Mathieu Desnoyers [Tue, 26 Dec 2017 14:47:22 +0000 (09:47 -0500)]
Update kvm instrumentation for 4.15
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 19 Dec 2017 20:06:42 +0000 (15:06 -0500)]
Fix: ACCESS_ONCE() removed in kernel 4.15
The ACCESS_ONCE() macro was removed in kernel 4.15 and should be
replaced by READ_ONCE and WRITE_ONCE which were introduced in kernel
3.19.
This commit replaces all calls to ACCESS_ONCE() with the appropriate
READ_ONCE or WRITE_ONCE and adds compatibility macros for kernels that
have them.
See this upstream commit:
commit
b03a0fe0c5e4b46dcd400d27395b124499554a71
Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Date: Mon Oct 23 14:07:25 2017 -0700
locking/atomics, mm: Convert ACCESS_ONCE() to READ_ONCE()/WRITE_ONCE()
For several reasons, it is desirable to use {READ,WRITE}_ONCE() in
preference to ACCESS_ONCE(), and new code is expected to use one of the
former. So far, there's been no reason to change most existing uses of
ACCESS_ONCE(), as these aren't currently harmful.
However, for some features it is necessary to instrument reads and
writes separately, which is not possible with ACCESS_ONCE(). This
distinction is critical to correct operation.
It's possible to transform the bulk of kernel code using the Coccinelle
script below. However, this doesn't handle comments, leaving references
to ACCESS_ONCE() instances which have been removed. As a preparatory
step, this patch converts the mm code and comments to use
{READ,WRITE}_ONCE() consistently.
----
virtual patch
@ depends on patch @
expression E1, E2;
@@
- ACCESS_ONCE(E1) = E2
+ WRITE_ONCE(E1, E2)
@ depends on patch @
expression E;
@@
- ACCESS_ONCE(E)
+ READ_ONCE(E)
----
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Mon, 18 Dec 2017 19:35:55 +0000 (14:35 -0500)]
Fix: sched instrumentation on stable RT kernels
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 29 Nov 2017 22:03:21 +0000 (17:03 -0500)]
timer API transition for kernel 4.15
The timer API changes starting from kernel 4.15.0.
There's an interresting LWN article on this subject:
https://lwn.net/Articles/735887/
Check these upstream commits for more details:
commit
686fef928bba6be13cabe639f154af7d72b63120
Author: Kees Cook <keescook@chromium.org>
Date: Thu Sep 28 06:38:17 2017 -0700
timer: Prepare to change timer callback argument type
Modern kernel callback systems pass the structure associated with a
given callback to the callback function. The timer callback remains one
of the legacy cases where an arbitrary unsigned long argument continues
to be passed as the callback argument. This has several problems:
- This bloats the timer_list structure with a normally redundant
.data field.
- No type checking is being performed, forcing callbacks to do
explicit type casts of the unsigned long argument into the object
that was passed, rather than using container_of(), as done in most
of the other callback infrastructure.
- Neighboring buffer overflows can overwrite both the .function and
the .data field, providing attackers with a way to elevate from a buffer
overflow into a simplistic ROP-like mechanism that allows calling
arbitrary functions with a controlled first argument.
- For future Control Flow Integrity work, this creates a unique function
prototype for timer callbacks, instead of allowing them to continue to
be clustered with other void functions that take a single unsigned long
argument.
This adds a new timer initialization API, which will ultimately replace
the existing setup_timer(), setup_{deferrable,pinned,etc}_timer() family,
named timer_setup() (to mirror hrtimer_setup(), making instances of its
use much easier to grep for).
In order to support the migration of existing timers into the new
callback arguments, timer_setup() casts its arguments to the existing
legacy types, and explicitly passes the timer pointer as the legacy
data argument. Once all setup_*timer() callers have been replaced with
timer_setup(), the casts can be removed, and the data argument can be
dropped with the timer expiration code changed to just pass the timer
to the callback directly.
:
Modern kernel callback systems pass the structure associated with a
given callback to the callback function. The timer callback remains one
of the legacy cases where an arbitrary unsigned long argument continues
to be passed as the callback argument. This has several problems:
- This bloats the timer_list structure with a normally redundant
.data field.
- No type checking is being performed, forcing callbacks to do
explicit type casts of the unsigned long argument into the object
that was passed, rather than using container_of(), as done in most
of the other callback infrastructure.
- Neighboring buffer overflows can overwrite both the .function and
the .data field, providing attackers with a way to elevate from a buffer
overflow into a simplistic ROP-like mechanism that allows calling
arbitrary functions with a controlled first argument.
- For future Control Flow Integrity work, this creates a unique function
prototype for timer callbacks, instead of allowing them to continue to
be clustered with other void functions that take a single unsigned long
argument.
This adds a new timer initialization API, which will ultimately replace
the existing setup_timer(), setup_{deferrable,pinned,etc}_timer() family,
named timer_setup() (to mirror hrtimer_setup(), making instances of its
use much easier to grep for).
In order to support the migration of existing timers into the new
callback arguments, timer_setup() casts its arguments to the existing
legacy types, and explicitly passes the timer pointer as the legacy
data argument. Once all setup_*timer() callers have been replaced with
timer_setup(), the casts can be removed, and the data argument can be
dropped with the timer expiration code changed to just pass the timer
to the callback directly.
Since the regular pattern of using container_of() during local variable
declaration repeats the need for the variable type declaration
to be included, this adds a helper modeled after other from_*()
helpers that wrap container_of(), named from_timer(). This helper uses
typeof(*variable), removing the type redundancy and minimizing the need
for line wraps in forthcoming conversions from "unsigned data long" to
"struct timer_list *" in the timer callbacks:
-void callback(unsigned long data)
+void callback(struct timer_list *t)
{
- struct some_data_structure *local = (struct some_data_structure *)data;
+ struct some_data_structure *local = from_timer(local, t, timer);
Finally, in order to support the handful of timer users that perform
open-coded assignments of the .function (and .data) fields, provide
cast macros (TIMER_FUNC_TYPE and TIMER_DATA_TYPE) that can be used
temporarily. Once conversion has been completed, these can be globally
trivially removed.
