This is an automated email from the ASF dual-hosted git repository.
gregfelice pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/age.git
The following commit(s) were added to refs/heads/master by this push:
new d84b4578 Add automatic header-dependency tracking to the Makefile
(#2454)
d84b4578 is described below
commit d84b4578fe7551211d8bf5fe8d6adf3bcc086e78
Author: John Gemignani <[email protected]>
AuthorDate: Thu Jul 2 10:23:39 2026 -0700
Add automatic header-dependency tracking to the Makefile (#2454)
AGE lists OBJS explicitly and relies on PGXS, whose built-in dependency
tracking only runs when the server was built with --enable-depend (often
disabled). Consequently, editing a header did not recompile the .c files
that include it, leaving stale .o files. This is especially dangerous for
node/struct headers: a stale ag_nodes.o keeps an outdated node_size, so
_readExtensibleNode under-allocates and readNode corrupts the heap
("unrecognized node type").
Emit a .d file beside each object via -MMD -MP and -include them, deriving
DEPFILES from OBJS. The mechanism is self-contained (independent of
--enable-depend): -MMD skips system headers and -MP tolerates deleted
headers. On servers built with --enable-depend, PGXS appends its own -MF
after CFLAGS (last -MF wins), so this degrades cleanly to PGXS's tracking.
Add DEPFILES to EXTRA_CLEAN and *.d to .gitignore.
Co-authored-by: Copilot <[email protected]>
modified: .gitignore
modified: Makefile
---
.gitignore | 1 +
Makefile | 24 +++++++++++++++++++++++-
2 files changed, 24 insertions(+), 1 deletion(-)
diff --git a/.gitignore b/.gitignore
index 1e2f8f67..98f4a7f9 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,4 +1,5 @@
*.o
+*.d
*.so
build.sh
.idea
diff --git a/Makefile b/Makefile
index 21c7f81f..32f34d6e 100644
--- a/Makefile
+++ b/Makefile
@@ -162,6 +162,10 @@ OBJS = src/backend/age.o \
src/backend/utils/name_validation.o \
src/backend/utils/ag_guc.o
+# Per-object header-dependency files (see "Automatic header-dependency
+# tracking" below the PGXS include). One .d is generated beside each .o.
+DEPFILES = $(OBJS:.o=.d)
+
# ===== Extension SQL & data files =====
EXTENSION = age
@@ -258,7 +262,8 @@ EXTRA_CLEAN = $(addprefix $(ag_regress_dir)/,
$(ag_regress_out)) \
$(all_age_sql) \
$(age_init_sql) \
$(age_upgrade_test_sql) \
- $(ag_regress_dir)/age_upgrade_cleanup.sh
+ $(ag_regress_dir)/age_upgrade_cleanup.sh \
+ $(DEPFILES)
GEN_KEYWORDLIST = $(PERL) -I ./tools/ ./tools/gen_keywordlist.pl
GEN_KEYWORDLIST_DEPS = ./tools/gen_keywordlist.pl tools/PerfectHash.pm
@@ -271,6 +276,23 @@ PG_CONFIG ?= pg_config
PGXS := $(shell $(PG_CONFIG) --pgxs)
include $(PGXS)
+# ===== Automatic header-dependency tracking =====
+#
+# AGE lists OBJS explicitly, and PGXS's built-in .deps tracking only runs when
+# the *server* was built with --enable-depend (often off). Without the lines
+# below, editing a header does NOT rebuild the .c files that include it,
leaving
+# STALE .o files. This is especially dangerous for node/struct headers: a stale
+# ag_nodes.o keeps an old node_size, so _readExtensibleNode under-allocates and
+# readNode corrupts the heap ("unrecognized node type: <garbage>").
+#
+# The compiler emits a .d file next to each object (-MMD = user headers only;
+# -MP adds phony targets so deleting a header does not break the build). With
+# "-o foo.o", -MMD writes "foo.d" automatically (no -MF, no basename clashes).
+# On servers that DO set --enable-depend, PGXS appends its own "-MF .deps/*.Po"
+# after $(CFLAGS) (last -MF wins), so this degrades cleanly to that mechanism.
+override CFLAGS += -MMD -MP
+-include $(DEPFILES)
+
# ===== Build rules =====
# 32-bit platform support: pass SIZEOF_DATUM=4 to enable (e.g., make
SIZEOF_DATUM=4)
