On 3/28/19 1:53 PM, Richard Biener wrote: > On Wed, Mar 27, 2019 at 10:52 AM marxin <mli...@suse.cz> wrote: >> > > Hmm, I don't think we want this - it looks too much like a hack to me. > > I guess for an easier way to figure out the min/max ranges (do we > handle anti-ranges?) would be to specify a function name the > debug counter would be enabled for. That might also help in your > situation.
Well, it's a bit hack, but a useful one. In my case I had a couple files which triggered the debug counter limit. Function filter will work, but note that you have also an IPA debug counters. Here function context can be not clear. > > What I've usually done is do the bisection via dbg-counter on > a single file, manually link the spec binary and then > do a runcpu w/o recompiling the binary. Having a complex build system, adjusting just -dbg-cnt is much more convenient than doing a manual setup. Martin > > Richard. > >> gcc/ChangeLog: >> >> 2019-03-27 Martin Liska <mli...@suse.cz> >> >> * dbgcnt.c (dbg_cnt_set_limit_by_name): Add new argument >> aux_base and filter based on aux_base_name. >> (dbg_cnt_process_single_pair): Parse aux_base. >> * doc/invoke.texi: Document new extended format. >> >> gcc/testsuite/ChangeLog: >> >> 2019-03-27 Martin Liska <mli...@suse.cz> >> >> * gcc.dg/dbg-cnt-1.c: New test. >> --- >> gcc/dbgcnt.c | 11 ++++++++--- >> gcc/doc/invoke.texi | 8 ++++++-- >> gcc/testsuite/gcc.dg/dbg-cnt-1.c | 6 ++++++ >> 3 files changed, 20 insertions(+), 5 deletions(-) >> create mode 100644 gcc/testsuite/gcc.dg/dbg-cnt-1.c >>
