If the perf_event has PERF_SAMPLE_CALLCHAIN, BPF can use it for stack trace.
The problematic cases like PEBS and IBS already handled in the PMU driver and
they filled the callchain info in the sample data. For others, we can call
perf_callchain() before the BPF handler.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220908214104.3851807-2-namhyung@kernel.org
So that it can call perf_callchain() only if needed. Historically it used
__PERF_SAMPLE_CALLCHAIN_EARLY but we can do that with sample_flags in the
struct perf_sample_data.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220908214104.3851807-1-namhyung@kernel.org
All the fixed counters share a fixed control register. The current
perf reads and re-writes the fixed control register for each fixed
counter disable/enable, which is unnecessary.
When changing the fixed control register, the entire PMU must be
disabled via the global control register. The changing cannot be taken
effect until the entire PMU is re-enabled. Only updating the fixed
control register once right before the entire PMU re-enabling is
enough.
The read of the fixed control register is not necessary either. The
value can be cached in the per CPU cpu_hw_events.
Test results:
Counting all the fixed counters with the perf bench sched pipe as below
on a SPR machine.
$perf stat -e cycles,instructions,ref-cycles,slots --no-inherit --
taskset -c 1 perf bench sched pipe
The Total elapsed time reduces from 5.36s (without the patch) to 4.99s
(with the patch), which is ~6.9% improvement.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220804140729.2951259-1-kan.liang@linux.intel.com
Now that it is all internal to the intel driver, remove
x86_pmu::update_topdown_event.
Assumes that is_topdown_count(event) can only be true when the
hardware has topdown stuff and the function is set.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220829101321.771635301@infradead.org
While auditing 6b959ba22d ("perf/core: Fix reentry problem in
perf_output_read_group()") a few spots were found that wanted
assertions.
Notable for_each_sibling_event() relies on exclusion from
modification. This would normally be holding either ctx->lock or
ctx->mutex, however due to how things are constructed disabling IRQs
is a valid and sufficient substitute for ctx->lock.
Another possible site to add assertions would be the various
pmu::{add,del,read,..}() methods, but that's not trivially expressable
in C -- the best option is wrappers, but those are easy enough to
forget.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Ensure all platform specific event flags are within PERF_EVENT_FLAG_ARCH.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220907091924.439193-5-anshuman.khandual@arm.com
Ensure all platform specific event flags are within PERF_EVENT_FLAG_ARCH.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220907091924.439193-4-anshuman.khandual@arm.com
This just ensures that PERF_EVENT_FLAG_ARCH does not overlap with generic
hardware event flags.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220907091924.439193-3-anshuman.khandual@arm.com
Two hardware event flags on x86 platform has overshot PERF_EVENT_FLAG_ARCH
(0x0000ffff). These flags are PERF_X86_EVENT_PEBS_LAT_HYBRID (0x20000) and
PERF_X86_EVENT_AMD_BRS (0x10000). Lets expand PERF_EVENT_FLAG_ARCH mask to
accommodate those flags, and also create room for two more in the future.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220907091924.439193-2-anshuman.khandual@arm.com
Besides the branch type filtering requests, 'event.attr.branch_sample_type'
also contains various flags indicating which additional information should
be captured, along with the base branch record. These flags help configure
the underlying hardware, and capture the branch records appropriately when
required e.g after PMU interrupt. But first, this moves an existing helper
perf_sample_save_hw_index() into the header before adding some more helpers
for other branch sample filter flags.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220906084414.396220-1-anshuman.khandual@arm.com
Use the new sample_flags to indicate whether the txn field is filled by
the PMU driver.
Remove the txn field from the perf_sample_data_init() to minimize the
number of cache lines touched.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-7-kan.liang@linux.intel.com
Use the new sample_flags to indicate whether the data_src field is
filled by the PMU driver.
Remove the data_src field from the perf_sample_data_init() to minimize
the number of cache lines touched.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-6-kan.liang@linux.intel.com
Use the new sample_flags to indicate whether the weight field is filled
by the PMU driver.
Remove the weight field from the perf_sample_data_init() to minimize the
number of cache lines touched.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-5-kan.liang@linux.intel.com
Use the new sample_flags to indicate whether the branch stack is filled
by the PMU driver.
