aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2011-11-02Merge git://git.kernel.org/pub/scm/linux/kernel/git/steve/linux-dmLinus Torvalds42-37/+12079
* git://git.kernel.org/pub/scm/linux/kernel/git/steve/linux-dm: dm: raid fix device status indicator when array initializing dm log userspace: add log device dependency dm log userspace: fix comment hyphens dm: add thin provisioning target dm: add persistent data library dm: add bufio dm: export dm get md dm table: add immutable feature dm table: add always writeable feature dm table: add singleton feature dm kcopyd: add dm_kcopyd_zero to zero an area dm: remove superfluous smp_mb dm: use local printk ratelimit dm table: propagate non rotational flag
2011-11-02Merge branch 'for-linus' of git://git.selinuxproject.org/~jmorris/linux-securityLinus Torvalds1-0/+30
* 'for-linus' of git://git.selinuxproject.org/~jmorris/linux-security: TOMOYO: Fix interactive judgment functionality.
2011-11-02ixgbevf: Update release versionGreg Rose1-1/+1
Signed-off-by: Greg Rose <[email protected]> Tested-by: Sibai Li <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02ixgbe: DCB, return max for IEEE traffic classesJohn Fastabend1-4/+6
Returning the max traffic classes on get requests simplifies user space configurations because applications will know explicitly how many traffic classes can be used. Typical switch implementations use 2 or 3 traffic classes so this not seen often today. And user space can learn the number of traffic classes by return codes but this allows user space to configure ixgbe correctly at the start. Signed-off-by: John Fastabend <[email protected]> Tested-by: Ross Brattain <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02ixgbe: fix reading of the buffer returned by the firmwareEmil Tantilov1-9/+11
This patch fixes some issues found in the buffer read portion of ixgbe_host_interface_command() - use `bi` as the buffer index counter instead of `i` - add conversion to native cpu byte ordering on register read - fix conversion from bytes to dword - use dword_len instead of buf_len when reading the register Signed-off-by: Emil Tantilov <[email protected]> Tested-by: Phil Schmitt <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02ixgbe: Fix compiler warningsGreg Rose1-0/+2
Wrap SR-IOV specific functions in CONFIG_PCI_IOV to avoid compiler warnings. Signed-off-by: Greg Rose <[email protected]> Tested-by: Sibai Li <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02ixgbe: fix smatch splat due to missing NULL checkJohn Fastabend1-21/+12
ixgbe_ieee_ets and ixgbe_ieee_pfc are intialized at the same time. Do a check for both before configuring IEEE802.1Qaz. Also max_frame was causing a sparse warning resolved here as well. Reported-by: Dan Carpenter <[email protected]> Signed-off-by: John Fastabend <[email protected]> Tested-by: Ross Brattain <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02ixgbe: fix disabling of Tx laser at probeEmil Tantilov1-7/+7
register_netdev() calls ndo_set_features() which may result in HW reset which in turn will bring the laser back up. This patch moves ixgbe_laser_tx_disable() below register_netdev() in ixgbe_probe() to make sure laser is shut off on load. Signed-off-by: Emil Tantilov <[email protected]> Tested-by: Phil Schmitt <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02Merge branch 'linux_next' of ↵Linus Torvalds11-525/+2671
git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-edac * 'linux_next' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-edac: (21 commits) MAINTAINERS: add an entry for Edac Sandy Bridge driver edac: tag sb_edac as EXPERIMENTAL, as it requires more testing EDAC: Fix incorrect edac mode reporting in sb_edac edac: sb_edac: Add it to the building system edac: Add an experimental new driver to support Sandy Bridge CPU's i7300_edac: Fix error cleanup logic i7core_edac: Initialize memory name with cpu, channel, bank i7core_edac: Fix compilation on 32 bits arch i7core_edac: scrubbing fixups EDAC: Correct Kconfig dependencies i7core_edac: return -ENODEV if no MC is found i7core_edac: use edac's own way to print errors MAINTAINERS: remove dropped edac_mce.* from the file i7core_edac: Drop the edac_mce facility x86, MCE: Use notifier chain only for MCE decoding EDAC i7core: Use mce socketid for better compatibility i7core_edac: Don't enable memory scrubbing for Xeon 35xx i7core_edac: Add scrubbing support edac: Move edac main structs to include/linux/edac.h i7core_edac: Fix oops when trying to inject errors ...
