There is a race condition in the USB hub code with regard to handling
TT clear requests that can get the HCD driver in a deadlock. Usually
when an TT clear request is scheduled it will be executed immediately:
<7>[ 6.077583] usb 2-1.3: unlink qh1-0e01/f4d4db00 start 0 [1/2 us]
<3>[ 6.078041] usb 2-1: clear tt buffer port 3, a3 ep2 t04048d82
<7>[ 6.078299] hub_tt_work:731
<7>[ 9.309089] usb 2-1.5: link qh1-0e01/f4d506c0 start 0 [1/2 us]
<7>[ 9.324526] ehci_hcd 0000:00:1d.0: reused qh f4d4db00 schedule
<7>[ 9.324539] usb 2-1.3: link qh1-0e01/f4d4db00 start 0 [1/2 us]
<7>[ 9.341530] usb 1-1.1: link qh4-0e01/f397aec0 start 2 [1/2 us]
<7>[ 10.116159] usb 2-1.3: unlink qh1-0e01/f4d4db00 start 0 [1/2 us]
<3>[ 10.116459] usb 2-1: clear tt buffer port 3, a3 ep2 t04048d82
<7>[ 10.116537] hub_tt_work:731
However, if a suspend operation is triggered before hub_tt_work is
scheduled, hub_quiesce will cancel the work without notifying the HCD
driver:
<3>[ 35.033941] usb 2-1: clear tt buffer port 3, a3 ep2 t04048d80
<5>[ 35.034022] sd 0:0:0:0: [sda] Stopping disk
<7>[ 35.034039] hub 2-1:1.0: hub_suspend
<7>[ 35.034067] usb 2-1: unlink qh256-0001/f3b1ab00 start 1 [1/0 us]
<7>[ 35.035085] hub 1-0:1.0: hub_suspend
<7>[ 35.035102] usb usb1: bus suspend, wakeup 0
<7>[ 35.035106] ehci_hcd 0000:00:1a.0: suspend root hub
<7>[ 35.035298] hub 2-0:1.0: hub_suspend
<7>[ 35.035313] usb usb2: bus suspend, wakeup 0
<7>[ 35.035315] ehci_hcd 0000:00:1d.0: suspend root hub
<6>[ 35.250017] PM: suspend of devices complete after 216.979 msecs
<6>[ 35.250822] PM: late suspend of devices complete after 0.799 msecs
<7>[ 35.252343] ehci_hcd 0000:00:1d.0: wakeup: 1
<7>[ 35.262923] ehci_hcd 0000:00:1d.0: --> PCI D3hot
<7>[ 35.263302] ehci_hcd 0000:00:1a.0: wakeup: 1
<7>[ 35.273912] ehci_hcd 0000:00:1a.0: --> PCI D3hot
<6>[ 35.274254] PM: noirq suspend of devices complete after 23.442 msecs
<6>[ 35.274975] ACPI: Preparing to enter system sleep state S3
<6>[ 35.292666] PM: Saving platform NVS memory
<7>[ 35.295030] Disabling non-boot CPUs ...
<6>[ 35.297351] CPU 1 is now offline
<6>[ 35.300345] CPU 2 is now offline
<6>[ 35.303929] CPU 3 is now offline
<7>[ 35.303931] lockdep: fixing up alternatives.
<6>[ 35.304825] Extended CMOS year: 2000
When the device will resume the EHCI driver will get stuck in
ehci_endpoint_disable waiting for the tt_clearing flag to reset:
This patch changes hub_quiesce behavior to flush the TT clear work
instead of canceling it, to make sure that no TT clear request remains
uncompleted before suspend.
Signed-off-by: Octavian Purdila <octavian.purdila@intel.com> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The warning check for duplicate sysfs entries can cause a buffer overflow
when printing the warning, as strcat() doesn't check buffer sizes.
Use strlcat() instead.
Since strlcat() doesn't return a pointer to the passed buffer, unlike
strcat(), I had to convert the nested concatenation in sysfs_add_one() to
an admittedly more obscure comma operator construct, to avoid emitting code
for the concatenation if CONFIG_BUG is disabled.
The call to xprt_disconnect_done() that is triggered by a successful
connection reset will trigger another automatic wakeup of all tasks
on the xprt->pending rpc_wait_queue. In particular it will cause an
early wake up of the task that called xprt_connect().
All we really want to do here is clear all the socket-specific state
flags, so we split that functionality out of xs_sock_mark_closed()
into a helper that can be called by xs_abort_connection()
Reported-by: Chris Perl <chris.perl@gmail.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Chris Perl <chris.perl@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
This reverts commit 55420c24a0d4d1fce70ca713f84aa00b6b74a70e.
Now that we clear the connected flag when entering TCP_CLOSE_WAIT,
the deadlock described in this commit is no longer possible.
Instead, the resulting call to xs_tcp_shutdown() can interfere
with pending reconnection attempts.
Reported-by: Chris Perl <chris.perl@gmail.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Chris Perl <chris.perl@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Chris Perl reports that we're seeing races between the wakeup call in
xs_error_report and the connect attempts. Basically, Chris has shown
that in certain circumstances, the call to xs_error_report causes the
rpc_task that is responsible for reconnecting to wake up early, thus
triggering a disconnect and retry.
Since the sk->sk_error_report() calls in the socket layer are always
followed by a tcp_done() in the cases where we care about waking up
the rpc_tasks, just let the state_change callbacks take responsibility
for those wake ups.
Reported-by: Chris Perl <chris.perl@gmail.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Tested-by: Chris Perl <chris.perl@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Fix possible overflow of the buffer used for expanding environment
variables when building file list.
In the extremely unlikely case of an attacker having control over the
environment variables visible to gen_init_cpio, control over the
contents of the file gen_init_cpio parses, and gen_init_cpio was built
without compiler hardening, the attacker can gain arbitrary execution
control via a stack buffer overflow.
According to xHCI spec section 4.6.1.1 and section 4.6.1.2,
after aborting a command on the command ring, xHC will
generate a command completion event with its completion
code set to Command Ring Stopped at least. If a command is
currently executing at the time of aborting a command, xHC
also generate a command completion event with its completion
code set to Command Abort. When the command ring is stopped,
software may remove, add, or rearrage Command Descriptors.
To cancel a command, software will initialize a command
descriptor for the cancel command, and add it into a
cancel_cmd_list of xhci. When the command ring is stopped,
software will find the command trbs described by command
descriptors in cancel_cmd_list and modify it to No Op
command. If software can't find the matched trbs, we can
think it had been finished.
This patch should be backported to kernels as old as 3.0, that contain
the commit 7ed603ecf8b68ab81f4c83097d3063d43ec73bb8 "xhci: Add an
assertion to check for virt_dev=0 bug." That commit papers over a NULL
pointer dereference, and this patch fixes the underlying issue that
caused the NULL pointer dereference.
