aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* Merge branch 'for-linus' of ↵Linus Torvalds2010-05-20
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/suspend-2.6: PM: PM QOS update fix Freezer / cgroup freezer: Update stale locking comments PM / platform_bus: Allow runtime PM by default i2c: Fix bus-level power management callbacks PM QOS update PM / Hibernate: Fix block_io.c printk warning PM / Hibernate: Group swap ops PM / Hibernate: Move the first_sector out of swsusp_write PM / Hibernate: Separate block_io PM / Hibernate: Snapshot cleanup FS / libfs: Implement simple_write_to_buffer PM / Hibernate: document open(/dev/snapshot) side effects PM / Runtime: Add sysfs debug files PM: Improve device power management document PM: Update device power management document PM: Allow runtime_suspend methods to call pm_schedule_suspend() PM: pm_wakeup - switch to using bool
| * PM: PM QOS update fixMark Gross2010-05-16
| | | | | | | | | | | | | | | | | | | | | | | | This update handles a use case where pm_qos update requests need to silently fail if the update is being sent to a handle that is NULL. The problem was that the original pm_qos silently fails when a request update is passed to a parameter that has not been added to the list yet. This update restores that behavior. Signed-off-by: markgross <markgross@thegnar.org> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * Freezer / cgroup freezer: Update stale locking commentsMatt Helsley2010-05-10
| | | | | | | | | | | | | | | | | | Update stale comments regarding locking order and add a little more detail so it's easier to follow the locking between the cgroup freezer and the power management freezer code. Signed-off-by: Matt Helsley <matthltc@us.ibm.com> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / platform_bus: Allow runtime PM by defaultMark Brown2010-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the default runtime PM callbacks for platform devices return -ENOSYS, preventing the use of runtime PM platforms until they have provided at least a default implementation. This hinders the use of runtime PM by devices which work with many platforms such as memory mapped devices, MFDs and on chip IPs shared by multiple architectures. Change the default implementation to the standard pm_generic_runtime one, allowing drivers to use runtime PM without per-architecture changes. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Greg Kroah-Hartman <gregkh@suse.de> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * i2c: Fix bus-level power management callbacksRafael J. Wysocki2010-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are three issues with the i2c bus type's power management callbacks at the moment. First, they don't include any hibernate callbacks, although they should at least include the .restore() callback (there's no guarantee that the driver will be present in memory before loading the image kernel and we must restore the pre-hibernation state of the device). Second, the "legacy" callbacks are not going to be invoked by the PM core since the bus type's pm object is not NULL. Finally, the system sleep PM (ie. suspend/resume) callbacks don't check if the device has been already suspended at run time, in which case they should skip suspending it. Also, it looks like the i2c bus type can use the generic subsystem-level runtime PM callbacks. For these reasons, rework the system sleep PM callbacks provided by the i2c bus type to handle hibernation correctly and to invoke the "legacy" callbacks for drivers that provide them. In addition to that make the i2c bus type use the generic subsystem-level runtime PM callbacks. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jean Delvare <khali@linux-fr.org>
| * PM QOS updateMark Gross2010-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch changes the string based list management to a handle base implementation to help with the hot path use of pm-qos, it also renames much of the API to use "request" as opposed to "requirement" that was used in the initial implementation. I did this because request more accurately represents what it actually does. Also, I added a string based ABI for users wanting to use a string interface. So if the user writes 0xDDDDDDDD formatted hex it will be accepted by the interface. (someone asked me for it and I don't think it hurts anything.) This patch updates some documentation input I got from Randy. Signed-off-by: markgross <mgross@linux.intel.com> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / Hibernate: Fix block_io.c printk warningRandy Dunlap2010-05-10
| | | | | | | | | | | | | | | | | | Fix printk format warning in block_io.c: kernel/power/block_io.c:41: warning: format '%ld' expects type 'long int', but argument 2 has type 'sector_t' Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / Hibernate: Group swap opsJiri Slaby2010-05-10
| | | | | | | | | | | | | | | | Move all the swap processing into one function. It will make swap calls from a non-swap code easier. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / Hibernate: Move the first_sector out of swsusp_writeJiri Slaby2010-05-10