2011-11-02ixgbe: Fix link issues caused by a reset while interface is downEmil Tantilov1-1/+0
Interface fails to obtain link on 82599 SFP in the following scenario: 1. Set advertised speed to GB: ethtool -s eth0 advertise 0x20 2. Bring interface down ip link set eth0 down 3. Issue any command that leads to a reset: ethtool -t eth0 4. Bring link back up: ip link set eth0 up Following patch makes sure that the driver flaps the Tx laser every time ixgbe_start_hw() is called, and not only when the speed is set. Signed-off-by: Emil Tantilov <[email protected]> Tested-by: Phil Schmitt <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02igb: Fix for I347AT4 PHY cable length unit detectionKantecki, Tomasz1-1/+1
The PHY cable length unit detection was not using the correct the correct PHY data variable for I347AT4. Signed-off-by: Tomasz Kantecki <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02e100: make sure vlan support isn't advertised on old adaptersJesse Brandeburg1-0/+4
e100 parts don't support vlan offload but they generally do allow use of vlans in higher software layers via the 8021q module. That said, there are a couple of really old revisions of e100 hardware that don't even allow the longer frame sizes required for vlan use with standard MTU. Use the VLAN_CHALLENGED flag to prevent vlan binding to these devices. Reported-by: Michael Tokarev <[email protected]> Signed-off-by: Jesse Brandeburg <[email protected]> CC: Michael Tokarev <[email protected]> CC: David Lamparter <[email protected]> Tested-by: Aaron Brown <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02e1000e: demote a debugging WARN to a debug log messageBruce Allan1-2/+1
This debugging message was recently added but it does not need to be as alarming as a WARN. Signed-off-by: Bruce Allan <[email protected]> Tested-by: Aaron Brown <[email protected]> Signed-off-by: Jeff Kirsher <[email protected]>
2011-11-02Merge branch 'for-3.2' of git://linux-nfs.org/~bfields/linuxLinus Torvalds1-1/+1
* 'for-3.2' of git://linux-nfs.org/~bfields/linux: nfsd4: typo logical vs bitwise negate in nfsd4_decode_share_access
2011-11-02Merge branch 'misc-3.2' of ↵Linus Torvalds9-9/+19
git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux * 'misc-3.2' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux: MAINTAINERS: Update entry for IA64 [IA64] gpio: GENERIC_GPIO default must be n [IA64[ add CONFIG_NET_VENDOR_INTEL=y to default config files where needed [IA64] agp/hp-agp: Allow binding user memory to the AGP GART [IA64] sn2: add missing put_cpu()
2011-11-02Merge branch 'akpm' (Andrew's incoming - part two)Linus Torvalds70-1035/+4909
Says Andrew: "60 patches. That's good enough for -rc1 I guess. I have quite a lot of detritus to be rechecked, work through maintainers, etc. - most of the remains of MM - rtc - various misc - cgroups - memcg - cpusets - procfs - ipc - rapidio - sysctl - pps - w1 - drivers/misc - aio" * akpm: (60 commits) memcg: replace ss->id_lock with a rwlock aio: allocate kiocbs in batches drivers/misc/vmw_balloon.c: fix typo in code comment drivers/misc/vmw_balloon.c: determine page allocation flag can_sleep outside loop w1: disable irqs in critical section drivers/w1/w1_int.c: multiple masters used same init_name drivers/power/ds2780_battery.c: fix deadlock upon insertion and removal drivers/power/ds2780_battery.c: add a nolock function to w1 interface drivers/power/ds2780_battery.c: create central point for calling w1 interface w1: ds2760 and ds2780, use ida for id and ida_simple_get() to get it pps gpio client: add missing dependency pps: new client driver using GPIO pps: default echo function include/linux/dma-mapping.h: add dma_zalloc_coherent() sysctl: make CONFIG_SYSCTL_SYSCALL default to n sysctl: add support for poll() RapidIO: documentation update drivers/net/rionet.c: fix ethernet address macros for LE platforms RapidIO: fix potential null deref in rio_setup_device() RapidIO: add mport driver for Tsi721 bridge ...