Note from Sarah: The TRB_TYPE_LINK_LE32 macro is not in the 3.0 stable
kernel, so I added it to this patch.
Signed-off-by: Elric Fu <elricfu1@gmail.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Tested-by: Miroslav Sabljic <miroslav.sabljic@avl.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The patch is used to cancel command when the command isn't
acknowledged and a timeout occurs.
This patch should be backported to kernels as old as 3.0, that contain
the commit 7ed603ecf8b68ab81f4c83097d3063d43ec73bb8 "xhci: Add an
assertion to check for virt_dev=0 bug." That commit papers over a NULL
pointer dereference, and this patch fixes the underlying issue that
caused the NULL pointer dereference.
Signed-off-by: Elric Fu <elricfu1@gmail.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Tested-by: Miroslav Sabljic <miroslav.sabljic@avl.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Software have to abort command ring and cancel command
when a command is failed or hang. Otherwise, the command
ring will hang up and can't handle the others. An example
of a command that may hang is the Address Device Command,
because waiting for a SET_ADDRESS request to be acknowledged
by a USB device is outside of the xHC's ability to control.
To cancel a command, software will initialize a command
descriptor for the cancel command, and add it into a
cancel_cmd_list of xhci.
Sarah: Fixed missing newline on "Have the command ring been stopped?"
debugging statement.
This patch should be backported to kernels as old as 3.0, that contain
the commit 7ed603ecf8b68ab81f4c83097d3063d43ec73bb8 "xhci: Add an
assertion to check for virt_dev=0 bug." That commit papers over a NULL
pointer dereference, and this patch fixes the underlying issue that
caused the NULL pointer dereference.
Signed-off-by: Elric Fu <elricfu1@gmail.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Tested-by: Miroslav Sabljic <miroslav.sabljic@avl.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Adding cmd_ring_state for command ring. It helps to verify
the current command ring state for controlling the command
ring operations.
This patch should be backported to kernels as old as 3.0. The commit 7ed603ecf8b68ab81f4c83097d3063d43ec73bb8 "xhci: Add an assertion to
check for virt_dev=0 bug." papers over the NULL pointer dereference that
I now believe is related to a timed out Set Address command. This (and
the four patches that follow it) contain the real fix that also allows
VIA USB 3.0 hubs to consistently re-enumerate during the plug/unplug
stress tests.
Signed-off-by: Elric Fu <elricfu1@gmail.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Tested-by: Miroslav Sabljic <miroslav.sabljic@avl.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
On a 2-node machine with 256GB of ram we get 512 lines of
console output, which is just too much.
This mimicks Yinghai Lu's x86 commit c2b91e2eec9678dbda274e906cc32ea8f711da3b
(x86_64/mm: check and print vmemmap allocation continuous) except that
we aren't ever going to get contiguous block pointers in between calls
so just print when the virtual address or node changes.
This decreases the output by an order of 16.
Also demote this to KERN_DEBUG.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
There are multiple errors in how sys_sparc64_personality() handles
personality flags stored in top three bytes.
- directly comparing current->personality against PER_LINUX32 doesn't work
in cases when any of the personality flags stored in the top three bytes
are used.
- directly forcefully setting personality to PER_LINUX32 or PER_LINUX
discards any flags stored in the top three bytes
Fix the first one by properly using personality() macro to compare only
PER_MASK bytes.
Fix the second one by setting only the bits that should be set, instead of
overwriting the whole value.
Signed-off-by: Jiri Kosina <jkosina@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
There was a serious disconnect in the logic happening in
sparc_pmu_disable_event() vs. sparc_pmu_enable_event().
Event disable is implemented by programming a NOP event into the PCR.
However, event enable was not reversing this operation. Instead, it
was setting the User/Priv/Hypervisor trace enable bits.
That's not sparc_pmu_enable_event()'s job, that's what
sparc_pmu_enable() and sparc_pmu_disable() do .
The intent of sparc_pmu_enable_event() is clear, since it first clear
out the event type encoding field. So fix this by OR'ing in the event
encoding rather than the trace enable bits.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
If the MM is not active, only report the top-level PC. Do not try to
access the address space.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
we want syscall_trace_leave() called on exit from any syscall;
skipping its call in case we'd done force_successful_syscall_return()
is broken...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
After commit e2446eaa ("tcp_v4_send_reset: binding oif to iif in no
sock case").. tcp resets are always lost, when routing is asymmetric.
Yes, backing out that patch will result in misrouting of resets for
dead connections which used interface binding when were alive, but we
actually cannot do anything here. What's died that's died and correct
handling normal unbound connections is obviously a priority.
Comment to comment:
> This has few benefits:
> 1. tcp_v6_send_reset already did that.
It was done to route resets for IPv6 link local addresses. It was a
mistake to do so for global addresses. The patch fixes this as well.
Actually, the problem appears to be even more serious than guaranteed
loss of resets. As reported by Sergey Soloviev <sol@eqv.ru>, those
misrouted resets create a lot of arp traffic and huge amount of
unresolved arp entires putting down to knees NAT firewalls which use
asymmetric routing.
This is the revised patch for fixing rds-ping spinlock recursion
according to Venkat's suggestions.
RDS ping/pong over TCP feature has been broken for years(2.6.39 to
3.6.0) since we have to set TCP cork and call kernel_sendmsg() between
ping/pong which both need to lock "struct sock *sk". However, this
lock has already been hold before rds_tcp_data_ready() callback is
triggerred. As a result, we always facing spinlock resursion which
would resulting in system panic.
Given that RDS ping is only used to test the connectivity and not for
serious performance measurements, we can queue the pong transmit to
rds_wq as a delayed response.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com> CC: Venkat Venkatsubra <venkat.x.venkatsubra@oracle.com> CC: David S. Miller <davem@davemloft.net> CC: James Morris <james.l.morris@oracle.com> Signed-off-by: Jie Liu <jeff.liu@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The retry loop in neigh_resolve_output() and neigh_connected_output()
call dev_hard_header() with out reseting the skb to network_header.
This causes the retry to fail with skb_under_panic. The fix is to
reset the network_header within the retry loop.
Signed-off-by: Ramesh Nagappa <ramesh.nagappa@ericsson.com> Reviewed-by: Shawn Lu <shawn.lu@ericsson.com> Reviewed-by: Robert Coulson <robert.coulson@ericsson.com> Reviewed-by: Billie Alsup <billie.alsup@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
On CougarPoint and PantherPoint PCH chips, the timing generator may fail
to start after DP training completes. This is due to a bug in the
FDI autotraining detect logic (which will stall the timing generator and
re-enable it once training completes), so disable it to avoid silent DP
mode setting failures.
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Keith Packard <keithp@keithp.com> Signed-off-by: Timo Aaltonen <timo.aaltonen@canonical.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
If none of the elements in scrubrates[] matches, this loop will cause
__amd64_set_scrub_rate() to incorrectly use the n+1th element.
As the function is designed to use the final scrubrates[] element in the
case of no match, we can fix this bug by simply terminating the array
search at the n-1th element.
