the next buffer, it will have to wait for the ACK response from the remote peer, so it will have to wait for both fragments to have been sent over the two PPP links, received by the remote peer and reconstructed. The resulting behaviour is that, rather than having a bundle working @1010KB/sec (the sum of the channels bandwidths), we'll have a bundle working @20KB/sec (the double of the slowest channels bandwidth). Problem Solution: The problem has been solved by redesigning the "ppp_mp_explode" function in such a way to make it split the sk_buff buffer according to the speeds of the underlying PPP channels (the speeds of the serial interfaces respectively attached to the PPP channels). Referring to the above example, the redesigned "ppp_mp_explode" function will now divide the 1000 Bytes buffer into two fragments whose sizes are set according to the speeds of the channels where they are going to be sent on (e.g . 10 Byets on 10KB/sec channel and 990 Bytes on 1000KB/sec channel). The reworked function grants the same performances of the original one in optimal working conditions (i.e. a bundle made up of PPP links all working at the same speed), while greatly improving performances on the bundles made up of channels working at different speeds. Signed-off-by: Gabriele Paoloni <gabriele.paoloni@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-03-13 23:09:12 +00:00
speed = tty_get_baud_rate(tty);
ap->chan.speed = speed;
err = ppp_register_channel(&ap->chan);
if (err)
goto out_free;
tty->disc_data = ap;
[PATCH] TTY layer buffering revamp The API and code have been through various bits of initial review by serial driver people but they definitely need to live somewhere for a while so the unconverted drivers can get knocked into shape, existing drivers that have been updated can be better tuned and bugs whacked out. This replaces the tty flip buffers with kmalloc objects in rings. In the normal situation for an IRQ driven serial port at typical speeds the behaviour is pretty much the same, two buffers end up allocated and the kernel cycles between them as before. When there are delays or at high speed we now behave far better as the buffer pool can grow a bit rather than lose characters. This also means that we can operate at higher speeds reliably. For drivers that receive characters in blocks (DMA based, USB and especially virtualisation) the layer allows a lot of driver specific code that works around the tty layer with private secondary queues to be removed. The IBM folks need this sort of layer, the smart serial port people do, the virtualisers do (because a virtualised tty typically operates at infinite speed rather than emulating 9600 baud). Finally many drivers had invalid and unsafe attempts to avoid buffer overflows by directly invoking tty methods extracted out of the innards of work queue structs. These are no longer needed and all go away. That fixes various random hangs with serial ports on overflow. The other change in here is to optimise the receive_room path that is used by some callers. It turns out that only one ldisc uses receive room except asa constant and it updates it far far less than the value is read. We thus make it a variable not a function call. I expect the code to contain bugs due to the size alone but I'll be watching and squashing them and feeding out new patches as it goes. Because the buffers now dynamically expand you should only run out of buffering when the kernel runs out of memory for real. That means a lot of the horrible hacks high performance drivers used to do just aren't needed any more. Description: tty_insert_flip_char is an old API and continues to work as before, as does tty_flip_buffer_push() [this is why many drivers dont need modification]. It does now also return the number of chars inserted There are also tty_buffer_request_room(tty, len) which asks for a buffer block of the length requested and returns the space found. This improves efficiency with hardware that knows how much to transfer. and tty_insert_flip_string_flags(tty, str, flags, len) to