Remove the br_stack from the perf_sample_data_init() to minimize the number
of cache lines touched.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-4-kan.liang@linux.intel.com
The PEBS TSC-based timestamps do not appear correctly in the final
perf.data output file from perf record.
The data->time field setup by PEBS in the setup_pebs_fixed_sample_data()
is later overwritten by perf_events generic code in
perf_prepare_sample(). There is an ordering problem.
Set the sample flags when the data->time is updated by PEBS.
The data->time field will not be overwritten anymore.
Reported-by: Andreas Kogler <andreas.kogler.0x@gmail.com>
Reported-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-3-kan.liang@linux.intel.com
On some platforms, some data e.g., timestamps, can be retrieved from
the PMU driver. Usually, the data from the PMU driver is more accurate.
The current perf kernel should output the PMU-filled sample data if
it's available.
To check the availability of the PMU-filled sample data, the current
perf kernel initializes the related fields in the
perf_sample_data_init(). When outputting a sample, the perf checks
whether the field is updated by the PMU driver. If yes, the updated
value will be output. If not, the perf uses an SW way to calculate the
value or just outputs the initialized value if an SW way is unavailable
either.
With more and more data being provided by the PMU driver, more fields
has to be initialized in the perf_sample_data_init(). That will
increase the number of cache lines touched in perf_sample_data_init()
and be harmful to the performance.
Add new "sample_flags" to indicate the PMU-filled sample data. The PMU
driver should set the corresponding PERF_SAMPLE_ flag when the field is
updated. The initialization of the corresponding field is not required
anymore. The following patches will make use of it and remove the
corresponding fields from the perf_sample_data_init(), which will
further minimize the number of cache lines touched.
Only clear the sample flags that have already been done by the PMU
driver in the perf_prepare_sample() for the PERF_RECORD_SAMPLE. For the
other PERF_RECORD_ event type, the sample data is not available.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-2-kan.liang@linux.intel.com
We can still see that a majority of the time is spent hashing task pointers:
...
16.98% [kernel] [k] rhashtable_jhash2
...
Doing the bookkeeping in toggle_bp_slots() is currently O(#cpus),
calling task_bp_pinned() for each CPU, even if task_bp_pinned() is
CPU-independent. The reason for this is to update the per-CPU
'tsk_pinned' histogram.
To optimize the CPU-independent case to O(1), keep a separate
CPU-independent 'tsk_pinned_all' histogram.
The major source of complexity are transitions between "all
CPU-independent task breakpoints" and "mixed CPU-independent and
CPU-dependent task breakpoints". The code comments list all cases that
require handling.
After this optimization:
| $> perf bench -r 100 breakpoint thread -b 4 -p 128 -t 512
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 100 threads with 4 breakpoints and 128 parallelism
| Total time: 1.758 [sec]
|
| 34.336621 usecs/op
| 4395.087500 usecs/op/cpu
38.08% [kernel] [k] queued_spin_lock_slowpath
10.81% [kernel] [k] smp_cfm_core_cond
3.01% [kernel] [k] update_sg_lb_stats
2.58% [kernel] [k] osq_lock
2.57% [kernel] [k] llist_reverse_order
1.45% [kernel] [k] find_next_bit
1.21% [kernel] [k] flush_tlb_func_common
1.01% [kernel] [k] arch_install_hw_breakpoint
Showing that the time spent hashing keys has become insignificant.
With the given benchmark parameters, that's an improvement of 12%
compared with the old O(#cpus) version.
And finally, using the less aggressive parameters from the preceding
changes, we now observe:
| $> perf bench -r 30 breakpoint thread -b 4 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 4 breakpoints and 64 parallelism
| Total time: 0.067 [sec]
|
| 35.292187 usecs/op
| 2258.700000 usecs/op/cpu
Which is an improvement of 12% compared to without the histogram
optimizations (baseline is 40 usecs/op). This is now on par with the
theoretical ideal (constraints disabled), and only 12% slower than no
breakpoints at all.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-15-elver@google.com
Running the perf benchmark with (note: more aggressive parameters vs.
preceding changes, but same 256 CPUs host):
| $> perf bench -r 100 breakpoint thread -b 4 -p 128 -t 512
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 100 threads with 4 breakpoints and 128 parallelism
| Total time: 1.989 [sec]
|
| 38.854160 usecs/op
| 4973.332500 usecs/op/cpu
20.43% [kernel] [k] queued_spin_lock_slowpath
18.75% [kernel] [k] osq_lock
16.98% [kernel] [k] rhashtable_jhash2
8.34% [kernel] [k] task_bp_pinned
4.23% [kernel] [k] smp_cfm_core_cond
3.65% [kernel] [k] bcmp
2.83% [kernel] [k] toggle_bp_slot
1.87% [kernel] [k] find_next_bit
1.49% [kernel] [k] __reserve_bp_slot
We can see that a majority of the time is now spent hashing task
pointers to index into task_bps_ht in task_bp_pinned().