| | | | | | | | | | | | | | | | | | | | The first sector knowledge is swap-only specific. Move it into the swap handle. This will be needed for later non-swap specific code moving into snapshot.c. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Acked-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: "Rafael J. Wysocki" <rjw@sisk.pl>
| * PM / Hibernate: Separate block_ioJiri Slaby2010-05-10
| | | | | | | | | | | | | | | | Move block I/O operations to a separate file. It is because it will be used later not only by the swap writer. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / Hibernate: Snapshot cleanupJiri Slaby2010-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | Remove support of reads with offset. This means snapshot_read/write_next now does not accept count parameter. It allows to clean up the functions and snapshot handle which no longer needs to care about offsets. /dev/snapshot handler is converted to simple_{read_from,write_to}_buffer which take care of offsets. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Acked-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * FS / libfs: Implement simple_write_to_bufferJiri Slaby2010-05-10
| | | | | | | | | | | | | | | | | | | | It will be used in suspend code and serves as an easy wrap around copy_from_user. Similar to simple_read_from_buffer, it takes care of transfers with proper lengths depending on available and count parameters and advances ppos appropriately. Signed-off-by: Jiri Slaby <jslaby@suse.cz> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / Hibernate: document open(/dev/snapshot) side effectsPavel Machek2010-05-10
| | | | | | | | | | | | | | | | | | | | Opening and closing /dev/snapshot causes the kernel to carry out some hibernate preparations that should be documented. [rjw: Added the changelog.] Signed-off-by: Pavel Machek <pavel@ucw.cz> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM / Runtime: Add sysfs debug filesDominik Brodowski2010-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a few sysfs files relating to runtime power management for advanced debug purposes: runtime_enabled: is runtime PM enabled for this device? States are "enabled", "disabled", "forbidden" or a combination of the latter two. runtime_status: what state is the device in currently? E.g., it reports "suspended" for runtime-suspended devices, and "active" for active devices. NOTE: if runtime_enabled returns "disabled", the value of this file may not reflect its physical state. runtime_usage: the runtime PM usage count of a device runtime_active_kids: the runtime PM children usage count of a device, or 0 if the ignore_children flag is set. Also, CONFIG_PM_SLEEP_ADVANCED_DEBUG is not defined in any Kconfig file, so replace it with CONFIG_PM_ADVANCED_DEBUG. Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM: Improve device power management documentAlan Stern2010-05-10
| | | | | | | | | | | | | | | | Improve the device power management document after it's been updated by the previous patch. Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM: Update device power management documentRafael J. Wysocki2010-05-10
| | | | | | | | | | | | | | | | | | The device PM document, Documentation/power/devices.txt, is badly outdated and requires total rework to fit the current design of the PM framework. Make it more up to date. Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl> Reviewed-by: Randy Dunlap <randy.dunlap@oracle.com>
| * PM: Allow runtime_suspend methods to call pm_schedule_suspend()Alan Stern2010-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch (as1361) changes the runtime PM interface slightly; it allows suspend requests to be scheduled while the runtime_suspend method is running. If the method succeeds then the scheduled request is cancelled, whereas if the method fails then an idle notification is sent only if no request was scheduled. Being able to schedule suspend requests from within a runtime_suspend method is useful for drivers that need to test for idleness and suspend the device all while holding a single spinlock, or for drivers that want to check for idleness by polling. Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
| * PM: pm_wakeup - switch to using boolDmitry Torokhov2010-05-10
| | | | | | | | | | | | | | | | | | Also change couple of stubs implemented as macros in !CONFIG_PM case in statinc inline functions to provide proper typechecking of arguments regardless of config. Signed-off-by: Dmitry Torokhov <dtor@mail.ru> Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
* | Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wqLinus Torvalds2010-05-20
|\ \ | | | | | | | | | | | | | | | * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq: workqueue: change cancel_work_sync() to clear work->data workqueue: warn about flush_scheduled_work()