2011-11-02memcg: replace ss->id_lock with a rwlockAndrew Bresticker2-10/+10
While back-porting Johannes Weiner's patch "mm: memcg-aware global reclaim" for an internal effort, we noticed a significant performance regression during page-reclaim heavy workloads due to high contention of the ss->id_lock. This lock protects idr map, and serializes calls to idr_get_next() in css_get_next() (which is used during the memcg hierarchy walk). Since idr_get_next() is just doing a look up, we need only serialize it with respect to idr_remove()/idr_get_new(). By making the ss->id_lock a rwlock, contention is greatly reduced and performance improves. Tested: cat a 256m file from a ramdisk in a 128m container 50 times on each core (one file + container per core) in parallel on a NUMA machine. Result is the time for the test to complete in 1 of the containers. Both kernels included Johannes' memcg-aware global reclaim patches. Before rwlock patch: 1710.778s After rwlock patch: 152.227s Signed-off-by: Andrew Bresticker <[email protected]> Cc: Paul Menage <[email protected]> Cc: Li Zefan <[email protected]> Acked-by: KAMEZAWA Hiroyuki <[email protected]> Cc: Ying Han <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02aio: allocate kiocbs in batchesJeff Moyer2-29/+108
In testing aio on a fast storage device, I found that the context lock takes up a fair amount of cpu time in the I/O submission path. The reason is that we take it for every I/O submitted (see __aio_get_req). Since we know how many I/Os are passed to io_submit, we can preallocate the kiocbs in batches, reducing the number of times we take and release the lock. In my testing, I was able to reduce the amount of time spent in _raw_spin_lock_irq by .56% (average of 3 runs). The command I used to test this was: aio-stress -O -o 2 -o 3 -r 8 -d 128 -b 32 -i 32 -s 16384 <dev> I also tested the patch with various numbers of events passed to io_submit, and I ran the xfstests aio group of tests to ensure I didn't break anything. Signed-off-by: Jeff Moyer <[email protected]> Cc: Daniel Ehrenberg <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/misc/vmw_balloon.c: fix typo in code commentRakib Mullick1-1/+1
Fix typo in code comment. Signed-off-by: Rakib Mullick <[email protected]> Acked-by: Dmitry Torokhov <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/misc/vmw_balloon.c: determine page allocation flag can_sleep outside ↵Rakib Mullick1-1/+1
loop In vmballoon_reserve_page(), flags has been passed from the callee function (vmballoon_inflate here). So, we can determine can_sleep outside the loop. Signed-off-by: Rakib Mullick <[email protected]> Acked-by: Dmitry Torokhov <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02w1: disable irqs in critical sectionJan Weitzel1-0/+5
Interrupting w1_delay() in w1_read_bit() results in missing the low level on the w1 line and receiving "1" instead of "0". Add local_irq_save()/local_irq_restore() around the critical section Signed-off-by: Jan Weitzel <[email protected]> Acked-by: Evgeniy Polyakov <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/w1/w1_int.c: multiple masters used same init_nameFlorian Faber1-0/+1
When using multiple masters, w1_int.c would use the .init_name from w1.c for all entities, which will fail when creating a corresponding sysfs entry. This patch uses the unique name previously generated. WARNING: at fs/sysfs/dir.c:451 sysfs_add_one+0x48/0x64() sysfs: cannot create duplicate filename '/devices/w1 bus master' Modules linked in: Call trace: [<9001a604>] warn_slowpath_common+0x34/0x44 [<9001a64c>] warn_slowpath_fmt+0x14/0x18 [<90078020>] sysfs_add_one+0x48/0x64 [<900784ec>] create_dir+0x40/0x68 [<9007857a>] sysfs_create_dir+0x66/0x78 [<900c1a8a>] kobject_add_internal+0x6e/0x104 [<900c1bc0>] kobject_add_varg+0x20/0x2c [<900c1c1c>] kobject_add+0x30/0x3c [<900dbd66>] device_add+0x6a/0x378 [<900dbb4a>] device_initialize+0x12/0x48 [<900dc080>] device_register+0xc/0x10 [<900f99be>] w1_add_master_device+0x162/0x274 [<90008e7a>] w1_gpio_probe+0x66/0xb4 [<9000030c>] kernel_init+0x0/0xe8 [<900dde54>] platform_drv_probe+0xc/0xe [<9000030c>] kernel_init+0x0/0xe8 [<900dd4f8>] driver_probe_device+0x6c/0xdc [<900dd5fc>] __driver_attach+0x34/0x48 [<900dcce8>] bus_for_each_dev+0x2c/0x48 [<900dd5c8>] __driver_attach+0x0/0x48 [<900dd38c>] driver_attach+0x10/0x14 [<900dd16a>] bus_add_driver+0x6a/0x18c [<900dd768>] driver_register+0x60/0xb8 [<90011594>] __initcall_w1_therm_init6+0x0/0x4 [<90008e00>] w1_gpio_init+0x0/0x14 [<9000030c>] kernel_init+0x0/0xe8 [<900ddf48>] platform_driver_register+0x30/0x38 [<90011594>] __initcall_w1_therm_init6+0x0/0x4 [<90008e00>] w1_gpio_init+0x0/0x14 [<9000030c>] kernel_init+0x0/0xe8 [<900ddf5e>] platform_driver_probe+0xe/0x3c [<90008e0c>] w1_gpio_init+0xc/0x14 [<90011594>] __initcall_w1_therm_init6+0x0/0x4 [<90008e00>] w1_gpio_init+0x0/0x14 [<900126d4>] do_one_initcall+0x34/0x130 [<90000372>] kernel_init+0x66/0xe8 [<90011594>] __initcall_w1_therm_init6+0x0/0x4 [<9001ca3e>] do_exit+0x0/0x3a6 [<9000030c>] kernel_init+0x0/0xe8 [<9001ca3e>] do_exit+0x0/0x3a6 ---[ end trace 5a9233884fead918 ]--- kobject_add_internal failed for w1 bus master with -EEXIST, don't try to register things with the same name in the same directory. Signed-off-by: Florian Faber <[email protected]> Cc: Evgeniy Polyakov <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/power/ds2780_battery.c: fix deadlock upon insertion and removalClifton Barnes1-0/+9
Fixes the deadlock when inserting and removing the ds2780. Signed-off-by: Clifton Barnes <[email protected]> Cc: Evgeniy Polyakov <[email protected]> Cc: <[email protected]> Cc: <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/power/ds2780_battery.c: add a nolock function to w1 interfaceClifton Barnes2-13/+37
Adds a nolock function to the w1 interface to avoid locking the mutex if needed. Signed-off-by: Clifton Barnes <[email protected]> Cc: Evgeniy Polyakov <[email protected]> Cc: <[email protected]> Cc: <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/power/ds2780_battery.c: create central point for calling w1 interfaceClifton Barnes1-35/+42
Simply creates one point to call the w1 interface. Signed-off-by: Clifton Barnes <[email protected]> Cc: Evgeniy Polyakov <[email protected]> Cc: <[email protected]> Cc: <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02w1: ds2760 and ds2780, use ida for id and ida_simple_get() to get itJonathan Cameron2-84/+12
Straightforward. As an aside, the ida_init calls are not needed as far as I can see needed. (DEFINE_IDA does the same already). Signed-off-by: Jonathan Cameron <[email protected]> Cc: Evgeniy Polyakov <[email protected]> Acked-by: Clifton Barnes <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02pps gpio client: add missing dependencyHeiko Carstens1-1/+1
Add "depends on GENERIC_HARDIRQS" to avoid compile breakage on s390: drivers/built-in.o: In function `pps_gpio_remove': linux-next/drivers/pps/clients/pps-gpio.c:189: undefined reference to `free_irq' Signed-off-by: Heiko Carstens <[email protected]> Cc: James Nuss <[email protected]> Cc: Rodolfo Giometti <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02pps: new client driver using GPIOJames Nuss4-0/+269
This client driver allows you to use a GPIO pin as a source for PPS signals. Platform data [1] are used to specify the GPIO pin number, label, assert event edge type, and whether clear events are captured. This driver is based on the work by Ricardo Martins who submitted an initial implementation [2] of a PPS IRQ client driver to the linuxpps mailing-list on Dec 3 2010. [1] include/linux/pps-gpio.h [2] http://ml.enneenne.com/pipermail/linuxpps/2010-December/004155.html [[email protected]: remove unneeded cast of void*] Signed-off-by: James Nuss <[email protected]> Cc: Ricardo Martins <[email protected]> Acked-by: Rodolfo Giometti <[email protected]> Signed-off-by: Ricardo Martins <[email protected]> Cc: Alexander Gordeev <[email protected]> Cc: Igor Plyatov <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02pps: default echo functionJames Nuss3-28/+13