insert a string of characters and flags For a smart interface the usual code is len = tty_request_buffer_room(tty, amount_hardware_says); tty_insert_flip_string(tty, buffer_from_card, len); More description! At the moment tty buffers are attached directly to the tty. This is causing a lot of the problems related to tty layer locking, also problems at high speed and also with bursty data (such as occurs in virtualised environments) I'm working on ripping out the flip buffers and replacing them with a pool of dynamically allocated buffers. This allows both for old style "byte I/O" devices and also helps virtualisation and smart devices where large blocks of data suddenely materialise and need storing. So far so good. Lots of drivers reference tty->flip.*. Several of them also call directly and unsafely into function pointers it provides. This will all break. Most drivers can use tty_insert_flip_char which can be kept as an API but others need more. At the moment I've added the following interfaces, if people think more will be needed now is a good time to say int tty_buffer_request_room(tty, size) Try and ensure at least size bytes are available, returns actual room (may be zero). At the moment it just uses the flipbuf space but that will change. Repeated calls without characters being added are not cumulative. (ie if you call it with 1, 1, 1, and then 4 you'll have four characters of space. The other functions will also try and grow buffers in future but this will be a more efficient way when you know block sizes. int tty_insert_flip_char(tty, ch, flag) As before insert a character if there is room. Now returns 1 for success, 0 for failure. int tty_insert_flip_string(tty, str, len) Insert a block of non error characters. Returns the number inserted. int tty_prepare_flip_string(tty, strptr, len) Adjust the buffer to allow len characters to be added. Returns a buffer pointer in strptr and the length available. This allows for hardware that needs to use functions like insl or mencpy_fromio. Signed-off-by: Alan Cox <alan@redhat.com> Cc: Paul Fulghum <paulkf@microgate.com> Signed-off-by: Hirokazu Takata <takata@linux-m32r.org> Signed-off-by: Serge Hallyn <serue@us.ibm.com> Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: John Hawkes <hawkes@sgi.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-10 04:54:13 +00:00
tty->receive_room = 65536;
return 0;
out_free:
kfree(ap);
out:
return err;
}
/*
* Called when the tty is put into another line discipline
* or it hangs up. We have to wait for any cpu currently
* executing in any of the other ppp_synctty_* routines to
* finish before we can call ppp_unregister_channel and free
* the syncppp struct. This routine must be called from
* process context, not interrupt or softirq context.
*/
static void
ppp_sync_close(struct tty_struct *tty)
{
struct syncppp *ap;
write_lock_irq(&disc_data_lock);
ap = tty->disc_data;
tty->disc_data = NULL;
write_unlock_irq(&disc_data_lock);
if (!ap)
return;
/*
* We have now ensured that nobody can start using ap from now
* on, but we have to wait for all existing users to finish.
* Note that ppp_unregister_channel ensures that no calls to
* our channel ops (i.e. ppp_sync_send/ioctl) are in progress
* by the time it returns.
*/
if (!atomic_dec_and_test(&ap->refcnt))
wait_for_completion(&ap->dead_cmp);
tasklet_kill(&ap->tsk);
ppp_unregister_channel(&ap->chan);
skb_queue_purge(&ap->rqueue);
kfree_skb(ap->tpkt);
kfree(ap);
}
/*
* Called on tty hangup in process context.
*
* Wait for I/O to driver to complete and unregister PPP channel.
* This is already done by the close routine, so just call that.
*/
static int ppp_sync_hangup(struct tty_struct *tty)
{
ppp_sync_close(tty);
return 0;
}
/*
* Read does nothing - no data is ever available this way.
* Pppd reads and writes packets via /dev/ppp instead.
*/
static ssize_t
ppp_sync_read(struct tty_struct *tty, struct file *file,
unsigned char __user *buf, size_t count)
{
return -EAGAIN;
}
/*
* Write on the tty does nothing, the packets all come in
* from the ppp generic stuff.
