On Tue, Oct 17, 2023 at 01:16:28PM -0700, Namhyung Kim wrote: > On Tue, Oct 17, 2023 at 1:43 AM Jiri Olsa <olsajiri@xxxxxxxxx> wrote: > > > > On Mon, Oct 16, 2023 at 06:36:10PM -0700, Namhyung Kim wrote: > > > On Thu, Oct 12, 2023 at 11:37 PM Jiri Olsa <olsajiri@xxxxxxxxx> wrote: > > > > > > > > On Thu, Oct 12, 2023 at 08:57:33PM -0700, Namhyung Kim wrote: > > > > > Hi Jiri, > > > > > > > > > > On Sun, Oct 8, 2023 at 2:23 PM Jiri Olsa <jolsa@xxxxxxxxxx> wrote: > > > > > > > > > > > > As Dmitry described in [1] changelog the current way of detecting > > > > > > -s option is broken for new make. > > > > > > > > > > I'm not sure what -s option does for perf (at least). > > > > > It doesn't seem much different whether I give it or not. > > > > > Am I missing something? > > > > > > > > what's your make version? the wrong output is visible when running > > > > with make version > 4.4 .. basicaly the -s is wrongly detected and > > > > you either get no output at all from some builds or overly verbose > > > > output > > > > > > > > it's mentioned in the [1] commit changelog, I can put it to the > > > > changelog in new version > > > > > > IIUC it's about detecting `make -s` properly and not being confused > > > by `make a=s` or something. I'm not objecting on it but I don't see > > > what `make -s` does actually. > > > > so the tools/build/Makefile.build and tools/scripts/Makefile.include detect > > make -s option, which puts make into silent mode, so both makefiles switch > > off the output by setting quiet=silent_ or silent=1 > > > > the problem is that the detection of make -s option changed in make > 4.4 > > and current code could be tricked to switch to silent mode just by having > > 's' persent on the command line, like with 'a=s' > > I think our talk is circulating :-). Anyway I'm ok with the change, so :) ok, thanks > > Acked-by: Namhyung Kim <namhyung@xxxxxxxxxx> > > Which tree do you want to route it? I think perf tree is the best one to route it thanks, jirka