Boris: this code is fragile anyway, see here why:
http://marc.info/?l=linux-kernel&m=135102834131236&w=2
notify_on_release must be triggered when the last process in a cgroup is
move to another. But if the first(and only) process in a cgroup is moved to
another, notify_on_release is not triggered.
# mkdir /cgroup/cpu/SRC
# mkdir /cgroup/cpu/DST
#
# echo 1 >/cgroup/cpu/SRC/notify_on_release
# echo 1 >/cgroup/cpu/DST/notify_on_release
#
# sleep 300 &
[1] 8629
#
# echo 8629 >/cgroup/cpu/SRC/tasks
# echo 8629 >/cgroup/cpu/DST/tasks
-> notify_on_release for /SRC must be triggered at this point,
but it isn't.
This is because put_css_set() is called before setting CGRP_RELEASABLE
in cgroup_task_migrate(), and is a regression introduce by the
commit:74a1166d(cgroups: make procs file writable), which was merged
into v3.0.
Acked-by: Li Zefan <lizefan@huawei.com> Cc: Ben Blum <bblum@andrew.cmu.edu> Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Signed-off-by: Tejun Heo <tj@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
If the write endpoint is interrupt type, usb_sndintpipe() should
be passed to usb_fill_int_urb() instead of usb_sndbulkpipe().
Signed-off-by: Ming Lei <ming.lei@canonical.com> Cc: Oliver Neukum <oneukum@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
In 32 bit guests, if a userspace process has %eax == -ERESTARTSYS
(-512) or -ERESTARTNOINTR (-513) when it is interrupted by an event
/and/ the process has a pending signal then %eip (and %eax) are
corrupted when returning to the main process after handling the
signal. The application may then crash with SIGSEGV or a SIGILL or it
may have subtly incorrect behaviour (depending on what instruction it
returned to).
The occurs because handle_signal() is incorrectly thinking that there
is a system call that needs to restarted so it adjusts %eip and %eax
to re-execute the system call instruction (even though user space had
not done a system call).
If %eax == -514 (-ERESTARTNOHAND (-514) or -ERESTART_RESTARTBLOCK
(-516) then handle_signal() only corrupted %eax (by setting it to
-EINTR). This may cause the application to crash or have incorrect
behaviour.
handle_signal() assumes that regs->orig_ax >= 0 means a system call so
any kernel entry point that is not for a system call must push a
negative value for orig_ax. For example, for physical interrupts on
bare metal the inverse of the vector is pushed and page_fault() sets
regs->orig_ax to -1, overwriting the hardware provided error code.
xen_hypervisor_callback() was incorrectly pushing 0 for orig_ax
instead of -1.
Classic Xen kernels pushed %eax which works as %eax cannot be both
non-negative and -RESTARTSYS (etc.), but using -1 is consistent with
other non-system call entry points and avoids some of the tests in
handle_signal().
There were similar bugs in xen_failsafe_callback() of both 32 and
64-bit guests. If the fault was corrected and the normal return path
was used then 0 was incorrectly pushed as the value for orig_ax.
Signed-off-by: David Vrabel <david.vrabel@citrix.com> Acked-by: Jan Beulich <JBeulich@suse.com> Acked-by: Ian Campbell <ian.campbell@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
On systems with very large memory (1 TB in our case), BIOS may report a
reserved region or a hole in the E820 map, even above the 4 GB range. Exclude
these from the direct mapping.
[ hpa: this should be done not just for > 4 GB but for everything above the legacy
region (1 MB), at the very least. That, however, turns out to require significant
restructuring. That work is well underway, but is not suitable for rc/stable. ]
Calling uname() with the UNAME26 personality set allows a leak of kernel
stack contents. This fixes it by defensively calculating the length of
copy_to_user() call, making the len argument unsigned, and initializing
the stack buffer to zero (now technically unneeded, but hey, overkill).
CVE-2012-0957
Reported-by: PaX Team <pageexec@freemail.hu> Signed-off-by: Kees Cook <keescook@chromium.org> Cc: Andi Kleen <ak@linux.intel.com> Cc: PaX Team <pageexec@freemail.hu> Cc: Brad Spengler <spender@grsecurity.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
If the filehandle is stale, or open access is denied for some reason,
nlm_fopen() may return one of the NLMv4-specific error codes nlm4_stale_fh
or nlm4_failed. These get passed right through nlm_lookup_file(),
and so when nlmsvc_retrieve_args() calls the latter, it needs to filter
the result through the cast_status() machinery.
Failure to do so, will trigger the BUG_ON() in encode_nlm_stat...
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Reported-by: Larry McVoy <lm@bitmover.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
tpm_write calls tpm_transmit without checking the return value and
assigns the return value unconditionally to chip->pending_data, even if
it's an error value.
This causes three bugs.
So if we write to /dev/tpm0 with a tpm_param_size bigger than
TPM_BUFSIZE=0x1000 (e.g. 0x100a)
and a bufsize also bigger than TPM_BUFSIZE (e.g. 0x100a)
tpm_transmit returns -E2BIG which is assigned to chip->pending_data as
-7, but tpm_write returns that TPM_BUFSIZE bytes have been successfully
been written to the TPM, altough this is not true (bug #1).
As we did write more than than TPM_BUFSIZE bytes but tpm_write reports
that only TPM_BUFSIZE bytes have been written the vfs tries to write
the remaining bytes (in this case 10 bytes) to the tpm device driver via
tpm_write which then blocks at
/* cannot perform a write until the read has cleared
either via tpm_read or a user_read_timer timeout */
while (atomic_read(&chip->data_pending) != 0)
msleep(TPM_TIMEOUT);
for 60 seconds, since data_pending is -7 and nobody is able to
read it (since tpm_read luckily checks if data_pending is greater than
0) (#bug 2).
After that the remaining bytes are written to the TPM which are
interpreted by the tpm as a normal command. (bug #3)
So if the last bytes of the command stream happen to be a e.g.
tpm_force_clear this gets accidentally sent to the TPM.
This patch fixes all three bugs, by propagating the error code of
tpm_write and returning -E2BIG if the input buffer is too big,
since the response from the tpm for a truncated value is bogus anyway.
Moreover it returns -EBUSY to userspace if there is a response ready to be
read.
Signed-off-by: Peter Huewe <peter.huewe@infineon.com> Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
ext3 users of data=journal mode with blocksize < pagesize were occasionally
hitting assertion failure in journal_commit_transaction() checking whether the
transaction has at least as many credits reserved as buffers attached. The
core of the problem is that when a file gets truncated, buffers that still need
checkpointing or that are attached to the committing transaction are left with
buffer_mapped set. When this happens to buffers beyond i_size attached to a
page stradding i_size, subsequent write extending the file will see these
buffers and as they are mapped (but underlying blocks were freed) things go
awry from here.
