ata_qc_complete() contains special handling for certain commands. For
example, it schedules EH for device revalidation after certain
configurations are changed. These shouldn't be applied to EH
commands but they were.
In most cases, it doesn't cause an actual problem because EH doesn't
issue any command which would trigger special handling; however, ACPI
can issue such commands via _GTF which can cause weird interactions.
Restructure ata_qc_complete() such that EH commands are always passed
on to __ata_qc_complete().
stable: Please apply to -stable only after 2.6.38 is released.
Add necessary alias to autoload ip6ip6 tunnel module.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
Since a8f80e8ff94ecba629542d9b4b5f5a8ee3eb565c any process with
CAP_NET_ADMIN may load any module from /lib/modules/. This doesn't mean
that CAP_NET_ADMIN is a superset of CAP_SYS_MODULE as modules are
limited to /lib/modules/**. However, CAP_NET_ADMIN capability shouldn't
allow anybody load any module not related to networking.
This patch restricts an ability of autoloading modules to netdev modules
with explicit aliases. This fixes CVE-2011-1019.
Arnd Bergmann suggested to leave untouched the old pre-v2.6.32 behavior
of loading netdev modules by name (without any prefix) for processes
with CAP_SYS_MODULE to maintain the compatibility with network scripts
that use autoloading netdev modules by aliases like "eth0", "wlan0".
Currently there are only three users of the feature in the upstream
kernel: ipip, ip_gre and sit.
root@albatros:~# capsh --drop=$(seq -s, 0 11),$(seq -s, 13 34) --
root@albatros:~# grep Cap /proc/$$/status
CapInh: 0000000000000000
CapPrm: fffffff800001000
CapEff: fffffff800001000
CapBnd: fffffff800001000
root@albatros:~# modprobe xfs
FATAL: Error inserting xfs
(/lib/modules/2.6.38-rc6-00001-g2bf4ca3/kernel/fs/xfs/xfs.ko): Operation not permitted
root@albatros:~# lsmod | grep xfs
root@albatros:~# ifconfig xfs
xfs: error fetching interface information: Device not found
root@albatros:~# lsmod | grep xfs
root@albatros:~# lsmod | grep sit
root@albatros:~# ifconfig sit
sit: error fetching interface information: Device not found
root@albatros:~# lsmod | grep sit
root@albatros:~# ifconfig sit0
sit0 Link encap:IPv6-in-IPv4
NOARP MTU:1480 Metric:1
root@albatros:~# lsmod | grep sit
sit 10457 0
tunnel4 2957 1 sit
For CAP_SYS_MODULE module loading is still relaxed:
I found that one of the 8168c chipsets (concretely XID 1c4000c0) starts
generating RxFIFO overflow errors. The result is an infinite loop in
interrupt handler as the RxFIFOOver is handled only for ...MAC_VER_11.
With the workaround everything goes fine.
Signed-off-by: Ivan Vecera <ivecera@redhat.com> Acked-by: Francois Romieu <romieu@fr.zoreil.com> Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Hayes <hayeswang@realtek.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Like many other places, we have to check that the array index is
within allowed limits, or otherwise, a kernel oops and other nastiness
can ensue when we access memory beyond the end of the array.
The problem goes back to v2.6.30-rc1~1372~1342~31 where nf_log_bind
was decoupled from nf_log_register.
Reported-by: Miguel Di Ciurcio Filho <miguel.filho@gmail.com>,
via irc.freenode.net/#netfilter Signed-off-by: Jan Engelhardt <jengelh@medozas.de> Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
When CPU hotplug is used, some CPUs may be offline at the time a kexec is
performed. The subsequent kernel may expect these CPUs to be already running,
and will declare them stuck. On pseries, there's also a soft-offline (cede)
state that CPUs may be in; this can also cause problems as the kexeced kernel
may ask RTAS if they're online -- and RTAS would say they are. The CPU will
either appear stuck, or will cause a crash as we replace its cede loop beneath
it.
This patch kicks each present offline CPU awake before the kexec, so that
none are forever lost to these assumptions in the subsequent kernel.
Now, the behaviour is that all available CPUs that were offlined are now
online & usable after the kexec. This mimics the behaviour of a full reboot
(on which all CPUs will be restarted).
Signed-off-by: Matt Evans <matt@ozlabs.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Kamalesh babulal <kamalesh@linux.vnet.ibm.com> Signed-off-by: Andi Kleen <ak@linux.intel.com>
cc: Anton Blanchard <anton@samba.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Robert Swiecki reported a BUG_ON(page_mapped) from a fuzzer, punching
a hole with madvise(,, MADV_REMOVE). That path is under mutex, and
cannot be explained by lack of serialization in unmap_mapping_range().
Reviewing the code, I found one place where vm_truncate_count handling
should have been updated, when I switched at the last minute from one
way of managing the restart_addr to another: mremap move changes the
virtual addresses, so it ought to adjust the restart_addr.
But rather than exporting the notion of restart_addr from memory.c, or
converting to restart_pgoff throughout, simply reset vm_truncate_count
to 0 to force a rescan if mremap move races with preempted truncation.
We have no confirmation that this fixes Robert's BUG,
but it is a fix that's worth making anyway.
We have found a hardware erratum on 82599 hardware that can lead to
unpredictable behavior when Header Splitting mode is enabled. So
we are no longer enabling this feature on affected hardware.
