percpu_ref has treated the dropping of the base reference and
switching to atomic mode as an integral operation; however, there's
nothing inherent tying the two together.
The use cases for percpu_ref have been expanding continuously. While
the current init/kill/reinit/exit model can cover a lot, the coupling
of kill/reinit with atomic/percpu mode switching is turning out to be
too restrictive for use cases where many percpu_refs are created and
destroyed back-to-back with only some of them reaching extended
operation. The coupling also makes implementing always-atomic debug
mode difficult.
This patch separates out percpu mode switching into
percpu_ref_switch_to_percpu() and reimplements percpu_ref_reinit() on
top of it.
* DEAD still requires ATOMIC. A dead ref can't be switched to percpu
mode w/o going through reinit.
v2: __percpu_ref_switch_to_percpu() was missing static. Fixed.
Reported by Fengguang aka kbuild test robot.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: kbuild test robot <fengguang.wu@intel.com>
percpu_ref has treated the dropping of the base reference and
switching to atomic mode as an integral operation; however, there's
nothing inherent tying the two together.
The use cases for percpu_ref have been expanding continuously. While
the current init/kill/reinit/exit model can cover a lot, the coupling
of kill/reinit with atomic/percpu mode switching is turning out to be
too restrictive for use cases where many percpu_refs are created and
destroyed back-to-back with only some of them reaching extended
operation. The coupling also makes implementing always-atomic debug
mode difficult.
This patch separates out atomic mode switching into
percpu_ref_switch_to_atomic() and reimplements
percpu_ref_kill_and_confirm() on top of it.
* The handling of __PERCPU_REF_ATOMIC and __PERCPU_REF_DEAD is now
differentiated. Among get/put operations, percpu_ref_tryget_live()
is the only one which cares about DEAD.
* percpu_ref_switch_to_atomic() can be called multiple times on the
same ref. This means that multiple @confirm_switch may get queued
up which we can't do reliably without extra memory area. This is
handled by making the later invocation synchronously wait for the
completion of the previous one. This isn't particularly desirable
but such synchronous waits shouldn't happen in most cases.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
percpu_ref will be restructured so that percpu/atomic mode switching
and reference killing are dedoupled. In preparation, add
PCPU_REF_DEAD and PCPU_REF_ATOMIC_DEAD which is OR of ATOMIC and DEAD.
For now, ATOMIC and DEAD are changed together and all PCPU_REF_ATOMIC
uses are converted to PCPU_REF_ATOMIC_DEAD without causing any
behavior changes.
percpu_ref_init() now specifies an explicit alignment when allocating
the percpu counters so that the pointer has enough unused low bits to
accomodate the flags. Note that one flag was fine as min alignment
for percpu memory is 2 bytes but two flags are already too many for
the natural alignment of unsigned longs on archs like cris and m68k.
v2: The original patch had BUILD_BUG_ON() which triggers if unsigned
long's alignment isn't enough to accomodate the flags, which
triggered on cris and m64k. percpu_ref_init() updated to specify
the required alignment explicitly. Reported by Fengguang.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
Cc: kbuild test robot <fengguang.wu@intel.com>
percpu_ref will be restructured so that percpu/atomic mode switching
and reference killing are dedoupled. In preparation, do the following
renames.
* percpu_ref->confirm_kill -> percpu_ref->confirm_switch
* __PERCPU_REF_DEAD -> __PERCPU_REF_ATOMIC
* __percpu_ref_alive() -> __ref_is_percpu()
This patch is pure rename and doesn't introduce any functional
changes.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
percpu_ref uses pcpu_ prefix for internal stuff and percpu_ for
externally visible ones. This is the same convention used in the
percpu allocator implementation. It works fine there but percpu_ref
doesn't have too much internal-only stuff and scattered usages of
pcpu_ prefix are confusing than helpful.
This patch replaces all pcpu_ prefixes with percpu_. This is pure
rename and there's no functional change. Note that PCPU_REF_DEAD is
renamed to __PERCPU_REF_DEAD to signify that the flag is internal.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
* Some comments became stale. Updated.