The assertion failure just coincidentally (and in this case luckily as we would
start corrupting filesystem) triggers due to journal_head not being properly
cleaned up as well.
Under some rare circumstances this bug could even hit data=ordered mode users.
There the assertion won't trigger and we would end up corrupting the
filesystem.
We fix the problem by unmapping buffers if possible (in lots of cases we just
need a buffer attached to a transaction as a place holder but it must not be
written out anyway). And in one case, we just have to bite the bullet and wait
for transaction commit to finish.
Reviewed-by: Josef Bacik <jbacik@fusionio.com> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Commit v2.6.19-rc1~1272^2~41 tells us that r->cost != 0 can happen when
a running state is saved to userspace and then reinstated from there.
Make sure that private xt_limit area is initialized with correct values.
Otherwise, random matchings due to use of uninitialized memory.
Signed-off-by: Jan Engelhardt <jengelh@inai.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
In __nf_ct_expect_check, the function refresh_timer returns 1
if a matching expectation is found and its timer is successfully
refreshed. This results in nf_ct_expect_related returning 0.
Note that at this point:
- the passed expectation is not inserted in the expectation table
and its timer was not initialized, since we have refreshed one
matching/existing expectation.
- nf_ct_expect_alloc uses kmem_cache_alloc, so the expectation
timer is in some undefined state just after the allocation,
until it is appropriately initialized.
This can be a problem for the SIP helper during the expectation
addition:
...
if (nf_ct_expect_related(rtp_exp) == 0) {
if (nf_ct_expect_related(rtcp_exp) != 0)
nf_ct_unexpect_related(rtp_exp);
...
Note that nf_ct_expect_related(rtp_exp) may return 0 for the timer refresh
case that is detailed above. Then, if nf_ct_unexpect_related(rtcp_exp)
returns != 0, nf_ct_unexpect_related(rtp_exp) is called, which does:
spin_lock_bh(&nf_conntrack_lock);
if (del_timer(&exp->timeout)) {
nf_ct_unlink_expect(exp);
nf_ct_expect_put(exp);
}
spin_unlock_bh(&nf_conntrack_lock);
Note that del_timer always returns false if the timer has been
initialized. However, the timer was not initialized since setup_timer
was not called, therefore, the expectation timer remains in some
undefined state. If I'm not missing anything, this may lead to the
removal an unexistent expectation.
To fix this, the optimization that allows refreshing an expectation
is removed. Now nf_conntrack_expect_related looks more consistent
to me since it always add the expectation in case that it returns
success.
Thanks to Patrick McHardy for participating in the discussion of
this patch.
I think this may be the source of the problem described by:
http://marc.info/?l=netfilter-devel&m=134073514719421&w=2
Reported-by: Rafal Fitt <rafalf@aplusc.com.pl> Acked-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Via-headers are parsed beginning at the first character after the Via-address.
When the address is translated first and its length decreases, the offset to
start parsing at is incorrect and header parameters might be missed.
Update the offset after translating the Via-address to fix this.
Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
1. On Host1, setup brige br0(192.168.1.106)
2. Boot a kvm guest(192.168.1.105) on Host1 and start httpd
3. Start IPVS service on Host1
ipvsadm -A -t 192.168.1.106:80 -s rr
ipvsadm -a -t 192.168.1.106:80 -r 192.168.1.105:80 -m
4. Run apache benchmark on Host2(192.168.1.101)
ab -n 1000 http://192.168.1.106/
Actually, IPVS wants in this case just to replace nfct
with untracked version. So replace the nf_reset(skb) call
in ip_vs_notrack() with a nf_conntrack_put(skb->nfct) call.
Signed-off-by: Lin Ming <mlin@ss.pku.edu.cn> Signed-off-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: Simon Horman <horms@verge.net.au> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
It was reported that the Linux kernel sometimes logs:
klogd: [2629147.402413] kernel BUG at net / netfilter /
nf_conntrack_proto_tcp.c: 447!
klogd: [1072212.887368] kernel BUG at net / netfilter /
nf_conntrack_proto_tcp.c: 392
ipv4_get_l4proto() in nf_conntrack_l3proto_ipv4.c and tcp_error() in
nf_conntrack_proto_tcp.c should catch malformed packets, so the errors
at the indicated lines - TCP options parsing - should not happen.
However, tcp_error() relies on the "dataoff" offset to the TCP header,
calculated by ipv4_get_l4proto(). But ipv4_get_l4proto() does not check
bogus ihl values in IPv4 packets, which then can slip through tcp_error()
and get caught at the TCP options parsing routines.
The patch fixes ipv4_get_l4proto() by invalidating packets with bogus
ihl value.
The patch closes netfilter bugzilla id 771.
Signed-off-by: Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Existing code assumes that del_timer returns true for alive conntrack
entries. However, this is not true if reliable events are enabled.
In that case, del_timer may return true for entries that were
just inserted in the dying list. Note that packets / ctnetlink may
hold references to conntrack entries that were just inserted to such
list.
This patch fixes the issue by adding an independent timer for
event delivery. This increases the size of the ecache extension.
Still we can revisit this later and use variable size extensions
to allocate this area on demand.
Tested-by: Oliver Smith <olipro@8.c.9.b.0.7.4.0.1.0.0.2.ip6.arpa> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
After commit 39f618b4fd95ae243d940ec64c961009c74e3333 (3.4)
"ipvs: reset ipvs pointer in netns" we can oops in
ip_vs_dst_event on rmmod ip_vs because ip_vs_control_cleanup
is called after the ipvs_core_ops subsys is unregistered and
net->ipvs is NULL. Fix it by exiting early from ip_vs_dst_event
if ipvs is NULL. It is safe because all services and dests
for the net are already freed.
Signed-off-by: Julian Anastasov <ja@ssi.bg> Signed-off-by: Simon Horman <horms@verge.net.au> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Acked-by: David Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
And, the check "if (datalen < sizeof(struct pktgen_hdr))" will be
passed as "datalen" is promoted to unsigned, therefore will cause
a crash later.
This is a quick fix by checking if "datalen" is negative. The following
patch will increase the default value of 'min_pkt_size' for IPv6.
This bug should exist for a long time, so Cc -stable too.
Signed-off-by: Cong Wang <amwang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Adding two (or more) timers with large values for "expires" (they have
to reside within tv5 in the same list) leads to endless looping
between cascade() and internal_add_timer() in case CONFIG_BASE_SMALL
is one and jiffies are crossing the value 1 << 18. The bug was
introduced between 2.6.11 and 2.6.12 (and survived for quite some
time).
This patch ensures that when cascade() is called timers within tv5 are
not added endlessly to their own list again, instead they are added to
the next lower tv level tv4 (as expected).
In autofs4_d_automount(), if a mount fail occurs the AUTOFS_INF_PENDING
mount pending flag is not cleared.
One effect of this is when using the "browse" option, directory entry
attributes show up with all "?"s due to the incorrect callback and
subsequent failure return (when in fact no callback should be made).