Please see the 82599 Specification Update for more information.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com> Tested-by: Stephen Ko <stephen.s.ko@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
However the code in rxrpc_instantiate strips it away:
data += sizeof(kver);
datalen -= sizeof(kver);
Removing kif_version fixes my problem.
Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
When we get oplock break notification we should set the appropriate
value of OplockLevel field in oplock break acknowledge according to
the oplock level held by the client in this time. As we only can have
level II oplock or no oplock in the case of oplock break, we should be
aware only about clientCanCacheRead field in cifsInodeInfo structure.
Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <sfrench@us.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
NETDEV_NOTIFY_PEER is an explicit request by the driver to send a link
notification while NETDEV_UP/NETDEV_CHANGEADDR generate link
notifications as a sort of side effect.
In the later cases the sysctl option is present because link
notification events can have undesired effects e.g. if the link is
flapping. I don't think this applies in the case of an explicit
request from a driver.
This patch makes NETDEV_NOTIFY_PEER unconditional, if preferred we
could add a new sysctl for this case which defaults to on.
This change causes Xen post-migration ARP notifications (which cause
switches to relearn their MAC tables etc) to be sent by default.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Andi Kleen <ak@linux.intel.com>
[reported to solve hyperv live migration problem - gkh] Cc: Haiyang Zhang <haiyangz@microsoft.com> Cc: Mike Surcouf <mike@surcouf.co.uk> Cc: Hank Janssen <hjanssen@microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
If the iowarrior devices in this case statement support more than 8 bytes
per report, it is possible to write past the end of a kernel heap allocation.
This will probably never be possible, but change the allocation to be more
defensive anyway.
For some time is known that ASPM is causing troubles on r8169, i.e. make
device randomly stop working without any errors in dmesg.
Currently Tomi Leppikangas reports that system with r8169 device hangs
with MCE errors when ASPM is enabled:
https://bugzilla.redhat.com/show_bug.cgi?id=642861#c4
Lets disable ASPM for r8169 devices at all, to avoid problems with
r8169 PCIe devices at least for some users.
Reported-by: Tomi Leppikangas <tomi.leppikangas@gmail.com> Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
When support for 82577/82578 was added[1] in 2.6.31, PHY wakeup was in-
advertently enabled (even though it does not function properly) on ICH10
LOMs. This patch makes it so that the ICH10 LOMs use MAC wakeup instead
as was done with the initial support for those devices (i.e. 82567LM-3,
82567LF-3 and 82567V-4).
This fixes a bug in the order of dccp_rcv_state_process() that still permitted
reception even after closing the socket. A Reset after close thus causes a NULL
pointer dereference by not preventing operations on an already torn-down socket.
dccp_v4_do_rcv()
|
| state other than OPEN
v
dccp_rcv_state_process()
|
| DCCP_PKT_RESET
v
dccp_rcv_reset()
|
v
dccp_time_wait()
WARNING: at net/ipv4/inet_timewait_sock.c:141 __inet_twsk_hashdance+0x48/0x128()
Modules linked in: arc4 ecb carl9170 rt2870sta(C) mac80211 r8712u(C) crc_ccitt ah
[<c0038850>] (unwind_backtrace+0x0/0xec) from [<c0055364>] (warn_slowpath_common)
[<c0055364>] (warn_slowpath_common+0x4c/0x64) from [<c0055398>] (warn_slowpath_n)
[<c0055398>] (warn_slowpath_null+0x1c/0x24) from [<c02b72d0>] (__inet_twsk_hashd)
[<c02b72d0>] (__inet_twsk_hashdance+0x48/0x128) from [<c031caa0>] (dccp_time_wai)
[<c031caa0>] (dccp_time_wait+0x40/0xc8) from [<c031c15c>] (dccp_rcv_state_proces)
[<c031c15c>] (dccp_rcv_state_process+0x120/0x538) from [<c032609c>] (dccp_v4_do_)
[<c032609c>] (dccp_v4_do_rcv+0x11c/0x14c) from [<c0286594>] (release_sock+0xac/0)
[<c0286594>] (release_sock+0xac/0x110) from [<c031fd34>] (dccp_close+0x28c/0x380)
[<c031fd34>] (dccp_close+0x28c/0x380) from [<c02d9a78>] (inet_release+0x64/0x70)
The fix is by testing the socket state first. Receiving a packet in Closed state
now also produces the required "No connection" Reset reply of RFC 4340, 8.3.1.
Reported-and-tested-by: Johan Hovold <jhovold@gmail.com> Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
Reported-by: Mark Davis Signed-off-by: Christian Lamparter <chunkeey@googlemail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
vfs_rename_other() does not lock renamed inode with i_mutex. Thus changing
i_nlink in a non-atomic manner (which happens in ext2_rename()) can corrupt
it as reported and analyzed by Josh.
In fact, there is no good reason to mess with i_nlink of the moved file.
We did it presumably to simulate linking into the new directory and unlinking
from an old one. But the practical effect of this is disputable because fsck
can possibly treat file as being properly linked into both directories without
writing any error which is confusing. So we just stop increment-decrement
games with i_nlink which also fixes the corruption.
CC: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Josh Hunt <johunt@akamai.com> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
When the per cpu timer is marked CLOCK_EVT_FEAT_C3STOP, then we only
can switch into oneshot mode, when the backup broadcast device
supports oneshot mode as well. Otherwise we would try to switch the
broadcast device into an unsupported mode unconditionally. This went
unnoticed so far as the current available broadcast devices support
oneshot mode. Seth unearthed this problem while debugging and working
around an hpet related BIOS wreckage.