*/
static ssize_t
ppp_sync_write(struct tty_struct *tty, struct file *file,
const unsigned char *buf, size_t count)
{
return -EAGAIN;
}
static int
ppp_synctty_ioctl(struct tty_struct *tty, struct file *file,
unsigned int cmd, unsigned long arg)
{
struct syncppp *ap = sp_get(tty);
int __user *p = (int __user *)arg;
int err, val;
if (!ap)
return -ENXIO;
err = -EFAULT;
switch (cmd) {
case PPPIOCGCHAN:
err = -EFAULT;
if (put_user(ppp_channel_index(&ap->chan), p))
break;
err = 0;
break;
case PPPIOCGUNIT:
err = -EFAULT;
if (put_user(ppp_unit_number(&ap->chan), p))
break;
err = 0;
break;
case TCFLSH:
/* flush our buffers and the serial port's buffer */
if (arg == TCIOFLUSH || arg == TCOFLUSH)
ppp_sync_flush_output(ap);
err = tty_perform_flush(tty, arg);
break;
case FIONREAD:
val = 0;
if (put_user(val, p))
break;
err = 0;
break;
default:
err = tty_mode_ioctl(tty, file, cmd, arg);
break;
}
sp_put(ap);
return err;
}
/* No kernel lock - fine */
static unsigned int
ppp_sync_poll(struct tty_struct *tty, struct file *file, poll_table *wait)
{
return 0;
}
/* May sleep, don't call from interrupt level or with interrupts disabled */
Revert "tty: make receive_buf() return the amout of bytes received" This reverts commit b1c43f82c5aa265442f82dba31ce985ebb7aa71c. It was broken in so many ways, and results in random odd pty issues. It re-introduced the buggy schedule_work() in flush_to_ldisc() that can cause endless work-loops (see commit a5660b41af6a: "tty: fix endless work loop when the buffer fills up"). It also used an "unsigned int" return value fo the ->receive_buf() function, but then made multiple functions return a negative error code, and didn't actually check for the error in the caller. And it didn't actually work at all. BenH bisected down odd tty behavior to it: "It looks like the patch is causing some major malfunctions of the X server for me, possibly related to PTYs. For example, cat'ing a large file in a gnome terminal hangs the kernel for -minutes- in a loop of what looks like flush_to_ldisc/workqueue code, (some ftrace data in the quoted bits further down). ... Some more data: It -looks- like what happens is that the flush_to_ldisc work queue entry constantly re-queues itself (because the PTY is full ?) and the workqueue thread will basically loop forver calling it without ever scheduling, thus starving the consumer process that could have emptied the PTY." which is pretty much exactly the problem we fixed in a5660b41af6a. Milton Miller pointed out the 'unsigned int' issue. Reported-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Reported-by: Milton Miller <miltonm@bga.com> Cc: Stefan Bigler <stefan.bigler@keymile.com> Cc: Toby Gray <toby.gray@realvnc.com> Cc: Felipe Balbi <balbi@ti.com> Cc: Greg Kroah-Hartman <gregkh@suse.de> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-06-03 21:33:24 +00:00
static void
ppp_sync_receive(struct tty_struct *tty, const unsigned char *buf,
char *cflags, int count)
{
struct syncppp *ap = sp_get(tty);
unsigned long flags;
if (!ap)
Revert "tty: make receive_buf() return the amout of bytes received" This reverts commit b1c43f82c5aa265442f82dba31ce985ebb7aa71c. It was broken in so many ways, and results in random odd pty issues. It re-introduced the buggy schedule_work() in flush_to_ldisc() that can cause endless work-loops (see commit a5660b41af6a: "tty: fix endless work loop when the buffer fills up"). It also used an "unsigned int" return value fo the ->receive_buf() function, but then made multiple functions return a negative error code, and didn't actually check for the error in the caller. And it didn't actually work at all. BenH bisected down odd tty behavior to it: "It looks like the patch is causing some major malfunctions