* percpu_ref_tryget() unnecessarily initializes @ret. Removed.
* A blank line removed from percpu_ref_kill_rcu().
* Explicit function name in a WARN format string replaced with __func__.
* WARN_ON() in percpu_ref_reinit() converted to WARN_ON_ONCE().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
percpu_ref is gonna go through restructuring. Move
percpu_ref_reinit() after percpu_ref_kill_and_confirm(). This will
make later changes easier to follow and result in cleaner
organization.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Kent Overstreet <kmo@daterainc.com>
This reverts commit 0a30288da1, which
was a temporary fix for SCSI blk-mq stall issue. The following
patches will fix the issue properly by introducing atomic mode to
percpu_ref.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Kent Overstreet <kmo@daterainc.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@lst.de>
This is to receive 0a30288da1 ("blk-mq, percpu_ref: implement a
kludge for SCSI blk-mq stall during probe") which implements
__percpu_ref_kill_expedited() to work around SCSI blk-mq stall. The
commit reverted and patches to implement proper fix will be added.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Kent Overstreet <kmo@daterainc.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Christoph Hellwig <hch@lst.de>
blk-mq uses percpu_ref for its usage counter which tracks the number
of in-flight commands and used to synchronously drain the queue on
freeze. percpu_ref shutdown takes measureable wallclock time as it
involves a sched RCU grace period. This means that draining a blk-mq
takes measureable wallclock time. One would think that this shouldn't
matter as queue shutdown should be a rare event which takes place
asynchronously w.r.t. userland.
Unfortunately, SCSI probing involves synchronously setting up and then
tearing down a lot of request_queues back-to-back for non-existent
LUNs. This means that SCSI probing may take more than ten seconds
when scsi-mq is used.
This will be properly fixed by implementing a mechanism to keep
q->mq_usage_counter in atomic mode till genhd registration; however,
that involves rather big updates to percpu_ref which is difficult to
apply late in the devel cycle (v3.17-rc6 at the moment). As a
stop-gap measure till the proper fix can be implemented in the next
cycle, this patch introduces __percpu_ref_kill_expedited() and makes
blk_mq_freeze_queue() use it. This is heavy-handed but should work
for testing the experimental SCSI blk-mq implementation.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Christoph Hellwig <hch@infradead.org>
Link: http://lkml.kernel.org/g/20140919113815.GA10791@lst.de
Fixes: add703fda9 ("blk-mq: use percpu_ref for mq usage count")
Cc: Kent Overstreet <kmo@daterainc.com>
Cc: Jens Axboe <axboe@kernel.dk>
Tested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@fb.com>
Conflicts:
arch/mips/net/bpf_jit.c
drivers/net/can/flexcan.c
Both the flexcan and MIPS bpf_jit conflicts were cases of simple
overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking fixes from David Miller:
1) If the user gives us a msg_namelen of 0, don't try to interpret
anything pointed to by msg_name. From Ani Sinha.
2) Fix some bnx2i/bnx2fc randconfig compilation errors.
The gist of the issue is that we firstly have drivers that span both
SCSI and networking. And at the top of that chain of dependencies
we have things like SCSI_FC_ATTRS and SCSI_NETLINK which are
selected.
But since select is a sledgehammer and ignores dependencies,
everything to select's SCSI_FC_ATTRS and/or SCSI_NETLINK has to also
explicitly select their dependencies and so on and so forth.
Generally speaking 'select' is supposed to only be used for child
nodes, those which have no dependencies of their own. And this
whole chain of dependencies in the scsi layer violates that rather
strongly.
So just make SCSI_NETLINK depend upon it's dependencies, and so on
and so forth for the things selecting it (either directly or
indirectly).
From Anish Bhatt and Randy Dunlap.
3) Fix generation of blackhole routes in IPSEC, from Steffen Klassert.
4) Actually notice netdev feature changes in rtl_open() code, from
Hayes Wang.
5) Fix divide by zero in bond enslaving, from Nikolay Aleksandrov.