A default echo function has been provided so it is no longer an error when you specify PPS_ECHOASSERT or PPS_ECHOCLEAR without an explicit echo function. This allows some code re-use and also makes it easier to write client drivers since the default echo function does not normally need to change. Signed-off-by: James Nuss <[email protected]> Reviewed-by: Ben Gardiner <[email protected]> Acked-by: Rodolfo Giometti <[email protected]> Cc: Ricardo Martins <[email protected]> Cc: Alexander Gordeev <[email protected]> Cc: Igor Plyatov <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02include/linux/dma-mapping.h: add dma_zalloc_coherent()Andrew Morton2-0/+17
Lots of driver code does a dma_alloc_coherent() and then zeroes out the memory with a memset. Make it easy for them. Cc: Alexandre Bounine <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02sysctl: make CONFIG_SYSCTL_SYSCALL default to nWANG Cong2-37/+2
When I tried to send a patch to remove it, Andi told me we still need to keep compabitlies for old libc, so we can't remove this completely. Then just make it default to n and remove the doc from feature-removal-schedule.txt. Signed-off-by: WANG Cong <[email protected]> Cc: Eric Biederman <[email protected]> Cc: Andi Kleen <[email protected]> Cc: Alexey Dobriyan <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02sysctl: add support for poll()Lucas De Marchi5-0/+108
Adding support for poll() in sysctl fs allows userspace to receive notifications of changes in sysctl entries. This adds a infrastructure to allow files in sysctl fs to be pollable and implements it for hostname and domainname. [[email protected]: s/declare/define/ for definitions] Signed-off-by: Lucas De Marchi <[email protected]> Cc: Greg KH <[email protected]> Cc: Kay Sievers <[email protected]> Cc: Al Viro <[email protected]> Cc: "Eric W. Biederman" <[email protected]> Cc: Alexey Dobriyan <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02RapidIO: documentation updateAlexandre Bounine1-1/+1
Update rapidio.txt to reflect changes from recent patch. See http://marc.info/?l=linux-kernel&m=131285620113589&w=2 for details. Signed-off-by: Alexandre Bounine <[email protected]> Cc: Liu Gang <[email protected]> Cc: Micha Nelissen <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/net/rionet.c: fix ethernet address macros for LE platformsAlexandre Bounine1-2/+2
Modify Ethernet addess macros to be compatible with BE/LE platforms Signed-off-by: Alexandre Bounine <[email protected]> Cc: Chul Kim <[email protected]> Cc: Kumar Gala <[email protected]> Cc: Matt Porter <[email protected]> Cc: Li Yang <[email protected]> Cc: <[email protected]> [2.6.39+] Cc: <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02RapidIO: fix potential null deref in rio_setup_device()Alexandre Bounine1-1/+1
The "goto cleanup" path can deference "rswitch" when it is NULL. Reported-by: Dan Carpenter <[email protected]> Signed-off-by: Alexandre Bounine <[email protected]> Cc: Dan Carpenter <[email protected]> Cc: Kumar Gala <[email protected]> Cc: Matt Porter <[email protected]> Cc: Chul Kim <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02RapidIO: add mport driver for Tsi721 bridgeAlexandre Bounine8-2/+3196
Add RapidIO mport driver for IDT TSI721 PCI Express-to-SRIO bridge device. The driver provides full set of callback functions defined for mport devices in RapidIO subsystem. It also is compatible with current version of RIONET driver (Ethernet over RapidIO messaging services). This patch is applicable to kernel versions starting from 2.6.39. Signed-off-by: Alexandre Bounine <[email protected]> Signed-off-by: Chul Kim <[email protected]> Cc: Kumar Gala <[email protected]> Cc: Matt Porter <[email protected]> Cc: Li Yang <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02arch/powerpc/sysdev/fsl_rio.c: release rapidio port I/O region resource if ↵Liu Gang1-0/+1
