[bug #47960] Adding variable assignment and new rule in else, break execution.

Update of bug #47960 (project make): Item Group:None => Bug Status:None => Fixed Assigned to:None => psmith Open/Closed:

[bug #44742] Double-dep with double-colon rule not built

Update of bug #44742 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #47880] Allow updates to .INCLUDE_DIRS to change search path

Update of bug #47880 (project make): Operating System: MS Windows => Any Summary: [Win] added MAKELIBPATH as default include directory => Allow updates to .INCLUDE_DIRS to change search path

[bug #47996] Make 4.2 does not build with cl version 15.00 (and older)

Update of bug #47996 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #48060] Recipe line flags leak between targets defined in one rule

URL: Summary: Recipe line flags leak between targets defined in one rule Project: make Submitted by: psmith Submitted on: Mon 30 May 2016 05:32:13 PM EDT Severity: 3 - Normal

[bug #48061] make check in targets/DELETE_ON_ERROR fails on Solaris 10

Update of bug #48061 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #48045] gmake-4.2 breaks firefox-46.0.1 build

Update of bug #48045 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #1: Duplicate of bug #4799

[bug #48057] make 4.2 breaks parallel builds of git, perl, dialog, xterm and probably more

Update of bug #48057 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #1: Duplicate of bug #4799

[bug #48009] Increased stack rlimit is inherited by the subprocesses to make when make restarts

Update of bug #48009 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #47995] Parallel double-colon breaks FORCE

Update of bug #47995 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Component Version:

[bug #46242] Race condition when input file is updated while compiling

Update of bug #46242 (project make): Item Group: Bug => Enhancement ___ Reply to this item at: ___ Messag

[bug #48070] "make update" fails at wget of translationproject translations

Follow-up Comment #1, bug #48070 (project make): I've sent a query to the translation project coordinator: I have no control over or insight into the management of that site. You can probably hack this, if you wanted to, by mucking with the po/LINGUAS file to remove translations you don't have.

[bug #48070] "make update" fails at wget of translationproject translations

Update of bug #48070 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #48276] Better errors for accidental spaces

Follow-up Comment #1, bug #48276 (project make): I'm not that excited about this idea. The problem is that lots of makefiles indent their variables, rules, etc. For example inside "ifeq" statements, or even just to right-align variable assignments. Having warnings about changing to TAB on most

[bug #49014] Zombies in parallel builds with pselect code

Follow-up Comment #4, bug #49014 (project make): I took a look through the code. I could to do a more comprehensive review but at first glance it looks like everything is OK: if PSELECT is enabled then we block SIGCHLD right at the start of the make process and we never unblock it except as a par

[bug #49014] Zombies in parallel builds with pselect code

Follow-up Comment #6, bug #49014 (project make): It can't be implemented in userland. What pselect() does is unblock the signal, call select, then have the signal blocked again on return. The point of using pselect() is that the signal unblock/block must be atomic with the select() system call.

[bug #49093] ifdef checks for non-empty value, not definition

Update of bug #49093 (project make): Item Group: Bug => Documentation Triage Status:None => Small Effort ___ Follow-up Comment #1: What the doc is trying

[bug #49093] ifdef checks for non-empty value, not definition

Update of bug #49093 (project make): Status:None => Works for me Open/Closed:Open => Closed ___ Follow-up Comment #2: Heh. Philip Guenther

[bug #49183] Macros set in MAKEFLAGS should not be added to the environment

URL: Summary: Macros set in MAKEFLAGS should not be added to the environment Project: make Submitted by: psmith Submitted on: Sun 25 Sep 2016 04:07:02 PM EDT Severity: 3 - Normal

[bug #49117] [read.c:710]: (warning) Possible null pointer dereference: collapsed

Update of bug #49117 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #1: This isn't a bug: the

[bug #49183] Macros set in MAKEFLAGS should not be added to the environment

Follow-up Comment #3, bug #49183 (project make): The current POSIX standard is available online for free and has been for many years now: http://www.opengroup.org/onlinepubs/9699919799/ The change seems pretty innocuous to me. The number of people who set the MAKEFLAGS variable in their local en

[bug #49590] wildcard no longer give sorted output in version 4.2

Update of bug #49590 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Reply to this item at:

[bug #49681] Make fails to glob lib/*.{o,a}

Update of bug #49681 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #1: Any commands inside a

[bug #49773] release/archive older make manuals

Update of bug #49773 (project make): Item Group:None => Documentation ___ Follow-up Comment #1: I've had discussions before with the GNU web site folks about doing this. I'd prefer to have some

[bug #48951] Grammar mistake in section 5.3.1

Update of bug #48951 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49111] w32: Removed unused telldir() function.