6) Missing memory barrier in sunvnet driver, from David Stevens.
7) Don't leave anycast addresses around when ipv6 interface is
destroyed, from Sabrina Dubroca.
8) Don't call efx_{arch}_filter_sync_rx_mode before addr_list_lock is
initialized in SFC driver, from Edward Cree.
9) Fix missing DMA error checking in 3c59x, from Neal Horman.
10) Openvswitch doesn't emit OVS_FLOW_CMD_NEW notifications accidently,
fix from Samuel Gauthier.
11) pch_gbe needs to select NET_PTP_CLASSIFY otherwise we can get a
build error.
12) Fix macvlan regression wherein we stopped emitting
broadcast/multicast frames over software devices. From Nicolas
Dichtel.
13) Fix infiniband bug due to unintended overflow of skb->cb[], from
Eric Dumazet. And add an assertion so this doesn't happen again.
14) dm9000_parse_dt() should return error pointers, not NULL. From
Tobias Klauser.
15) IP tunneling code uses this_cpu_ptr() in preemptible contexts, fix
from Eric Dumazet.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (87 commits)
net: bcmgenet: call bcmgenet_dma_teardown in bcmgenet_fini_dma
net: bcmgenet: fix TX reclaim accounting for fragments
ipv4: do not use this_cpu_ptr() in preemptible context
dm9000: Return an ERR_PTR() in all error conditions of dm9000_parse_dt()
r8169: fix an if condition
r8152: disable ALDPS
ipoib: validate struct ipoib_cb size
net: sched: shrink struct qdisc_skb_cb to 28 bytes
tg3: Work around HW/FW limitations with vlan encapsulated frames
macvlan: allow to enqueue broadcast pkt on virtual device
pch_gbe: 'select' NET_PTP_CLASSIFY.
scsi: Use 'depends' with LIBFC instead of 'select'.
openvswitch: restore OVS_FLOW_CMD_NEW notifications
genetlink: add function genl_has_listeners()
lib: rhashtable: remove second linux/log2.h inclusion
net: allow macvlans to move to net namespace
3c59x: Fix bad offset spec in skb_frag_dma_map
3c59x: Add dma error checking and recovery
sparc: bpf_jit: fix support for ldx/stx mem and SKF_AD_VLAN_TAG
can: at91_can: add missing prepare and unprepare of the clock
...
old gcc 4.2 used by avr32 architecture produces warnings:
lib/test_bpf.c:1741: warning: integer constant is too large for 'long' type
lib/test_bpf.c:1741: warning: integer constant is too large for 'long' type
lib/test_bpf.c: In function '__run_one':
lib/test_bpf.c:1897: warning: 'ret' may be used uninitialized in this function
silence these warnings.
Fixes: 02ab695bb3 ("net: filter: add "load 64-bit immediate" eBPF instruction")
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Acked-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
percpu_ref is currently based on ints and the number of refs it can
cover is (1 << 31). This makes it impossible to use a percpu_ref to
count memory objects or pages on 64bit machines as it may overflow.
This forces those users to somehow aggregate the references before
contributing to the percpu_ref which is often cumbersome and sometimes
challenging to get the same level of performance as using the
percpu_ref directly.
While using ints for the percpu counters makes them pack tighter on
64bit machines, the possible gain from using ints instead of longs is
extremely small compared to the overall gain from per-cpu operation.
This patch makes percpu_ref based on longs so that it can be used to
directly count memory objects or pages.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Kent Overstreet <kmo@daterainc.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
percpu_ref's WARN messages can be a lot more helpful by indicating
who's the culprit. Make them report the release function that the
offending percpu-refcount is associated with. This should make it a
lot easier to track down the reported invalid refcnting operations.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Kent Overstreet <kmo@daterainc.com>
Provide a function to convert a buffer of binary data into an unterminated
ascii hex string representation of that data.
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
It used to be an ad-hoc hack defined by the x86 version of
<asm/bitops.h> that enabled a couple of library routines to know whether
an integer multiply is faster than repeated shifts and additions.