port failed to initialize The "struct rio_mport" contains a member of master port I/O memory resource structure "struct resource iores". This resource will be read from device tree and be used for rapidio R/W transaction memory space. Rapidio requests the port I/O memory resource under the root resource "iomem_resource". struct rio_mport *port; port = kzalloc(sizeof(struct rio_mport), GFP_KERNEL); request_resource(&iomem_resource, &port->iores); When port failed to initialize, allocated "rio_mport" structure memory will be freed, and the port I/O memory resource structure pointer "&port->iores" will be invalid. If other requests resource under "iomem_resource", "&port->iores" node may be operated in the child resources list and this will cause the system to crash. So the requested port I/O memory resource should be released before freeing allocated "rio_mport" structure. Signed-off-by: Liu Gang <[email protected]> Acked-by: Alexandre Bounine <[email protected]> Cc: Benjamin Herrenschmidt <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Grant Likely <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02drivers/rapidio/rio-scan.c: use discovered bit to test if enumeration is ↵Liu Gang1-2/+2
complete The discovered bit in PGCCSR register indicates if the device has been discovered by system host. In Rapidio systems, some agent devices can also be master devices. They can issue requests into the system. Signed-off-by: Liu Gang <[email protected]> Acked-by: Alexandre Bounine <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02init: add root=PARTUUID=UUID/PARTNROFF=%d supportWill Drewry1-5/+43
Expand root=PARTUUID=UUID syntax to support selecting a root partition by integer offset from a known, unique partition. This approach provides similar properties to specifying a device and partition number, but using the UUID as the unique path prior to evaluating the offset. For example, root=PARTUUID=99DE9194-FC15-4223-9192-FC243948F88B/PARTNROFF=1 selects the partition with UUID 99DE.. then select the next partition. This change is motivated by a particular usecase in Chromium OS where the bootloader can easily determine what partition it is on (by UUID) but doesn't perform general partition table walking. That said, support for this model provides a direct mechanism for the user to modify the root partition to boot without specifically needing to extract each UUID or update the bootloader explicitly when the root partition UUID is changed (if it is recreated to be larger, for instance). Pinning to a /boot-style partition UUID allows the arbitrary root partition reconfiguration/modifications with slightly less ambiguity than just [dev][partition] and less stringency than the specific root partition UUID. [[email protected]: fix init sections warning] Signed-off-by: Will Drewry <[email protected]> Cc: Kay Sievers <[email protected]> Cc: Randy Dunlap <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Trond Myklebust <[email protected]> Cc: Jens Axboe <[email protected]> Signed-off-by: Stephen Rothwell <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02include/linux/sem.h: make sysv_sem empty if SYSVIPC is disabledManfred Spraul1-2/+7
For the sysvsem undo, each task struct contains a sysv_sem structure with a pointer to the undo information. This pointer is only necessary if sysvipc is enabled - thus the pointer can be made conditional on CONFIG_SYSVIPC. Signed-off-by: Manfred Spraul <[email protected]> Acked-by: Peter Zijlstra <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Mike Galbraith <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02ipc/sem.c: remove private structures from public header fileManfred Spraul2-42/+46
include/linux/sem.h contains several structures that are only used within ipc/sem.c. The patch moves them into ipc/sem.c - there is no need to expose the structures to the whole kernel. No functional changes, only whitespace cleanups and 80-char per line fixes. Signed-off-by: Manfred Spraul <[email protected]> Acked-by: Peter Zijlstra <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Mike Galbraith <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02ipc/sem.c: handle spurious wakeupsManfred Spraul1-0/+9