Update of bug #49111 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49112] arscan.c: Fixed printf conversion specifier in TEST mode

Update of bug #49112 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49113] expand.c: Fixed possible null pointer dereference

Update of bug #49113 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49116] w32/subproc/sub_proc.c: Fixed potential null pointer dereference.

Update of bug #49116 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49115] read.c: Fixed invariant condition.

Update of bug #49115 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Operating System:

[bug #45477] cygwin: regression in VPATH handling.

Update of bug #45477 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49114] job.c: Fixed conversion specifier.

Update of bug #49114 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49865] read.c:2681: possible missing break ?

Update of bug #49865 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #48809] renaming --jobserver-fds to --jobserver-auth breaks jobserver interoperability

Update of bug #48809 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #40234] Ignored missing include in makefile causes "no makefile found" message.

Update of bug #40234 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #20513] # still needs to be escaped in a function like $(shell)

Update of bug #20513 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #40236] include, -include and sinclude ignores failures to read the included file

Update of bug #40236 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #43378] $(guile ...) must be coaxed to accept (or does not accept) valid guile code.

Update of bug #43378 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #5: The issue about # not

[bug #48360] Environment variable doesn't loose specialness, although redefined

Update of bug #48360 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #2: It seems this is a dup

[bug #46585] variables being auto-exported to the environment if it was set in the environment

Update of bug #46585 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #3: See also duplicate bug

[bug #49935] num_mkfiles uninitialized

Update of bug #49935 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #49938] fdin might be (but isn't?) clobbered by vfork

Update of bug #49938 (project make): Status:None => Fixed Assigned to:None => psmith Open/Closed:Open => Closed Fixed Release:

[bug #35323] When the environment is modified with export or undefine, $(shell ...) still inherits the original environment

Update of bug #35323 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #1: This is a duplicate of

[bug #10593] $(shell) doesn't honor export but this is undocumented?

Follow-up Comment #6, bug #10593 (project make): See also duplicate bug #35323 ___ Reply to this item at: ___ Message sent via/by Savannah http://sava

[bug #10708] target names are canonicalized (./ removed, etc.)

Follow-up Comment #2, bug #10708 (project make): See also bug #17230 ___ Reply to this item at: ___ Message sent via/by Savannah http://savannah.gnu.o

[bug #17230] Leading ./ is stripped from targets and prerequisites

Update of bug #17230 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #2: I don't think bug #153

[bug #24487] "foo" should not be identified with "./foo" if foo is phony

Follow-up Comment #4, bug #24487 (project make): The documentation for .PHONY states ( https://www.gnu.org/software/make/manual/html_node/Phony-Targets.html ): _The implicit rule search (see Implicit Rules) is skipped for .PHONY targets. This is why declaring a target as .PHONY is good for perfor

[bug #49661] syntax anomaly with order-only prereqs and pattern rules

Update of bug #49661 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #3: I'm going to close thi

[bug #50062] Variable needed to check that Makefile is run by GNU make?

Follow-up Comment #1, bug #50062 (project make): I don't really know of any way to determine whether or not you're running a particular version of make, in a makefile portable enough that it would provide useful output if you _weren't_ using that version of make. For instance in your example: i

[bug #50062] Variable needed to check that Makefile is run by GNU make?

Follow-up Comment #2, bug #50062 (project make): Similarly, the $(error ...) function and indeed functions like this themselves are features only found in GNU make... so even if you were using some other implementation of make that happened to support "ifndef" the same way GNU make did, it could v

[bug #42125] Explicit rules do not support multiple targets

Follow-up Comment #5, bug #42125 (project make): I understand why you may consider it counterintuitive based on the naming, but I don't understand why you consider it unfortunate. Suppose that we made the change in the implementation that you suggest: that would mean that the targets listed in th

[bug #50492] configure missing / autoconf fails

Update of bug #50492 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #1: All of the files you m

[bug #42125] Explicit rules do not support multiple targets

Follow-up Comment #8, bug #42125 (project make): I'm afraid I don't understand: how can a static pattern rule have multiple targets? Each pattern in the target needs to match each of the elements of the static list, so clearly you can't have multiple patterns. Unless you're saying that the targe

[bug #50754] Archive member rules and GNU ar's deterministic (-D) option

Update of bug #50754 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #1: Unfortunately there's

[bug #50300] Set CFLAGS to “-O1” with .POSIX target

Update of bug #50300 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #2: I'm not sure I underst

[bug #50062] Variable needed to check that Makefile is run by GNU make?

