From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
Paul Fulghum <paulkf@microgate.com>,
Greg Kroah-Hartman <gregkh@suse.de>
Subject: [46/46] tty: Make flush_to_ldisc() locking more robust
Date: Fri, 16 Oct 2009 10:10:39 -0700 [thread overview]
Message-ID: <20091016171210.987917756@linux.site> (raw)
In-Reply-To: <20091016171422.GA13339@kroah.com>
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: tty-make-flush_to_ldisc-locking-more-robust.patch --]
[-- Type: text/plain, Size: 4345 bytes --]
2.6.31-stable review patch. If anyone has any objections, please let us know.
------------------
From: Linus Torvalds <torvalds@linux-foundation.org>
commit c8e33141911bf8fe87dc6c92793b9a59b2be0130 upstream.
The locking logic in this function is extremely subtle, and it broke
when we started doing potentially concurrent 'flush_to_ldisc()' calls in
commit e043e42bdb66885b3ac10d27a01ccb9972e2b0a3 ("pty: avoid forcing
'low_latency' tty flag").
The code in flush_to_ldisc() used to set 'tty->buf.head' to NULL, with
the intention that this would then cause any other concurrent calls to
not do anything (locking note: we have to drop the buf.lock over the
call to ->receive_buf that can block, which is why we can have
concurrency here at all in the first place).
It also used to set the TTY_FLUSHING bit, which would then cause any
concurrent 'tty_buffer_flush()' to not free all the tty buffers and
clear 'tty->buf.tail'. And with 'buf.head' being NULL, and 'buf.tail'
being non-NULL, new data would never touch 'buf.head'.
Does that sound a bit too subtle? It was. If another concurrent call to
'flush_to_ldisc()' were to come in, the NULL buf.head would indeed cause
it to not process the buffer list, but it would still clear TTY_FLUSHING
afterwards, making the buffer protection against 'tty_buffer_flush()' no
longer work.
So this clears it all up. We depend purely on TTY_FLUSHING for handling
re-entrancy, and stop playing games with the buffer list entirely. In
fact, the buffer list handling is now robust enough that we could
probably stop doing the whole "protect against 'tty_buffer_flush()'"
thing entirely.
However, Alan also points out that we would probably be better off
simplifying the locking even further, and just take the tty ldisc_mutex
around all the buffer flushing calls. That seems like a good idea, but
in the meantime this is a conceptually minimal fix (with the patch
itself being bigger than required just to clean the code up and make it
readable).
This fixes keyboard trouble under X:
http://bugzilla.kernel.org/show_bug.cgi?id=14388
Reported-and-tested-by: Frédéric Meunier <fredlwm@gmail.com>
Reported-and-tested-by: Boyan <btanastasov@yahoo.co.uk>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Paul Fulghum <paulkf@microgate.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
drivers/char/tty_buffer.c | 29 +++++++++++++----------------
1 file changed, 13 insertions(+), 16 deletions(-)
--- a/drivers/char/tty_buffer.c
+++ b/drivers/char/tty_buffer.c
@@ -402,28 +402,26 @@ static void flush_to_ldisc(struct work_s
container_of(work, struct tty_struct, buf.work.work);
unsigned long flags;
struct tty_ldisc *disc;
- struct tty_buffer *tbuf, *head;
- char *char_buf;
- unsigned char *flag_buf;
disc = tty_ldisc_ref(tty);
if (disc == NULL) /* !TTY_LDISC */
return;
spin_lock_irqsave(&tty->buf.lock, flags);
- /* So we know a flush is running */
- set_bit(TTY_FLUSHING, &tty->flags);
- head = tty->buf.head;
- if (head != NULL) {
- tty->buf.head = NULL;
- for (;;) {
- int count = head->commit - head->read;
+
+ if (!test_and_set_bit(TTY_FLUSHING, &tty->flags)) {
+ struct tty_buffer *head;
+ while ((head = tty->buf.head) != NULL) {
+ int count;
+ char *char_buf;
+ unsigned char *flag_buf;
+
+ count = head->commit - head->read;
if (!count) {
if (head->next == NULL)
break;
- tbuf = head;
- head = head->next;
- tty_buffer_free(tty, tbuf);
+ tty->buf.head = head->next;
+ tty_buffer_free(tty, head);
continue;
}
/* Ldisc or user is trying to flush the buffers
@@ -445,9 +443,9 @@ static void flush_to_ldisc(struct work_s
flag_buf, count);
spin_lock_irqsave(&tty->buf.lock, flags);
}
- /* Restore the queue head */
- tty->buf.head = head;
+ clear_bit(TTY_FLUSHING, &tty->flags);
}
+
/* We may have a deferred request to flush the input buffer,
if so pull the chain under the lock and empty the queue */
if (test_bit(TTY_FLUSHPENDING, &tty->flags)) {
@@ -455,7 +453,6 @@ static void flush_to_ldisc(struct work_s
clear_bit(TTY_FLUSHPENDING, &tty->flags);
wake_up(&tty->read_wait);
}
- clear_bit(TTY_FLUSHING, &tty->flags);
spin_unlock_irqrestore(&tty->buf.lock, flags);
tty_ldisc_deref(disc);
