| Message ID | 20230427234833.1576130-2-namhyung@kernel.org (mailing list archive) |
|---|---|
| State | Not Applicable |
| Delegated to: | BPF |
| Headers | show |
| Series | [1/2] perf lock contention: Fix struct rq lock access | expand |
| Context | Check | Description |
|---|---|---|
| netdev/tree_selection | success | Not a local patch |
| bpf/vmtest-bpf-PR | fail | merge-conflict |
On Thu, Apr 27, 2023 at 4:48 PM Namhyung Kim <namhyung@kernel.org> wrote: > > It seems BPF CO-RE reloc doesn't work well with the pattern that gets > the field-offset only. Use offsetof() to make it explicit so that > the compiler would generate the correct code. > > Fixes: 0c1228486bef ("perf lock contention: Support pre-5.14 kernels") > Co-developed-by: Andrii Nakryiko <andrii.nakryiko@gmail.com> > Signed-off-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Ian Rogers <irogers@google.com> Thanks, Ian > --- > tools/perf/util/bpf_skel/lock_contention.bpf.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) > > diff --git a/tools/perf/util/bpf_skel/lock_contention.bpf.c b/tools/perf/util/bpf_skel/lock_contention.bpf.c > index 30c193078bdb..8d3cfbb3cc65 100644 > --- a/tools/perf/util/bpf_skel/lock_contention.bpf.c > +++ b/tools/perf/util/bpf_skel/lock_contention.bpf.c > @@ -429,21 +429,21 @@ struct rq___new { > SEC("raw_tp/bpf_test_finish") > int BPF_PROG(collect_lock_syms) > { > - __u64 lock_addr; > + __u64 lock_addr, lock_off; > __u32 lock_flag; > > + if (bpf_core_field_exists(struct rq___new, __lock)) > + lock_off = offsetof(struct rq___new, __lock); > + else > + lock_off = offsetof(struct rq___old, lock); > + > for (int i = 0; i < MAX_CPUS; i++) { > struct rq *rq = bpf_per_cpu_ptr(&runqueues, i); > - struct rq___new *rq_new = (void *)rq; > - struct rq___old *rq_old = (void *)rq; > > if (rq == NULL) > break; > > - if (bpf_core_field_exists(rq_new->__lock)) > - lock_addr = (__u64)&rq_new->__lock; > - else > - lock_addr = (__u64)&rq_old->lock; > + lock_addr = (__u64)(void *)rq + lock_off; > lock_flag = LOCK_CLASS_RQLOCK; > bpf_map_update_elem(&lock_syms, &lock_addr, &lock_flag, BPF_ANY); > } > -- > 2.40.1.495.gc816e09b53d-goog >
diff --git a/tools/perf/util/bpf_skel/lock_contention.bpf.c b/tools/perf/util/bpf_skel/lock_contention.bpf.c index 30c193078bdb..8d3cfbb3cc65 100644 --- a/tools/perf/util/bpf_skel/lock_contention.bpf.c +++ b/tools/perf/util/bpf_skel/lock_contention.bpf.c @@ -429,21 +429,21 @@ struct rq___new { SEC("raw_tp/bpf_test_finish") int BPF_PROG(collect_lock_syms) { - __u64 lock_addr; + __u64 lock_addr, lock_off; __u32 lock_flag; + if (bpf_core_field_exists(struct rq___new, __lock)) + lock_off = offsetof(struct rq___new, __lock); + else + lock_off = offsetof(struct rq___old, lock); + for (int i = 0; i < MAX_CPUS; i++) { struct rq *rq = bpf_per_cpu_ptr(&runqueues, i); - struct rq___new *rq_new = (void *)rq; - struct rq___old *rq_old = (void *)rq; if (rq == NULL) break; - if (bpf_core_field_exists(rq_new->__lock)) - lock_addr = (__u64)&rq_new->__lock; - else - lock_addr = (__u64)&rq_old->lock; + lock_addr = (__u64)(void *)rq + lock_off; lock_flag = LOCK_CLASS_RQLOCK; bpf_map_update_elem(&lock_syms, &lock_addr, &lock_flag, BPF_ANY); }
It seems BPF CO-RE reloc doesn't work well with the pattern that gets the field-offset only. Use offsetof() to make it explicit so that the compiler would generate the correct code. Fixes: 0c1228486bef ("perf lock contention: Support pre-5.14 kernels") Co-developed-by: Andrii Nakryiko <andrii.nakryiko@gmail.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> --- tools/perf/util/bpf_skel/lock_contention.bpf.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)