Glauber Costa [Tue, 31 Jul 2012 23:43:07 +0000 (16:43 -0700)]
memcg: fix bad behavior in use_hierarchy file
I have an application that does the following:
* copy the state of all controllers attached to a hierarchy
* replicate it as a child of the current level.
I would expect writes to the files to mostly succeed, since they are
inheriting sane values from parents.
But that is not the case for use_hierarchy. If it is set to 0, we succeed
ok. If we're set to 1, the value of the file is automatically set to 1 in
the children, but if userspace tries to write the very same 1, it will
fail. That same situation happens if we set use_hierarchy, create a
child, and then try to write 1 again.
Now, there is no reason whatsoever for failing to write a value that is
already there. It doesn't even match the comments, that states:
/* If parent's use_hierarchy is set, we can't make any modifications
* in the child subtrees...
since we are not changing anything.
So test the new value against the one we're storing, and automatically
return 0 if we're not proposing a change.
Minchan Kim [Tue, 31 Jul 2012 23:42:59 +0000 (16:42 -0700)]
mm: do not use page_count() without a page pin
d179e84ba ("mm: vmscan: do not use page_count without a page pin") fixed
this problem in vmscan.c but same problem is in __count_immobile_pages().
I copy and paste d179e84ba's contents for description.
"It is unsafe to run page_count during the physical pfn scan because
compound_head could trip on a dangling pointer when reading
page->first_page if the compound page is being freed by another CPU."
David Rientjes [Tue, 31 Jul 2012 23:42:56 +0000 (16:42 -0700)]
mm, oom: replace some information in tasklist dump
The number of ptes and swap entries are used in the oom killer's badness
heuristic, so they should be shown in the tasklist dump.
This patch adds those fields and replaces cpu and oom_adj values that are
currently emitted. Cpu isn't interesting and oom_adj is deprecated and
will be removed later this year, the same information is already displayed
as oom_score_adj which is used internally.
At the same time, make the documentation a little more clear to state this
information is helpful to determine why the oom killer chose the task it
did to kill.
David Rientjes [Tue, 31 Jul 2012 23:42:55 +0000 (16:42 -0700)]
mm, oom: fix potential killing of thread that is disabled from oom killing
/proc/sys/vm/oom_kill_allocating_task will immediately kill current when
the oom killer is called to avoid a potentially expensive tasklist scan
for large systems.
Currently, however, it is not checking current's oom_score_adj value which
may be OOM_SCORE_ADJ_MIN, meaning that it has been disabled from oom
killing.
This patch avoids killing current in such a condition and simply falls
back to the tasklist scan since memory still needs to be freed.
mm: clear pages_scanned only if draining a pcp adds pages to the buddy allocator again
commit 2ff754fa8f ("mm: clear pages_scanned only if draining a pcp adds
pages to the buddy allocator again") fixed one free_pcppages_bulk()
misuse. But two another miuse still exist.
mm, fadvise: don't return -EINVAL when filesystem cannot implement fadvise()
Eric Wong reported his test suite failex when /tmp is tmpfs.
https://lkml.org/lkml/2012/2/24/479
Currentlt the input check of POSIX_FADV_WILLNEED has two problems.
- requires a_ops->readpage. But in fact, force_page_cache_readahead()
requires that the target filesystem has either ->readpage or ->readpages.
- returns -EINVAL when the filesystem doesn't have ->readpage. But
posix says that fadvise is merely a hint. Thus fadvise() should return
0 if filesystem has no means of implementing fadvise(). The userland
application should not know nor care whcih type of filesystem backs the
TMPDIR directory, as Eric pointed out. There is nothing which userspace
can do to solve this error.
So change the return value to 0 when filesytem doesn't support readahead.
memcg: make mem_cgroup_force_empty_list() return bool
mem_cgroup_force_empty_list() just returns 0 or -EBUSY and -EBUSY
indicates 'you need to retry'. Make mem_cgroup_force_empty_list() return
a bool to simplify the logic.
memcg: clean up force_empty_list() return value check
After bf544fdc241da8 "memcg: move charges to root cgroup if
use_hierarchy=0 in mem_cgroup_move_hugetlb_parent()"
mem_cgroup_move_parent() returns only -EBUSY or -EINVAL. So we can remove
the -ENOMEM and -EINTR checks.
memcg: remove check for signal_pending() during rmdir()
After bf544fdc241da8 "memcg: move charges to root cgroup if
use_hierarchy=0 in mem_cgroup_move_hugetlb_parent()", no memory reclaim
will occur when removing a memory cgroup. If -EINTR is returned here,
cgroup will show a warning.
We don't need to handle any user interruption signal. Remove this.
David Rientjes [Tue, 31 Jul 2012 23:42:37 +0000 (16:42 -0700)]
mm, oom: do not schedule if current has been killed
The oom killer currently schedules away from current in an uninterruptible
sleep if it does not have access to memory reserves. It's possible that
current was killed because it shares memory with the oom killed thread or
because it was killed by the user in the interim, however.