...
commit
e99e88a9d2b067465adaa9c111ada99a041bef9a
Author: Kees Cook <keescook@chromium.org>
Date: Mon Oct 16 14:43:17 2017 -0700
treewide: setup_timer() -> timer_setup()
This converts all remaining cases of the old setup_timer() API into using
timer_setup(), where the callback argument is the structure already
holding the struct timer_list. These should have no behavioral changes,
since they just change which pointer is passed into the callback with
the same available pointers after conversion. It handles the following
examples, in addition to some other variations.
...
commit
185981d54a60ae90942c6ba9006b250f3348cef2
Author: Kees Cook <keescook@chromium.org>
Date: Wed Oct 4 16:26:58 2017 -0700
timer: Remove init_timer_pinned() in favor of timer_setup()
This refactors the only users of init_timer_pinned() to use
the new timer_setup() and from_timer(). Drops the definition of
init_timer_pinned().
...
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Wed, 13 Dec 2017 18:40:42 +0000 (13:40 -0500)]
Fix: Don't nest get online cpus
Since the cpu hotplug refactoring in the Linux kernel, CPU hotplug
"online cpus" read lock cannot be nested anymore.
Fix this by disabling preemption around the section instead.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 8 Dec 2017 19:17:21 +0000 (14:17 -0500)]
Fix: lttng_channel_syscall_mask() bool use in bitfield
gcc 7 warns about using ~ on a bool. Pass a char as input type instead.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 28 Nov 2017 21:02:45 +0000 (16:02 -0500)]
Fix: update kmem instrumentation for kernel 4.15
See upstream commit:
commit
2d4894b5d2ae0fe1725ea7abd57b33bfbbe45492
Author: Mel Gorman <mgorman@techsingularity.net>
Date: Wed Nov 15 17:37:59 2017 -0800
mm: remove cold parameter from free_hot_cold_page*
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Wed, 8 Nov 2017 19:10:27 +0000 (14:10 -0500)]
Version 2.9.7
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 7 Nov 2017 21:44:36 +0000 (16:44 -0500)]
Fix: lttng_kvmalloc helper NULL pointer OOPS
The static function __vmalloc_node is not visible by KALLSYMS_ALL on at
least some kernels, which leads to a call to a NULL function when trying
to perform allocation of lttng buffer memory under memory fragmentation
conditions (kmalloc_node failure).
Use __vmalloc_node_range instead, and check that the returned pointer
is non-NULL to ensure this type of failure does not happen in any
condition.
Fallback to __vmalloc(), even though it is not NUMA-aware, in case
we fail to find __vmalloc_node_range, and print an explicit warning
to the user console about the need to enable KALLSYMS_ALL.
This affects kernels < 4.12. Later kernels provide kvmalloc(), which
we use.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 3 Nov 2017 21:27:11 +0000 (17:27 -0400)]
Version 2.9.6
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 31 Oct 2017 22:23:59 +0000 (18:23 -0400)]
Fix: lttng-logger get_user_pages_fast error handling
Comparing a signed return value against an unsigned nr_pages performs
the comparison as "unsigned", and therefore mistakenly considers
get_user_pages_fast() errors as success.
By passing an invalid pointer to write() to the /proc/lttng-logger
interface, unprivileged user-space processes can trigger a kernel OOPS.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 5 Oct 2017 21:05:53 +0000 (17:05 -0400)]
Version 2.9.5
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 5 Oct 2017 18:52:15 +0000 (14:52 -0400)]
Fix: update block instrumentation for 4.14 kernel
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 5 Oct 2017 18:45:43 +0000 (14:45 -0400)]
Revert "Fix: update block instrumentation for kernel 4.14"
This reverts commit
49447902967115fe5a07ee7a1df3d17fbf4b1ab8.
It introduces a NULL pointer dereference:
[ 37.862398] BUG: unable to handle kernel NULL pointer dereference at
0000000000000008
181.3 [ 37.864108] IP: [<
ffffffffa01c41b7>] __event_probe__block_get_rq+0x127/0x4b0 [lttng_probe_block]
181.4 [ 37.864108] PGD
7a402067 PUD
7a4c7067 PMD 0
181.5 [ 37.864108] Oops: 0000 [#1] SMP
181.6 [ 37.864108] Modules linked in: lttng_probe_x86_exceptions(OE) lttng_probe_x86_irq_vectors(OE) lttng_probe_writeback(OE) lttng_probe_workqueue(OE) lttng_probe_vmscan(OE) lttng_probe_udp(OE) lttng_probe_timer(OE) lttng_probe_sunrpc(OE) lttng_probe_statedump(OE) lttng_probe_sock(OE) lttng_probe_skb(OE) lttng_probe_signal(OE) lttng_probe_scsi(OE) lttng_probe_sched(OE) lttng_probe_regulator(OE) lttng_probe_regmap(OE) lttng_probe_rcu(OE) lttng_probe_random(OE) lttng_probe_printk(OE) lttng_probe_power(OE) lttng_probe_net(OE) lttng_probe_napi(OE) lttng_probe_module(OE) lttng_probe_kvm_x86_mmu(OE) lttng_probe_kvm_x86(OE) lttng_probe_kvm(OE) lttng_probe_kmem(OE) lttng_probe_jbd2(OE) lttng_probe_irq(OE) lttng_probe_i2c(OE) lttng_probe_gpio(OE) lttng_probe_ext4(OE) lttng_probe_compaction(OE) lttng_probe_btrfs(OE) lttng_probe_block(OE) lttng_ring_buffer_metadata_mmap_client(OE) lttng_ring_buffer_client_mmap_overwrite(OE) lttng_ring_buffer_client_mmap_discard(OE) lttng_ring_buffer_metadata_client(OE) lttng_ring_buffer_client_overwrite(OE) lttng_ring_buffer_client_discard(OE) lttng_tracer(OE) lttng_statedump(OE) lttng_ftrace(OE) lttng_kprobes(OE) lttng_clock(OE) lttng_lib_ring_buffer(OE) lttng_kretprobes(OE)