| * | workqueue: change cancel_work_sync() to clear work->dataOleg Nesterov2010-04-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In short: change cancel_work_sync(work) to mark this work as "never queued" upon return. When cancel_work_sync(work) succeeds, we know that this work can't be queued or running, and since we own WORK_STRUCT_PENDING nobody can change the bits in work->data under us. This means we can also clear the "cwq" part along with _PENDING bit lockless before return, unless the work is queued nobody can assume get_wq_data() is stable even under cwq->lock. This change can speedup the subsequent cancel/flush requests, and as Dmitry pointed out this simplifies the usage of work_struct's which can be queued on different workqueues. Consider this pseudo code from the input subsystem: struct workqueue_struct *WQ; struct work_struct *WORK; for (;;) { WQ = create_workqueue(); ... if (condition()) queue_work(WQ, WORK); ... cancel_work_sync(WORK); destroy_workqueue(WQ); } If condition() returns T and then F, cancel_work_sync() will crash the kernel because WORK->data still points to the already destroyed workqueue. With this patch the code like above becomes correct. Suggested-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Oleg Nesterov <oleg@redhat.com> Signed-off-by: Tejun Heo <tj@kernel.org>
| * | workqueue: warn about flush_scheduled_work()Alan Stern2010-04-30
| | | | | | | | | | | | | | | | | | | | | | | | This patch (as1319) adds kerneldoc and a pointed warning to flush_scheduled_work(). Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Tejun Heo <tj@kernel.org>
* | | Merge branch 'for-linus' of ↵Linus Torvalds2010-05-20
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu: ia64: add sparse annotation to __ia64_per_cpu_var() percpu: implement kernel memory based chunk allocation percpu: move vmalloc based chunk management into percpu-vm.c percpu: misc preparations for nommu support percpu: reorganize chunk creation and destruction percpu: factor out pcpu_addr_in_first/reserved_chunk() and update per_cpu_ptr_to_phys()
| * | | ia64: add sparse annotation to __ia64_per_cpu_var()Tejun Heo2010-05-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ia64 has its own optimized percpu accessor - __ia64_per_cpu_var(). Add percpu sparse annotations to it. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: Christoph Lameter <cl@linux-foundation.org>
| * | | percpu: implement kernel memory based chunk allocationTejun Heo2010-05-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Implement an alternate percpu chunk management based on kernel memeory for nommu SMP architectures. Instead of mapping into vmalloc area, chunks are allocated as a contiguous kernel memory using alloc_pages(). As such, percpu allocator on nommu will have the following restrictions. * It can't fill chunks on-demand page-by-page. It has to allocate each chunk fully upfront. * It can't support sparse chunk for NUMA configurations. SMP w/o mmu is crazy enough. Let's hope no one does NUMA w/o mmu. :-P * If chunk size isn't power-of-two multiple of PAGE_SIZE, the unaligned amount will be wasted on each chunk. So, archs which use this better align chunk size. For instructions on how to use this, read the comment on top of mm/percpu-km.c. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: David Howells <dhowells@redhat.com> Cc: Graff Yang <graff.yang@gmail.com> Cc: Sonic Zhang <sonic.adi@gmail.com>
| * | | percpu: move vmalloc based chunk management into percpu-vm.cTejun Heo2010-05-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Separate out and move chunk management (creation/desctruction and [de]population) code into percpu-vm.c which is included by percpu.c and compiled together. The interface for chunk management is defined as follows. * pcpu_populate_chunk - populate the specified range of a chunk * pcpu_depopulate_chunk - depopulate the specified range of a chunk * pcpu_create_chunk - create a new chunk * pcpu_destroy_chunk - destroy a chunk, always preceded by full depop * pcpu_addr_to_page - translate address to physical address * pcpu_verify_alloc_info - check alloc_info is acceptable during init Other than wrapping vmalloc_to_page() inside pcpu_addr_to_page() and dummy pcpu_verify_alloc_info() implementation, this patch only moves code around. This separation is to allow alternate chunk management implementation. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: David Howells <dhowells@redhat.com> Cc: Graff Yang <graff.yang@gmail.com> Cc: Sonic Zhang <sonic.adi@gmail.com>
| * | | percpu: misc preparations for nommu supportTejun Heo2010-05-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make the following misc preparations for percpu nommu support. * Remove refernces to vmalloc in common comments as nommu percpu won't use it. * Rename chunk->vms to chunk->data and make it void *. Its use is determined by chunk management implementation. * Relocate utility functions and add __maybe_unused to functions which might not be used by different chunk management implementations. This patch doesn't cause any functional change. This is to allow alternate chunk management implementation for percpu nommu support. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: David Howells <dhowells@redhat.com> Cc: Graff Yang <graff.yang@gmail.com> Cc: Sonic Zhang <sonic.adi@gmail.com>