This patch only schedules away from current if it does not have a pending
kill, i.e. if it does not share memory with the oom killed thread. It's
possible that it will immediately retry its memory allocation and fail,
but it will immediately be given access to memory reserves if it calls the
oom killer again.
This prevents the delay of memory freeing when threads that share memory
with the oom killed thread get unnecessarily scheduled.
hugetlb/cgroup: remove exclude and wakeup rmdir calls from migrate
We already hold the hugetlb_lock. That should prevent a parallel cgroup
rmdir from touching page's hugetlb cgroup. So remove the exclude and
wakeup calls.
hugetlb/cgroup: assign the page hugetlb cgroup when we move the page to active list.
A page's hugetlb cgroup assignment and movement to the active list should
occur with hugetlb_lock held. Otherwise when we remove the hugetlb cgroup
we will iterate the active list and find pages with NULL hugetlb cgroup
values.
When we fail to allocate pages from the reserve pool, hugetlb tries to
allocate huge pages using alloc_buddy_huge_page. Add these to the active
list. We also need to add the huge page we allocate when we soft offline
the oldpage to active list.
hugetlb/cgroup: migrate hugetlb cgroup info from oldpage to new page during migration
With HugeTLB pages, hugetlb cgroup is uncharged in compound page
destructor. Since we are holding a hugepage reference, we can be sure
that old page won't get uncharged till the last put_page().
hugetlb/cgroup: add charge/uncharge routines for hugetlb cgroup
Add the charge and uncharge routines for hugetlb cgroup. We do cgroup
charging in page alloc and uncharge in compound page destructor.
Assigning page's hugetlb cgroup is protected by hugetlb_lock.
hugetlb/cgroup: add the cgroup pointer to page lru
Add the hugetlb cgroup pointer to 3rd page lru.next. This limit the usage
to hugetlb cgroup to only hugepages with 3 or more normal pages. I guess
that is an acceptable limitation.
Implement a new controller that allows us to control HugeTLB allocations.
The extension allows to limit the HugeTLB usage per control group and
enforces the controller limit during page fault. Since HugeTLB doesn't
support page reclaim, enforcing the limit at page fault time implies that,
the application will get SIGBUS signal if it tries to access HugeTLB pages
beyond its limit. This requires the application to know beforehand how
much HugeTLB pages it would require for its use.
The charge/uncharge calls will be added to HugeTLB code in later patch.
Support for cgroup removal will be added in later patches.
hugetlb: add a list for tracking in-use HugeTLB pages
hugepage_activelist will be used to track currently used HugeTLB pages.
We need to find the in-use HugeTLB pages to support HugeTLB cgroup removal.
On cgroup removal we update the page's HugeTLB cgroup to point to parent
cgroup.
Since we migrate only one hugepage, don't use linked list for passing the
page around. Directly pass the page that need to be migrated as argument.
This also removes the usage of page->lru in the migrate path.
The current use of VM_FAULT_* codes with ERR_PTR requires us to ensure
VM_FAULT_* values will not exceed MAX_ERRNO value. Decouple the
VM_FAULT_* values from MAX_ERRNO.
This patchset implements a cgroup resource controller for HugeTLB pages.
The controller allows to limit the HugeTLB usage per control group and
enforces the controller limit during page fault. Since HugeTLB doesn't
support page reclaim, enforcing the limit at page fault time implies that,
the application will get SIGBUS signal if it tries to access HugeTLB pages
beyond its limit. This requires the application to know beforehand how
much HugeTLB pages it would require for its use.
The goal is to control how many HugeTLB pages a group of task can
allocate. It can be looked at as an extension of the existing quota
interface which limits the number of HugeTLB pages per hugetlbfs
superblock. HPC job scheduler requires jobs to specify their resource
requirements in the job file. Once their requirements can be met, job
schedulers like (SLURM) will schedule the job. We need to make sure that
the jobs won't consume more resources than requested. If they do we
should either error out or kill the application.
This patch:
Rename max_hstate to hugetlb_max_hstate. We will be using this from other
subsystems like hugetlb controller in later patches.
Wanpeng Li [Tue, 31 Jul 2012 23:41:52 +0000 (16:41 -0700)]
mm: prepare for removal of obsolete /proc/sys/vm/nr_pdflush_threads
Since per-BDI flusher threads were introduced in 2.6, the pdflush
mechanism is not used any more. But the old interface exported through
/proc/sys/vm/nr_pdflush_threads still exists and is obviously useless.
For back-compatibility, printk warning information and return 2 to notify
the users that the interface is removed.
Currently, function should_fail() has "bool" for its return value, so it's
reasonable to change the return value of function should_fail_alloc_page()
into "bool" as well.
The patch does cleanup on function should_fail_alloc_page() to have "bool"
for its return value.
vm_stat_account() accounts the shared_vm, stack_vm and reserved_vm now.