semtimedop() does not handle spurious wakeups, it returns -EINTR to user space. Most other schedule() users would just loop and not return to user space. The patch adds such a loop to semtimedop() Signed-off-by: Manfred Spraul <[email protected]> Reported-by: Peter Zijlstra <[email protected]> Acked-by: Peter Zijlstra <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Mike Galbraith <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02ipc/sem.c: fix return code race with semop vs. semop +semctl(IPC_RMID)Manfred Spraul1-1/+0
sys_semtimedop() may return -EIDRM although the semaphore operation completed successfully: thread 1: thread 2: semtimedop(), sleeps semop(): * acquires sem_lock() semtimedop() woken up due to timeout sem_lock() loops * notices that thread 2 could be completed. * performs the operations that thread 2 is sleeping on. * marks the semaphore operation as IN_WAKEUP * drops sem_lock(), does wakeup, sets return code to 0 * thread delayed due to interrupt, whatever * returns to user space * thread still delayed semctl(IPC_RMID) * acquires sem_lock() * ipc_rmid(), ipcp->deleted=1 * drops sem_lock() * thread finally continues - but seem_lock() now fails due to ipcp->deleted == 1 * returns -EIDRM instead of 0 The fix is trivial: Always use the return code in queue.status. In real world, the race probably doesn't matter: If the semaphore array is destroyed, the app is probably not interested if the last operation succeeded or was already cancelled. Signed-off-by: Manfred Spraul <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Mike Galbraith <[email protected]> Acked-by: Peter Zijlstra <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02ida: make ida_simple_get/put() IRQ safeTejun Heo1-4/+7
It's often convenient to be able to release resource from IRQ context. Make ida_simple_*() use irqsave/restore spin ops so that they are IRQ safe. Signed-off-by: Tejun Heo <[email protected]> Acked-by: Rusty Russell <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02proc: fix races against execve() of /proc/PID/fd**Vasiliy Kulikov1-43/+103
fd* files are restricted to the task's owner, and other users may not get direct access to them. But one may open any of these files and run any setuid program, keeping opened file descriptors. As there are permission checks on open(), but not on readdir() and read(), operations on the kept file descriptors will not be checked. It makes it possible to violate procfs permission model. Reading fdinfo/* may disclosure current fds' position and flags, reading directory contents of fdinfo/ and fd/ may disclosure the number of opened files by the target task. This information is not sensible per se, but it can reveal some private information (like length of a password stored in a file) under certain conditions. Used existing (un)lock_trace functions to check for ptrace_may_access(), but instead of using EPERM return code from it use EACCES to be consistent with existing proc_pid_follow_link()/proc_pid_readlink() return code. If they differ, attacker can guess what fds exist by analyzing stat() return code. Patched handlers: stat() for fd/*, stat() and read() for fdindo/*, readdir() and lookup() for fd/ and fdinfo/. Signed-off-by: Vasiliy Kulikov <[email protected]> Cc: Cyrill Gorcunov <[email protected]> Cc: <[email protected]> Cc: <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02procfs: report EISDIR when reading sysctl dirs in procPavel Emelyanov1-0/+1
On reading sysctl dirs we should return -EISDIR instead of -EINVAL. Signed-off-by: Pavel Emelyanov <[email protected]> Signed-off-by: Cyrill Gorcunov <[email protected]> Cc: Alexey Dobriyan <[email protected]> Cc: Al Viro <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02cpusets: avoid looping when storing to mems_allowed if one node remains setDavid Rientjes1-3/+6