181.7 [ 37.864108] CPU: 1 PID: 6 Comm: kworker/u4:0 Tainted: G OE 4.4.90 #1
181.8 [ 37.864108] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Ubuntu-1.8.2-1ubuntu1 04/01/2014
181.9 [ 37.864108] Workqueue: events_freezable_power_ disk_events_workfn
181.10 [ 37.864108] task:
ffff88007c861bc0 ti:
ffff88007c868000 task.ti:
ffff88007c868000
181.11 [ 37.864108] RIP: 0010:[<
ffffffffa01c41b7>] [<
ffffffffa01c41b7>] __event_probe__block_get_rq+0x127/0x4b0 [lttng_probe_block]
181.12 [ 37.864108] RSP: 0018:
ffff88007c86ba98 EFLAGS:
00010246
181.13 [ 37.864108] RAX:
0000000000000000 RBX:
ffff880073683348 RCX:
ffff8800747d0000
181.14 [ 37.864108] RDX:
00000008d0c5bde9 RSI:
00000000000009f2 RDI:
0000000000400000
181.15 [ 37.864108] RBP:
ffff88007c86bba8 R08:
00000000001789ed R09:
0000000000100000
181.16 [ 37.864108] R10:
ffffe8ffffd02460 R11:
0000000000000000 R12:
0000000000000000
181.17 [ 37.864108] R13:
0000000000017fe0 R14:
ffff88007363c6e8 R15:
ffff88007bef83c0
181.18 [ 37.864108] FS:
0000000000000000(0000) GS:
ffff88007fd00000(0000) knlGS:
0000000000000000
181.19 [ 37.864108] CS: 0010 DS: 0000 ES: 0000 CR0:
000000008005003b
181.20 [ 37.864108] CR2:
0000000000000008 CR3:
000000007a4d0000 CR4:
00000000000006e0
181.21 [ 37.864108] Stack:
181.22 [ 37.864108]
0000000000000000 ffffffff8115a46b ffff88007c86bbe8 ffff88007bc67e30
181.23 [ 37.864108]
ffff880073683348 00000000ffffff01 ffff88007a7a1000 ffff88007c86bab8
181.24 [ 37.864108]
0000000000000028 0000000100000001 ffffe8ffffd02460 0000000000000035
181.25 [ 37.864108] Call Trace:
181.26 [ 37.864108] [<
ffffffff8115a46b>] ? ktime_get_mono_fast_ns+0x4b/0x90
181.27 [ 37.864108] [<
ffffffff81532849>] ? alloc_request_struct+0x19/0x20
181.28 [ 37.864108] [<
ffffffff811e8d8f>] ? mempool_alloc+0x5f/0x150
181.29 [ 37.864108] [<
ffffffffa021815c>] ? __event_probe__kmem_alloc+0x1dc/0x2c0 [lttng_probe_kmem]
181.30 [ 37.864108] [<
ffffffff810ad85e>] ? kvm_clock_read+0x1e/0x20
181.31 [ 37.864108] [<
ffffffff81535f4f>] get_request+0x4af/0x760
181.32 [ 37.864108] [<
ffffffff8112c270>] ? wake_atomic_t_function+0x60/0x60
181.33 [ 37.864108] [<
ffffffff81536283>] blk_get_request+0x83/0xe0
181.34 [ 37.864108] [<
ffffffff81773b5d>] scsi_execute+0x3d/0x1d0
181.35 [ 37.864108] [<
ffffffff817758fe>] scsi_execute_req_flags+0x8e/0xf0
181.36 [ 37.864108] [<
ffffffff81788f4d>] sr_check_events+0x8d/0x2a0
181.37 [ 37.864108] [<
ffffffff81547590>] ? disk_check_events+0x130/0x130
181.38 [ 37.864108] [<
ffffffff8181b618>] cdrom_check_events+0x18/0x30
181.39 [ 37.864108] [<
ffffffff8178935a>] sr_block_check_events+0x2a/0x30
181.40 [ 37.864108] [<
ffffffff815474b1>] disk_check_events+0x51/0x130
181.41 [ 37.864108] [<
ffffffff815475a6>] disk_events_workfn+0x16/0x20
181.42 [ 37.864108] [<
ffffffff81102b85>] process_one_work+0x165/0x480
181.43 [ 37.864108] [<
ffffffff81102eeb>] worker_thread+0x4b/0x4c0
181.44 [ 37.864108] [<
ffffffff81102ea0>] ? process_one_work+0x480/0x480
181.45 [ 37.864108] [<
ffffffff81108d86>] kthread+0xd6/0xf0
181.46 [ 37.864108] [<
ffffffff81108cb0>] ? kthread_create_on_node+0x180/0x180
181.47 [ 37.864108] [<
ffffffff81aa690f>] ret_from_fork+0x3f/0x70
181.48 [ 37.864108] [<
ffffffff81108cb0>] ? kthread_create_on_node+0x180/0x180
181.49 [ 37.864108] Code: 00 00 00 00 48 89 85 20 ff ff ff 48 8d 85 10 ff ff ff 8b 73 04 48 89 85 28 ff ff ff 49 8b 47 48 ff 50 28 85 c0 0f 88 5d 01 00 00 <49> 8b 44 24 08 48 85 c0 0f 84 3d 03 00 00 8b 00 89 85 08 ff ff
181.50 [ 37.864108] RIP [<
ffffffffa01c41b7>] __event_probe__block_get_rq+0x127/0x4b0 [lttng_probe_block]
181.51 [ 37.864108] RSP <
ffff88007c86ba98>
181.52 [ 37.864108] CR2:
0000000000000008
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 3 Oct 2017 18:37:51 +0000 (14:37 -0400)]
Version 2.9.4
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Fri, 29 Sep 2017 20:40:36 +0000 (16:40 -0400)]
Fix: version check error in btrfs instrumentation
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 20 Sep 2017 16:12:41 +0000 (12:12 -0400)]
Fix: update btrfs instrumentation for kernel 4.14
See upstream commit:
Author: Jeff Mahoney <jeffm@suse.com>
Date: Wed Jun 28 21:56:54 2017 -0600
btrfs: constify tracepoint arguments
Tracepoint arguments are all read-only. If we mark the arguments
as const, we're able to keep or convert those arguments to const
where appropriate.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 20 Sep 2017 16:12:40 +0000 (12:12 -0400)]
Fix: update writeback instrumentation for kernel 4.14
See upstream commits:
commit
11fb998986a72aa7e997d96d63d52582a01228c5
Author: Mel Gorman <mgorman@techsingularity.net>
Date: Thu Jul 28 15:46:20 2016 -0700
mm: move most file-based accounting to the node
There are now a number of accounting oddities such as mapped file pages
being accounted for on the node while the total number of file pages are
accounted on the zone. This can be coped with to some extent but it's
confusing so this patch moves the relevant file-based accounted. Due to
throttling logic in the page allocator for reliable OOM detection, it is
still necessary to track dirty and writeback pages on a per-zone basis.