Obtaining the max_bp_pinned_slots() for CPU-independent task targets
currently is O(#cpus), and calls task_bp_pinned() for each CPU, even if
the result of task_bp_pinned() is CPU-independent.
The loop in max_bp_pinned_slots() wants to compute the maximum slots
across all CPUs. If task_bp_pinned() is CPU-independent, we can do so by
obtaining the max slots across all CPUs and adding task_bp_pinned().
To do so in O(1), use a bp_slots_histogram for CPU-pinned slots.
After this optimization:
| $> perf bench -r 100 breakpoint thread -b 4 -p 128 -t 512
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 100 threads with 4 breakpoints and 128 parallelism
| Total time: 1.930 [sec]
|
| 37.697832 usecs/op
| 4825.322500 usecs/op/cpu
19.13% [kernel] [k] queued_spin_lock_slowpath
18.21% [kernel] [k] rhashtable_jhash2
15.46% [kernel] [k] osq_lock
6.27% [kernel] [k] toggle_bp_slot
5.91% [kernel] [k] task_bp_pinned
5.05% [kernel] [k] smp_cfm_core_cond
1.78% [kernel] [k] update_sg_lb_stats
1.36% [kernel] [k] llist_reverse_order
1.34% [kernel] [k] find_next_bit
1.19% [kernel] [k] bcmp
Suggesting that time spent in task_bp_pinned() has been reduced.
However, we're still hashing too much, which will be addressed in the
subsequent change.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-14-elver@google.com
Factor out the existing `atomic_t count[N]` into its own struct called
'bp_slots_histogram', to generalize and make its intent clearer in
preparation of reusing elsewhere. The basic idea of bucketing "total
uses of N slots" resembles a histogram, so calling it such seems most
intuitive.
No functional change.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-13-elver@google.com
While optimizing task_bp_pinned()'s runtime complexity to O(1) on
average helps reduce time spent in the critical section, we still suffer
due to serializing everything via 'nr_bp_mutex'. Indeed, a profile shows
that now contention is the biggest issue:
95.93% [kernel] [k] osq_lock
0.70% [kernel] [k] mutex_spin_on_owner
0.22% [kernel] [k] smp_cfm_core_cond
0.18% [kernel] [k] task_bp_pinned
0.18% [kernel] [k] rhashtable_jhash2
0.15% [kernel] [k] queued_spin_lock_slowpath
when running the breakpoint benchmark with (system with 256 CPUs):
| $> perf bench -r 30 breakpoint thread -b 4 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 4 breakpoints and 64 parallelism
| Total time: 0.207 [sec]
|
| 108.267188 usecs/op
| 6929.100000 usecs/op/cpu
The main concern for synchronizing the breakpoint constraints data is
that a consistent snapshot of the per-CPU and per-task data is observed.
The access pattern is as follows:
1. If the target is a task: the task's pinned breakpoints are counted,
checked for space, and then appended to; only bp_cpuinfo::cpu_pinned
is used to check for conflicts with CPU-only breakpoints;
bp_cpuinfo::tsk_pinned are incremented/decremented, but otherwise
unused.
2. If the target is a CPU: bp_cpuinfo::cpu_pinned are counted, along
with bp_cpuinfo::tsk_pinned; after a successful check, cpu_pinned is
incremented. No per-task breakpoints are checked.
Since rhltable safely synchronizes insertions/deletions, we can allow
concurrency as follows:
1. If the target is a task: independent tasks may update and check the
constraints concurrently, but same-task target calls need to be
serialized; since bp_cpuinfo::tsk_pinned is only updated, but not
checked, these modifications can happen concurrently by switching
tsk_pinned to atomic_t.
