From c5dfd78eb79851e278b7973031b9ca363da87a7e Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 21 Apr 2016 12:28:50 -0300 Subject: perf core: Allow setting up max frame stack depth via sysctl The default remains 127, which is good for most cases, and not even hit most of the time, but then for some cases, as reported by Brendan, 1024+ deep frames are appearing on the radar for things like groovy, ruby. And in some workloads putting a _lower_ cap on this may make sense. One that is per event still needs to be put in place tho. The new file is: # cat /proc/sys/kernel/perf_event_max_stack 127 Chaging it: # echo 256 > /proc/sys/kernel/perf_event_max_stack # cat /proc/sys/kernel/perf_event_max_stack 256 But as soon as there is some event using callchains we get: # echo 512 > /proc/sys/kernel/perf_event_max_stack -bash: echo: write error: Device or resource busy # Because we only allocate the callchain percpu data structures when there is a user, which allows for changing the max easily, its just a matter of having no callchain users at that point. Reported-and-Tested-by: Brendan Gregg Reviewed-by: Frederic Weisbecker Acked-by: Alexei Starovoitov Acked-by: David Ahern Cc: Adrian Hunter Cc: Alexander Shishkin Cc: He Kuang Cc: Jiri Olsa Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Milian Wolff Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: Wang Nan Cc: Zefan Li Link: http://lkml.kernel.org/r/20160426002928.GB16708@kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- kernel/bpf/stackmap.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/bpf') diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c index 499d9e933f8e..f5a19548be12 100644 --- a/kernel/bpf/stackmap.c +++ b/kernel/bpf/stackmap.c @@ -66,7 +66,7 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr) /* check sanity of attributes */ if (attr->max_entries == 0 || attr->key_size != 4 || value_size < 8 || value_size % 8 || - value_size / 8 > PERF_MAX_STACK_DEPTH) + value_size / 8 > sysctl_perf_event_max_stack) return ERR_PTR(-EINVAL); /* hash table size must be power of 2 */ @@ -124,8 +124,8 @@ static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5) struct perf_callchain_entry *trace; struct stack_map_bucket *bucket, *new_bucket, *old_bucket; u32 max_depth = map->value_size / 8; - /* stack_map_alloc() checks that max_depth <= PERF_MAX_STACK_DEPTH */ - u32 init_nr = PERF_MAX_STACK_DEPTH - max_depth; + /* stack_map_alloc() checks that max_depth <= sysctl_perf_event_max_stack */ + u32 init_nr = sysctl_perf_event_max_stack - max_depth; u32 skip = flags & BPF_F_SKIP_FIELD_MASK; u32 hash, id, trace_nr, trace_len; bool user = flags & BPF_F_USER_STACK; @@ -143,7 +143,7 @@ static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5) return -EFAULT; /* get_perf_callchain() guarantees that trace->nr >= init_nr - * and trace-nr <= PERF_MAX_STACK_DEPTH, so trace_nr <= max_depth + * and trace-nr <= sysctl_perf_event_max_stack, so trace_nr <= max_depth */ trace_nr = trace->nr - init_nr; -- cgit v1.2.3