commit
c4a25635b60d08853a3e4eaae3ab34419a36cfa2
Author: Mel Gorman <mgorman@techsingularity.net>
Date: Thu Jul 28 15:46:23 2016 -0700
mm: move vmscan writes and file write accounting to the node
As reclaim is now node-based, it follows that page write activity due to
page reclaim should also be accounted for on the node. For consistency,
also account page writes and page dirtying on a per-node basis.
After this patch, there are a few remaining zone counters that may appear
strange but are fine. NUMA stats are still per-zone as this is a
user-space interface that tools consume. NR_MLOCK, NR_SLAB_*,
NR_PAGETABLE, NR_KERNEL_STACK and NR_BOUNCE are all allocations that
potentially pin low memory and cannot trivially be reclaimed on demand.
This information is still useful for debugging a page allocation failure
warning.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 20 Sep 2017 16:12:39 +0000 (12:12 -0400)]
Fix: update block instrumentation for kernel 4.14
See upstream commit:
commit
74d46992e0d9dee7f1f376de0d56d31614c8a17a
Author: Christoph Hellwig <hch@lst.de>
Date: Wed Aug 23 19:10:32 2017 +0200
block: replace bi_bdev with a gendisk pointer and partitions index
This way we don't need a block_device structure to submit I/O. The
block_device has different life time rules from the gendisk and
request_queue and is usually only available when the block device node
is open. Other callers need to explicitly create one (e.g. the lightnvm
passthrough code, or the new nvme multipathing code).
For the actual I/O path all that we need is the gendisk, which exists
once per block device. But given that the block layer also does
partition remapping we additionally need a partition index, which is
used for said remapping in generic_make_request.
Note that all the block drivers generally want request_queue or
sometimes the gendisk, so this removes a layer of indirection all
over the stack.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 26 Sep 2017 18:16:47 +0000 (14:16 -0400)]
Fix: vmalloc wrapper on kernel < 2.6.38
Ensure that all probes end up including the vmalloc wrapper through the
lttng-tracer.h header so the trace_*() static inlines are generated
through inclusion of include/trace/events/kmem.h before we define
CREATE_TRACE_POINTS.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 26 Sep 2017 17:46:30 +0000 (13:46 -0400)]
Fix: vmalloc wrapper on kernel >= 4.12
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Mon, 25 Sep 2017 14:56:20 +0000 (10:56 -0400)]
Add kmalloc failover to vmalloc
This patch is based on the kvmalloc helpers introduced in kernel 4.12.
It will gracefully failover memory allocations of more than one page to
vmalloc for systems under high memory pressure or fragmentation.
See Linux kernel commit:
commit
a7c3e901a46ff54c016d040847eda598a9e3e653
Author: Michal Hocko <mhocko@suse.com>
Date: Mon May 8 15:57:09 2017 -0700
mm: introduce kv[mz]alloc helpers
Patch series "kvmalloc", v5.
There are many open coded kmalloc with vmalloc fallback instances in the
tree. Most of them are not careful enough or simply do not care about
the underlying semantic of the kmalloc/page allocator which means that
a) some vmalloc fallbacks are basically unreachable because the kmalloc
part will keep retrying until it succeeds b) the page allocator can
invoke a really disruptive steps like the OOM killer to move forward
which doesn't sound appropriate when we consider that the vmalloc
fallback is available.
As it can be seen implementing kvmalloc requires quite an intimate
knowledge if the page allocator and the memory reclaim internals which
strongly suggests that a helper should be implemented in the memory
subsystem proper.
Most callers, I could find, have been converted to use the helper
instead. This is patch 6. There are some more relying on __GFP_REPEAT
in the networking stack which I have converted as well and Eric Dumazet
was not opposed [2] to convert them as well.
[1] http://lkml.kernel.org/r/
20170130094940.13546-1-mhocko@kernel.org
[2] http://lkml.kernel.org/r/
1485273626.16328.301.camel@edumazet-glaptop3.roam.corp.google.com
This patch (of 9):
Using kmalloc with the vmalloc fallback for larger allocations is a
common pattern in the kernel code. Yet we do not have any common helper
for that and so users have invented their own helpers. Some of them are
really creative when doing so. Let's just add kv[mz]alloc and make sure
it is implemented properly. This implementation makes sure to not make
a large memory pressure for > PAGE_SZE requests (__GFP_NORETRY) and also
to not warn about allocation failures. This also rules out the OOM
killer as the vmalloc is a more approapriate fallback than a disruptive
user visible action.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 19 Sep 2017 16:16:58 +0000 (12:16 -0400)]
Fix: mmap: caches aliased on virtual addresses
Some architectures (e.g. implementations of arm64) implement their
caches based on the virtual addresses (rather than physical address).
It has the upside of making the cache access faster (no TLB lookup
required to access the cache line), but the downside of requiring
virtual mappings (e.g. kernel vs user-space) to be aligned on the number
of bits used for cache aliasing.
Perform dcache flushing for the entire sub-buffer in the get_subbuf
operation on those architectures, thus ensuring we don't end up with
cache aliasing issues.
An alternative approach we could eventually take would be to create a
kernel mapping for the ring buffer that is aligned with the user-space
mapping.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Mon, 21 Aug 2017 18:47:08 +0000 (14:47 -0400)]
Fix: update ext4 instrumentation for kernel 4.13
See this upstream commit :
commit
a627b0a7c15ee4d2c87a86d5be5c8167382e8d0d
Author: Eric Whitney <enwlinux@gmail.com>
Date: Sun Jul 30 22:30:11 2017 -0400
ext4: remove unused metadata accounting variables
Two variables in ext4_inode_info, i_reserved_meta_blocks and
i_allocated_meta_blocks, are unused. Removing them saves a little
memory per in-memory inode and cleans up clutter in several tracepoints.