Add the necessary check to tick_is_oneshot_available().
A customer of ours, complained that when setting the reset
vector back to 0, it trashed other data and hung their box.
They noticed when only 4 bytes were set to 0 instead of 8,
everything worked correctly.
Mathew pointed out:
|
| We're supposed to be resetting trampoline_phys_low and
| trampoline_phys_high here, which are two 16-bit values.
| Writing 64 bits is definitely going to overwrite space
| that we're not supposed to be touching.
|
Current refcounttree codes actually didn't writeback the new pages out in
write-back mode, due to a bug of always passing a ZERO number of clusters
to 'ocfs2_cow_sync_writeback', the patch tries to pass a proper one in.
Signed-off-by: Tristan Ye <tristan.ye@oracle.com> Signed-off-by: Joel Becker <jlbec@evilplan.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
In x25_link_free(), we destroy 'nb' before dereferencing
'nb->dev'. Don't do this, because 'nb' might be freed
by then.
Reported-by: Randy Dunlap <randy.dunlap@oracle.com> Tested-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
does not take into account that the remaining data length can be less
than sg_dma_len(sg). In that case, running_total can end up being
greater than the total data length, so an extra TRB is counted.
Changing the expression to
while (running_total < sg_dma_len(sg) && running_total < temp)
fixes that.
This patch should be queued for stable kernels back to 2.6.31.
Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
are incorrect, because running_total can never be zero, so the if()
expression will never be true. I think the intention was that
running_total be in the range of 0 to TRB_MAX_BUFF_SIZE-1, not 1
to TRB_MAX_BUFF_SIZE. So adding a
running_total &= TRB_MAX_BUFF_SIZE - 1;
fixes the problem.
This patch should be queued for stable kernels back to 2.6.31.
Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
This makes it easier to spot some problems, which will be fixed by the
next patch in the series. Also change dev_dbg to dev_err in
check_trb_math(), so any math errors will be visible even when running
with debug disabled.
Note: This patch changes the expressions containing
"((1 << TRB_MAX_BUFF_SHIFT) - 1)" to use the equivalent
"(TRB_MAX_BUFF_SIZE - 1)". No change in behavior is intended for
those expressions.
This patch should be queued for stable kernels back to 2.6.31.
Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
On some SB800 systems polarity for IOAPIC pin2 is wrongly
specified as low active by BIOS. This caused system hangs after
resume from S3 when HPET was used in one-shot mode on such
systems because a timer interrupt was missed (HPET signal is
high active).
'mdp' devices are md devices with preallocated device numbers
for partitions. As such it is possible to mknod and open a partition
before opening the whole device.
this causes md_probe() to be called with a device number of a
partition, which in-turn calls mddev_find with such a number.
However mddev_find expects the number of a 'whole device' and
does the wrong thing with partition numbers.
So add code to mddev_find to remove the 'partition' part of
a device number and just work with the 'whole device'.
This patch addresses https://bugzilla.kernel.org/show_bug.cgi?id=28652
The kernel automatically evaluates partition tables of storage devices.
The code for evaluating LDM partitions (in fs/partitions/ldm.c) contains
a bug that causes a kernel oops on certain corrupted LDM partitions. A
kernel subsystem seems to crash, because, after the oops, the kernel no
longer recognizes newly connected storage devices.
The patch changes ldm_parse_vmdb() to Validate the value of vblk_size.
Signed-off-by: Timo Warns <warns@pre-sense.de> Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Eugene Teo <eugeneteo@kernel.sg> Acked-by: Richard Russon <ldm@flatcap.org> Cc: Harvey Harrison <harvey.harrison@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@suse.de>
In several places, an epoll fd can call another file's ->f_op->poll()
method with ep->mtx held. This is in general unsafe, because that other
file could itself be an epoll fd that contains the original epoll fd.
The code defends against this possibility in its own ->poll() method using
ep_call_nested, but there are several other unsafe calls to ->poll
elsewhere that can be made to deadlock. For example, the following simple
program causes the call in ep_insert recursively call the original fd's
->poll, leading to deadlock:
#include <unistd.h>
#include <sys/epoll.h>
int main(void) {
int e1, e2, p[2];
struct epoll_event evt = {
.events = EPOLLIN
};
On insertion, check whether the inserted file is itself a struct epoll,
and if so, do a recursive walk to detect whether inserting this file would
create a loop of epoll structures, which could lead to deadlock.
[nelhage@ksplice.com: Use epmutex to serialize concurrent inserts] Signed-off-by: Davide Libenzi <davidel@xmailserver.org> Signed-off-by: Nelson Elhage <nelhage@ksplice.com> Signed-off-by: Andi Kleen <ak@linux.intel.com> Reported-by: Nelson Elhage <nelhage@ksplice.com> Tested-by: Nelson Elhage <nelhage@ksplice.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@suse.de>
My Galaxy Spica needs this quirk when in modem mode, otherwise
it causes endless USB bus resets and is unusable in this mode.
Unfortunately Samsung decided to reuse ID of its old CDMA phone SGH-I500
for the modem part.
That's why in addition to this patch the visor driver must be prevented
from binding to SPH-I500 ID, so ACM driver can do that.