Signed-off-by: Ian Kent <ikent@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The hypervisor will trap it. However without this patch,
we would crash as the .read_tscp is set to NULL. This patch
fixes it and sets it to the native_read_tscp call.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Fix two bugs of the /dev/fw* character device concerning the
FW_CDEV_IOC_GET_INFO ioctl with nonzero fw_cdev_get_info.bus_reset.
(Practically all /dev/fw* clients issue this ioctl right after opening
the device.)
Both bugs are caused by sizeof(struct fw_cdev_event_bus_reset) being 36
without natural alignment and 40 with natural alignment.
1) Memory corruption, affecting i386 userland on amd64 kernel:
Userland reserves a 36 bytes large buffer, kernel writes 40 bytes.
This has been first found and reported against libraw1394 if
compiled with gcc 4.7 which happens to order libraw1394's stack such
that the bug became visible as data corruption.
2) Information leak, affecting all kernel architectures except i386:
4 bytes of random kernel stack data were leaked to userspace.
Hence limit the respective copy_to_user() to the 32-bit aligned size of
struct fw_cdev_event_bus_reset.
Reported-by: Simon Kirby <sim@hostway.ca> Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Fuzzing with trinity oopsed on the 1st instruction of shmem_fh_to_dentry(),
u64 inum = fid->raw[2];
which is unhelpfully reported as at the end of shmem_alloc_inode():
Right, tmpfs is being stupid to access fid->raw[2] before validating that
fh_len includes it: the buffer kmalloc'ed by do_sys_name_to_handle() may
fall at the end of a page, and the next page not be present.
But some other filesystems (ceph, gfs2, isofs, reiserfs, xfs) are being
careless about fh_len too, in fh_to_dentry() and/or fh_to_parent(), and
could oops in the same way: add the missing fh_len checks to those.
Reported-by: Sasha Levin <levinsasha928@gmail.com> Signed-off-by: Hugh Dickins <hughd@google.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Sage Weil <sage@inktank.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Christoph Hellwig <hch@infradead.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Eric Dumazet [Fri, 26 Oct 2012 18:55:11 +0000 (11:55 -0700)]
net: add a truesize parameter to skb_add_rx_frag()
skb_add_rx_frag() API is misleading.
Network skbs built with this helper can use uncharged kernel memory and
eventually stress/crash machine in OOM.
Add a 'truesize' parameter and then fix drivers in followup patches.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Cc: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Maxim Uvarov <maxim.uvarov@oracle.com>
Chuck Anderson [Wed, 26 Sep 2012 21:47:28 +0000 (14:47 -0700)]
CVE-2012-2313: dl2k: Clean up rio_ioctl
The dl2k driver's rio_ioctl call has a few issues:
- No permissions checking
- Implements SIOCGMIIREG and SIOCGMIIREG using the SIOCDEVPRIVATE numbers
- Has a few ioctls that may have been used for debugging at one point
but have no place in the kernel proper.
This patch removes all but the MII ioctls, renumbers them to use the
standard ones, and adds the proper permission check for SIOCSMIIREG.
We can also get rid of the dl2k-specific struct mii_data in favor of
the generic struct mii_ioctl_data.
Since we have the phyid on hand, we can add the SIOCGMIIPHY ioctl too.
Most of the MII code for the driver could probably be converted to use
the generic MII library but I don't have a device to test the results.
Reported-by: Stephan Mueller <stephan.mueller@atsec.com> Signed-off-by: Jeff Mahoney <jeffm@suse.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Orabug: 14680245 Signed-off-by: Chuck Anderson <chuck.anderson@oracle.com>
Maxim Uvarov [Thu, 18 Oct 2012 17:43:58 +0000 (10:43 -0700)]
[net/sfc] limit number of segments per skb on tx
Orabug: 14769994
This fixes: CVE-2012-3412
The kernels have 5b6262d0ccf759a16fabe11d904a2531125a4b71 (sfc: Replace
some literal constants with EFX_PAGE_SIZE/EFX_BUF_SIZE) and 7e6d06f0de3f74ca929441add094518ae332257c (sfc: Fix maximum number of TSO
segments and minimum TX queue size) from upstream but the upstream
version relies on the TCP layer limiting the maximum number of GSO
segments which isn't in UEK. 30b678d844af3305cda5953467005cebb5d7b687
(net: Allow driver to limit number of GSO segments per skb) provided
this in mainline. Instead of modifing network stack we fix add additional
check to efx_enqueue_skb_tso(). Signed-off-by: Maxim Uvarov <maxim.uvarov@oracle.com>
Srinivas Eeda [Wed, 17 Oct 2012 23:23:41 +0000 (16:23 -0700)]
ocfs2: submit disk heartbeat bio using WRITE_SYNC
Under heavy I/O load, writing the disk heartbeat can be forced
to wait for minutes, and this causes the node to be fenced.
This patch tries to use WRITE_SYNC in submitting the heartbeat
bio, so that writing the heartbeat will have a priority over
other requests.
Signed-off-by: Noboru Iwamatsu <n_iwamatsu at jp.fujitsu.com> Acked-by: Tao Ma <boyu.mt at taobao.com> Acked-by: Sunil Mushran <sunil.mushran at oracle.com> Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com>
In the event of CPU hotplug, the kernel modifies the cpusets' cpus_allowed
masks as and when necessary to ensure that the tasks belonging to the cpusets
have some place (online CPUs) to run on. And regular CPU hotplug is
destructive in the sense that the kernel doesn't remember the original cpuset
configurations set by the user, across hotplug operations.
However, suspend/resume (which uses CPU hotplug) is a special case in which
the kernel has the responsibility to restore the system (during resume), to
exactly the same state it was in before suspend.
In order to achieve that, do the following:
1. Don't modify cpusets during suspend/resume. At all.
In particular, don't move the tasks from one cpuset to another, and
don't modify any cpuset's cpus_allowed mask. So, simply ignore cpusets
during the CPU hotplug operations that are carried out in the
suspend/resume path.
2. However, cpusets and sched domains are related. We just want to avoid
altering cpusets alone. So, to keep the sched domains updated, build
a single sched domain (containing all active cpus) during each of the
CPU hotplug operations carried out in s/r path, effectively ignoring
the cpusets' cpus_allowed masks.
(Since userspace is frozen while doing all this, it will go unnoticed.)
3. During the last CPU online operation during resume, build the sched
domains by looking up the (unaltered) cpusets' cpus_allowed masks.
That will bring back the system to the same original state as it was in
before suspend.
Ultimately, this will not only solve the cpuset problem related to suspend
resume (ie., restores the cpusets to exactly what it was before suspend, by
not touching it at all) but also speeds up suspend/resume because we avoid
running cpuset update code for every CPU being offlined/onlined.