Adjust tracepoint output from ext4_alloc_da_blocks() for consistency
and fix a typo and whitespace near these changes.
Signed-off-by: Eric Whitney <enwlinux@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 21 Jul 2017 12:22:04 +0000 (08:22 -0400)]
Fix: Sleeping function called from invalid context
It affects system call instrumentation for accept, accept4 and connect,
only on the x86-64 architecture.
We need to use the LTTng accessing functions to touch user-space memory,
which take care of disabling the page fault handler, so we don't preempt
while in preempt-off context (tracepoints disable preemption).
Fixes #1111
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Mon, 10 Jul 2017 22:13:11 +0000 (18:13 -0400)]
Fix: sched for v4.11.5-rt1
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Fri, 23 Jun 2017 18:29:42 +0000 (14:29 -0400)]
Fix: handle missing ftrace header on v4.12
Properly handle the case where we build against the distro headers of a
kernel >= 4.12 and ftrace is enabled but the private header is
unavailable.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 12 Jun 2017 21:47:32 +0000 (17:47 -0400)]
Version 2.9.3
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 1 Jun 2017 18:24:11 +0000 (14:24 -0400)]
Fix: pid tracker should track "pgid"
The "pid" notion exposed by LTTng translates to the "pgid" notion in the
Linux kernel. Therefore using "current->pid" as argument to the PID
tracker actually ends up behaving as a "tid" tracker, which does not
match the intent nor the user-space tracer behavior.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 30 May 2017 13:36:31 +0000 (09:36 -0400)]
Fix: Build ftrace probe on kernels prior to 4.12
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Thu, 25 May 2017 20:56:52 +0000 (16:56 -0400)]
Fix: update ftrace probe for kernel 4.12
Follow changes introduced by Linux upstream commits:
ec19b85913486993d7d6f747beed1a711afd47d8
bca6c8d0480a8aa5c86f8f416db96c71f6b79e29
b5f081b563a6cdcb85a543df8c851951a8978275
6e4443199e5354255e8a4c1e8e5cfc8ef064c3ce
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Thu, 25 May 2017 20:56:51 +0000 (16:56 -0400)]
Fix: update block instrumentation for kernel 4.12
Follow changes introduced by Linux upstream commits:
48b77ad6084481ef9330a5d2bee289966da0975b
cee4b7ce3f9161c88f7255a3d73c1c4d5bbabea7
caf7df12272118e0274c8353bcfeaf60c7743a47
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 24 May 2017 15:19:50 +0000 (11:19 -0400)]
Fix: Add support for 4.9.27-rt18 kernel
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 23 May 2017 19:46:41 +0000 (15:46 -0400)]
Fix: update btrfs instrumentation for kernel 4.12
See upstream commit
490b54d6fb75f6ffd0471ec58bb38a992e2b40cd
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 23 May 2017 19:45:47 +0000 (15:45 -0400)]
Fix: update ringbuffer for kernel 4.12
flags removed from splice_pipe_desc in 4.12.
See upstream commit
f81dc7d7d5a2528f98f26a0b9406e822d0b35011
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 23 May 2017 19:45:18 +0000 (15:45 -0400)]
Fix: update sched instrumentation for kernel 4.12
See upstream commit
b91473ff6e979c0028f02f90e40c844959c736d8
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Tue, 23 May 2017 19:43:25 +0000 (15:43 -0400)]
Fix: ext3 was completely removed from the kernel in v4.3
Don't display the warning about missing ext3 headers on kernels >= 4.3
See upstream commit
e31fb9e00543e5d3c5b686747d3c862bc09b59f3
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Wed, 17 May 2017 21:09:12 +0000 (17:09 -0400)]
Fix: NULL pointer dereference of THIS_MODULE with built-in modules
THIS MODULE is defined to 0 when a module is built-in the kernel [1].
This caused NULL pointer dereference when booting a kernel with the
lttng-modules built-in.
To fix this issue, add #if guard around the wrapper_lttng_fixup_sig
function checking if the MODULE macro is defined to confirm that this
piece of code will end up in a module and not in the kernel itself.
[1]: linux/include/linux/export.h:32
Fixes: #1107
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 11 May 2017 20:50:50 +0000 (16:50 -0400)]
Fix: add "flush empty" ioctl for stream intersection
Changing the behavior of the "snapshot" lttng command to implicitly do a
buffer "flush" (even when current packet is empty) had unwanted
side-effects: for instance, the snapshot ABI is used by the live timer
to grab the buffer positions, and we don't want to generate useless
empty packets in that scenario.
Therefore, add the "flush empty" behavior as a new ioctl to the ring
buffer. This allows lttng-tools to perform buffer flush (even for empty
packets) when it needs to. Given that this new ioctl is added within
stable branches as well, lttng-tools always need to handle "-ENOSYS"
gracefully.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 11 May 2017 20:42:46 +0000 (16:42 -0400)]
Revert "Fix: flush empty packets on snapshot channel"
This reverts commit
dc5cd5702b74d72f0db0141c6d888a1d820aed9c.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 11 May 2017 20:42:34 +0000 (16:42 -0400)]
Revert "Fix: don't perform extra flush on metadata channel"
This reverts commit
7cf44d034bdda1896f6b0c6374c90c06d45ee4fd.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Fri, 5 May 2017 16:08:07 +0000 (12:08 -0400)]
Fix: remove CONFIG_KALLSYMS_ALL warning on clean
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 25 Apr 2017 13:56:53 +0000 (09:56 -0400)]
Version 2.9.2
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Thu, 20 Apr 2017 19:23:25 +0000 (15:23 -0400)]
Fix: Always build vmscan probe
The mm/vmscan.c compile unit is a obj-y, even on an old 2.6.36 kernel,
always build the vmscan probe regardless of kernel configuration.