2. If the target is a CPU: access to the per-CPU constraints needs to
be serialized with other CPU-target and task-target callers (to
stabilize the bp_cpuinfo::tsk_pinned snapshot).
We can allow the above concurrency by introducing a per-CPU constraints
data reader-writer lock (bp_cpuinfo_sem), and per-task mutexes (reuses
task_struct::perf_event_mutex):
1. If the target is a task: acquires perf_event_mutex, and acquires
bp_cpuinfo_sem as a reader. The choice of percpu-rwsem minimizes
contention in the presence of many read-lock but few write-lock
acquisitions: we assume many orders of magnitude more task target
breakpoints creations/destructions than CPU target breakpoints.
2. If the target is a CPU: acquires bp_cpuinfo_sem as a writer.
With these changes, contention with thousands of tasks is reduced to the
point where waiting on locking no longer dominates the profile:
| $> perf bench -r 30 breakpoint thread -b 4 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 4 breakpoints and 64 parallelism
| Total time: 0.077 [sec]
|
| 40.201563 usecs/op
| 2572.900000 usecs/op/cpu
21.54% [kernel] [k] task_bp_pinned
20.18% [kernel] [k] rhashtable_jhash2
6.81% [kernel] [k] toggle_bp_slot
5.47% [kernel] [k] queued_spin_lock_slowpath
3.75% [kernel] [k] smp_cfm_core_cond
3.48% [kernel] [k] bcmp
On this particular setup that's a speedup of 2.7x.
We're also getting closer to the theoretical ideal performance through
optimizations in hw_breakpoint.c -- constraints accounting disabled:
| perf bench -r 30 breakpoint thread -b 4 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 4 breakpoints and 64 parallelism
| Total time: 0.067 [sec]
|
| 35.286458 usecs/op
| 2258.333333 usecs/op/cpu
Which means the current implementation is ~12% slower than the
theoretical ideal.
For reference, performance without any breakpoints:
| $> bench -r 30 breakpoint thread -b 0 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 0 breakpoints and 64 parallelism
| Total time: 0.060 [sec]
|
| 31.365625 usecs/op
| 2007.400000 usecs/op/cpu
On a system with 256 CPUs, the theoretical ideal is only ~12% slower
than no breakpoints at all; the current implementation is ~28% slower.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-12-elver@google.com
Internal data structures (cpu_bps, task_bps) of powerpc's hw_breakpoint
implementation have relied on nr_bp_mutex serializing access to them.
Before overhauling synchronization of kernel/events/hw_breakpoint.c,
introduce 2 spinlocks to synchronize cpu_bps and task_bps respectively,
thus avoiding reliance on callers synchronizing powerpc's hw_breakpoint.
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-10-elver@google.com
Flexible breakpoints have never been implemented, with
bp_cpuinfo::flexible always being 0. Unfortunately, they still occupy 4
bytes in each bp_cpuinfo and bp_busy_slots, as well as computing the max
flexible count in fetch_bp_busy_slots().
This again causes suboptimal code generation, when we always know that
`!!slots.flexible` will be 0.
Just get rid of the flexible "placeholder" and remove all real code
related to it. Make a note in the comment related to the constraints
algorithm but don't remove them from the algorithm, so that if in future
flexible breakpoints need supporting, it should be trivial to revive
them (along with reverting this change).
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-9-elver@google.com
Due to being a __weak function, hw_breakpoint_weight() will cause the
compiler to always emit a call to it. This generates unnecessarily bad
code (register spills etc.) for no good reason; in fact it appears in
profiles of `perf bench -r 100 breakpoint thread -b 4 -p 128 -t 512`:
...
0.70% [kernel] [k] hw_breakpoint_weight
...
While a small percentage, no architecture defines its own
hw_breakpoint_weight() nor are there users outside hw_breakpoint.c,
which makes the fact it is currently __weak a poor choice.
Change hw_breakpoint_weight()'s definition to follow a similar protocol
to hw_breakpoint_slots(), such that if <asm/hw_breakpoint.h> defines
hw_breakpoint_weight(), we'll use it instead.
The result is that it is inlined and no longer shows up in profiles.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-8-elver@google.com
Optimize internal hw_breakpoint state if the architecture's number of
breakpoint slots is constant. This avoids several kmalloc() calls and
potentially unnecessary failures if the allocations fail, as well as
subtly improves code generation and cache locality.