next prev parent reply other threads:[~2009-10-16 17:18 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20091016170953.128828149@linux.site>
2009-10-16 17:14 ` [00/46] 2.6.31.5-stable review Greg KH
2009-10-16 17:09 ` [01/46] SCSI: Retry ADD_TO_MLQUEUE return value for EH commands Greg KH
2009-10-16 17:09 ` [02/46] SCSI: Fix protection scsi_data_buffer leak Greg KH
2009-10-16 17:09 ` [03/46] SCSI: sg: Free data buffers after calling blk_rq_unmap_user Greg KH
2009-10-16 17:09 ` [04/46] ARM: pxa: workaround errata #37 by not using half turbo switching Greg KH
2009-10-16 17:09 ` [05/46] tracing/filters: Fix memory leak when setting a filter Greg KH
2009-10-16 17:09 ` [06/46] x86/paravirt: Use normal calling sequences for irq enable/disable Greg KH
2009-10-16 17:10 ` [07/46] USB: ftdi_sio: remove tty->low_latency Greg KH
2009-10-16 17:10 ` [08/46] USB: ftdi_sio: remove unused rx_byte counter Greg KH
2009-10-16 17:10 ` [09/46] USB: ftdi_sio: clean up read completion handler Greg KH
2009-10-16 17:10 ` [10/46] USB: ftdi_sio: re-implement read processing Greg KH
2009-10-16 17:10 ` [11/46] USB: pl2303: fix error characters not being reported to ldisc Greg KH
2009-10-16 17:10 ` [12/46] USB: digi_acceleport: Fix broken unthrottle Greg KH
2009-10-16 17:10 ` [13/46] USB: serial: dont call release without attach Greg KH
2009-10-16 17:10 ` [14/46] USB: option: Toshiba G450 device id Greg KH
2009-10-16 17:10 ` [15/46] USB: ipaq: fix oops when device is plugged in Greg KH
2009-10-16 17:10 ` [16/46] USB: cp210x: Add support for the DW700 UART Greg KH
2009-10-16 17:10 ` [17/46] USB: Fix throttling in generic usbserial driver Greg KH
2009-10-16 17:10 ` [18/46] USB: storage: When a device returns no sense data, call it a Hardware Error Greg KH
2009-10-16 17:10 ` [19/46] arm, cris, mips, sparc, powerpc, um, xtensa: fix build with bash 4.0 Greg KH
2009-10-16 17:10 ` [20/46] intel-iommu: Cope with broken HP DC7900 BIOS Greg KH
2009-10-16 17:10 ` [21/46] futex: Detect mismatched requeue targets Greg KH
2009-10-16 17:10 ` [22/46] futex: Fix wakeup race by setting TASK_INTERRUPTIBLE before queue_me() Greg KH
2009-10-16 17:10 ` [23/46] tpm-fixup-pcrs-sysfs-file-update Greg KH
2009-10-16 17:10 ` [24/46] TPM: fix pcrread Greg KH
2009-10-16 17:10 ` [25/46] Bluetooth: Disconnect HIDRAW devices on disconnect Greg KH
2009-10-16 17:10 ` [26/46] Bluetooth: Add extra device reference counting for connections Greg KH
2009-10-16 17:10 ` [27/46] Bluetooth: Let HIDP grab the device reference " Greg KH
2009-10-16 17:10 ` [28/46] connector: Keep the skb in cn_callback_data Greg KH
2009-10-16 17:10 ` [29/46] connector: Provide the senders credentials to the callback Greg KH
2009-10-16 17:10 ` [30/46] connector: Removed the destruct_data callback since it is always kfree_skb() Greg KH
2009-10-16 17:10 ` [31/46] dm/connector: Only process connector packages from privileged processes Greg KH
2009-10-16 17:10 ` [32/46] dst/connector: Disallow unpliviged users to configure dst Greg KH
2009-10-16 17:10 ` [33/46] pohmelfs/connector: Disallow unpliviged users to configure pohmelfs Greg KH
2009-10-16 17:10 ` [34/46] uvesafb/connector: Disallow unpliviged users to send netlink packets Greg KH
2009-10-16 17:10 ` [35/46] e1000e: swap max hw supported frame size between 82574 and 82583 Greg KH
2009-10-16 17:10 ` [36/46] MAINTAINERS: Fix Riku Voipios address Greg KH
2009-10-16 17:10 ` [37/46] macintosh: Dont assume i2c device probing always succeeds Greg KH
2009-10-16 17:10 ` [38/46] i2c: Hide probe errors caused by ACPI resource conflicts Greg KH
2009-10-16 17:10 ` [39/46] ALSA: Dont assume i2c device probing always succeeds Greg KH
2009-10-16 17:10 ` [40/46] bsdacct: switch credentials for writing to the accounting file Greg KH
2009-10-16 17:10 ` [41/46] sysfs: Allow sysfs_notify_dirent to be called from interrupt context Greg KH
2009-10-16 17:10 ` [42/46] Staging: rt2860sta: prevent a panic when disabling when associated Greg KH
2009-10-16 17:10 ` [43/46] usb-storage: Workaround devices with bogus sense size Greg KH
2009-10-16 17:10 ` [44/46] iwlwifi: incorrect method used for finding valid OTP blocks Greg KH
2009-10-16 17:10 ` [45/46] mac80211: fix vlan and optimise RX Greg KH
2009-10-16 17:10 ` Greg KH [this message]
2009-10-16 17:57 ` [00/46] 2.6.31.5-stable review Massimo Cetra
2009-10-16 18:03 ` Greg KH
2009-10-17 5:20 ` Massimo Cetra
2009-11-05 23:12 ` Greg KH
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20091016171210.987917756@linux.site \
--to=gregkh@suse.de \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=linux-kernel@vger.kernel.org \
--cc=paulkf@microgate.com \
--cc=stable-review@kernel.org \
--cc=stable@kernel.org \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox