On 01.08.2018 09:27, Christos Zoulas wrote: > In article <a0316562-2fd3-1509-aca2-76fde6ab6989%gmx.com@localhost>, > Kamil Rytarowski <n54%gmx.com@localhost> wrote: >> -=-=-=-=-=- >> -=-=-=-=-=- >> >> On 31.07.2018 10:28, Christos Zoulas wrote: >>> In article <2c408f23-6eae-da00-dfb2-ebc7b66e689c%gmx.com@localhost>, >>> Kamil Rytarowski <n54%gmx.com@localhost> wrote: >>>> >>>> I'm for removal of gprof and deduplication of .a files in the base. >>>> >>>> While there, can we get rid of moncontrol(3), profil(2) and all the >>>> corresponding code? >>> >>> I constantly see proposals to remove things from base without replacing >>> the functionality first. Yes, the current state is poor but the >>> "hand hurts, cut hand" approach is worse. >>> >>> christos >>> >> >> The original proposal contained an analysis of gperftools as an >> immediate replacement through pkgsrc (it requires Perl). > > yes, gperftools could fit the bill. > >> And while there, there was a process of analysis of the compiler-rt >> profile [coverage] feature during GSoC (while porting libFuzzer, we were >> researching whether we need it at that point - but the conclusion was >> that it's not required for fuzzing). > We will get 'LLVM profile' once the compiler-rt project will be merged with base (but probably with just Clang/LLVM). It could be treated as a modern -pg replacement. I've ported it with local patches, but there were few failing tests around left. I keep it on my TODO queues maybe to be picked by a GSoC student ( http://wiki.netbsd.org/users/kamil/proposals/ ). > christos >
Attachment:
signature.asc
Description: OpenPGP digital signature