I decided this feature was too impacting to make the permanent default
behavior. This set of changes makes the default behavior of make the
old behavior (no second expansion). If you want second expansion, you
must define the .SECONDEXPANSION: special target before the first target
that needs it.
This set of changes ONLY fixes explicit and static pattern rules to work
like this. Implicit rules still have second expansion enabled all the
time: I'll work on that next.
Note that there is still a backward-incompatibility: now to get the old
SysV behavior using $$@ etc. in the prerequisites list you need to set
.SECONDEXPANSION: as well.
Taylor. There are two forms of this: first, it was possible to lose
tokens when using -j and -l at the same time, because waiting jobs were
not checked when determining whether any jobs were outstanding. Second,
if you had an exported recursive variable that contained a $(shell ...)
function there is a possibility to lose tokens, since a token was taken
but the child list was not updated until after the shell function was
complete.
To resolve this I introduced a new variable that counted the number of
tokens we have obtained, rather than checking whether there were any
children on the list. I also added some sanity checks to make sure we
weren't writing back too many or not enough tokens. And, the master
make will drain the token pipe before exiting and compare the count of
tokens at the end to what was written there at the beginning.
Also:
* Ensure a bug in the environment (missing "=") doesn't cause make to core.
* Rename the .DEFAULT_TARGET variable to .DEFAULT_GOAL, to match the
terminology in the documentation and other variables like MAKECMDGOALS.
* Add documentation of the .DEFAULT_GOAL special variable.
Still need to document the secondary expansion stuff...
I did this by adding intelligence into the algorithm such that the
second expansion was only actually performed when the prerequisite list
contained at least one "$", so we knew it is actually needed.
Without this we were using up a LOT more memory, since every single
target (even ones never used by make) had their file variables
initialized. This also used a lot more CPU, since we needed to create
and populate a new variable hash table for every target.
There is one issue remaining with this feature: it leaks memory. In
pattern_search() we now initialize the file variables for every pattern
target, which allocates a hash table, etc. However, sometimes we
recursively invoke pattern_search() (for intermediate files) with an
automatic variable (alloca() I believe) as the file. When that function
returns, obviously, the file variable hash memory is lost.
* New function: $(info ...)
* Disallow $(eval ...) to create prereq relationships inside command scripts
(caused core dumps)
* Try to allow more tests to succeed in Windows/DOS by sanitizing CRLF and \
* Various bug fixes and code cleanups (see the ChangeLog entry)
reported by Markus Mauhart <qwe123@chello.at>. One was a simple typo; to
fix the other we call patsubst_expand() for all instances of variable
substitution, even when there is no '%'. We used to call subst_expand()
with a special flag set in the latter case, but it didn't work properly
in all situations. Easier to just use patsubst_expand() since that's
what it is.
- Apply a fix for the "thundering herd" problem when using "-j -l".
This also fixes bug #4693.
- Fix bug #7257: allow functions as ifdef arguments
- Fix bug #4518: make sure we print all double-colon rules with -p.
- Upgrade to autconf 2.58/automake 1.8/gettext 0.13.1
- Various doc cleanups, etc.
Implement a fix for bug # 2169: too many OSs, even major OSs like Solaris,
don't properly implement SA_RESTART: important system calls like stat() can
still fail when SA_RESTART is set. So, forget the BROKEN_RESTART config
check and get rid of atomic_stat() and atomic_readdir(), and implement
permanent wrappers for EINTR checking on various system calls (stat(),
fstat(), opendir(), and readdir() so far).
enable the automake ansi2knr capability.
Right now this doesn't quite build using a K&R compiler because of a
problem with the loadavg test program, but the rest of the code works. I'm
asking the automake list about this problem.
Implemented enhancement #1391: allow "export" in target-specific
variable definitions.
Change the Info name of the "Automatic" node to "Automatic Variables".
Add text clarifying the scope of automatic variables to that section.
Invent a new macro HAVE_DOS_PATHS and change various instances of:
#if defined(WINDOWS) || defined(__MSDOS__)
to use the new macro instead. This should help make the OS/2 port
cleaner, as well.
Invent a cvs-clean maintainer target that tries to get the workspace
back to the state it was in after a CVS checkout.
New language.
GNU make. Also he provides some other performance fixups after doing
some profiling of make on large makefiles.
Modify the test suite to allow the use of Valgrind to find memory problems.
Incorporate "order-only" prerequisites patch. Wrote a test for it.
The test shows what might be a bug in the code; I need to look at it
more closely (anyway it doesn't behave as I expected). Also I haven't
done the docs yet.
New version of the manual, put into the doc subdir.
Enhancements: $(eval ...) and $(value ...) functions, various bug
fixes, etc. See the ChangeLog.
More to come.