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/20120524141611.3692.20155.stgit@srivatsabhat.in.ibm.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Preeti U Murthy <preeti@linux.vnet.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Commit cc9a6c877661 ("cpuset: mm: reduce large amounts of memory barrier
related damage v3") introduced a potential memory corruption.
shmem_alloc_page() uses a pseudo vma and it has one significant unique
combination, vma->vm_ops=NULL and vma->policy->flags & MPOL_F_SHARED.
get_vma_policy() does NOT increase a policy ref when vma->vm_ops=NULL
and mpol_cond_put() DOES decrease a policy ref when a policy has
MPOL_F_SHARED. Therefore, when a cpuset update race occurs,
alloc_pages_vma() falls in 'goto retry_cpuset' path, decrements the
reference count and frees the policy prematurely.
When shared_policy_replace() fails to allocate new->policy is not freed
correctly by mpol_set_shared_policy(). The problem is that shared
mempolicy code directly call kmem_cache_free() in multiple places where
it is easy to make a mistake.
This patch creates an sp_free wrapper function and uses it. The bug was
introduced pre-git age (IOW, before 2.6.12-rc2).
shared_policy_replace() use of sp_alloc() is unsafe. 1) sp_node cannot
be dereferenced if sp->lock is not held and 2) another thread can modify
sp_node between spin_unlock for allocating a new sp node and next
spin_lock. The bug was introduced before 2.6.12-rc2.
Kosaki's original patch for this problem was to allocate an sp node and
policy within shared_policy_replace and initialise it when the lock is
reacquired. I was not keen on this approach because it partially
duplicates sp_alloc(). As the paths were sp->lock is taken are not that
performance critical this patch converts sp->lock to sp->mutex so it can
sleep when calling sp_alloc().
[kosaki.motohiro@jp.fujitsu.com: Original patch] Signed-off-by: Mel Gorman <mgorman@suse.de> Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Reviewed-by: Christoph Lameter <cl@linux.com> Cc: Josh Boyer <jwboyer@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The problem is that the structure is being prematurely freed due to a
reference count imbalance. In the following case mbind(addr, len) should
replace the memory policies of both vma1 and vma2 and thus they will
become to share the same mempolicy and the new mempolicy will have the
MPOL_F_SHARED flag.
alloc_pages_vma() uses get_vma_policy() and mpol_cond_put() pair for
maintaining the mempolicy reference count. The current rule is that
get_vma_policy() only increments refcount for shmem VMA and
mpol_conf_put() only decrements refcount if the policy has
MPOL_F_SHARED.
In above case, vma1 is not shmem vma and vma->policy has MPOL_F_SHARED!
The reference count will be decreased even though was not increased
whenever alloc_page_vma() is called. This has been broken since commit
[52cd3b07: mempolicy: rework mempolicy Reference Counting] in 2008.
There is another serious bug with the sharing of memory policies.
Currently, mempolicy rebind logic (it is called from cpuset rebinding)
ignores a refcount of mempolicy and override it forcibly. Thus, any
mempolicy sharing may cause mempolicy corruption. The bug was
introduced by commit [68860ec1: cpusets: automatic numa mempolicy
rebinding].
Ideally, the shared policy handling would be rewritten to either
properly handle COW of the policy structures or at least reference count
MPOL_F_SHARED based exclusively on information within the policy.
However, this patch takes the easier approach of disabling any policy
sharing between VMAs. Each new range allocated with sp_alloc will
allocate a new policy, set the reference count to 1 and drop the
reference count of the old policy. This increases the memory footprint
but is not expected to be a major problem as mbind() is unlikely to be
used for fine-grained ranges. It is also inefficient because it means
we allocate a new policy even in cases where mbind_range() could use the
new_policy passed to it. However, it is more straight-forward and the
change should be invisible to the user.
[mgorman@suse.de: Edited changelog] Reported-by: Dave Jones <davej@redhat.com> Cc: Christoph Lameter <cl@linux.com> Reviewed-by: Christoph Lameter <cl@linux.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: Josh Boyer <jwboyer@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Each grace period is supposed to have at least one callback waiting
for that grace period to complete. However, if CONFIG_NO_HZ=n, an
extra callback-free grace period is no big problem -- it will chew up
a tiny bit of CPU time, but it will complete normally. In contrast,
CONFIG_NO_HZ=y kernels have the potential for all the CPUs to go to
sleep indefinitely, in turn indefinitely delaying completion of the
callback-free grace period. Given that nothing is waiting on this grace
period, this is also not a problem.
That is, unless RCU CPU stall warnings are also enabled, as they are
in recent kernels. In this case, if a CPU wakes up after at least one
minute of inactivity, an RCU CPU stall warning will result. The reason
that no one noticed until quite recently is that most systems have enough
OS noise that they will never remain absolutely idle for a full minute.
But there are some embedded systems with cut-down userspace configurations
that consistently get into this situation.
All this begs the question of exactly how a callback-free grace period
gets started in the first place. This can happen due to the fact that
CPUs do not necessarily agree on which grace period is in progress.
If a CPU still believes that the grace period that just completed is
still ongoing, it will believe that it has callbacks that need to wait for
another grace period, never mind the fact that the grace period that they
were waiting for just completed. This CPU can therefore erroneously
decide to start a new grace period. Note that this can happen in
TREE_RCU and TREE_PREEMPT_RCU even on a single-CPU system: Deadlock
considerations mean that the CPU that detected the end of the grace
period is not necessarily officially informed of this fact for some time.
Once this CPU notices that the earlier grace period completed, it will
invoke its callbacks. It then won't have any callbacks left. If no
other CPU has any callbacks, we now have a callback-free grace period.
This commit therefore makes CPUs check more carefully before starting a
new grace period. This new check relies on an array of tail pointers
into each CPU's list of callbacks. If the CPU is up to date on which
grace periods have completed, it checks to see if any callbacks follow
the RCU_DONE_TAIL segment, otherwise it checks to see if any callbacks
follow the RCU_WAIT_TAIL segment. The reason that this works is that
the RCU_WAIT_TAIL segment will be promoted to the RCU_DONE_TAIL segment
as soon as the CPU is officially notified that the old grace period
has ended.
This change is to cpu_needs_another_gp(), which is called in a number
of places. The only one that really matters is in rcu_start_gp(), where
the root rcu_node structure's ->lock is held, which prevents any
other CPU from starting or completing a grace period, so that the
comparison that determines whether the CPU is missing the completion
of a grace period is stable.
Reported-by: Becky Bruce <bgillbruce@gmail.com> Reported-by: Subodh Nijsure <snijsure@grid-net.com> Reported-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Tested-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
In many places !pmd_present has been converted to pmd_none. For pmds
that's equivalent and pmd_none is quicker so using pmd_none is better.
However (unless we delete pmd_present) we should provide an accurate
pmd_present too. This will avoid the risk of code thinking the pmd is non
present because it's under __split_huge_page_map, see the pmd_mknotpresent
there and the comment above it.
