Andrii Nakryiko <andrii.nakryiko@xxxxxxxxx> wrote:
> On Wed, Feb 23, 2022 at 2:20 PM Hao Luo <haoluo@xxxxxxxxxx> wrote:
>>
>> For binaries that are statically linked, consecutive stack frames are
>> likely to be in the same VMA and therefore have the same build id.
>> As an optimization for this case, we can cache the previous frame's
>> VMA, if the new frame has the same VMA as the previous one, reuse the
>> previous one's build id. We are holding the MM locks as reader across
>> the entire loop, so we don't need to worry about VMA going away.
>>
>> Tested through "stacktrace_build_id" and "stacktrace_build_id_nmi" in
>> test_progs.
>>
>> Suggested-by: Greg Thelen <gthelen@xxxxxxxxxx>
>> Signed-off-by: Hao Luo <haoluo@xxxxxxxxxx>
>> ---
>> kernel/bpf/stackmap.c | 11 ++++++++++-
>> 1 file changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
>> index 22c8ae94e4c1..280b9198af27 100644
>> --- a/kernel/bpf/stackmap.c
>> +++ b/kernel/bpf/stackmap.c
>> @@ -132,7 +132,8 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
>> int i;
>> struct mmap_unlock_irq_work *work = NULL;
>> bool irq_work_busy = bpf_mmap_unlock_get_irq_work(&work);
>> - struct vm_area_struct *vma;
>> + struct vm_area_struct *vma, *prev_vma = NULL;
>> + const char *prev_build_id;
>>
>> /* If the irq_work is in use, fall back to report ips. Same
>> * fallback is used for kernel stack (!user) on a stackmap with
>> @@ -151,6 +152,11 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
>>
>> for (i = 0; i < trace_nr; i++) {
>> vma = find_vma(current->mm, ips[i]);
>
> as a further optimization, shouldn't we first check if ips[i] is
> within prev_vma and avoid rbtree walk altogether? Would this work:
>
> if (prev_vma && range_in_vma(prev_vma, ips[i])) {
> /* reuse build_id */
> }
> vma = find_vma(current->mm, ips[i]);
>
>
> ?
Yes, that's a nice addition. Good idea.
>> + if (vma && vma == prev_vma) {
>> + memcpy(id_offs[i].build_id, prev_build_id,
>> + BUILD_ID_SIZE_MAX);
>> + goto build_id_valid;
>> + }
>> if (!vma || build_id_parse(vma, id_offs[i].build_id, NULL)) {
>> /* per entry fall back to ips */
>> id_offs[i].status = BPF_STACK_BUILD_ID_IP;
>> @@ -158,9 +164,12 @@ static void stack_map_get_build_id_offset(struct bpf_stack_build_id *id_offs,
>> memset(id_offs[i].build_id, 0, BUILD_ID_SIZE_MAX);
>> continue;
>> }
>> +build_id_valid:
>> id_offs[i].offset = (vma->vm_pgoff << PAGE_SHIFT) + ips[i]
>> - vma->vm_start;
>> id_offs[i].status = BPF_STACK_BUILD_ID_VALID;
>> + prev_vma = vma;
>> + prev_build_id = id_offs[i].build_id;
>> }
>> bpf_mmap_unlock_mm(work, current->mm);
>> }
>> --
>> 2.35.1.473.g83b2b277ed-goog
>>