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Wed, 8 Mar 2017 19:32:31 +0000 (14:32 -0500)]
Fix: use of uninitialized ret value in lttng_abi_open_metadata_stream
Fixes the following compiler warning:
lttng-abi.c: In function ‘lttng_metadata_ioctl’:
lttng-abi.c:971:6: warning: ‘ret’ may be used uninitialized in this function [-Wmaybe-uninitialized]
int ret;
^
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Wed, 8 Mar 2017 04:37:30 +0000 (23:37 -0500)]
Fix: kref changes for kernel 4.11
The underlying type of `struct kref` changed in kernel 4.11 from an
atomic_t to a refcount_t. This change was introduced in kernel
commit:
10383ae. This commit also added a builtin overflow checks to
`kref_get()` so we use it.
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Wed, 8 Mar 2017 16:50:38 +0000 (11:50 -0500)]
Fix: atomic_add_unless() returns true/false rather than prior value
The previous implementation assumed that `atomic_add_unless` returned
the prior value of the atomic counter when in fact it returned if the
addition was performed (true) or not performed (false).
Since `atomic_add_unless` can not return INT_MAX, the `lttng_kref_get`
always returned that the call was successful.
This issue had a low likelihood of being triggered since the two refcounts
of the counters used with this call are both bounded by the maximum
number of file descriptors on the system.
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 7 Mar 2017 16:21:59 +0000 (11:21 -0500)]
Fix: timers cputime_t arguments replaced by ull in kernel 4.11
cputime_t was changed to ull in the kernel commit:
858cf3a
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 7 Mar 2017 16:16:47 +0000 (11:16 -0500)]
Fix: update scsi instrumentation for kernel 4.11
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 7 Mar 2017 15:35:21 +0000 (10:35 -0500)]
Fix: changes to the vm_op fault cb prototype in libringbuffer
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 7 Mar 2017 15:14:19 +0000 (10:14 -0500)]
Fix: update btrfs instrumentation for kernel 4.11
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 7 Mar 2017 14:48:08 +0000 (09:48 -0500)]
Fix: update mm_vmscan instrumentation for kernel 4.11
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Francis Deslauriers [Tue, 7 Mar 2017 14:12:31 +0000 (09:12 -0500)]
Fix: section mismatch warning caused by __exit annotation
lttng_logger_exit is used in a non-exit function so it should not be
annotated with `__exit`.
Signed-off-by: Francis Deslauriers <francis.deslauriers@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 17 Feb 2017 17:16:41 +0000 (12:16 -0500)]
Version 2.9.1
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 10 Feb 2017 01:46:44 +0000 (20:46 -0500)]
Fix: nmi-safe clock on 32-bit systems
On 32-bit systems, the algorithm within lttng-modules that ensures the
nmi-safe clock increases monotonically on a CPU assumes to have one
clock read per 32-bit LSB overflow period, which is not guaranteed. It
also has an issue on the first clock reads after module load, because
the initial value for the last LSB is 0. It can cause the time to stay
stuck at the same value for a few seconds at the beginning of the trace,
which is unfortunate for the first trace after module load, because this
is where the offset between realtime and trace_clock is sampled, which
prevents correlation of kernel and user-space traces for that session.
It only affects 32-bit systems with kernels >= 3.17.
Fix this by using the non-nmi-safe clock source on 32-bit systems.
While we are there, remove an implementation-defined c99 behavior
regarding casting u64 to long by using unsigned arithmetic instead:
turn:
if (((long) now - (long) last) < 0)
into:
if (U64_MAX / 2 < now - last)
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 23 Jan 2017 20:16:22 +0000 (15:16 -0500)]
Fix: only include linux/cpuhotplug.h for kernels >= 4.10
Kernels at least <= 4.4 did not have this header file.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 23 Jan 2017 17:34:07 +0000 (12:34 -0500)]
Fix: 4.10 hotplug adaptation backward compat
from /home/compudj/git/lttng-modules/lttng-context-perf-counters.c:23:
/home/compudj/git/lttng-modules/lttng-context-perf-counters.c: In function ‘lttng_add_perf_counter_to_ctx’:
/home/compudj/git/lttng-modules/lttng-context-perf-counters.c:353:22: error: ‘cpu’ undeclared (first use in this function)
for_each_online_cpu(cpu) {
^
./include/linux/cpumask.h:223:8: note: in definition of macro ‘for_each_cpu’
for ((cpu) = -1; \
^
/home/compudj/git/lttng-modules/lttng-context-perf-counters.c:353:2: note: in expansion of macro ‘for_each_online_cpu’
for_each_online_cpu(cpu) {
^
/home/compudj/git/lttng-modules/lttng-context-perf-counters.c:353:22: note: each undeclared identifier is reported only once for each function it appears in
for_each_online_cpu(cpu) {
^
./include/linux/cpumask.h:223:8: note: in definition of macro ‘for_each_cpu’
for ((cpu) = -1; \
^
/home/compudj/git/lttng-modules/lttng-context-perf-counters.c:353:2: note: in expansion of macro ‘for_each_online_cpu’
for_each_online_cpu(cpu) {
^
./include/linux/cpumask.h:224:38: warning: left-hand operand of comma expression has no effect [-Wunused-value]
(cpu) = cpumask_next((cpu), (mask)), \
^
./include/linux/cpumask.h:717:36: note: in expansion of macro ‘for_each_cpu’
#define for_each_online_cpu(cpu) for_each_cpu((cpu), cpu_online_mask)
^
/home/compudj/git/lttng-modules/lttng-context-perf-counters.c:353:2: note: in expansion of macro ‘for_each_online_cpu’
for_each_online_cpu(cpu) {
^
scripts/Makefile.build:289: recipe for target '/home/compudj/git/lttng-modules/lttng-context-perf-counters.o' failed
make[2]: *** [/home/compudj/git/lttng-modules/lttng-context-perf-counters.o] Error 1
make[2]: *** Waiting for unfinished jobs....