If the page has been mprotected as PROT_NONE, it would also lead to a
pmd_present false negative in the same way as the race with
split_huge_page.
Because the PSE bit stays on at all times (both during split_huge_page and
when the _PAGE_PROTNONE bit get set), we could only check for the PSE bit,
but checking the PROTNONE bit too is still good to remember pmd_present
must always keep PROT_NONE into account.
This explains a not reproducible BUG_ON that was seldom reported on the
lists.
The same issue is in pmd_large, it would go wrong with both PROT_NONE and
if it races with split_huge_page.
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Acked-by: Rik van Riel <riel@redhat.com> Cc: Johannes Weiner <jweiner@redhat.com> Cc: Hugh Dickins <hughd@google.com> Cc: Mel Gorman <mgorman@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
In fuzzing with trinity, lockdep protested "possible irq lock inversion
dependency detected" when isolate_lru_page() reenabled interrupts while
still holding the supposedly irq-safe tree_lock:
isolate_lru_page() is correct to enable interrupts unconditionally:
invalidate_complete_page2() is incorrect to call clear_page_mlock() while
holding tree_lock, which is supposed to nest inside lru_lock.
Both truncate_complete_page() and invalidate_complete_page() call
clear_page_mlock() before taking tree_lock to remove page from radix_tree.
I guess invalidate_complete_page2() preferred to test PageDirty (again)
under tree_lock before committing to the munlock; but since the page has
already been unmapped, its state is already somewhat inconsistent, and no
worse if clear_page_mlock() moved up.
Reported-by: Sasha Levin <levinsasha928@gmail.com> Deciphered-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Hugh Dickins <hughd@google.com> Acked-by: Mel Gorman <mel@csn.ul.ie> Cc: Rik van Riel <riel@redhat.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michel Lespinasse <walken@google.com> Cc: Ying Han <yinghan@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Code tracking when transaction needs to be committed on fdatasync(2) forgets
to handle a situation when only inode's i_size is changed. Thus in such
situations fdatasync(2) doesn't force transaction with new i_size to disk
and that can result in wrong i_size after a crash.
Fix the issue by updating inode's i_datasync_tid whenever its size is
updated.
Reported-by: Kristian Nielsen <knielsen@knielsen-hq.org> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Proper block swap for inodes with full journaling enabled is
truly non obvious task. In order to be on a safe side let's
explicitly disable it for now.
For bridges with "secondary > subordinate", i.e., invalid bus number
apertures, we don't enumerate anything behind the bridge unless the
user specified "pci=assign-busses".
This patch makes us automatically try to reassign the downstream bus
numbers in this case (just for that bridge, not for all bridges as
"pci=assign-busses" does).
We don't discover all the devices on the Intel DP43BF motherboard
without this change (or "pci=assign-busses") because its BIOS configures
a bridge as:
pci 0000:00:1e.0: PCI bridge to [bus 20-08] (subtractive decode)
[bhelgaas: changelog, change message to dev_info]
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=18412
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=625754 Reported-by: Brian C. Huffman <bhuffman@graze.net> Reported-by: VL <vl.homutov@gmail.com> Tested-by: VL <vl.homutov@gmail.com> Signed-off-by: Yinghai Lu <yinghai@kernel.org> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
In order for the network layer to see that AoE requires
no checksumming in a generic way, the packets must be
marked as requiring no checksum, so we make this requirement
explicit with the assertion.
Signed-off-by: Ed Cashin <ecashin@coraid.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
A change in a series of VLAN-related changes appears to have
inadvertently disabled the use of the scatter gather feature of
network cards for transmission of non-IP ethernet protocols like ATA
over Ethernet (AoE). Below is a reference to the commit that
introduces a "harmonize_features" function that turns off scatter
gather when the NIC does not support hardware checksumming for the
ethernet protocol of an sk buff.
net offloading: Generalize netif_get_vlan_features().
The can_checksum_protocol function is not equipped to consider a
protocol that does not require checksumming. Calling it for a
protocol that requires no checksum is inappropriate.
The patch below has harmonize_features call can_checksum_protocol when
the protocol needs a checksum, so that the network layer is not forced
to perform unnecessary skb linearization on the transmission of AoE
packets. Unnecessary linearization results in decreased performance
and increased memory pressure, as reported here:
The problem has probably not been widely experienced yet, because
only recently has the kernel.org-distributed aoe driver acquired the
ability to use payloads of over a page in size, with the patchset
recently included in the mm tree:
https://lkml.org/lkml/2012/8/28/140
The coraid.com-distributed aoe driver already could use payloads of
greater than a page in size, but its users generally do not use the
newest kernels.
Signed-off-by: Ed Cashin <ecashin@coraid.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
While investigating l2tp bug, I hit a bug in eth_type_trans(),
because not enough bytes were pulled in skb head.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
mip6_mh_filter() should not modify its input, or else its caller
would need to recompute ipv6_hdr() if skb->head is reallocated.
Use skb_header_pointer() instead of pskb_may_pull()
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
icmpv6_filter() should not modify its input, or else its caller
would need to recompute ipv6_hdr() if skb->head is reallocated.
Use skb_header_pointer() instead of pskb_may_pull() and
change the prototype to make clear both sk and skb are const.
Also, if icmpv6 header cannot be found, do not deliver the packet,
as we do in IPv4.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
icmp_filter() should not modify its input, or else its caller
would need to recompute ip_hdr() if skb->head is reallocated.
Use skb_header_pointer() instead of pskb_may_pull() and
change the prototype to make clear both sk and skb are const.
Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Its possible to use RAW sockets to get a crash in
tcp_set_keepalive() / sk_reset_timer()
Fix is to make sure socket is a SOCK_STREAM one.
Reported-by: Dave Jones <davej@redhat.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
In the current rxhash calculation function, while the
sorting of the ports/addrs is coherent (you get the
same rxhash for packets sharing the same 4-tuple, in
both directions), ports and addrs are sorted
independently. This implies packets from a connection
between the same addresses but crossed ports hash to
the same rxhash.
For example, traffic between A=S:l and B=L:s is hashed
(in both directions) from {L, S, {s, l}}. The same
rxhash is obtained for packets between C=S:s and D=L:l.
This patch ensures that you either swap both addrs and ports,
or you swap none. Traffic between A and B, and traffic
between C and D, get their rxhash from different sources
({L, S, {l, s}} for A<->B, and {L, S, {s, l}} for C<->D)
The patch is co-written with Eric Dumazet <edumazet@google.com>
Signed-off-by: Chema Gonzalez <chema@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
When PPPOE is running over a virtual ethernet interface (e.g., a
bonding interface) and the user tries to delete the interface in case
the PPPOE state is ZOMBIE, the kernel will loop forever while
unregistering net_device for the reference count is not decreased to
zero which should have been done with dev_put().