{get,put}_mems_allowed() exist so that general kernel code may locklessly access a task's set of allowable nodes without having the chance that a concurrent write will cause the nodemask to be empty on configurations where MAX_NUMNODES > BITS_PER_LONG. This could incur a significant delay, however, especially in low memory conditions because the page allocator is blocking and reclaim requires get_mems_allowed() itself. It is not atypical to see writes to cpuset.mems take over 2 seconds to complete, for example. In low memory conditions, this is problematic because it's one of the most imporant times to change cpuset.mems in the first place! The only way a task's set of allowable nodes may change is through cpusets by writing to cpuset.mems and when attaching a task to a generic code is not reading the nodemask with get_mems_allowed() at the same time, and then clearing all the old nodes. This prevents the possibility that a reader will see an empty nodemask at the same time the writer is storing a new nodemask. If at least one node remains unchanged, though, it's possible to simply set all new nodes and then clear all the old nodes. Changing a task's nodemask is protected by cgroup_mutex so it's guaranteed that two threads are not changing the same task's nodemask at the same time, so the nodemask is guaranteed to be stored before another thread changes it and determines whether a node remains set or not. Signed-off-by: David Rientjes <[email protected]> Cc: Miao Xie <[email protected]> Cc: KOSAKI Motohiro <[email protected]> Cc: Nick Piggin <[email protected]> Cc: Paul Menage <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02mm/page_cgroup.c: quiet sparse noiseH Hartley Sweeten1-1/+1
warning: symbol 'swap_cgroup_ctrl' was not declared. Should it be static? Signed-off-by: H Hartley Sweeten <[email protected]> Cc: Paul Menage <[email protected]> Cc: Li Zefan <[email protected]> Acked-by: Balbir Singh <[email protected]> Cc: Daisuke Nishimura <[email protected]> Acked-by: KAMEZAWA Hiroyuki <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02memcg: Fix race condition in memcg_check_events() with this_cpu usageSteven Rostedt1-4/+6
Various code in memcontrol.c () calls this_cpu_read() on the calculations to be done from two different percpu variables, or does an open-coded read-modify-write on a single percpu variable. Disable preemption throughout these operations so that the writes go to the correct palces. [[email protected]: added this_cpu to __this_cpu conversion] Signed-off-by: Johannes Weiner <[email protected]> Signed-off-by: Steven Rostedt <[email protected]> Cc: Greg Thelen <[email protected]> Cc: KAMEZAWA Hiroyuki <[email protected]> Cc: Balbir Singh <[email protected]> Cc: Daisuke Nishimura <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Christoph Lameter <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2011-11-02memcg: close race between charge and putbackJohannes Weiner1-1/+20
There is a potential race between a thread charging a page and another thread putting it back to the LRU list: charge: putback: SetPageCgroupUsed SetPageLRU PageLRU && add to memcg LRU PageCgroupUsed && add to memcg LRU The order of setting one flag and checking the other is crucial, otherwise the charge may observe !PageLRU while the putback observes !PageCgroupUsed and the page is not linked to the memcg LRU at all. Global memory pressure may fix this by trying to isolate and putback the page for reclaim, where that putback would link it to the memcg LRU again. Without that, the memory cgroup is undeletable due to a charge whose physical page can not be found and moved out. Signed-off-by: Johannes Weiner <[email protected]> Cc: Ying Han <[email protected]> Acked-by: KAMEZAWA Hiroyuki <[email protected]> Cc: Daisuke Nishimura <[email protected]> Cc: Balbir Singh <[email protected]> Cc: Michal Hocko <[email protected]> Signed-off-by: Andrew Morton <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.