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 10 Jan 2017 16:19:51 +0000 (11:19 -0500)]
Adapt lttng-modules to Linux 4.10 cpu hotplug state machine
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 23 Jan 2017 17:32:17 +0000 (12:32 -0500)]
Fix: 4.10 btrfs instrumentation update backward compat
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 23 Jan 2017 17:18:35 +0000 (12:18 -0500)]
Update btrfs instrumentation for 4.10 kernel
Based on commit
92a1bf76 "Btrfs: add 'inode' for extent map tracepoint"
in the upstream Linux kernel.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 10 Jan 2017 16:41:11 +0000 (11:41 -0500)]
btrfs instrumentation: update to 4.10 kernel
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 10 Jan 2017 16:29:49 +0000 (11:29 -0500)]
timer instrumentation: adapt to ktime_t without union
Introduced in Linux upstream in 4.10.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 7 Dec 2016 19:17:33 +0000 (14:17 -0500)]
Fix: asoc instrumentation for RHEL 7.3
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 7 Dec 2016 16:09:31 +0000 (11:09 -0500)]
Fix: SCSI instrumentation for SLES12 SP2
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Michael Jeanson [Wed, 7 Dec 2016 16:09:30 +0000 (11:09 -0500)]
Add SUSE Linux Enterprise kernel version tests
Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 29 Nov 2016 22:26:02 +0000 (17:26 -0500)]
Version 2.9.0
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 21 Nov 2016 21:08:22 +0000 (16:08 -0500)]
Performance: add missing unlikely in reserve
Add missing branch prediction hints within lttng_event_reserve().
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 24 Oct 2016 17:27:01 +0000 (13:27 -0400)]
Fix: preemptible and migratable context error handling
When built against preempt-rt and preempt kernels, the "return 0" case
means success, but lttng-modules incorrectly prints an error in the
kernel log.
Given that we handle the -ENOSYS error in lttng_context_init, there is
no need to keep the ifdefs in that function.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 13 Oct 2016 14:18:07 +0000 (16:18 +0200)]
Version 2.9.0-rc2
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 13 Oct 2016 13:50:21 +0000 (15:50 +0200)]
Fix: bump stable kernel version ranges for clock work-around
Linux commit
27727df240c7 ("Avoid taking lock in NMI path with
CONFIG_DEBUG_TIMEKEEPING"), changed the logic to open-code
the timekeeping_get_ns() function, but forgot to include
the unit conversion from cycles to nanoseconds, breaking the
function's output, which impacts LTTng.
We expected Linux commit
58bfea9532 "timekeeping: Fix
__ktime_get_fast_ns() regression" to make its way into stable
kernels promptly, but it appears new stable kernel releases were
done before the fix was cherry-picked from the master branch.
We therefore need to bump the version ranges for the work-around
in lttng-modules.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
CC: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CC: John Stultz <john.stultz@linaro.org>
Mathieu Desnoyers [Fri, 7 Oct 2016 19:19:52 +0000 (15:19 -0400)]
Version 2.9.0-rc1
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 7 Oct 2016 14:55:16 +0000 (10:55 -0400)]
Fix: i2c: support kernels < 3.15
i2c instrumentation has only been added in kernel 3.15.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 6 Oct 2016 11:45:35 +0000 (07:45 -0400)]
Fix: show warning for broken clock work-around
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Wed, 5 Oct 2016 16:47:58 +0000 (12:47 -0400)]
Bump minor ABI version
Command added: LTTNG_KERNEL_SESSION_STATEDUMP
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Wed, 5 Oct 2016 11:20:32 +0000 (07:20 -0400)]
Fix: work-around upstream Linux timekeeping bug
Linux commit
27727df240c7 ("Avoid taking lock in NMI path with
CONFIG_DEBUG_TIMEKEEPING"), changed the logic to open-code
the timekeeping_get_ns() function, but forgot to include
the unit conversion from cycles to nanoseconds, breaking the
function's output, which impacts LTTng.
The following kernel versions are affected: 4.8, 4.7.4+, 4.4.20+,
4.1.32+
We expect that the upstream fix will reach the master and stable
branches timely before the next releases, so we use 4.8.1, 4.7.7,
4.4.24, and 4.1.34 as upper bounds (exclusive).
Fall-back to the non-NMI-safe trace clock for those kernel versions.
We simply discard events from NMI context with a in_nmi() check,
as we did before Linux 3.17.
Link: http://lkml.kernel.org/r/1475636148-26539-1-git-send-email-john.stultz@linaro.org
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Simon Marchi [Tue, 4 Oct 2016 21:07:05 +0000 (17:07 -0400)]
Add support for i2c tracepoints
This patch teaches lttng-modules about the i2c tracepoints in the Linux
kernel.
It contains the following tracepoints:
* i2c_write
* i2c_read
* i2c_reply
* i2c_result
I translated the fields and assignments from the kernel's
include/trace/events/i2c.h as well as I could. I also tried building
this module against a kernel without CONFIG_I2C, and it built fine (the
required types are unconditionally defined). So I don't think any "#if
CONFIG_I2C" or similar are required.
A module parameter (extract_sensitive_payload) controls the extraction
of possibly sensitive data from events.
[ With edit by Mathieu Desnoyers. ]
Signed-off-by: Simon Marchi <simon.marchi@ericsson.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 3 Oct 2016 21:35:27 +0000 (17:35 -0400)]
Cleanup: makefile version checks with single "ge"
Version checks in makefiles should always be a disjunctive normal form
where the conjunctions consist of one or more "equals" comparisons and
at most a single greater-or-equal comparison.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Mon, 26 Sep 2016 17:37:50 +0000 (13:37 -0400)]
Performance: special-case NULL in lttng_strlen_user_inatomic
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 16:30:00 +0000 (12:30 -0400)]
Fix: lttng_inline_memcpy does not take a __user argument
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 16:27:01 +0000 (12:27 -0400)]
Performance: implement lttng_inline_memcpy
Because all length parameters received for serializing data coming from
applications go through a callback, they are never constant, and it
hurts performance to perform a call to memcpy each time.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 16:02:25 +0000 (12:02 -0400)]
Performance: cache the backend pages pointer in context
Getting the backend pages pointer requires pointer chasing through the
ring buffer backend tables. Cache the current value so it can be re-used
for all backend write operations writing fields for the same event.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 15:11:48 +0000 (11:11 -0400)]
Cleanup: libringbuffer: remove duplicate pointer chasing in slow paths
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 15:06:10 +0000 (11:06 -0400)]
Performance: Only dereference commit index once
The commit fast path should not dereference the commit counter index
repeatedly for performance reasons.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 15:01:29 +0000 (11:01 -0400)]
Performance: Mark channel and buffer event disabled check unlikely
Channel and buffer are typically enabled when tracing, else the probes
would not be connected. Ensure the compiler optimizes for the right
usual case.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 15:00:08 +0000 (11:00 -0400)]
Performance: Relax atomicity constraints for crash handling
Use a store rather than a cmpxchg() for the update of the
sequential commit counter. This speeds up commit. The downside
is that short race windows between the if() check to see if the
counter is larger than the new value and the update could result
in the counter going backwards, in unlikely preemption or signal
delivery scenarios.