Update of bug #50062 (project make): Status:None => Not A Bug Open/Closed:Open => Closed ___ Follow-up Comment #4: I'm closing this as I

[bug #50300] Set CFLAGS to “-O1” with .POSIX target

Follow-up Comment #6, bug #50300 (project make): I have this change already queued locally along with a bunch of others. I've been away for a while on vacation but I'll try to push those to Savannah this week/weekend. ___ Reply to this ite

[bug #50300] Set CFLAGS to “-O1” with .POSIX target

Update of bug #50300 (project make): Status: Not A Bug => None Open/Closed: Closed => Open ___ Reply to this item at:

[bug #50790] Some kind of memory corruption in error messages with gcc-6.3.0 -flto=4

Update of bug #50790 (project make): Status:None => Duplicate Open/Closed:Open => Closed ___ Follow-up Comment #3: This is a duplicate of

Re: Target-specific variable corruption

This has already been fixed in the latest version of GNU make (currently 3.78.1). ftp://ftp.gnu.org/gnu/make/ Or mirrors. -- --- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development &

Re: Problem with case insensitive build on Win32

for inclusion (or, I have to start keeping my own local patched copy, which I'd like to avoid). Thx. -- --- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I

RE: Problem with case insensitive build on Win32

: 0) ! #if defined _AMIGA || defined VMS || defined HAVE_CASE_INSENSITIVE_FS | FNM_CASEFOLD #endif ); -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Manageme

Re: Bug in target-specific variables

given target, _not_ when expanding the target or prerequisite lists. See the GNU make manual section describing target-specific variables ``Target-Specific Variable Values''; specifically the second paragraph. -- --

Re: Bug in pattern specific variable values

ces (I tried your example and it worked) and will be in the next release. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a pr

Re: $(if ) function crashes executing 'else' part if it is not last in the expression

report; this bug has been reported and is fixed in the development sources for the next version of GNU make. http://www-gnats.gnu.org:8080/cgi-bin/wwwgnats.pl/full/1429 -- ------- Paul D. Smith <[EMAIL PROTECTED]>

Re: [PATCH]Small fixes in job.c and main.c

in that free(0); is legal and considered a no-op, so the if-statements are superfluous anyway. But, the code is obviously wrong and I fixed it. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management D

Re: Compiliance

ic pattern rules instead: $(COBJS) : %$(EXT).o : %.c -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a professional." --Mad Scientist -

Re: mksh: Fatal error in reader: = missing from replacement macro reference

from mksh, whatever that is, not make. I don't even see where this makefile might print "MAKEFLAGS value:" or what command invokes mksh. PS. What version of make are you using? -- ----------- Paul D. Smith <

Re: mksh: Fatal error in reader: = missing from replacement macro reference

d by a program make invoked or you're using the wrong version of make" or something. Thx. -- --- Paul D. Smith <[EMAIL PROTECTED]> Netw

Re: Variable assignment in Make 3.77 on i686, Red Hat

Please upgrade to a new version of GNU make. This bug has been fixed for a while now. The current latest version is 3.78.1. Thanks. -- --- Paul D. Smith <[EMAIL PROTECTED]> Network Management Devel

Re: Cannot use computed %

e... so whenever the Makefile changes you have to assume the value of those variables changed. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management

Re: Variable assignment in Make 3.77 on i686, Red Hat

%% Jussi Piitulainen <[EMAIL PROTECTED]> writes: jp> "Paul D. Smith" <[EMAIL PROTECTED]> writes: >> The current latest version is 3.78.1. jp> I tried to find this information, and a list of known bugs, at jp> www.gnu.org but failed. Perhaps it i

Re: Variable assignment in Make 3.77 on i686, Red Hat

Sorry, I forgot: the GNU BTS is at: http://www-gnats.gnu.org:8080/cgi-bin/wwwgnats.pl -- --- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad,

Re: Cannot use computed %

sn't really any way that a rule could be run with a _target_ that was vpath expanded, given the way it works now, so $* should never expand to a vpath pathname. Well, I take that back; you could do it if you also defined GPATH. Phew! Dodged another bullet! :) --

Re: .SECONDARY and implicit rule chains

/1423 HTH... -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a professional." --Mad Scientist --

Re: gmake -j on DYNIX/ptx

e both your client and server systems the same OS/version, or are you using a NAC or Auspex or something as an NFS server? You might try asking on comp.protocols.nfs. -- ------- Paul D. Smith <[EMAIL PROTECTED]>

Re: bug report (v3.77): target specific variable values

oblem. -- --- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a professional." --Mad Scientist --- These are my opini

Re: MAKEFLAGS bug?

