On Fri, Dec 6, 2024 at 11:09 AM Kumar Kartikeya Dwivedi <memxor@xxxxxxxxx> wrote: > > On Fri, 6 Dec 2024 at 19:37, Alexei Starovoitov > <alexei.starovoitov@xxxxxxxxx> wrote: > > > > On Fri, Dec 6, 2024 at 10:11 AM Kumar Kartikeya Dwivedi > > <memxor@xxxxxxxxx> wrote: > > > > I think we need to revert the raw_tp masking hack and > > > > go with denylist the way Jiri proposed: > > > > https://lore.kernel.org/bpf/ZrIj9jkXqpKXRuS7@krava/ > > > > ... > > > > > Jiri, do you have the diff around for that attempt? Could you post a > > > revert of the patches and then the diff you shared? > > > > the link above. > > The link only has information about one tracepoint, but further down > the thread Jiri found more examples and the case of IS_ERR. > https://lore.kernel.org/bpf/Zr3q8ihbe8cUdpfp@krava > It would probably be good to add all of them? Or did I misunderstand > and you just add PTR_MAYBE_NULL for the scheduler tracepoint in the > report? I mean Jiri's patch fixes one tracepoint, but the same approach can be used for the rest of tp-s Jiri found in his 2nd link. IS_ERR can be handled as well as info->reg_type = SCALAR; So revert plus one patch to fix them all.