This just makes it use the real Kconfig system instead, and makes x86
(which was the only architecture that did this) select the option.
NOTE! Even for x86, this really is kind of wrong. If we cared, we would
probably not enable this for builds optimized for netburst (P4), where
shifts-and-adds are generally faster than multiplies. This patch does
*not* change that kind of logic, though, it is purely a syntactic change
with no code changes.
This was triggered by the fact that we have other places that really
want to know "do I want to expand multiples by constants by hand or
not", particularly the hash generation code.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This fixes CVE-2014-3631.
It is possible for an associative array to end up with a shortcut node at the
root of the tree if there are more than fan-out leaves in the tree, but they
all crowd into the same slot in the lowest level (ie. they all have the same
first nibble of their index keys).
When assoc_array_gc() returns back up the tree after scanning some leaves, it
can fall off of the root and crash because it assumes that the back pointer
from a shortcut (after label ascend_old_tree) must point to a normal node -
which isn't true of a shortcut node at the root.
Should we find we're ascending rootwards over a shortcut, we should check to
see if the backpointer is zero - and if it is, we have completed the scan.
This particular bug cannot occur if the root node is not a shortcut - ie. if
you have fewer than 17 keys in a keyring or if you have at least two keys that
sit into separate slots (eg. a keyring and a non keyring).
This can be reproduced by:
ring=`keyctl newring bar @s`
for ((i=1; i<=18; i++)); do last_key=`keyctl newring foo$i $ring`; done
keyctl timeout $last_key 2
Doing this:
echo 3 >/proc/sys/kernel/keys/gc_delay
first will speed things up.
If we do fall off of the top of the tree, we get the following oops:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
IP: [<ffffffff8136cea7>] assoc_array_gc+0x2f7/0x540
PGD dae15067 PUD cfc24067 PMD 0
Oops: 0000 [#1] SMP
Modules linked in: xt_nat xt_mark nf_conntrack_netbios_ns nf_conntrack_broadcast ip6t_rpfilter ip6t_REJECT xt_conntrack ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6table_ni
CPU: 0 PID: 26011 Comm: kworker/0:1 Not tainted 3.14.9-200.fc20.x86_64 #1
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
Workqueue: events key_garbage_collector
task: ffff8800918bd580 ti: ffff8800aac14000 task.ti: ffff8800aac14000
RIP: 0010:[<ffffffff8136cea7>] [<ffffffff8136cea7>] assoc_array_gc+0x2f7/0x540
RSP: 0018:ffff8800aac15d40 EFLAGS: 00010206
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff8800aaecacc0
RDX: ffff8800daecf440 RSI: 0000000000000001 RDI: ffff8800aadc2bc0
RBP: ffff8800aac15da8 R08: 0000000000000001 R09: 0000000000000003
R10: ffffffff8136ccc7 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000070 R15: 0000000000000001
FS: 0000000000000000(0000) GS:ffff88011fc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000018 CR3: 00000000db10d000 CR4: 00000000000006f0
Stack:
ffff8800aac15d50 0000000000000011 ffff8800aac15db8 ffffffff812e2a70
ffff880091a00600 0000000000000000 ffff8800aadc2bc3 00000000cd42c987
ffff88003702df20 ffff88003702dfa0 0000000053b65c09 ffff8800aac15fd8
Call Trace:
[<ffffffff812e2a70>] ? keyring_detect_cycle_iterator+0x30/0x30
[<ffffffff812e3e75>] keyring_gc+0x75/0x80
[<ffffffff812e1424>] key_garbage_collector+0x154/0x3c0
[<ffffffff810a67b6>] process_one_work+0x176/0x430
[<ffffffff810a744b>] worker_thread+0x11b/0x3a0
[<ffffffff810a7330>] ? rescuer_thread+0x3b0/0x3b0
[<ffffffff810ae1a8>] kthread+0xd8/0xf0
[<ffffffff810ae0d0>] ? insert_kthread_work+0x40/0x40
[<ffffffff816ffb7c>] ret_from_fork+0x7c/0xb0
[<ffffffff810ae0d0>] ? insert_kthread_work+0x40/0x40
Code: 08 4c 8b 22 0f 84 bf 00 00 00 41 83 c7 01 49 83 e4 fc 41 83 ff 0f 4c 89 65 c0 0f 8f 5a fe ff ff 48 8b 45 c0 4d 63 cf 49 83 c1 02 <4e> 8b 34 c8 4d 85 f6 0f 84 be 00 00 00 41 f6 c6 01 0f 84 92
RIP [<ffffffff8136cea7>] assoc_array_gc+0x2f7/0x540
RSP <ffff8800aac15d40>
CR2: 0000000000000018
---[ end trace 1129028a088c0cbd ]---
Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: James Morris <james.l.morris@oracle.com>
add BPF_LD_IMM64 instruction to load 64-bit immediate value into a register.