Accept that we may lose a few events in a crash dump for the
benefit of tracing speed.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 14:46:44 +0000 (10:46 -0400)]
Performance: mark ring buffer do_copy callers always inline
The underlying copy operation is more efficient if the size is a
constant, which only happens if this function is inlined in the caller.
Otherwise, we end up calling memcpy for each field.
Force inlining for performance reasons for:
- lib_ring_buffer_do_strcpy,
- lib_ring_buffer_do_strcpy_from_user_inatomic,
- lib_ring_buffer_copy_from_user_inatomic.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Sun, 25 Sep 2016 14:43:22 +0000 (10:43 -0400)]
Performance: mark lib_ring_buffer_write always inline
The underlying copy operation is more efficient if the size is a
constant, which only happens if this function is inlined in the caller.
Otherwise, we end up calling memcpy for each field.
Force inlining for performance reasons.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 20 Sep 2016 14:29:33 +0000 (10:29 -0400)]
Kconfig: select TRACEPOINTS when built-in
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Fri, 9 Sep 2016 21:01:18 +0000 (17:01 -0400)]
Performance: disable event counting by default
Performance improvement changelog from lttng-ust, ported back to
lttng-modules:
Disable event counting in the ring buffer, which can count the number of
events produced per ring-buffer, as well as the number of events
overwritten in overwrite mode.
This feature is currently unused anyway: it is not saved in the ring
buffer header, nor made available to lttng-tools.
This saves 70 ns/event in lttng-ust on the ARM32 Cubietruck.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Thu, 1 Sep 2016 22:08:15 +0000 (18:08 -0400)]
Fix: handle large number of pages or subbuffers per buffer
Do no trigger kernel console warnings when we try to allocate too many
pages, or a too large kmalloc area for page array (within a subbuffer),
or a sub-buffer array (within a buffer).
Use vmalloc/vfree for the "pages" local variable used only during
allocation, which is an array of nr_subbuf * nr_pages_per_subbuf
pointers. This ensures we do not limit the overall buffer size due to
kmalloc limitations.
Fixes #1031
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Jonathan Rajotte [Thu, 1 Sep 2016 21:52:09 +0000 (17:52 -0400)]
Fix: unregister cpu hotplug notifier on buffer alloc error
The cpu hotplug notifier needs to be unregistered in the error path of
buffer allocation, else it eventually causes kernel OOPS when the kernel
accesses freed memory of the notifier block.
Fixes #1031
Signed-off-by: Jonathan Rajotte <jonathan.rajotte-julien@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Jérémie Galarneau [Wed, 31 Aug 2016 04:21:14 +0000 (00:21 -0400)]
Fix: sa_family is of type unsigned short
sa_family is an unsigned short in sockaddr definitions. For instance,
the kernel's unix_getname() function sets addrlen to sizeof(short) as it
only returns the socket's family.
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Jérémie Galarneau [Wed, 31 Aug 2016 04:18:29 +0000 (00:18 -0400)]
Fix: check for sizeof sa_family to save sa_family in accept and connect
The check of addrlen >= sizeof(struct sockaddr) is too restrictive
and causes sa_family to not be saved in the case of AF_UNIX sockets
as the addrlen returned by the syscall may be only sizeof(short).
Individual checks per socket family are performed anyhow in the
switch case, making this safe.
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 30 Aug 2016 22:42:10 +0000 (18:42 -0400)]
Fix: use printk_once() for wrapper warning messages
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 30 Aug 2016 21:32:35 +0000 (17:32 -0400)]
Fix: btrfs instrumentation for 4.8 kernel
Note: there are missing events.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 30 Aug 2016 21:13:37 +0000 (17:13 -0400)]
Fix: update mm_vmscan instrumentation for kernel 4.8
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 30 Aug 2016 21:02:09 +0000 (17:02 -0400)]
Fix: napi instrumentation for 4.8 kernels
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Mathieu Desnoyers [Tue, 30 Aug 2016 20:59:13 +0000 (16:59 -0400)]
Fix: update block instrumentation to compile on 4.8 kernels
These 2 upstream Linux commits break the block instrumentation in lttng
modules master:
https://github.com/torvalds/linux/commit/
4e1b2d52a80d79296a5d899d73249748dea71a53
https://github.com/torvalds/linux/commit/
288dab8a35a0bde426a09870943c8d3ee3a50dab
The errors are :
lttng-modules/probes/../instrumentation/events/lttng-module/block.h:35:13:
error: ‘REQ_DISCARD’ undeclared (first use in this function)
lttng-modules/probes/../instrumentation/events/lttng-module/block.h:41:14:
error: ‘REQ_SECURE’ undeclared (first use in this function)
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Jérémie Galarneau [Mon, 29 Aug 2016 20:02:10 +0000 (16:02 -0400)]
Cleanup: reuse code in accept() and accept4() instrumentation
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Jérémie Galarneau [Mon, 29 Aug 2016 20:02:09 +0000 (16:02 -0400)]
Add x86-64 override for accept4 syscall
This patch adds an instrumentation override for the accept4() syscall
which is almost identical to accept(), except for an additional
"flags" parameter.
A follow-up patch refactors both overrides to minimize code
duplication as is done for the select/pselect6 overrides.
Signed-off-by: Jérémie Galarneau <jeremie.galarneau@efficios.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
This page took 0.048564 seconds and 4 git commands to generate.