Jim Meyering wrote: > Running "make -j25 check" on a nominal-12-core F14 system would > cause serious difficulty leading to an OOM kill -- and this is brand new. > It worked fine yesterday. I tracked it down to all of the make processes > working on the "built_programs.list" (in src/Makefile.am) rule ... > -built_programs = \ > - (cd $(top_builddir)/src && MAKEFLAGS= $(MAKE) -s built_programs.list) > +CLEANFILES = > +CLEANFILES += .built-programs > +check-am: .built-programs > +.built-programs: > + $(AM_V_GEN)cd $(top_builddir)/src \ > + && MAKEFLAGS= $(MAKE) -s built_programs.list \ > + > $@-t && mv $@-t $@
I fixed the above rule last night (it puts the .built-programs file in ../src rather than in "." where it's required), but somehow I pushed the original. Here's the fix: >From 15571e0c8f4be1170184388f9acd8259ba75b22f Mon Sep 17 00:00:00 2001 From: Jim Meyering <meyer...@redhat.com> Date: Thu, 10 Feb 2011 10:01:23 +0100 Subject: [PATCH] tests: fix bug in preceding check.mk change * tests/check.mk (.built-programs): Run cd'd submake in a subshell so the redirected output ends up in the current directory, not ../src. --- tests/check.mk | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/check.mk b/tests/check.mk index b80dce8..1e3ca85 100644 --- a/tests/check.mk +++ b/tests/check.mk @@ -44,8 +44,8 @@ CLEANFILES = CLEANFILES += .built-programs check-am: .built-programs .built-programs: - $(AM_V_GEN)cd $(top_builddir)/src \ - && MAKEFLAGS= $(MAKE) -s built_programs.list \ + $(AM_V_GEN)(cd $(top_builddir)/src \ + && MAKEFLAGS= $(MAKE) -s built_programs.list) \ > $@-t && mv $@-t $@ # Note that the first lines are statements. They ensure that environment -- 1.7.4.2.g597a6