[USB]Add Samsung SGH-I500/Android modem ID switch to visor driver
Samsung decided to reuse USB ID of its old CDMA phone SGH-I500 for the
modem part of some of their Android phones. At least Galaxy Spica
is affected.
This modem needs ACM driver and does not work with visor driver which
binds the conflicting ID for SGH-I500.
Because SGH-I500 is pretty an old hardware its best to add switch to
visor
driver in cause somebody still wants to use that phone with Linux.
Note that this is needed only when using the Android phone as modem,
not in USB storage or ADB mode.
This patch (as1448) adds a quirks entry for the Keytouch QWERTY Panel
firmware, used in the IEC 60945 keyboard. This device crashes during
enumeration when the computer asks for its configuration string
descriptor.
The idle timer could trigger after clock had been disabled leading to
kernel panic when MUSB_DEVCTL is accessed in musb_do_idle on 2.6.37.
The fault below is no longer triggered on 2.6.38-rc4 (clock is disabled
later, and only if compiled as a module, and the offending memory access
has moved) but the timer should be cancelled nonetheless.
With CONFIG_SHIRQ_DEBUG=y we call a newly installed interrupt handler
in request_threaded_irq().
The original implementation (commit a304e1b8) called the handler
_BEFORE_ it was installed, but that caused problems with handlers
calling disable_irq_nosync(). See commit 377bf1e4.
It's braindead in the first place to call disable_irq_nosync in shared
handlers, but ....
Moving this call after we installed the handler looks innocent, but it
is very subtle broken on SMP.
Interrupt handlers rely on the fact, that the irq core prevents
reentrancy.
Now this debug call violates that promise because we run the handler
w/o the IRQ_INPROGRESS protection - which we cannot apply here because
that would result in a possibly forever masked interrupt line.
A concurrent real hardware interrupt on a different CPU results in
handler reentrancy and can lead to complete wreckage, which was
unfortunately observed in reality and took a fricking long time to
debug.
Leave the code here for now. We want this debug feature, but that's
not easy to fix. We really should get rid of those
disable_irq_nosync() abusers and remove that function completely.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Anton Vorontsov <avorontsov@ru.mvista.com> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Don't allow everybody to change ACPI settings. The comment says that it
is done deliberatelly, however, the comment before disp_proc_write()
says that at least one of these setting is experimental.
The lower filesystem may do some type of inode revalidation during a
getattr call. eCryptfs should take advantage of that by copying the
lower inode attributes to the eCryptfs inode after a call to
vfs_getattr() on the lower inode.
I originally wrote this fix while working on eCryptfs on nfsv3 support,
but discovered it also fixed an eCryptfs on ext4 nanosecond timestamp
bug that was reported.
Ensure a predictable endian state when entering signal handlers. This
avoids programs which use SETEND to momentarily switch their endian
state from having their signal handlers entered with an unpredictable
endian state.
Acked-by: Dave Martin <dave.martin@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
| drivers/media/radio/radio-aimslab.c: In function ‘rt_decvol’:
| drivers/media/radio/radio-aimslab.c:76: error: implicit declaration of function ‘msleep’
Currently we return 0 in swsusp_alloc() when alloc_image_page() fails.
Fix that. Also remove unneeded "error" variable since the only
useful value of error is -ENOMEM.
[rjw: Fixed up the changelog and changed subject.]
Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
task_show_regs used to be a debugging aid in the early bringup days
of Linux on s390. /proc/<pid>/status is a world readable file, it
is not a good idea to show the registers of a process. The only
correct fix is to remove task_show_regs.
Reported-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
Adds IBM Power Virtual SCSI ALUA devices to the ALUA device handler.
Signed-off-by: Brian King <brking@linux.vnet.ibm.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: maximilian attems <max@stro.at> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
By the commit af7fa16 2010-08-03 NFS: Fix up the fsync code
close(2) became returning the non-zero value even if it went well.
nfs_file_fsync() should return 0 when "status" is positive.
Signed-off-by: J. R. Okajima <hooanon05@yahoo.co.jp> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Tim Gardner <tim.gardner@canonical.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
In prepare_kernel_cred() since 2.6.29, put_cred(new) is called without
assigning new->usage when security_prepare_creds() returned an error. As a
result, memory for new and refcount for new->{user,group_info,tgcred} are
leaked because put_cred(new) won't call __put_cred() unless old->usage == 1.
Fix these leaks by assigning new->usage (and new->subscribers which was added
in 2.6.32) before calling security_prepare_creds().
In cred_alloc_blank() since 2.6.32, abort_creds(new) is called with
new->security == NULL and new->magic == 0 when security_cred_alloc_blank()
returns an error. As a result, BUG() will be triggered if SELinux is enabled
or CONFIG_DEBUG_CREDENTIALS=y.
If CONFIG_DEBUG_CREDENTIALS=y, BUG() is called from __invalid_creds() because
cred->magic == 0. Failing that, BUG() is called from selinux_cred_free()
because selinux_cred_free() is not expecting cred->security == NULL. This does
not affect smack_cred_free(), tomoyo_cred_free() or apparmor_cred_free().
Fix these bugs by
(1) Set new->magic before calling security_cred_alloc_blank().
(2) Handle null cred->security in creds_are_invalid() and selinux_cred_free().
In get_empty_filp() since 2.6.29, file_free(f) is called with f->f_cred == NULL
when security_file_alloc() returned an error. As a result, kernel will panic()
due to put_cred(NULL) call within RCU callback.