But we can also account for total_vm in the vm_stat_account() which makes
the code tidy.
Even for mprotect_fixup(), we can get the right result in the end.
documentation: update how page-cluster affects swap I/O
Fix of the documentation of /proc/sys/vm/page-cluster to match the
behavior of the code and add some comments about what the tunable will
change in that behavior.
Swap readahead works fine, but the I/O to disk is almost always done in
page size requests, despite the fact that readahead submits
1<<page-cluster pages at a time.
On older kernels the old per device plugging behavior might have captured
this and merged the requests, but currently all comes down to much more
I/Os than required.
On a single device this might not be an issue, but as soon as a server
runs on shared san resources savin I/Os not only improves swapin
throughput but also provides a lower resource utilization.
With a load running KVM in a lot of memory overcommitment (the hot memory
is 1.5 times the host memory) swapping throughput improves significantly
and the lead feels more responsive as well as achieves more throughput.
In a test setup with 16 swap disks running blocktrace on one of those disks
shows the improved merging:
Prior:
Reads Queued: 560,888, 2,243MiB Writes Queued: 226,242, 904,968KiB
Read Dispatches: 544,701, 2,243MiB Write Dispatches: 159,318, 904,968KiB
Reads Requeued: 0 Writes Requeued: 0
Reads Completed: 544,716, 2,243MiB Writes Completed: 159,321, 904,980KiB
Read Merges: 16,187, 64,748KiB Write Merges: 61,744, 246,976KiB
IO unplugs: 149,614 Timer unplugs: 2,940
I got ~10% to ~40% more throughput in my cases and at the same time much
lower cpu consumption when broken down per transferred kilobyte (the
majority of that due to saved interrupts and better cache handling). In a
shared SAN others might get an additional benefit as well, because this
now causes less protocol overhead.
Jan Kara [Tue, 31 Jul 2012 23:41:37 +0000 (16:41 -0700)]
mm: make vb_alloc() more foolproof
If someone calls vb_alloc() (or vm_map_ram() for that matter) to allocate
0 bytes (0 pages), get_order() returns BITS_PER_LONG - PAGE_CACHE_SHIFT
and interesting stuff happens. So make debugging such problems easier and
warn about 0-size allocation.
vmalloc: walk vmap_areas by sorted list instead of rb_next()
There's a walk by repeating rb_next to find a suitable hole. Could be
simply replaced by walk on the sorted vmap_area_list. More simpler and
efficient.
Mutation of the list and tree only happens in pair within
__insert_vmap_area and __free_vmap_area, under protection of
vmap_area_lock. The patch code is also under vmap_area_lock, so the list
walk is safe, and consistent with the tree walk.
Tested on SMP by repeating batch of vmalloc anf vfree for random sizes and
rounds for hours.
Andrew Morton [Tue, 31 Jul 2012 23:41:34 +0000 (16:41 -0700)]
drivers/media/video/v4l2-ioctl.c: fix build
Fix zillions of these:
drivers/media/video/v4l2-ioctl.c:1848: error: unknown field 'func' specified in initializer
drivers/media/video/v4l2-ioctl.c:1848: warning: missing braces around initializer
drivers/media/video/v4l2-ioctl.c:1848: warning: (near initialization for 'v4l2_ioctls[0].<anonymous>')
drivers/media/video/v4l2-ioctl.c:1848: warning: initialization makes integer from pointer without a cast
drivers/media/video/v4l2-ioctl.c:1848: error: initializer element is not computable at load time
drivers/media/video/v4l2-ioctl.c:1848: error: (near initialization for 'v4l2_ioctls[0].<anonymous>.offset')
block/blk-cgroup.c:609:2: error: unknown type name 'atomic64_t'
block/blk-cgroup.c:609:2: error: implicit declaration of function 'ATOMIC64_INIT' [-Werror=implicit-function-declaration]
Akinobu Mita [Tue, 31 Jul 2012 23:41:31 +0000 (16:41 -0700)]
fault-injection: fix failcmd.sh warning
"fault-injection: add tool to run command with failslab or
fail_page_alloc" added tools/testing/fault-injection/failcmd.sh to make it
easier to inject slab/page allocation failures by fault injection.
failcmd.sh prints the following warning when running with arguments
for command.
Linus Torvalds [Wed, 1 Aug 2012 01:08:25 +0000 (18:08 -0700)]
Merge tag 'for-v3.6' of git://git.infradead.org/battery-2.6
Pull battery updates from Anton Vorontsov:
"The tag contains just a few battery-related changes for v3.6. It's is
all pretty straightforward, except one thing.
One of our patches added thermal support for power supply class, but
thermal/ subsystem changed under our feet. We (well, Stephen, that
is) caught the issue and it was decided[1] that I'd just delay the
battery pull request, and then will fix it up by merging upstream back
into battery tree at the specific commit.
