linux-hardened/kernel/trace
Joel Fernandes 848618857d tracing/ring_buffer: Try harder to allocate
ftrace can fail to allocate per-CPU ring buffer on systems with a large
number of CPUs coupled while large amounts of cache happening in the
page cache. Currently the ring buffer allocation doesn't retry in the VM
implementation even if direct-reclaim made some progress but still
wasn't able to find a free page. On retrying I see that the allocations
almost always succeed. The retry doesn't happen because __GFP_NORETRY is
used in the tracer to prevent the case where we might OOM, however if we
drop __GFP_NORETRY, we risk destabilizing the system if OOM killer is
triggered. To prevent this situation, use the __GFP_RETRY_MAYFAIL flag
introduced recently [1].

Tested the following still succeeds without destabilizing a system with
1GB memory.
echo 300000 > /sys/kernel/debug/tracing/buffer_size_kb

[1] https://marc.info/?l=linux-mm&m=149820805124906&w=2

Link: http://lkml.kernel.org/r/20170713021416.8897-1-joelaf@google.com

Cc: Tim Murray <timmurray@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Joel Fernandes <joelaf@google.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2017-07-19 08:22:12 -04:00
..
blktrace.c block: switch bios to blk_status_t 2017-06-09 09:27:32 -06:00
bpf_trace.c bpf: extend bpf_trace_printk to support %i 2017-07-03 02:22:52 -07:00
ftrace.c ftrace: Fix uninitialized variable in match_records() 2017-07-12 09:48:31 -04:00
Kconfig tracing: Rename update the enum_map file 2017-06-13 17:13:06 -04:00
Makefile
power-traces.c
ring_buffer.c tracing/ring_buffer: Try harder to allocate 2017-07-19 08:22:12 -04:00
ring_buffer_benchmark.c ring-buffer: Return reader page back into existing ring buffer 2017-05-01 10:26:40 -04:00
rpm-traces.c
trace.c A few more minor updates: 2017-07-13 13:17:19 -07:00
trace.h tracing: Add support for recording tgid of tasks 2017-06-27 13:30:28 -04:00
trace_benchmark.c
trace_benchmark.h
trace_branch.c
trace_clock.c
trace_entries.h trace: make trace_hwlat timestamp y2038 safe 2017-05-08 17:15:15 -07:00
trace_event_perf.c
trace_events.c tracing: Add support for recording tgid of tasks 2017-06-27 13:30:28 -04:00
trace_events_filter.c
trace_events_filter_test.h
trace_events_hist.c
trace_events_trigger.c
trace_export.c
trace_functions.c ftrace: Fix regression with module command in stack_trace_filter 2017-06-29 10:05:45 -04:00
trace_functions_graph.c
trace_hwlat.c trace: make trace_hwlat timestamp y2038 safe 2017-05-08 17:15:15 -07:00
trace_irqsoff.c
trace_kdb.c
trace_kprobe.c A few more minor updates: 2017-07-13 13:17:19 -07:00
trace_mmiotrace.c
trace_nop.c
trace_output.c tracing: Add support for display of tgid in trace output 2017-06-27 13:30:28 -04:00
trace_output.h
trace_printk.c
trace_probe.c
trace_probe.h
trace_sched_switch.c tracing: Add support for recording tgid of tasks 2017-06-27 13:30:28 -04:00
trace_sched_wakeup.c
trace_selftest.c
trace_selftest_dynamic.c
trace_seq.c
trace_stack.c tracing: Do note expose stack_trace_filter without DYNAMIC_FTRACE 2017-07-11 19:21:04 -04:00
trace_stat.c
trace_stat.h
trace_syscalls.c
trace_uprobe.c
tracing_map.c
tracing_map.h