of the X server for me, possibly related to PTYs. For example, cat'ing a large file in a gnome terminal hangs the kernel for -minutes- in a loop of what looks like flush_to_ldisc/workqueue code, (some ftrace data in the quoted bits further down). ... Some more data: It -looks- like what happens is that the flush_to_ldisc work queue entry constantly re-queues itself (because the PTY is full ?) and the workqueue thread will basically loop forver calling it without ever scheduling, thus starving the consumer process that could have emptied the PTY." which is pretty much exactly the problem we fixed in a5660b41af6a. Milton Miller pointed out the 'unsigned int' issue. Reported-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Reported-by: Milton Miller <miltonm@bga.com> Cc: Stefan Bigler <stefan.bigler@keymile.com> Cc: Toby Gray <toby.gray@realvnc.com> Cc: Felipe Balbi <balbi@ti.com> Cc: Greg Kroah-Hartman <gregkh@suse.de> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-06-03 21:33:24 +00:00
return;
spin_lock_irqsave(&ap->recv_lock, flags);
ppp_sync_input(ap, buf, cflags, count);
spin_unlock_irqrestore(&ap->recv_lock, flags);
if (!skb_queue_empty(&ap->rqueue))
tasklet_schedule(&ap->tsk);
sp_put(ap);
tty_unthrottle(tty);
}
static void
ppp_sync_wakeup(struct tty_struct *tty)
{
struct syncppp *ap = sp_get(tty);
clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
if (!ap)
return;
set_bit(XMIT_WAKEUP, &ap->xmit_flags);
tasklet_schedule(&ap->tsk);
sp_put(ap);
}
static struct tty_ldisc_ops ppp_sync_ldisc = {
.owner = THIS_MODULE,
.magic = TTY_LDISC_MAGIC,
.name = "pppsync",
.open = ppp_sync_open,
.close = ppp_sync_close,
.hangup = ppp_sync_hangup,
.read = ppp_sync_read,
.write = ppp_sync_write,
.ioctl = ppp_synctty_ioctl,
.poll = ppp_sync_poll,
.receive_buf = ppp_sync_receive,
.write_wakeup = ppp_sync_wakeup,
};
static int __init
ppp_sync_init(void)
{
int err;
err = tty_register_ldisc(N_SYNC_PPP, &ppp_sync_ldisc);
if (err != 0)
printk(KERN_ERR "PPP_sync: error %d registering line disc.\n",
err);
return err;
}
/*
* The following routines provide the PPP channel interface.
*/
static int
ppp_sync_ioctl(struct ppp_channel *chan, unsigned int cmd, unsigned long arg)
{
struct syncppp *ap = chan->private;
int err, val;
u32 accm[8];
void __user *argp = (void __user *)arg;
u32 __user *p = argp;
err = -EFAULT;
switch (cmd) {
case PPPIOCGFLAGS:
val = ap->flags | ap->rbits;
if (put_user(val, (int __user *) argp))
break;
err = 0;
break;
case PPPIOCSFLAGS:
if (get_user(val, (int __user *) argp))
break;
ap->flags = val & ~SC_RCV_BITS;
spin_lock_irq(&ap->recv_lock);
ap->rbits = val & SC_RCV_BITS;
spin_unlock_irq(&ap->recv_lock);
err = 0;
break;
case PPPIOCGASYNCMAP:
if (put_user(ap->xaccm[0], p))
break;
err = 0;
break;
case PPPIOCSASYNCMAP:
if (get_user(ap->xaccm[0], p))
break;
err = 0;
break;
case PPPIOCGRASYNCMAP:
if (put_user(ap->raccm, p))
break;
err = 0;
break;
case PPPIOCSRASYNCMAP:
if (get_user(ap->raccm, p))
break;
err = 0;
break;
case PPPIOCGXASYNCMAP:
if (copy_to_user(argp, ap->xaccm, sizeof(ap->xaccm)))
break;
err = 0;
break;
case PPPIOCSXASYNCMAP:
if (copy_from_user(accm, argp, sizeof(accm)))
break;
accm[2] &= ~0x40000000U; /* can't escape 0x5e */
accm[3] |= 0x60000000U; /* must escape 0x7d, 0x7e */
memcpy(ap->xaccm, accm, sizeof(ap->xaccm));
err = 0;
break;
case PPPIOCGMRU:
if (put_user(ap->mru, (int __user *) argp))
break;
err = 0;
break;
case PPPIOCSMRU:
if (get_user(val, (int __user *) argp))
break;
if (val < PPP_MRU)
val = PPP_MRU;
ap->mru = val;
err = 0;
break;
default:
err = -ENOTTY;
}
return err;
}
/*
* This is called at softirq level to deliver received packets
* to the ppp_generic code, and to tell the ppp_generic code
* if we can accept more output now.