That's not all though: another[2] small fixup for thermal subsystem
was needed to get rid of a warning in power supply subsystem (the
warning was not drivers/power's "fault", the thermal registration
function just needed a proper const annotation, which is also done by
a small commit on top of the merge.
So, to sum this up:
- The 'master' branch of the battery tree was in the -next tree for
weeks, was never rebased, altered etc. It should be all OK;
- Although, for-v3.6 tag contains the 'master' branch + merge + the
warning fix.
* tag 'for-v3.6' of git://git.infradead.org/battery-2.6: (23 commits)
thermal: Constify 'type' argument for the registration routine
olpc-battery: update CHARGE_FULL_DESIGN property for BYD LiFe batteries
olpc-battery: Add VOLTAGE_MAX_DESIGN property
charger-manager: Fix build break related to EXTCON
lp8727_charger: Move header file into platform_data directory
power_supply: Add min/max alert properties for CAPACITY, TEMP, TEMP_AMBIENT
bq27x00_battery: Add support for BQ27425 chip
charger-manager: Set current limit of regulator for over current protection
charger-manager: Use EXTCON Subsystem to detect charger cables for charging
test_power: Add VOLTAGE_NOW and BATTERY_TEMP properties
test_power: Add support for USB AC source
gpio-charger: Use cansleep version of gpio_set_value
bq27x00_battery: Add support for power average and health properties
sbs-battery: Don't trigger false supply_changed event
twl4030_charger: Allow charger to control the regulator that feeds it
twl4030_charger: Add backup-battery charging
twl4030_charger: Fix some typos
max17042_battery: Support CHARGE_COUNTER power supply attribute
smb347-charger: Add constant charge and current properties
power_supply: Add constant charge_current and charge_voltage properties
...
Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf updates from Ingo Molnar:
"The biggest changes are Intel Nehalem-EX PMU uncore support, uprobes
updates/cleanups/fixes from Oleg and diverse tooling updates (mostly
fixes) now that Arnaldo is back from vacation."
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (40 commits)
uprobes: __replace_page() needs munlock_vma_page()
uprobes: Rename vma_address() and make it return "unsigned long"
uprobes: Fix register_for_each_vma()->vma_address() check
uprobes: Introduce vaddr_to_offset(vma, vaddr)
uprobes: Teach build_probe_list() to consider the range
uprobes: Remove insert_vm_struct()->uprobe_mmap()
uprobes: Remove copy_vma()->uprobe_mmap()
uprobes: Fix overflow in vma_address()/find_active_uprobe()
uprobes: Suppress uprobe_munmap() from mmput()
uprobes: Uprobe_mmap/munmap needs list_for_each_entry_safe()
uprobes: Clean up and document write_opcode()->lock_page(old_page)
uprobes: Kill write_opcode()->lock_page(new_page)
uprobes: __replace_page() should not use page_address_in_vma()
uprobes: Don't recheck vma/f_mapping in write_opcode()
perf/x86: Fix missing struct before structure name
perf/x86: Fix format definition of SNB-EP uncore QPI box
perf/x86: Make bitfield unsigned
perf/x86: Fix LLC-* and node-* events on Intel SandyBridge
perf/x86: Add Intel Nehalem-EX uncore support
perf/x86: Fix typo in format definition of uncore PCU filter
...
Merge branch 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc
Pull powerpc updates from Benjamin Herrenschmidt:
"Kumar sent me a handful of Freescale related fixes and I added another
regression fix to the pile.
PS. I -will- eventually learn about that signed tag business :-)"
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
powerpc/kvm/book3s_32: Fix MTMSR_EERI macro
powerpc/85xx: p1022ds: fix DIU/LBC switching with NAND enabled
powerpc/85xx: p1022ds: disable the NAND flash node if video is enabled
powerpc/85xx: Fix sram_offset parameter type
powerpc/85xx: P3041DS - change espi input-clock from 40MHz to 35MHz
powerpc/85xx: Fix pci base address error for p2020rdb-pc in dts
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Martin Schwidefsky:
"This it the second batch of s390 patches for the 3.6 merge window.
Included is enablement for two common code changes, killable page
faults and sorted exception tables. And the regular set of cleanup
and bug fix patches."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
s390: make use of user_mode() macro where possible
s390/mm: rename user_mode variable to addressing_mode
s390/mm: fix fault handling for page table walk case
s390/mm: make page faults killable
s390: update defconfig
s390/mm: downgrade page table after fork of a 31 bit process
s390/ipl: Use diagnose 8 command separation
s390/linker script: use RO_DATA_SECTION
s390/exceptions: sort exception table at build time
s390/debug: remove module_exit function / move EXPORT_SYMBOLs
Merge branch 'nfsd-next' of git://linux-nfs.org/~bfields/linux
Pull nfsd changes from J. Bruce Fields:
"This has been an unusually quiet cycle--mostly bugfixes and cleanup.
The one large piece is Stanislav's work to containerize the server's
grace period--but that in itself is just one more step in a
not-yet-complete project to allow fully containerized nfs service.
There are a number of outstanding delegation, container, v4 state, and
gss patches that aren't quite ready yet; 3.7 may be wilder."
* 'nfsd-next' of git://linux-nfs.org/~bfields/linux: (35 commits)
NFSd: make boot_time variable per network namespace
NFSd: make grace end flag per network namespace
Lockd: move grace period management from lockd() to per-net functions
LockD: pass actual network namespace to grace period management functions
LockD: manage grace list per network namespace
SUNRPC: service request network namespace helper introduced
NFSd: make nfsd4_manager allocated per network namespace context.
LockD: make lockd manager allocated per network namespace
LockD: manage grace period per network namespace
Lockd: add more debug to host shutdown functions
Lockd: host complaining function introduced
LockD: manage used host count per networks namespace
LockD: manage garbage collection timeout per networks namespace
LockD: make garbage collector network namespace aware.
LockD: mark host per network namespace on garbage collect
nfsd4: fix missing fault_inject.h include
locks: move lease-specific code out of locks_delete_lock
locks: prevent side-effects of locks_release_private before file_lock is initialized
NFSd: set nfsd_serv to NULL after service destruction
NFSd: introduce nfsd_destroy() helper
...
Eric Dumazet [Tue, 31 Jul 2012 05:45:30 +0000 (05:45 +0000)]
ipv4: percpu nh_rth_output cache
Input path is mostly run under RCU and doesnt touch dst refcnt
But output path on forwarding or UDP workloads hits
badly dst refcount, and we have lot of false sharing, for example
in ipv4_mtu() when reading rt->rt_pmtu
Using a percpu cache for nh_rth_output gives a nice performance
increase at a small cost.
24 udpflood test on my 24 cpu machine (dummy0 output device)
(each process sends 1.000.000 udp frames, 24 processes are started)
before : 5.24 s
after : 2.06 s
For reference, time on linux-3.5 : 6.60 s
Eric Dumazet [Tue, 31 Jul 2012 01:08:23 +0000 (01:08 +0000)]
ipv4: Restore old dst_free() behavior.
commit 404e0a8b6a55 (net: ipv4: fix RCU races on dst refcounts) tried
to solve a race but added a problem at device/fib dismantle time :
We really want to call dst_free() as soon as possible, even if sockets
still have dst in their cache.
dst_release() calls in free_fib_info_rcu() are not welcomed.
Root of the problem was that now we also cache output routes (in
nh_rth_output), we must use call_rcu() instead of call_rcu_bh() in
rt_free(), because output route lookups are done in process context.
Based on feedback and initial patch from David Miller (adding another
call_rcu_bh() call in fib, but it appears it was not the right fix)
I left the inet_sk_rx_dst_set() helper and added __rcu attributes
to nh_rth_output and nh_rth_input to better document what is going on in
this code.
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
Pull Ceph changes from Sage Weil:
"Lots of stuff this time around:
- lots of cleanup and refactoring in the libceph messenger code, and
many hard to hit races and bugs closed as a result.
- lots of cleanup and refactoring in the rbd code from Alex Elder,
mostly in preparation for the layering functionality that will be
coming in 3.7.
- some misc rbd cleanups from Josh Durgin that are finally going
upstream
- support for CRUSH tunables (used by newer clusters to improve the
data placement)
- some cleanup in our use of d_parent that Al brought up a while back
- a random collection of fixes across the tree
There is another patch coming that fixes up our ->atomic_open()
behavior, but I'm going to hammer on it a bit more before sending it."
Fix up conflicts due to commits that were already committed earlier in
drivers/block/rbd.c, net/ceph/{messenger.c, osd_client.c}
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (132 commits)
rbd: create rbd_refresh_helper()
rbd: return obj version in __rbd_refresh_header()
rbd: fixes in rbd_header_from_disk()
rbd: always pass ops array to rbd_req_sync_op()
rbd: pass null version pointer in add_snap()
rbd: make rbd_create_rw_ops() return a pointer
rbd: have __rbd_add_snap_dev() return a pointer
libceph: recheck con state after allocating incoming message
libceph: change ceph_con_in_msg_alloc convention to be less weird
libceph: avoid dropping con mutex before fault
libceph: verify state after retaking con lock after dispatch
libceph: revoke mon_client messages on session restart
libceph: fix handling of immediate socket connect failure
ceph: update MAINTAINERS file
libceph: be less chatty about stray replies
libceph: clear all flags on con_close
libceph: clean up con flags
libceph: replace connection state bits with states
libceph: drop unnecessary CLOSED check in socket state change callback
libceph: close socket directly from ceph_con_close()
...
nfs: increase number of permitted callback connections.
By default a sunrpc service is limited to (N+3)*20 connections
where N is the number of threads. This is 80 when N==1.
If this number is exceeded a warning is printed suggesting that
the number of threads be increased. However with services which
run a single thread, this is impossible.
