On Thu, Sep 4, 2025 at 1:35 PM Peter Zijlstra [off-list ref] wrote:

On Thu, Sep 04, 2025 at 11:27:45AM -0700, Andrii Nakryiko wrote:

quoted

quoted

quoted

So I've been thinking what's the simplest and most reliable way to
feature-detect support for this sys_uprobe (e.g., for libbpf to know
whether we should attach at nop5 vs nop1), and clearly that would be

wrt nop5/nop1.. so the idea is to have USDT macro emit both nop1,nop5
and store some info about that in the usdt's elf note, right?

Wait, what? You're doing to emit 6 bytes and two nops? Why? Surely the
old kernel can INT3 on top of a NOP5?

Yes it can, but it's 2x slower in terms of uprobe triggering compared
to nop1. So while it will work to use just nop5 on old kernels, it's
going to be a performance regression if we do this unconditionally.

So the idea was to have nop1 + nop5, stick to nop1 for old kernels,
attach to nop5 for newer ones (that's where feature detection I was
asking about is important, libbpf will do this automatically and
transparently).

I know it's messy, but I think we have to do that.