| * | | percpu: reorganize chunk creation and destructionTejun Heo2010-05-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Reorganize alloc/free_pcpu_chunk() such that chunk struct alloc/free live in pcpu_alloc/free_chunk() and the rest in pcpu_create/destroy_chunk(). While at it, add missing error handling for chunk->map allocation failure. This is to allow alternate chunk management implementation for percpu nommu support. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: David Howells <dhowells@redhat.com> Cc: Graff Yang <graff.yang@gmail.com> Cc: Sonic Zhang <sonic.adi@gmail.com>
| * | | percpu: factor out pcpu_addr_in_first/reserved_chunk() and update ↵Tejun Heo2010-05-01
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | per_cpu_ptr_to_phys() Factor out pcpu_addr_in_first/reserved_chunk() from pcpu_chunk_addr_search() and use it to update per_cpu_ptr_to_phys() such that it handles first chunk differently from the rest. This patch doesn't cause any functional change and is to prepare for percpu nommu support. Signed-off-by: Tejun Heo <tj@kernel.org> Reviewed-by: David Howells <dhowells@redhat.com> Cc: Graff Yang <graff.yang@gmail.com> Cc: Sonic Zhang <sonic.adi@gmail.com>
* | | | Merge branch 'for-linus' of ↵Linus Torvalds2010-05-20
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: IB/core: Use kmemdup() instead of kmalloc()+memcpy() IB/iser: Fix error flow in iser_create_ib_conn_res() IB/iser: Enhance disconnection logic for multi-pathing IB/iser: Remove buggy back-pointer setting IB/iser: Add asynchronous event handler MAINTAINERS: Add cxgb4 and iw_cxgb4 entries RDMA/cxgb3: Shrink .text with compile-time init of handlers arrays IPoIB: Allow disabling/enabling TSO on the fly through ethtool IB/mlx4: Add support for masked atomic operations IB/core: Add support for masked atomic operations RDMA/cma: Randomize local port allocation RDMA/nes: Make unnecessarily global functions static RDMA/nes: Make nesadapter->phy_lock usage consistent RDMA/cxgb4: Add driver for Chelsio T4 RNIC IB/mthca: Use the dma state API instead of pci equivalents RDMA/amso1100: Use the dma state API instead of pci equivalents RDMA/cxgb3: Don't free skbs on NET_XMIT_* indications from LLD RDMA/cxgb3: Use the dma state API instead of pci equivalents IB: Explicitly rule out llseek to avoid BKL in default_llseek()
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| | \ \ \
| *-----------------. \ \ \ Merge branches 'amso1100', 'bkl', 'cma', 'cxgb3', 'cxgb4', 'ipoib', 'iser', ↵Roland Dreier2010-05-15
| |\ \ \ \ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 'masked-atomics', 'misc', 'mthca' and 'nes' into for-next
| | | | | | | | | | | * | | | RDMA/nes: Make unnecessarily global functions staticRoland Dreier2010-04-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This allows the compiler to do a bit better; on my x86-64 build: add/remove: 0/2 grow/shrink: 1/0 up/down: 2288/-2365 (-77) function old new delta nes_init_phy 273 2561 +2288 nes_init_1g_phy 469 - -469 nes_init_2025_phy 1896 - -1896 Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | | | | | * | | | RDMA/nes: Make nesadapter->phy_lock usage consistentChien Tung2010-04-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | nes_{read,write}_1G_phy_reg() are using phy_lock while nes_{read,write}_10G_phy_reg() leave that to the caller. Remove phy_lock from 1G routines and leave the locking to the caller. Add additional phy_lock calls around 1G read/write. Signed-off-by: Chien Tung <chien.tin.tung@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | | | | * | | | | IB/mthca: Use the dma state API instead of pci equivalentsFUJITA Tomonori2010-04-21
| | | | | | | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The DMA API is preferred; no functional change. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | | | * / / / / IB/core: Use kmemdup() instead of kmalloc()+memcpy()Julia Lawall2010-05-15
| | | | | | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use kmemdup when some other buffer is immediately copied into the allocated region. A simplified version of the semantic patch that makes this change is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression from,to,size,flag; statement S; @@ - to = \(kmalloc\|kzalloc\)(size,flag); + to = kmemdup(from,size,flag); if (to==NULL || ...) S - memcpy(to, from, size); // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | | * | | | | IB/mlx4: Add support for masked atomic operationsVladimir Sokolovsky2010-04-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for masked atomic operations (masked compare and swap, masked fetch and add). Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | | * | | | | IB/core: Add support for masked atomic operationsVladimir Sokolovsky2010-04-21
| | | | | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Add new IB_WR_MASKED_ATOMIC_CMP_AND_SWP and IB_WR_MASKED_ATOMIC_FETCH_AND_ADD send opcodes that can be used to post "masked atomic compare and swap" and "masked atomic fetch and add" work request respectively. - Add masked_atomic_cap capability. - Add mask fields to atomic struct of ib_send_wr - Add new opcodes to ib_wc_opcode The new operations are described more precisely below: * Masked Compare and Swap (MskCmpSwap) The MskCmpSwap atomic operation is an extension to the CmpSwap operation defined in the IB spec. MskCmpSwap allows the user to select a portion of the 64 bit target data for the “compare” check as well as to restrict the swap to a (possibly different) portion. The pseudo code below describes the operation: | atomic_response = *va | if (!((compare_add ^ *va) & compare_add_mask)) then | *va = (*va & ~(swap_mask)) | (swap & swap_mask) | | return atomic_response The additional operands are carried in the Extended Transport Header. Atomic response generation and packet format for MskCmpSwap is as for standard IB Atomic operations. * Masked Fetch and Add (MFetchAdd) The MFetchAdd Atomic operation extends the functionality of the standard IB FetchAdd by allowing the user to split the target into multiple fields of selectable length. The atomic add is done independently on each one of this fields. A bit set in the field_boundary parameter specifies the field boundaries. The pseudo code below describes the operation: | bit_adder(ci, b1, b2, *co) | { | value = ci + b1 + b2 | *co = !!(value & 2) | | return value & 1 | } | | #define MASK_IS_SET(mask, attr) (!!((mask)&(attr))) | bit_position = 1 | carry = 0 | atomic_response = 0 | | for i = 0 to 63 | { | if ( i != 0 ) | bit_position = bit_position << 1 | | bit_add_res = bit_adder(carry, MASK_IS_SET(*va, bit_position), | MASK_IS_SET(compare_add, bit_position), &new_carry) | if (bit_add_res) | atomic_response |= bit_position | | carry = ((new_carry) && (!MASK_IS_SET(compare_add_mask, bit_position))) | } | | return atomic_response Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | * | | | | IB/iser: Fix error flow in iser_create_ib_conn_res()Dan Carpenter2010-05-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We shouldn't free things here because we free them later. The call tree looks like this: iser_connect() ==> initiating the connection establishment and later iser_cma_handler() => iser_route_handler() => iser_create_ib_conn_res() if we fail here, eventually iser_conn_release() is called, resulting in a double free. Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | * | | | | IB/iser: Enhance disconnection logic for multi-pathingOr Gerlitz2010-05-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The iser connection teardown flow isn't over until the underlying Connection Manager (e.g the IB CM) delivers a disconnected or timeout event through the RDMA-CM. When the remote (target) side isn't reachable, e.g when some HW e.g port/hca/switch isn't functioning or taken down administratively, the CM timeout flow is used and the event may be generated only after relatively long time -- on the order of tens of seconds. The current iser code exposes this possibly long delay to higher layers, specifically to the iscsid daemon and iscsi kernel stack. As a result, the iscsi stack doesn't respond well: this low-level CM delay is added to the fail-over time under HA schemes such as the one provided by DM multipath through the multipathd(8) service. This patch enhances the reference counting scheme on iser's IB connections so that the disconnect flow initiated by iscsid from user space (ep_disconnect) doesn't wait for the CM to deliver the disconnect/timeout event. (The connection teardown isn't done from iser's view point until the event is delivered) The iser ib (rdma) connection object is destroyed when its reference count reaches zero. When this happens on the RDMA-CM callback context, extra care is taken so that the RDMA-CM does the actual destroying of the associated ID, since doing it in the callback is prohibited. The reference count of iser ib connection normally reaches three, where the <ref, deref> relations are 1. conn <init, terminate> 2. conn <bind, stop/destroy> 3. cma id <create, disconnect/error/timeout callbacks> With this patch, multipath fail-over time is about 30 seconds, while without this patch, multipath fail-over time is about 130 seconds. Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | * | | | | IB/iser: Remove buggy back-pointer settingOr Gerlitz2010-05-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The iscsi connection object life cycle includes binding and unbinding (conn_stop) to/from the iscsi transport connection object. Since iscsi connection objects are recycled, at the time the transport connection (e.g iser's IB connection) is released, it is not valid to touch the iscsi connection tied to the transport back-pointer since it may already point to a different transport connection. Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | | * | | | | IB/iser: Add asynchronous event handlerOr Gerlitz2010-05-12
| | | | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add handler to handle events such as port up and down. This is useful when testing high-availability schemes such as multi-pathing. Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | | * / / / / IPoIB: Allow disabling/enabling TSO on the fly through ethtoolOr Gerlitz2010-04-21
| | | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | * | | | | MAINTAINERS: Add cxgb4 and iw_cxgb4 entriesRoland Dreier2010-05-05
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | | * | | | | RDMA/cxgb4: Add driver for Chelsio T4 RNICSteve Wise2010-04-21
| | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add an RDMA/iWARP driver for Chelsio T4 Ethernet adapters. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | * | | | | RDMA/cxgb3: Shrink .text with compile-time init of handlers arraysRoland Dreier2010-04-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Using compile-time designated initializers for the handler arrays instead of open-coding the initialization in iwch_cm_init() is (IMHO) cleaner, and leads to substantially smaller code: on my x86-64 build, bloat-o-meter shows: add/remove: 0/1 grow/shrink: 4/3 up/down: 4/-1682 (-1678) function old new delta tx_ack 167 168 +1 state_set 55 56 +1 start_ep_timer 99 100 +1 pass_establish 177 178 +1 act_open_req_arp_failure 39 38 -1 sched 84 82 -2 iwch_cm_init 442 91 -351 work_handlers 1328 - -1328 Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | * | | | | RDMA/cxgb3: Don't free skbs on NET_XMIT_* indications from LLDSteve Wise2010-04-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The low level cxgb3 driver can return NET_XMIT_CN and friends. The iw_cxgb3 driver should _not_ treat these as errors. Signed-off-by: Steve Wise <swise@opengridcomputing.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | | * | | | | RDMA/cxgb3: Use the dma state API instead of pci equivalentsFUJITA Tomonori2010-04-21
| | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The DMA API is preferred; no functional change. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | | * / / / / RDMA/cma: Randomize local port allocationTetsuo Handa2010-04-21
| | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Randomize local port allocation in the way sctp_get_port_local() does. Update rover at the end of loop since we're likely to pick a valid port on the first try. Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Reviewed-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <rolandd@cisco.com>
| | * / / / / IB: Explicitly rule out llseek to avoid BKL in default_llseek()Roland Dreier2010-04-21
| | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Several RDMA user-access drivers have file_operations structures with no .llseek method set. None of the drivers actually do anything with f_pos, so this means llseek is essentially a NOP, instead of returning an error as leaving other file_operations methods unimplemented would do. This is mostly harmless, except that a NULL .llseek means that default_llseek() is used, and this function grabs the BKL, which we would like to avoid. Since llseek does nothing useful on these files, we would like it to return an error to userspace instead of silently grabbing the BKL and succeeding. For nearly all of the file types, we take the belt-and-suspenders approach of setting the .llseek method to no_llseek and also calling nonseekable_open(); the exception is the uverbs_event files, which are created with anon_inode_getfile(), which already sets f_mode the same way as nonseekable_open() would. This work is motivated by Arnd Bergmann's bkl-removal tree. Signed-off-by: Roland Dreier <rolandd@cisco.com>
| * / / / / RDMA/amso1100: Use the dma state API instead of pci equivalentsFUJITA Tomonori2010-04-21
| |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The DMA API is preferred; no functional change. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Roland Dreier <rolandd@cisco.com>
* | | | | Merge branch 'for-linus' of ↵Linus Torvalds2010-05-20
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6 * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6: (61 commits) KEYS: Return more accurate error codes LSM: Add __init to fixup function. TOMOYO: Add pathname grouping support. ima: remove ACPI dependency TPM: ACPI/PNP dependency removal security/selinux/ss: Use kstrdup TOMOYO: Use stack memory for pending entry. Revert "ima: remove ACPI dependency" Revert "TPM: ACPI/PNP dependency removal" KEYS: Do preallocation for __key_link() TOMOYO: Use mutex_lock_interruptible. KEYS: Better handling of errors from construct_alloc_key() KEYS: keyring_serialise_link_sem is only needed for keyring->keyring links TOMOYO: Use GFP_NOFS rather than GFP_KERNEL. ima: remove ACPI dependency TPM: ACPI/PNP dependency removal selinux: generalize disabling of execmem for plt-in-heap archs LSM Audit: rename LSM_AUDIT_NO_AUDIT to LSM_AUDIT_DATA_NONE CRED: Holding a spinlock does not imply the holding of RCU read lock SMACK: Don't #include Ext2 headers ...
generated by cgit v1.2.2 (git 2.25.0) at 2024-12-28 00:34:15 -0500