Signed-off-by: Xiaodong Xu <stid.smth@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
SCTP charges wmem_alloc via sctp_set_owner_w() in sctp_sendmsg() and via
skb_set_owner_w() in sctp_packet_transmit(). If a sender runs out of
sndbuf it will sleep in sctp_wait_for_sndbuf() and expects to be waken up
by __sctp_write_space().
Buffer space charged via sctp_set_owner_w() is released in sctp_wfree()
which calls __sctp_write_space() directly.
Buffer space charged via skb_set_owner_w() is released via sock_wfree()
which calls sk->sk_write_space() _if_ SOCK_USE_WRITE_QUEUE is not set.
sctp_endpoint_init() sets SOCK_USE_WRITE_QUEUE on all sockets.
Therefore if sctp_packet_transmit() manages to queue up more than sndbuf
bytes, sctp_wait_for_sndbuf() will never be woken up again unless it is
interrupted by a signal.
This could be fixed by clearing the SOCK_USE_WRITE_QUEUE flag but ...
Charging for the data twice does not make sense in the first place, it
leads to overcharging sndbuf by a factor 2. Therefore this patch only
charges a single byte in wmem_alloc when transmitting an SCTP packet to
ensure that the socket stays alive until the packet has been released.
This means that control chunks are no longer accounted for in wmem_alloc
which I believe is not a problem as skb->truesize will typically lead
to overcharging anyway and thus compensates for any control overhead.
Signed-off-by: Thomas Graf <tgraf@suug.ch> CC: Vlad Yasevich <vyasevic@redhat.com> CC: Neil Horman <nhorman@tuxdriver.com> CC: David Miller <davem@davemloft.net> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
If recv() syscall is called for a TCP socket so that
- IOAT DMA is used
- MSG_WAITALL flag is used
- requested length is bigger than sk_rcvbuf
- enough data has already arrived to bring rcv_wnd to zero
then when tcp_recvmsg() gets to calling sk_wait_data(), receive
window can be still zero while sk_async_wait_queue exhausts
enough space to keep it zero. As this queue isn't cleaned until
the tcp_service_net_dma() call, sk_wait_data() cannot receive
any data and blocks forever.
If zero receive window and non-empty sk_async_wait_queue is
detected before calling sk_wait_data(), process the queue first.
Signed-off-by: Michal Kubecek <mkubecek@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
as we hold dst_entry before we call __ip6_del_rt,
so we should alse call dst_release not only return
-ENOENT when the rt6_info is ip6_null_entry.
and we already hold the dst entry, so I think it's
safe to call dst_release out of the write-read lock.
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
skb_reset_mac_len() relies on the value of the skb->network_header pointer,
therefore we must wait for such pointer to be recalculated before computing
the new mac_len value.
Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
I discovered I couldn't get sierra_net to work on a powerpc. Turns out
the firmware attribute check assumes the system is little endian and
hence fails because the attributes is a 16 bit value.
Signed-off-by: Len Sorensen <lsorense@csclub.uwaterloo.ca> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
If the old timestamps of a class, say cl, are stale when the class
becomes active, then QFQ may assign to cl a much higher start time
than the maximum value allowed. This may happen when QFQ assigns to
the start time of cl the finish time of a group whose classes are
characterized by a higher value of the ratio
max_class_pkt/weight_of_the_class with respect to that of
cl. Inserting a class with a too high start time into the bucket list
corrupts the data structure and may eventually lead to crashes.
This patch limits the maximum start time assigned to a class.
Signed-off-by: Paolo Valente <paolo.valente@unimore.it> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Its possible to setup a bad cbq configuration leading to
an infinite loop in cbq_classify()
DEV_OUT=eth0
ICMP="match ip protocol 1 0xff"
U32="protocol ip u32"
DST="match ip dst"
tc qdisc add dev $DEV_OUT root handle 1: cbq avpkt 1000 \
bandwidth 100mbit
tc class add dev $DEV_OUT parent 1: classid 1:1 cbq \
rate 512kbit allot 1500 prio 5 bounded isolated
tc filter add dev $DEV_OUT parent 1: prio 3 $U32 \
$ICMP $DST 192.168.3.234 flowid 1:
Reported-by: Denys Fedoryschenko <denys@visp.net.lb> Tested-by: Denys Fedoryschenko <denys@visp.net.lb> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
Add a check if pdev->bus->self == NULL (root bus). When attaching
a netxen NIC to a VM it can be on the root bus and the guest would
crash in netxen_mask_aer_correctable() because of a NULL pointer
dereference if CONFIG_PCIEAER is present.
Signed-off-by: Nikolay Aleksandrov <nikolay@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The current code fails to ensure that the netlink message actually
contains as many bytes as the header indicates. If a user creates a new
state or updates an existing one but does not supply the bytes for the
whole ESN replay window, the kernel copies random heap bytes into the
replay bitmap, the ones happen to follow the XFRMA_REPLAY_ESN_VAL
netlink attribute. This leads to following issues:
1. The replay window has random bits set confusing the replay handling
code later on.
2. A malicious user could use this flaw to leak up to ~3.5kB of heap
memory when she has access to the XFRM netlink interface (requires
CAP_NET_ADMIN).
Known users of the ESN replay window are strongSwan and Steffen's
iproute2 patch (<http://patchwork.ozlabs.org/patch/85962/>). The latter
uses the interface with a bitmap supplied while the former does not.
strongSwan is therefore prone to run into issue 1.
To fix both issues without breaking existing userland allow using the
XFRMA_REPLAY_ESN_VAL netlink attribute with either an empty bitmap or a
fully specified one. For the former case we initialize the in-kernel
bitmap with zero, for the latter we copy the user supplied bitmap. For
state updates the full bitmap must be supplied.
To prevent overflows in the bitmap length calculation the maximum size
of bmp_len is limited to 128 by this patch -- resulting in a maximum
replay window of 4096 packets. This should be sufficient for all real
life scenarios (RFC 4303 recommends a default replay window size of 64).
Signed-off-by: Mathias Krause <minipli@googlemail.com> Cc: Steffen Klassert <steffen.klassert@secunet.com> Cc: Martin Willi <martin@revosec.ch> Cc: Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The memory used for the template copy is a local stack variable. As
struct xfrm_user_tmpl contains multiple holes added by the compiler for
alignment, not initializing the memory will lead to leaking stack bytes
to userland. Add an explicit memset(0) to avoid the info leak.
Initial version of the patch by Brad Spengler.
Signed-off-by: Mathias Krause <minipli@googlemail.com> Cc: Brad Spengler <spender@grsecurity.net> Acked-by: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
The memory reserved to dump the xfrm policy includes multiple padding
bytes added by the compiler for alignment (padding bytes in struct
xfrm_selector and struct xfrm_userpolicy_info). Add an explicit
memset(0) before filling the buffer to avoid the heap info leak.
Signed-off-by: Mathias Krause <minipli@googlemail.com> Acked-by: Steffen Klassert <steffen.klassert@secunet.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
projects / users / jedix / linux-maple.git / log