Fix this bug by assigning f->f_cred before calling security_file_alloc().
Ben Hutchings [Mon, 7 Feb 2011 19:20:55 +0000 (19:20 +0000)]
bonding/vlan: Avoid mangled NAs on slaves without VLAN tag insertion
This is related to commit f88a4a9b65a6f3422b81be995535d0e69df11bb8
upstream, but the bug cannot be properly fixed without the other
changes to VLAN tagging in 2.6.37.
bond_na_send() attempts to insert a VLAN tag in between building and
sending packets of the respective formats. If the slave does not
implement hardware VLAN tag insertion then vlan_put_tag() will mangle
the network-layer header because the Ethernet header is not present at
this point (unlike in bond_arp_send()).
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
If the guest domain has been suspend/resumed or migrated, then the
system clock backing the pvclock clocksource may revert to a smaller
value (ie, can be non-monotonic across the migration/save-restore).
Make sure we zero last_value in that case so that the domain
continues to see clock updates.
If ocfs2_live_connection_list is empty, ocfs2_connection_find() will return
a pointer to the LIST_HEAD, cast as a ocfs2_live_connection. This can cause
an oops when ocfs2_control_send_down() dereferences c->oc_conn:
The sctp_asoc_get_hmac() function iterates through a peer's hmac_ids
array and attempts to ensure that only a supported hmac entry is
returned. The current code fails to do this properly - if the last id
in the array is out of range (greater than SCTP_AUTH_HMAC_ID_MAX), the
id integer remains set after exiting the loop, and the address of an
out-of-bounds entry will be returned and subsequently used in the parent
function, causing potentially ugly memory corruption. This patch resets
the id integer to 0 on encountering an invalid id so that NULL will be
returned after finishing the loop if no valid ids are found.
Signed-off-by: Dan Rosenberg <drosenberg@vsecurity.com> Acked-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
There's a branch at the end of this function that
is supposed to normalize the return value with what
the mid-layer expects. In this one case, we get it wrong.
Also increase the verbosity of the INFO level printk
at the end of mptscsih_abort to include the actual return value
and the scmd->serial_number. The reason being success
or failure is actually determined by the state of
the internal tag list when a TMF is issued, and not the
return value of the TMF cmd. The serial_number is also
used in this decision, thus it's useful to know for debugging
purposes.
Reported-by: Peter M. Petrakis <peter.petrakis@canonical.com> Signed-off-by: Kashyap Desai <kashyap.desai@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
Added missing release callback for file_operations mptctl_fops.
Without release callback there will be never freed. It remains on
mptctl's eent list even after the file is closed and released.
If nfsd fails to find an exported via NFS file in the readahead cache, it
should increment corresponding nfsdstats counter (ra_depth[10]), but due to a
bug it may instead write to ra_depth[11], corrupting the following field.
In a kernel with NFSDv4 compiled in the corruption takes the form of an
increment of a counter of the number of NFSv4 operation 0's received; since
there is no operation 0, this is harmless.
In a kernel with NFSDv4 disabled it corrupts whatever happens to be in the
memory beyond nfsdstats.
Signed-off-by: Konstantin Khorenko <khorenko@openvz.org> Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Andi Kleen <ak@linux.intel.com>
Currently we consider a sched domain to be well balanced when the imbalance
is less than the domain's imablance_pct. As the number of cores and threads
are increasing, current values of imbalance_pct (for example 25% for a
NUMA domain) are not enough to detect imbalances like:
a) On a WSM-EP system (two sockets, each having 6 cores and 12 logical threads),
24 cpu-hogging tasks get scheduled as 13 on one socket and 11 on another
socket. Leading to an idle HT cpu.
b) On a hypothetial 2 socket NHM-EX system (each socket having 8 cores and
16 logical threads), 16 cpu-hogging tasks can get scheduled as 9 on one
socket and 7 on another socket. Leaving one core in a socket idle
whereas in another socket we have a core having both its HT siblings busy.
While this issue can be fixed by decreasing the domain's imbalance_pct
(by making it a function of number of logical cpus in the domain), it
can potentially cause more task migrations across sched groups in an
overloaded case.
Fix this by using imbalance_pct only during newly_idle and busy
load balancing. And during idle load balancing, check if there
is an imbalance in number of idle cpu's across the busiest and this
sched_group or if the busiest group has more tasks than its weight that
the idle cpu in this_group can pull.
When CPU is idle and on first interrupt, irq_enter calls tick_check_idle()
to notify interruption from idle. But, there is a problem if this call
is done after __irq_enter, as all routines in __irq_enter may find
stale time due to yet to be done tick_check_idle.
Specifically, trace calls in __irq_enter when they use global clock and also
account_system_vtime change in this patch as it wants to use sched_clock_cpu()
to do proper irq timing.
But, tick_check_idle was moved after __irq_enter intentionally to
prevent problem of unneeded ksoftirqd wakeups by the commit ee5f80a:
irq: call __irq_enter() before calling the tick_idle_check
Impact: avoid spurious ksoftirqd wakeups
Moving tick_check_idle() before __irq_enter and wrapping it with
local_bh_enable/disable would solve both the problems.
irq time is technically not available to the tasks running on the CPU.
This patch removes irq time from CPU power piggybacking on
sched_rt_avg_update().