*/
static void ppp_sync_process(unsigned long arg)
{
struct syncppp *ap = (struct syncppp *) arg;
struct sk_buff *skb;
/* process received packets */
while ((skb = skb_dequeue(&ap->rqueue)) != NULL) {
if (skb->len == 0) {
/* zero length buffers indicate error */
ppp_input_error(&ap->chan, 0);
kfree_skb(skb);
}
else
ppp_input(&ap->chan, skb);
}
/* try to push more stuff out */
if (test_bit(XMIT_WAKEUP, &ap->xmit_flags) && ppp_sync_push(ap))
ppp_output_wakeup(&ap->chan);
}
/*
* Procedures for encapsulation and framing.
*/
static struct sk_buff*
ppp_sync_txmunge(struct syncppp *ap, struct sk_buff *skb)
{
int proto;
unsigned char *data;
int islcp;
data = skb->data;
proto = get_unaligned_be16(data);
/* LCP packets with codes between 1 (configure-request)
* and 7 (code-reject) must be sent as though no options
* have been negotiated.
*/
islcp = proto == PPP_LCP && 1 <= data[2] && data[2] <= 7;
/* compress protocol field if option enabled */
if (data[0] == 0 && (ap->flags & SC_COMP_PROT) && !islcp)
skb_pull(skb,1);
/* prepend address/control fields if necessary */
if ((ap->flags & SC_COMP_AC) == 0 || islcp) {
if (skb_headroom(skb) < 2) {
struct sk_buff *npkt = dev_alloc_skb(skb->len + 2);
if (npkt == NULL) {
kfree_skb(skb);
return NULL;
}
skb_reserve(npkt,2);
skb_copy_from_linear_data(skb,
skb_put(npkt, skb->len), skb->len);
kfree_skb(skb);
skb = npkt;
}
skb_push(skb,2);
skb->data[0] = PPP_ALLSTATIONS;
skb->data[1] = PPP_UI;
}
ap->last_xmit = jiffies;
if (skb && ap->flags & SC_LOG_OUTPKT)
ppp_print_buffer ("send buffer", skb->data, skb->len);
return skb;
}
/*
* Transmit-side routines.
*/
/*
* Send a packet to the peer over an sync tty line.
* Returns 1 iff the packet was accepted.
* If the packet was not accepted, we will call ppp_output_wakeup
* at some later time.
*/
static int
ppp_sync_send(struct ppp_channel *chan, struct sk_buff *skb)
{
struct syncppp *ap = chan->private;
ppp_sync_push(ap);
if (test_and_set_bit(XMIT_FULL, &ap->xmit_flags))
return 0; /* already full */
skb = ppp_sync_txmunge(ap, skb);
if (skb != NULL)
ap->tpkt = skb;
else
clear_bit(XMIT_FULL, &ap->xmit_flags);
ppp_sync_push(ap);
return 1;
}
/*
* Push as much data as possible out to the tty.