The protocol is that if an architecture defines hw_breakpoint_slots via
the preprocessor, it must be constant and the same for all types.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-7-elver@google.com
Mark read-only data after initialization as __ro_after_init.
While we are here, turn 'constraints_initialized' into a bool.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-6-elver@google.com
On a machine with 256 CPUs, running the recently added perf breakpoint
benchmark results in:
| $> perf bench -r 30 breakpoint thread -b 4 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 4 breakpoints and 64 parallelism
| Total time: 236.418 [sec]
|
| 123134.794271 usecs/op
| 7880626.833333 usecs/op/cpu
The benchmark tests inherited breakpoint perf events across many
threads.
Looking at a perf profile, we can see that the majority of the time is
spent in various hw_breakpoint.c functions, which execute within the
'nr_bp_mutex' critical sections which then results in contention on that
mutex as well:
37.27% [kernel] [k] osq_lock
34.92% [kernel] [k] mutex_spin_on_owner
12.15% [kernel] [k] toggle_bp_slot
11.90% [kernel] [k] __reserve_bp_slot
The culprit here is task_bp_pinned(), which has a runtime complexity of
O(#tasks) due to storing all task breakpoints in the same list and
iterating through that list looking for a matching task. Clearly, this
does not scale to thousands of tasks.
Instead, make use of the "rhashtable" variant "rhltable" which stores
multiple items with the same key in a list. This results in average
runtime complexity of O(1) for task_bp_pinned().
With the optimization, the benchmark shows:
| $> perf bench -r 30 breakpoint thread -b 4 -p 64 -t 64
| # Running 'breakpoint/thread' benchmark:
| # Created/joined 30 threads with 4 breakpoints and 64 parallelism
| Total time: 0.208 [sec]
|
| 108.422396 usecs/op
| 6939.033333 usecs/op/cpu
On this particular setup that's a speedup of ~1135x.
While one option would be to make task_struct a breakpoint list node,
this would only further bloat task_struct for infrequently used data.
Furthermore, after all optimizations in this series, there's no evidence
it would result in better performance: later optimizations make the time
spent looking up entries in the hash table negligible (we'll reach the
theoretical ideal performance i.e. no constraints).
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-5-elver@google.com
Clean up headers:
- Remove unused <linux/kallsyms.h>
- Remove unused <linux/kprobes.h>
- Remove unused <linux/module.h>
- Remove unused <linux/smp.h>
- Add <linux/export.h> for EXPORT_SYMBOL_GPL().
- Add <linux/mutex.h> for mutex.
- Sort alphabetically.
- Move <linux/hw_breakpoint.h> to top to test it compiles on its own.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-4-elver@google.com
Provide hw_breakpoint_is_used() to check if breakpoints are in use on
the system.
Use it in the KUnit test to verify the global state before and after a
test case.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-3-elver@google.com
Add KUnit test for hw_breakpoint constraints accounting, with various
interesting mixes of breakpoint targets (some care was taken to catch
interesting corner cases via bug-injection).
The test cannot be built as a module because it requires access to
hw_breakpoint_slots(), which is not inlinable or exported on all
architectures.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20220829124719.675715-2-elver@google.com
BRBE captured branch types will overflow perf_branch_entry.type and generic
branch types in perf_branch_entry.new_type. So override each available arch
specific branch type in the following manner to comprehensively process all
reported branch types in BRBE.
PERF_BR_ARM64_FIQ PERF_BR_NEW_ARCH_1
PERF_BR_ARM64_DEBUG_HALT PERF_BR_NEW_ARCH_2
PERF_BR_ARM64_DEBUG_EXIT PERF_BR_NEW_ARCH_3
PERF_BR_ARM64_DEBUG_INST PERF_BR_NEW_ARCH_4
PERF_BR_ARM64_DEBUG_DATA PERF_BR_NEW_ARCH_5
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220824044822.70230-5-anshuman.khandual@arm.com
Platforms like arm64 could capture privilege level information for all the
branch records. Hence this adds a new element in the struct branch_entry to
record the privilege level information, which could be requested through a
new event.attr.branch_sample_type based flag PERF_SAMPLE_BRANCH_PRIV_SAVE.