Tested this by keeping CPU X busy with a network intensive task having 75%
oa a single CPU irq processing (hard+soft) on a 4-way system. And start seven
cycle soakers on the system. Without this change, there will be two tasks on
each CPU. With this change, there is a single task on irq busy CPU X and
remaining 7 tasks are spread around among other 3 CPUs.
Signed-off-by: Venkatesh Pallipadi <venki@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1286237003-12406-8-git-send-email-venki@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Scheduler accounts both softirq and interrupt processing times to the
currently running task. This means, if the interrupt processing was
for some other task in the system, then the current task ends up being
penalized as it gets shorter runtime than otherwise.
Change sched task accounting to acoount only actual task time from
currently running task. Now update_curr(), modifies the delta_exec to
depend on rq->clock_task.
Note that this change only handles CONFIG_IRQ_TIME_ACCOUNTING case. We can
extend this to CONFIG_VIRT_CPU_ACCOUNTING with minimal effort. But, thats
for later.
This change will impact scheduling behavior in interrupt heavy conditions.
Tested on a 4-way system with eth0 handled by CPU 2 and a network heavy
task (nc) running on CPU 3 (and no RSS/RFS). With that I have CPU 2
spending 75%+ of its time in irq processing. CPU 3 spending around 35%
time running nc task.
Now, if I run another CPU intensive task on CPU 2, without this change
/proc/<pid>/schedstat shows 100% of time accounted to this task. With this
change, it rightly shows less than 25% accounted to this task as remaining
time is actually spent on irq processing.
Signed-off-by: Venkatesh Pallipadi <venki@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1286237003-12406-7-git-send-email-venki@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
s390/powerpc/ia64 have support for CONFIG_VIRT_CPU_ACCOUNTING which does
the fine granularity accounting of user, system, hardirq, softirq times.
Adding that option on archs like x86 will be challenging however, given the
state of TSC reliability on various platforms and also the overhead it will
add in syscall entry exit.
Instead, add a lighter variant that only does finer accounting of
hardirq and softirq times, providing precise irq times (instead of timer tick
based samples). This accounting is added with a new config option
CONFIG_IRQ_TIME_ACCOUNTING so that there won't be any overhead for users not
interested in paying the perf penalty.
This accounting is based on sched_clock, with the code being generic.
So, other archs may find it useful as well.
This patch just adds the core logic and does not enable this logic yet.
Signed-off-by: Venkatesh Pallipadi <venki@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1286237003-12406-5-git-send-email-venki@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
To account softirq time cleanly in scheduler, we need to identify whether
softirq is invoked in ksoftirqd context or softirq at hardirq tail context.
Add PF_KSOFTIRQD for that purpose.
As all PF flag bits are currently taken, create space by moving one of the
infrequently used bits (PF_THREAD_BOUND) down in task_struct to be along
with some other state fields.
Signed-off-by: Venkatesh Pallipadi <venki@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1286237003-12406-4-git-send-email-venki@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The problem is, softirq processing uses local_bh_disable internally. There
is no way, later in the flow, to differentiate between whether softirq is
being processed or is it just that bh has been disabled. So, a hardirq when bh
is disabled results in time being wrongly accounted as softirq.
Looking at the code a bit more, the problem exists in !VIRT_CPU_ACCOUNTING
as well. As account_system_time() in normal tick based accouting also uses
softirq_count, which will be set even when not in softirq with bh disabled.
Peter also suggested solution of using 2*SOFTIRQ_OFFSET as irq count
for local_bh_{disable,enable} and using just SOFTIRQ_OFFSET while softirq
processing. The patch below does that and adds API in_serving_softirq() which
returns whether we are currently processing softirq or not.
Also changes one of the usages of softirq_count in net/sched/cls_cgroup.c
to in_serving_softirq.
Looks like many usages of in_softirq really want in_serving_softirq. Those
changes can be made individually on a case by case basis.
Signed-off-by: Venkatesh Pallipadi <venki@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1286237003-12406-2-git-send-email-venki@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When SD_PREFER_SIBLING is set on a sched domain, drop group_capacity to 1
only if the local group has extra capacity. The extra check prevents the case
where you always pull from the heaviest group when it is already under-utilized
(possible with a large weight task outweighs the tasks on the system).
For example, consider a 16-cpu quad-core quad-socket machine with MC and NUMA
scheduling domains. Let's say we spawn 15 nice0 tasks and one nice-15 task,
and each task is running on one core. In this case, we observe the following
events when balancing at the NUMA domain:
- find_busiest_group() will always pick the sched group containing the niced
task to be the busiest group.
- find_busiest_queue() will then always pick one of the cpus running the
nice0 task (never picks the cpu with the nice -15 task since
weighted_cpuload > imbalance).
- The load balancer fails to migrate the task since it is the running task
and increments sd->nr_balance_failed.
- It repeats the above steps a few more times until sd->nr_balance_failed > 5,
at which point it kicks off the active load balancer, wakes up the migration
thread and kicks the nice 0 task off the cpu.
The load balancer doesn't stop until we kick out all nice 0 tasks from
the sched group, leaving you with 3 idle cpus and one cpu running the
nice -15 task.
When balancing at the NUMA domain, we drop sgs.group_capacity to 1 if the child
domain (in this case MC) has SD_PREFER_SIBLING set. Subsequent load checks are
not relevant because the niced task has a very large weight.