All previous instructions were 8-byte. This is first 16-byte instruction.
Two consecutive 'struct bpf_insn' blocks are interpreted as single instruction:
insn[0].code = BPF_LD | BPF_DW | BPF_IMM
insn[0].dst_reg = destination register
insn[0].imm = lower 32-bit
insn[1].code = 0
insn[1].imm = upper 32-bit
All unused fields must be zero.
Classic BPF has similar instruction: BPF_LD | BPF_W | BPF_IMM
which loads 32-bit immediate value into a register.
x64 JITs it as single 'movabsq %rax, imm64'
arm64 may JIT as sequence of four 'movk x0, #imm16, lsl #shift' insn
Note that old eBPF programs are binary compatible with new interpreter.
It helps eBPF programs load 64-bit constant into a register with one
instruction instead of using two registers and 4 instructions:
BPF_MOV32_IMM(R1, imm32)
BPF_ALU64_IMM(BPF_LSH, R1, 32)
BPF_MOV32_IMM(R2, imm32)
BPF_ALU64_REG(BPF_OR, R1, R2)
User space generated programs will use this instruction to load constants only.
To tell kernel that user space needs a pointer the _pseudo_ variant of
this instruction may be added later, which will use extra bits of encoding
to indicate what type of pointer user space is asking kernel to provide.
For example 'off' or 'src_reg' fields can be used for such purpose.
src_reg = 1 could mean that user space is asking kernel to validate and
load in-kernel map pointer.
src_reg = 2 could mean that user space needs readonly data section pointer
src_reg = 3 could mean that user space needs a pointer to per-cpu local data
All such future pseudo instructions will not be carrying the actual pointer
as part of the instruction, but rather will be treated as a request to kernel
to provide one. The kernel will verify the request_for_a_pointer, then
will drop _pseudo_ marking and will store actual internal pointer inside
the instruction, so the end result is the interpreter and JITs never
see pseudo BPF_LD_IMM64 insns and only operate on generic BPF_LD_IMM64 that
loads 64-bit immediate into a register. User space never operates on direct
pointers and verifier can easily recognize request_for_pointer vs other
instructions.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fix spelling typo found in DocBook/kernel-api.xml.
It is because the file is generated from the source comments,
I have to fix the comments in source codes.
Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Percpu allocator now supports allocation mask. Add @gfp to
percpu_ref_init() so that !GFP_KERNEL allocation masks can be used
with percpu_refs too.
This patch doesn't make any functional difference.
v2: blk-mq conversion was missing. Updated.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Kent Overstreet <koverstreet@google.com>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Nicholas A. Bellinger <nab@linux-iscsi.org>
Cc: Jens Axboe <axboe@kernel.dk>
Percpu allocator now supports allocation mask. Add @gfp to
[flex_]proportions init functions so that !GFP_KERNEL allocation masks
can be used with them too.
This patch doesn't make any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Peter Zijlstra <peterz@infradead.org>
Tejun Heo
David S. Miller
Linus Torvalds
Alexei Starovoitov
Fabian Frederick
Aaron Tomlin
David Howells
Masanari Iida