This flag helps user choose whether privilege information is captured.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220824044822.70230-4-anshuman.khandual@arm.com
branch_entry.type now has ran out of space to accommodate more branch types
classification. This will prevent perf branch stack implementation on arm64
(via BRBE) to capture all available branch types. Extending this bit field
i.e branch_entry.type [4 bits] is not an option as it will break user space
ABI both for little and big endian perf tools.
Extend branch classification with a new field branch_entry.new_type via a
new branch type PERF_BR_EXTEND_ABI in branch_entry.type. Perf tools which
could decode PERF_BR_EXTEND_ABI, will then parse branch_entry.new_type as
well.
branch_entry.new_type is a 4 bit field which can hold upto 16 branch types.
The first three branch types will hold various generic page faults followed
by five architecture specific branch types, which can be overridden by the
platform for specific use cases. These architecture specific branch types
gets overridden on arm64 platform for BRBE implementation.
New generic branch types
- PERF_BR_NEW_FAULT_ALGN
- PERF_BR_NEW_FAULT_DATA
- PERF_BR_NEW_FAULT_INST
New arch specific branch types
- PERF_BR_NEW_ARCH_1
- PERF_BR_NEW_ARCH_2
- PERF_BR_NEW_ARCH_3
- PERF_BR_NEW_ARCH_4
- PERF_BR_NEW_ARCH_5
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220824044822.70230-3-anshuman.khandual@arm.com
This expands generic branch type classification by adding two more entries
there in i.e system error and not in transaction. This also updates the x86
implementation to process X86_BR_NO_TX records as appropriate. This changes
branch types reported to user space on x86 platform but it should not be a
problem. The possible scenarios and impacts are enumerated here.
--------------------------------------------------------------------------
| kernel | perf tool | Impact |
--------------------------------------------------------------------------
| old | old | Works as before |
--------------------------------------------------------------------------
| old | new | PERF_BR_UNKNOWN is processed |
--------------------------------------------------------------------------
| new | old | PERF_BR_NO_TX is blocked via old PERF_BR_MAX |
--------------------------------------------------------------------------
| new | new | PERF_BR_NO_TX is recognized |
--------------------------------------------------------------------------
When PERF_BR_NO_TX is blocked via old PERF_BR_MAX (new kernel with old perf
tool) the user space might throw up an warning complaining about an
unrecognized branch types being reported, but it's expected. PERF_BR_SERROR
& PERF_BR_NO_TX branch types will be used for BRBE implementation on arm64
platform.
PERF_BR_NO_TX complements 'abort' and 'in_tx' elements in perf_branch_entry
which represent other transaction states for a given branch record. Because
this completes the transaction state classification.
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: James Clark <james.clark@arm.com>
Link: https://lkml.kernel.org/r/20220824044822.70230-2-anshuman.khandual@arm.com
Provide branch speculation information captured via AMD Last Branch Record
Extension Version 2 (LbrExtV2) by setting the speculation info in branch
records. The info is based on the "valid" and "spec" bits in the Branch To
registers.
Suggested-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/ddc02f6320464cad0e3ff5bdb2314531568a91bc.1660211399.git.sandipan.das@amd.com
Add a new "spec" bitfield to branch entries for providing speculation
information. This will be populated using hints provided by branch sampling
features on supported hardware. The following cases are covered:
* No branch speculation information is available
* Branch is speculative but taken on the wrong path
* Branch is non-speculative but taken on the correct path
* Branch is speculative and taken on the correct path
Suggested-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/834088c302faf21c7b665031dd111f424e509a64.1660211399.git.sandipan.das@amd.com
AMD Last Branch Record Extension Version 2 (LbrExtV2) can report a branch
from address that points to an instruction preceding the actual branch by
several bytes due to branch fusion and further optimizations in Zen4
processors.
In such cases, software should move forward sequentially in the instruction
stream from the reported address and the address of the first branch
encountered should be used instead. Hence, use the fusion-aware branch
classifier to determine the correct branch type and get the offset for
adjusting the branch from address.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/c324d2d0a9c3976da30b9563d09e50bfee0f264d.1660211399.git.sandipan.das@amd.com
With branch fusion and other optimizations, branch sampling hardware in
some processors can report a branch from address that points to an
instruction preceding the actual branch by several bytes.
In such cases, the classifier cannot determine the branch type which leads
to failures such as with the recently added test from commit b55878c90a
("perf test: Add test for branch stack sampling"). Branch information is
also easier to consume and annotate if branch from addresses always point
to branch instructions.