In this patch, we add an extra condition to the "if(prefer_sibling)" check in
update_sd_lb_stats(). We drop the capacity of a group only if the local group
has extra capacity, ie. nr_running < group_capacity. This patch preserves the
original intent of the prefer_siblings check (to spread tasks across the system
in low utilization scenarios) and fixes the case above.
It helps in the following ways:
- In low utilization cases (where nr_tasks << nr_cpus), we still drop
group_capacity down to 1 if we prefer siblings.
- On very busy systems (where nr_tasks >> nr_cpus), sgs.nr_running will most
likely be > sgs.group_capacity.
- When balancing large weight tasks, if the local group does not have extra
capacity, we do not pick the group with the niced task as the busiest group.
This prevents failed balances, active migration and the under-utilization
described above.
Signed-off-by: Nikhil Rao <ncrao@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1287173550-30365-5-git-send-email-ncrao@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch forces a load balance on a newly idle cpu when the local group has
extra capacity and the busiest group does not have any. It improves system
utilization when balancing tasks with a large weight differential.
Under certain situations, such as a niced down task (i.e. nice = -15) in the
presence of nr_cpus NICE0 tasks, the niced task lands on a sched group and
kicks away other tasks because of its large weight. This leads to sub-optimal
utilization of the machine. Even though the sched group has capacity, it does
not pull tasks because sds.this_load >> sds.max_load, and f_b_g() returns NULL.
With this patch, if the local group has extra capacity, we shortcut the checks
in f_b_g() and try to pull a task over. A sched group has extra capacity if the
group capacity is greater than the number of running tasks in that group.
Thanks to Mike Galbraith for discussions leading to this patch and for the
insight to reuse SD_NEWIDLE_BALANCE.
Signed-off-by: Nikhil Rao <ncrao@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1287173550-30365-4-git-send-email-ncrao@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When cycling through sched groups to determine the busiest group, set
group_imb only if the busiest cpu has more than 1 runnable task. This patch
fixes the case where two cpus in a group have one runnable task each, but there
is a large weight differential between these two tasks. The load balancer is
unable to migrate any task from this group, and hence do not consider this
group to be imbalanced.
Signed-off-by: Nikhil Rao <ncrao@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1286996978-7007-3-git-send-email-ncrao@google.com>
[ small code readability edits ] Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch adds a check in task_hot to return if the task has SCHED_IDLE
policy. SCHED_IDLE tasks have very low weight, and when run with regular
workloads, are typically scheduled many milliseconds apart. There is no
need to consider these tasks hot for load balancing.
Signed-off-by: Nikhil Rao <ncrao@google.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1287173550-30365-2-git-send-email-ncrao@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Here a newly created task is having its runqueue assigned. The new task
is not yet on the tasklist, so cannot go away. This is therefore a false
positive, suppress with an RCU read-side critical section.
Reported-by: Ben Greear <greearb@candelatech.com Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Andi Kleen <ak@linux.intel.com> Tested-by: Ben Greear <greearb@candelatech.com Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
If a high priority task is waking up on a CPU that is running a
lower priority task that is bound to a CPU, see if we can move the
high RT task to another CPU first. Note, if all other CPUs are
running higher priority tasks than the CPU bounded current task,
then it will be preempted regardless.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Gregory Haskins <ghaskins@novell.com>
LKML-Reference: <20100921024138.888922071@goodmis.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When first working on the RT scheduler design, we concentrated on
keeping all CPUs running RT tasks instead of having multiple RT
tasks on a single CPU waiting for the migration thread to move
them. Instead we take a more proactive stance and push or pull RT
tasks from one CPU to another on wakeup or scheduling.
When an RT task wakes up on a CPU that is running another RT task,
instead of preempting it and killing the cache of the running RT
task, we look to see if we can migrate the RT task that is waking
up, even if the RT task waking up is of higher priority.
This may sound a bit odd, but RT tasks should be limited in
migration by the user anyway. But in practice, people do not do
this, which causes high prio RT tasks to bounce around the CPUs.
This becomes even worse when we have priority inheritance, because
a high prio task can block on a lower prio task and boost its
priority. When the lower prio task wakes up the high prio task, if
it happens to be on the same CPU it will migrate off of it.
But in reality, the above does not happen much either, because the
wake up of the lower prio task, which has already been boosted, if
it was on the same CPU as the higher prio task, it would then
migrate off of it. But anyway, we do not want to migrate them
either.
To examine the scheduling, I created a test program and examined it
under kernelshark. The test program created CPU * 2 threads, where
each thread had a different priority. The program takes different
options. The options used in this change log was to have priority
inheritance mutexes or not.
All threads did the following loop:
static void grab_lock(long id, int iter, int l)
{
ftrace_write("thread %ld iter %d, taking lock %d\n",
id, iter, l);
pthread_mutex_lock(&locks[l]);
ftrace_write("thread %ld iter %d, took lock %d\n",
id, iter, l);
busy_loop(nr_tasks - id);
ftrace_write("thread %ld iter %d, unlock lock %d\n",
id, iter, l);
pthread_mutex_unlock(&locks[l]);
}
void *start_task(void *id)
{
[...]
while (!done) {
for (l = 0; l < nr_locks; l++) {
grab_lock(id, i, l);
ftrace_write("thread %ld iter %d sleeping\n",
id, i);
ms_sleep(id);
}
i++;
}
[...]
}
The busy_loop(ms) keeps the CPU spinning for ms milliseconds. The
ms_sleep(ms) sleeps for ms milliseconds. The ftrace_write() writes
to the ftrace buffer to help analyze via ftrace.
