16:57 <khaled> Hi All, Is there a documentation link or bug number to describe why do systemtap causes a 32bit process to sigfault in uretprobes on 64bit kernels?

17:23 <fche> khaled, that sounds a little bit familiar

17:23 <fche> have you tried running the equivalent perf probe against that same process ?

17:24 <fche> just to factor out kernel bugs rather than systemtap buts

17:24 <khaled> Yes and it failed until a friend of mine provided a fix that works on perf but not with systemtap

17:24 <khaled> https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?id=9212ec7d8357ea630031e89d0d399c761421c83b

17:27 <fche> ok, we'd need a fixed kernel before spending too much time on the stap side (if any)

17:27 <fche> will need to wait until e.g. fedora rawhide spins up a kernel with that patch in it

17:28 <fche> if you like, please feel free to open a bug in sourceware.org/bugzilla so we don't forget

17:30 <fche> but it could be worth a quick experiment in the stap runtime to s/is_32bit_app(current)/!user_64bit_mode(context->uregs)/ or similiarish

17:32 <khaled> OK, will give it a try and keep you updated, Thanks Frank :)

17:41 <fche> righto

19:21 <irker468> systemtap: sapatel systemtap.git:refs/heads/sapatel/pr23285 * release-4.1-70-g8d12a4a / bpf-internal.h bpf-translate.cxx stapbpf/bpfinterp.cxx: enabled the concatenation operator for stapbpf userspace probes http://tinyurl.com/y3rv5xap

19:21 <irker468> systemtap: sapatel systemtap.git:refs/heads/sapatel/pr23285 * release-4.1-71-g4bf390b / bpf-translate.cxx stapbpf/bpfinterp.cxx: removed redundant string copying and added comments about validity of c_str() http://tinyurl.com/y5338mza

19:21 <irker468> systemtap: sapatel systemtap.git:refs/heads/sapatel/pr23285 * release-4.1-72-g1d4d319 / bpf-internal.h bpf-translate.cxx stapbpf/bpfinterp.cxx tapset/string.stp: enabled character escaping macros (in progress) http://tinyurl.com/y2u2zots