For such services there is a ->sv_maxconn setting that can be
used to forcibly increase the limit, and silence the message.
This is used by lockd.
The nfs client uses a sunrpc service to handle callbacks and
it too is single-threaded, so to avoid the useless messages,
and to allow a reasonable number of concurrent connections,
we need to set ->sv_maxconn. 1024 seems like a good number.
Alex Williamson [Tue, 31 Jul 2012 14:16:24 +0000 (08:16 -0600)]
vfio: Add PCI device driver
Add PCI device support for VFIO. PCI devices expose regions
for accessing config space, I/O port space, and MMIO areas
of the device. PCI config access is virtualized in the kernel,
allowing us to ensure the integrity of the system, by preventing
various accesses while reducing duplicate support across various
userspace drivers. I/O port supports read/write access while
MMIO also supports mmap of sufficiently sized regions. Support
for INTx, MSI, and MSI-X interrupts are provided using eventfds to
userspace.
Alex Williamson [Tue, 31 Jul 2012 14:16:23 +0000 (08:16 -0600)]
vfio: Type1 IOMMU implementation
This VFIO IOMMU backend is designed primarily for AMD-Vi and Intel
VT-d hardware, but is potentially usable by anything supporting
similar mapping functionality. We arbitrarily call this a Type1
backend for lack of a better name. This backend has no IOVA
or host memory mapping restrictions for the user and is optimized
for relatively static mappings. Mapped areas are pinned into system
memory.
Alex Williamson [Tue, 31 Jul 2012 14:16:22 +0000 (08:16 -0600)]
vfio: VFIO core
VFIO is a secure user level driver for use with both virtual machines
and user level drivers. VFIO makes use of IOMMU groups to ensure the
isolation of devices in use, allowing unprivileged user access. It's
intended that VFIO will replace KVM device assignment and UIO drivers
(in cases where the target platform includes a sufficiently capable
IOMMU).
New in this version of VFIO is support for IOMMU groups managed
through the IOMMU core as well as a rework of the API, removing the
group merge interface. We now go back to a model more similar to
original VFIO with UIOMMU support where the file descriptor obtained
from /dev/vfio/vfio allows access to the IOMMU, but only after a
group is added, avoiding the previous privilege issues with this type
of model. IOMMU support is also now fully modular as IOMMUs have
vastly different interface requirements on different platforms. VFIO
users are able to query and initialize the IOMMU model of their
choice.
Please see the follow-on Documentation commit for further description
and usage example.
Anton Vorontsov [Tue, 31 Jul 2012 11:39:30 +0000 (04:39 -0700)]
thermal: Constify 'type' argument for the registration routine
thermal_zone_device_register() does not modify 'type' argument, so it is
safe to declare it as const. Otherwise, if we pass a const string, we are
getting the ugly warning:
CC drivers/power/power_supply_core.o
drivers/power/power_supply_core.c: In function 'psy_register_thermal':
drivers/power/power_supply_core.c:204:6: warning: passing argument 1 of 'thermal_zone_device_register' discards 'const' qualifier from pointer target type [enabled by default]
include/linux/thermal.h:140:29: note: expected 'char *' but argument is of type 'const char *'
Anton Vorontsov [Tue, 31 Jul 2012 11:59:42 +0000 (04:59 -0700)]
Merge with upstream to accommodate with thermal changes
This merge is performed to take commit c56f5c0342dfee11a1 ("Thermal: Make
Thermal trip points writeable") out of Linus' tree and then fixup power
supply class. This is needed since thermal stuff added a new argument:
CC drivers/power/power_supply_core.o
drivers/power/power_supply_core.c: In function ‘psy_register_thermal’:
drivers/power/power_supply_core.c:204:6: warning: passing argument 3 of ‘thermal_zone_device_register’ makes integer from pointer without a cast [enabled by default]
include/linux/thermal.h:154:29: note: expected ‘int’ but argument is of type ‘struct power_supply *’
drivers/power/power_supply_core.c:204:6: error: too few arguments to function ‘thermal_zone_device_register’
include/linux/thermal.h:154:29: note: declared here
make[1]: *** [drivers/power/power_supply_core.o] Error 1
make: *** [drivers/power/] Error 2
Alexander Graf [Mon, 30 Jul 2012 12:32:59 +0000 (12:32 +0000)]
powerpc/kvm/book3s_32: Fix MTMSR_EERI macro
Commit b38c77d82e4 moved the MTMSR_EERI macro from the KVM code to generic
ppc_asm.h code. However, while adding it in the headers for the ppc32 case,
it missed out to remove the former definition in the KVM code.
This patch fixes compilation on server type PPC32 targets with CONFIG_KVM
enabled.
Merge tag 'writeback-proportions' of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/linux
Pull writeback updates from Wu Fengguang:
"Use time based periods to age the writeback proportions, which can
adapt equally well to fast/slow devices."