The higher the id, the higher the prio, the shorter it does the
busy loop, but the longer it spins. This is usually the case with
RT tasks, the lower priority tasks usually run longer than higher
priority tasks.
At the end of the test, it records the number of loops each thread
took, as well as the number of voluntary preemptions, non-voluntary
preemptions, and number of migrations each thread took, taking the
information from /proc/$$/sched and /proc/$$/status.
Running this on a 4 CPU processor, the results without changes to
the kernel looked like this:
The total # of migrations did not change (several runs showed the
difference all within the noise). But we now see a dramatic
improvement to the higher priority tasks. (kernelshark showed that
the watchdog timer bumped the highest priority task to give it the
2 count. This was actually consistent with every run).
Notice that the # of iterations did not change either.
The above was with priority inheritance mutexes. That is, when the
higher prority task blocked on a lower priority task, the lower
priority task would inherit the higher priority task (which shows
why task 6 was bumped so many times). When not using priority
inheritance mutexes, the current kernel shows this:
Which shows a even bigger change. The big difference between task 3
and task 4 is because we have only 4 CPUs on the machine, causing
the 4 highest prio tasks to always have preference.
Although I did not measure cache misses, and I'm sure there would
be little to measure since the test was not data intensive, I could
imagine large improvements for higher priority tasks when dealing
with lower priority tasks. Thus, I'm satisfied with making the
change and agreeing with what Gregory Haskins argued a few years
ago when we first had this discussion.
One final note. All tasks in the above tests were RT tasks. Any RT
task will always preempt a non RT task that is running on the CPU
the RT task wants to run on.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Gregory Haskins <ghaskins@novell.com>
LKML-Reference: <20100921024138.605460343@goodmis.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
scheduler uses cache_nice_tries as an indicator to do cache_hot and
active load balance, when normal load balance fails. Currently,
this value is changed on any failed load balance attempt. That ends
up being not so nice to workloads that enter/exit idle often, as
they do more frequent new_idle balance and that pretty soon results
in cache hot tasks being pulled in.
Making the cache_nice_tries ignore failed new_idle balance seems to
make better sense. With that only the failed load balance in
periodic load balance gets accounted and the rate of accumulation
of cache_nice_tries will not depend on idle entry/exit (short
running sleep-wakeup kind of tasks). This reduces movement of
cache_hot tasks.
schedstat diff (after-before) excerpt from a workload that has
frequent and short wakeup-idle pattern (:2 in cpu col below refers
to NEWIDLE idx) This snapshot was across ~400 seconds.
Currently sched_avg_update() (which updates rt_avg stats in the rq)
is getting called from scale_rt_power() (in the load balance context)
which doesn't take rq->lock.
Fix it by moving the sched_avg_update() to more appropriate
update_cpu_load() where the CFS load gets updated as well.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <1282596171.2694.3.camel@sbsiddha-MOBL3> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Mike Galbraith <efault@gmx.de> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When there's an xHCI host power loss after a suspend from memory, the USB
core attempts to reset and verify the USB devices that are attached to the
system. The xHCI driver has to reallocate those devices, since the
hardware lost all knowledge of them during the power loss.
When a hub is plugged in, and the host loses power, the xHCI hardware
structures are not updated to say the device is a hub. This is usually
done in hub_configure() when the USB hub is detected. That function is
skipped during a reset and verify by the USB core, since the core restores
the old configuration and alternate settings, and the hub driver has no
idea this happened. This bug makes the xHCI host controller reject the
enumeration of low speed devices under the resumed hub.
Therefore, make the USB core re-setup the internal xHCI hub device
information by calling update_hub_device() when hub_activate() is called
for a hub reset resume. After a host power loss, all devices under the
roothub get a reset-resume or a disconnect.
This patch should be queued for the 2.6.37 stable tree.
Clearing the cpu in prev's mm_cpumask early will avoid the flush tlb
IPI's while the cr3 is still pointing to the prev mm. And this window
can lead to the possibility of bogus TLB fills resulting in strange
failures. One such problematic scenario is mentioned below.
T1. CPU-1 is context switching from mm1 to mm2 context and got a NMI
etc between the point of clearing the cpu from the mm_cpumask(mm1)
and before reloading the cr3 with the new mm2.
T2. CPU-2 is tearing down a specific vma for mm1 and will proceed with
flushing the TLB for mm1. It doesn't send the flush TLB to CPU-1
as it doesn't see that cpu listed in the mm_cpumask(mm1).
T3. After the TLB flush is complete, CPU-2 goes ahead and frees the
page-table pages associated with the removed vma mapping.
T4. CPU-2 now allocates those freed page-table pages for something
else.
T5. As the CR3 and TLB caches for mm1 is still active on CPU-1, CPU-1
can potentially speculate and walk through the page-table caches
and can insert new TLB entries. As the page-table pages are
already freed and being used on CPU-2, this page walk can
potentially insert a bogus global TLB entry depending on the
(random) contents of the page that is being used on CPU-2.
T6. This bogus TLB entry being global will be active across future CR3
changes and can result in weird memory corruption etc.
To avoid this issue, for the prev mm that is handing over the cpu to
another mm, clear the cpu from the mm_cpumask(prev) after the cr3 is
changed.
Marking it for -stable, though we haven't seen any reported failure that
can be attributed to this.