Hi, On Thu, 26 Sep 2019, Denton Liu wrote: > On Thu, Sep 26, 2019 at 02:49:55PM +0200, Johannes Schindelin wrote: > > > > On Mon, 23 Sep 2019, Denton Liu wrote: > > > > > On platforms that can run `make hdr-check` but require custom flags, > > > this target was failing because none of them were being passed to the > > > compiler. For example, on MacOS, the NO_OPENSSL flag was being set but > > > it was not being passed into compiler so the check was failing. > > > > > > Pass $(ALL_CFLAGS) into the compiler for the $(HCO) target so that these > > > custom flags can be used. > > > > > > Signed-off-by: Denton Liu <liu.denton@xxxxxxxxx> > > > --- > > > Makefile | 2 +- > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > diff --git a/Makefile b/Makefile > > > index f879697ea3..d8df4e316b 100644 > > > --- a/Makefile > > > +++ b/Makefile > > > @@ -2773,7 +2773,7 @@ CHK_HDRS = $(filter-out $(EXCEPT_HDRS),$(patsubst ./%,%,$(LIB_H))) > > > HCO = $(patsubst %.h,%.hco,$(CHK_HDRS)) > > > > > > $(HCO): %.hco: %.h FORCE > > > - $(QUIET_HDR)$(CC) -include git-compat-util.h -I. -o /dev/null -c -xc $< > > > + $(QUIET_HDR)$(CC) -include git-compat-util.h -I. -o /dev/null -c -xc $(ALL_CFLAGS) $< > > > > Isn't this essentially identical to what I sent in > > https://public-inbox.org/git/pull.129.git.gitgitgadget@xxxxxxxxx/ > > ? > > Whoops, didn't notice this. I heard there's some quote about great minds > or something ;). Hahaha! ;-) > Anyway, this patch is superceded by v2 4/4 which should emulate the > compilation process even better, as suggested by Peff. The only > problem with that patch is that it splashes temporary *.hcc files > everywhere but I think that if it's only a developer-run target, it > should be fine. I agree that Peff's improvement is nice. The only problem is that my patch made it into `next` already, according to https://github.com/gitgitgadget/git/pull/129 (look for the labels on the right side, or for the comments on the bottom), and https://github.com/gitster/git/commit/a3f332f4fb10 agrees (look for names of the branches containing that commit, under the commit message). So we should explicitly as for my patch to be backed out from `next` and be stopped from advancing to `master`. Junio, would you kindly do that? According to the PR, you gave this branch the name `js/honor-cflags-in-hdr-check`. Thanks, Dscho