*/
static int
ppp_sync_push(struct syncppp *ap)
{
int sent, done = 0;
struct tty_struct *tty = ap->tty;
int tty_stuffed = 0;
if (!spin_trylock_bh(&ap->xmit_lock))
return 0;
for (;;) {
if (test_and_clear_bit(XMIT_WAKEUP, &ap->xmit_flags))
tty_stuffed = 0;
if (!tty_stuffed && ap->tpkt) {
set_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
sent = tty->ops->write(tty, ap->tpkt->data, ap->tpkt->len);
if (sent < 0)
goto flush; /* error, e.g. loss of CD */
if (sent < ap->tpkt->len) {
tty_stuffed = 1;
} else {
kfree_skb(ap->tpkt);
ap->tpkt = NULL;
clear_bit(XMIT_FULL, &ap->xmit_flags);
done = 1;
}
continue;
}
/* haven't made any progress */
spin_unlock_bh(&ap->xmit_lock);
if (!(test_bit(XMIT_WAKEUP, &ap->xmit_flags) ||
(!tty_stuffed && ap->tpkt)))
break;
if (!spin_trylock_bh(&ap->xmit_lock))
break;
}
return done;
flush:
if (ap->tpkt) {
kfree_skb(ap->tpkt);
ap->tpkt = NULL;
clear_bit(XMIT_FULL, &ap->xmit_flags);
done = 1;
}
spin_unlock_bh(&ap->xmit_lock);
return done;
}
/*
* Flush output from our internal buffers.
* Called for the TCFLSH ioctl.
*/
static void
ppp_sync_flush_output(struct syncppp *ap)
{
int done = 0;
spin_lock_bh(&ap->xmit_lock);
if (ap->tpkt != NULL) {
kfree_skb(ap->tpkt);
ap->tpkt = NULL;
clear_bit(XMIT_FULL, &ap->xmit_flags);
done = 1;
}
spin_unlock_bh(&ap->xmit_lock);
if (done)
ppp_output_wakeup(&ap->chan);
}
/*
* Receive-side routines.
*/
/* called when the tty driver has data for us.
*
* Data is frame oriented: each call to ppp_sync_input is considered
* a whole frame. If the 1st flag byte is non-zero then the whole
* frame is considered to be in error and is tossed.
*/
static void
ppp_sync_input(struct syncppp *ap, const unsigned char *buf,
char *flags, int count)
{
struct sk_buff *skb;
unsigned char *p;
if (count == 0)
return;
if (ap->flags & SC_LOG_INPKT)
ppp_print_buffer ("receive buffer", buf, count);
/* stuff the chars in the skb */
skb = dev_alloc_skb(ap->mru + PPP_HDRLEN + 2);
if (!skb) {
printk(KERN_ERR "PPPsync: no memory (input pkt)\n");
goto err;
}
/* Try to get the payload 4-byte aligned */
if (buf[0] != PPP_ALLSTATIONS)
skb_reserve(skb, 2 + (buf[0] & 1));
if (flags && *flags) {
/* error flag set, ignore frame */
goto err;
} else if (count > skb_tailroom(skb)) {
/* packet overflowed MRU */
goto err;
}
p = skb_put(skb, count);
memcpy(p, buf, count);
/* strip address/control field if present */
p = skb->data;
if (p[0] == PPP_ALLSTATIONS && p[1] == PPP_UI) {
/* chop off address/control */
if (skb->len < 3)
goto err;
p = skb_pull(skb, 2);
}
/* decompress protocol field if compressed */
if (p[0] & 1) {
/* protocol is compressed */
skb_push(skb, 1)[0] = 0;
} else if (skb->len < 2)
goto err;
/* queue the frame to be processed */
skb_queue_tail(&ap->rqueue, skb);
return;
err:
/* queue zero length packet as error indication */
if (skb || (skb = dev_alloc_skb(0))) {
skb_trim(skb, 0);
skb_queue_tail(&ap->rqueue, skb);
}
}
static void __exit
ppp_sync_cleanup(void)
{
if (tty_unregister_ldisc(N_SYNC_PPP) != 0)
printk(KERN_ERR "failed to unregister Sync PPP line discipline\n");
}
module_init(ppp_sync_init);
module_exit(ppp_sync_cleanup);
MODULE_LICENSE("GPL");
MODULE_ALIAS_LDISC(N_SYNC_PPP);