Add a new variant of the branch classifier that can account for instruction
fusion. If fusion is expected and the current branch from address does not
point to a branch instruction, it attempts to find the first branch within
the next (MAX_INSN_SIZE - 1) bytes and if found, additionally provides the
offset between the reported branch from address and the address of the
expected branch instruction.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/b6bb0abaa8a54c0b6d716344700ee11a1793d709.1660211399.git.sandipan.das@amd.com
With AMD Last Branch Record Extension Version 2 (LbrExtV2), it is necessary
to process the branch records further as hardware filtering is not granular
enough for identifying certain types of branches. E.g. to record system
calls, one should record far branches. The filter captures both far calls
and far returns but the irrelevant records are filtered out based on the
branch type as seen by the branch classifier.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/e51de057517f77788abd393c832e8dea616d489c.1660211399.git.sandipan.das@amd.com
Commit 3e702ff6d1 ("perf/x86: Add LBR software filter support for Intel
CPUs") introduces a software branch filter which complements the hardware
branch filter and adds an x86 branch classifier.
Move the branch classifier to arch/x86/events/ so that it can be utilized
by other vendors for branch record filtering.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/bae5b95470d6bd49f40954bd379f414f5afcb965.1660211399.git.sandipan.das@amd.com
If AMD Last Branch Record Extension Version 2 (LbrExtV2) is detected,
convert the requested branch filter (PERF_SAMPLE_BRANCH_* flags) to the
corresponding hardware filter value and stash it in the event data when
a branch stack is requested. The hardware filter value is also saved in
per-CPU areas for use during event scheduling.
Hardware filtering is provided by the LBR Branch Select register. It has
bits which when set, suppress recording of the following types of branches:
* CPL = 0 (Kernel only)
* CPL > 0 (Userspace only)
* Conditional Branches
* Near Relative Calls
* Near Indirect Calls
* Near Returns
* Near Indirect Jumps (excluding Near Indirect Calls and Near Returns)
* Near Relative Jumps (excluding Near Relative Calls)
* Far Branches
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/9336af5c9785b8e14c62220fc0e6cfb10ab97de3.1660211399.git.sandipan.das@amd.com
If AMD Last Branch Record Extension Version 2 (LbrExtV2) is detected,
enable it alongside LBR Freeze on PMI when an event requests branch stack
i.e. PERF_SAMPLE_BRANCH_STACK.
Each branch record is represented by a pair of registers, LBR From and LBR
To. The freeze feature prevents any updates to these registers once a PMC
overflows. The contents remain unchanged until the freeze bit is cleared by
the PMI handler.
The branch records are read and copied to sample data before unfreezing.
However, only valid entries are copied. There is no additional register to
denote which of the register pairs represent the top of the stack (TOS)
since internal register renaming always ensures that the first pair (i.e.
index 0) is the one representing the most recent branch and so on.
The LBR registers are per-thread resources and are cleared explicitly
whenever a new task is scheduled in. There are no special implications on
the contents of these registers when transitioning to deep C-states.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/d3b8500a3627a0d4d0259b005891ee248f248d91.1660211399.git.sandipan.das@amd.com
AMD Last Branch Record Extension Version 2 (LbrExtV2) is driven by Core PMC
overflows. It records recently taken branches up to the moment when the PMC
overflow occurs.
Detect the feature during PMU initialization and set the branch stack depth
using CPUID leaf 0x80000022 EBX.
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/fc6e45378ada258f1bab79b0de6e05c393a8f1dd.1660211399.git.sandipan.das@amd.com
CPUID leaf 0x80000022 i.e. ExtPerfMonAndDbg advertises some new performance
monitoring features for AMD processors.
Bit 1 of EAX indicates support for Last Branch Record Extension Version 2
(LbrExtV2) features. If found to be set during PMU initialization, the EBX
bits of the same leaf can be used to determine the number of available LBR
entries.
For better utilization of feature words, LbrExtV2 is added as a scattered
feature bit.
[peterz: Rename to AMD_LBR_V2]
Signed-off-by: Sandipan Das <sandipan.das@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/172d2b0df39306ed77221c45ee1aa62e8ae0548d.1660211399.git.sandipan.das@amd.com