Fix up trivial conflict in comment in fs/sync.c
* tag 'writeback-proportions' of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/linux:
writeback: Fix some comment errors
block: Convert BDI proportion calculations to flexible proportions
lib: Fix possible deadlock in flexible proportion code
lib: Proportions with flexible period
Merge tag 'nfs-for-3.6-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client updates from Trond Myklebust:
"Features include:
- More preparatory patches for modularising NFSv2/v3/v4. Split out
the various NFSv2/v3/v4-specific code into separate files
- More preparation for the NFSv4 migration code
- Ensure that OPEN(O_CREATE) observes the pNFS mds threshold
parameters
- pNFS fast failover when the data servers are down
- Various cleanups and debugging patches"
* tag 'nfs-for-3.6-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (67 commits)
nfs: fix fl_type tests in NFSv4 code
NFS: fix pnfs regression with directio writes
NFS: fix pnfs regression with directio reads
sunrpc: clnt: Add missing braces
nfs: fix stub return type warnings
NFS: exit_nfs_v4() shouldn't be an __exit function
SUNRPC: Add a missing spin_unlock to gss_mech_list_pseudoflavors
NFS: Split out NFS v4 client functions
NFS: Split out the NFS v4 filesystem types
NFS: Create a single nfs_clone_super() function
NFS: Split out NFS v4 server creating code
NFS: Initialize the NFS v4 client from init_nfs_v4()
NFS: Move the v4 getroot code to nfs4getroot.c
NFS: Split out NFS v4 file operations
NFS: Initialize v4 sysctls from nfs_init_v4()
NFS: Create an init_nfs_v4() function
NFS: Split out NFS v4 inode operations
NFS: Split out NFS v3 inode operations
NFS: Split out NFS v2 inode operations
NFS: Clean up nfs4_proc_setclientid() and friends
...
Dan Carpenter [Fri, 20 Jul 2012 10:11:57 +0000 (07:11 -0300)]
[media] dib8000: move dereference after check for NULL
My static checker complains that we dereference "state" inside the call
to fft_to_mode() before checking for NULL. The comments say that it is
possible for "state" to be NULL so I have moved the dereference after
the check.
Ezequiel García [Wed, 18 Jul 2012 16:41:11 +0000 (13:41 -0300)]
[media] cx25821: Remove bad strcpy to read-only char*
The strcpy was being used to set the name of the board.
This was both wrong and redundant,
since the destination char* was read-only and
the name is set statically at compile time.
The type of the name field is changed to const char*
to prevent future errors.
Merge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
Pull media updates from Mauro Carvalho Chehab:
"This is the first part of the media patches for v3.6.
This patch series contain:
- new DVB frontend: rtl2832
- new video drivers: adv7393
- some unused files got removed
- a selection API cleanup between V4L2 and V4L2 subdev API's
- a major redesign at v4l-ioctl2, in order to clean it up
- several driver fixes and improvements."
* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media: (174 commits)
v4l: Export v4l2-common.h in include/linux/Kbuild
media: Revert "[media] Terratec Cinergy S2 USB HD Rev.2"
[media] media: Use pr_info not homegrown pr_reg macro
[media] Terratec Cinergy S2 USB HD Rev.2
[media] v4l: Correct conflicting V4L2 subdev selection API documentation
[media] Feature removal: V4L2 selections API target and flag definitions
[media] v4l: Unify selection flags documentation
[media] v4l: Unify selection flags
[media] v4l: Common documentation for selection targets
[media] v4l: Unify selection targets across V4L2 and V4L2 subdev interfaces
[media] v4l: Remove "_ACTUAL" from subdev selection API target definition names
[media] V4L: Remove "_ACTIVE" from the selection target name definitions
[media] media: dvb-usb: print mac address via native %pM
[media] s5p-tv: Use module_i2c_driver in sii9234_drv.c file
[media] media: gpio-ir-recv: add allowed_protos for platform data
[media] s5p-jpeg: Use module_platform_driver in jpeg-core.c file
[media] saa7134: fix spelling of detach in label
[media] cx88-blackbird: replace ioctl by unlocked_ioctl
[media] cx88: don't use current_norm
[media] cx88: fix a number of v4l2-compliance violations
...
Alex Elder [Wed, 25 Jul 2012 14:32:41 +0000 (09:32 -0500)]
rbd: return obj version in __rbd_refresh_header()
Add a new parameter to __rbd_refresh_header() through which the
version of the header object is passed back to the caller. In most
cases this isn't needed. The main motivation is to normalize
(almost) all calls to __rbd_refresh_header() so they are all
wrapped immediately by mutex_lock()/mutex_unlock().
Alex Elder [Wed, 11 Jul 2012 01:30:10 +0000 (20:30 -0500)]
rbd: fixes in rbd_header_from_disk()
This fixes a few issues in rbd_header_from_disk():
- There is a check intended to catch overflow, but it's wrong in
two ways.
- First, the type we don't want to overflow is size_t, not
unsigned int, and there is now a SIZE_MAX we can use for
use with that type.
- Second, we're allocating the snapshot ids and snapshot
image sizes separately (each has type u64; on disk they
grouped together as a rbd_image_header_ondisk structure).
So we can use the size of u64 in this overflow check.
- If there are no snapshots, then there should be no snapshot
names. Enforce this, and issue a warning if we encounter a
header with no snapshots but a non-zero snap_names_len.
- When saving the snapshot names into the header, be more direct
in defining the offset in the on-disk structure from which
they're being copied by using "snap_count" rather than "i"
in the array index.
- If an error occurs, the "snapc" and "snap_names" fields are
freed at the end of the function. Make those fields be null
pointers after they're freed, to be explicit that they are
no longer valid.
- Finally, move the definition of the local variable "i" to the
innermost scope in which it's needed.
Alex Elder [Tue, 26 Jun 2012 19:57:03 +0000 (12:57 -0700)]
rbd: always pass ops array to rbd_req_sync_op()
All of the callers of rbd_req_sync_op() except one pass a non-null
"ops" pointer. The only one that does not is rbd_req_sync_read(),
which passes CEPH_OSD_OP_READ as its "opcode" and, CEPH_OSD_FLAG_READ
for "flags".
By allocating the ops array in rbd_req_sync_read() and moving the
special case code for the null ops pointer into it, it becomes
clear that much of that code is not even necessary.
In addition, the "opcode" argument to rbd_req_sync_op() is never
actually used, so get rid of that.
Alex Elder [Tue, 26 Jun 2012 19:57:03 +0000 (12:57 -0700)]
rbd: make rbd_create_rw_ops() return a pointer
Either rbd_create_rw_ops() will succeed, or it will fail because a
memory allocation failed. Have it just return a valid pointer or
null rather than stuffing a pointer into a provided address and
returning an errno.
Alex Elder [Wed, 11 Jul 2012 01:30:10 +0000 (20:30 -0500)]
rbd: have __rbd_add_snap_dev() return a pointer
It's not obvious whether the snapshot pointer whose address is
provided to __rbd_add_snap_dev() will be assigned by that function.
Change it to return the snapshot, or a pointer-coded errno in the
event of a failure.
Sage Weil [Tue, 31 Jul 2012 01:19:45 +0000 (18:19 -0700)]
libceph: recheck con state after allocating incoming message
We drop the lock when calling the ->alloc_msg() con op, which means
we need to (a) not clobber con->in_msg without the mutex held, and (b)
we need to verify that we are still in the OPEN state when we retake
it to avoid causing any mayhem. If the state does change, -EAGAIN
will get us back to con_work() and loop.
Sage Weil [Tue, 31 Jul 2012 01:19:30 +0000 (18:19 -0700)]
libceph: change ceph_con_in_msg_alloc convention to be less weird
This function's calling convention is very limiting. In particular,
we can't return any error other than ENOMEM (and only implicitly),
which is a problem (see next patch).
Instead, return an normal 0 or error code, and make the skip a pointer
output parameter. Drop the useless in_hdr argument (we have the con
pointer).
Sage Weil [Tue, 31 Jul 2012 01:17:13 +0000 (18:17 -0700)]
libceph: avoid dropping con mutex before fault
The ceph_fault() function takes the con mutex, so we should avoid
dropping it before calling it. This fixes a potential race with
another thread calling ceph_con_close(), or _open(), or similar (we
don't reverify con->state after retaking the lock).
Add annotation so that lockdep realizes we will drop the mutex before
returning.
Sage Weil [Tue, 31 Jul 2012 01:16:56 +0000 (18:16 -0700)]
libceph: verify state after retaking con lock after dispatch
We drop the con mutex when delivering a message. When we retake the
lock, we need to verify we are still in the OPEN state before
preparing to read the next tag, or else we risk stepping on a
connection that has been closed.
Sage Weil [Tue, 31 Jul 2012 01:16:40 +0000 (18:16 -0700)]
libceph: revoke mon_client messages on session restart
Revoke all mon_client messages when we shut down the old connection.
This is mostly moot since we are re-using the same ceph_connection,
but it is cleaner.
Sage Weil [Mon, 30 Jul 2012 23:26:13 +0000 (16:26 -0700)]
libceph: be less chatty about stray replies
There are many (normal) conditions that can lead to us getting
unexpected replies, include cluster topology changes, osd failures,
and timeouts. There's no need to spam the console about it.
Sage Weil [Sat, 21 Jul 2012 00:24:40 +0000 (17:24 -0700)]
libceph: replace connection state bits with states
Use a simple set of 6 enumerated values for the socket states (CON_STATE_*)
and use those instead of the state bits. All of the con->state checks are
now under the protection of the con mutex, so this is safe. It also
simplifies many of the state checks because we can check for anything other
than the expected state instead of various bits for races we can think of.
This appears to hold up well to stress testing both with and without socket
failure injection on the server side.
projects / linux.git / log