S: with no prerequisites. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a professional." --Mad Scientist --- These a

Re: This is not a bug!

he shared and non-shared (PIC and non-PIC compiled) .o's around at the same time. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a professional." --Mad Scientist --

Re: auto-dependencies and deleted files

ing to a better method of dependency generation solves the problem more generically without it anyway. See here: http://www.ultranet.com/~pauld/gmake/autodep.html Have fun! -- ------- Paul D. Smith <[EMAIL PROTECTED

Re: Nanoseconds Enhancement does not work for UnixWare 7

Thanks for your work on this; I'll check out the issues you raise and let you know if I need more details, etc. -- --- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please

Re: Compilation error on OpenSteep 4.0

ompile line by hand with the -v option and check that there aren't any other odd include directories that get passed to the compiler. Then, you should run it with the -E output and see if you can tell why __SIZE_TYPE__ isn't defined where it should be. --

Re: Bug on the windows 32 port of gnu make

This isn't a bug. This is how VPATH is designed to work. Please see the discussion of "How Not to Use VPATH" at my web site below: http://www.ultranet.com/~pauld/gmake/ HTH. -- ------- Paul D. Smith &l

Re: Timestamp is not being checked on NT + other

between directories and files: if the target exists, it exists. What if you _wanted_ to write a target that created a directory: $(OBJDIR): ; mkdir $(OBJDIR) ? This is actually not that uncommon (although not the best way to do it, IMO). Look up the .PHONY rule in the GNU make manual for h

Re: Timestamp is not being checked on NT + other

eir version entirely. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management Development "Please remain calm...I may be mad, but I am a professional." --Mad Scientist

Re: conflict between RCS and VPATH

do is add a "GPATH := foo" to your makefile. That will tell make that "foo" is a "global directory" and that make shouldn't throw out pathnames when targets are found in that directory. See the make manual docs for GPATH. -- --- Paul D. Smith <[EMAIL PR

Re: ["Paul D. Smith" ] FWD: Compaq/Tandem NSK port of make-3.78.1

been dropped off the loop here. What's the resolution to this issue? Were changes made to autoconf? Why won't they help make users? Thx. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Network Management

Re: COFLAGS undefined

made this change. Thanks. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org http://www.ultranet.com/~pauld/gmake/ "Please remain calm...I may be

Re: FORCE works when .PHONY doesn't

e seeing. Using FORCE rules or making module.ot an explicit instead of an implicit target will solve this problem. -- --- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org http://www.ultranet.com/~pauld/gmake/ "Please remain calm...I may be mad, but I am a professional." --Mad Scientist

Re: Problem with .INTERMEDIATE

I do think this is incorrect behavior. Obviously it doesn't make sense for make to delete a file you explicitly asked it to build, irrespective of whether it's marked .INTERMEDIATE. -- --- Paul D. Smith <[EMAIL

Re: Troubles with .PHONY and pattern rules

you want to use it in an implicit rule search; make won't do implicit rule searches on .PHONY targets. You can use the FORCE alternative to .PHONY, or you can use static pattern rules instead of implicit rules. HTH... -- ----------

Re: problem with the -w flag when doing recursive makes

most unusual situations. HTH... -- ----------- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org http://www.ultranet.com/~pauld/gmake/ "Please remain calm...I ma

Re: Make 3.77, Solaris 7 (or 2.7 or whatever the &!@!%# they call it now)

e the percents in the target, like this: test\%1: file%1 file%2 ; cat $? > $@ Now you'll get: make: *** No rule to make target `file%1', needed by `test%1'. Stop. -- --- Paul D. Smith <[EM

Re: Sequencing of sub-makefile generation

t; > $@ %.d: % ; echo "$@: $<" > $@ a: ; echo "$@" > $@ $ make 13.mk:4: a.d: No such file or directory 13.mk:5: b.mk: No such file or directory echo "a" > a echo "b: a" > b.mk; echo "include b.d"

Re: VPATH

tory names in `VPATH', since the colon can be used in the pathname itself, after the drive letter.) -- --- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org

Re: Suggestion, not a bug

e other things of higher importance on the list for me right now. Contributions always welcome, of course! :). Let me know if you want to try this, so we can get the legal details straight. Thanks. -- ------- Paul D. Sm

Re: Assertion failed

ct nameseq)); - *colonp = ':'; + *p2 = ':'; if (!filenames) { -- ----------- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://w

Re: Bug in make-3.79

st put j> #define LOCALEDIR "" j> in config.h and it works. OK. Thx. -- ------- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org ht

Re: make-3.79 on BeOS

ry afterwards should fix the problem. The ordering of these targets is handled automatically by automake. I'm not aware of any way for automake packages to affect this reliably. I'll ask the automake folks what they think. -- ----

Re: ask for help

bash, there is a shortcut ">&foo" that does the same thing (send both stdout and stderr to file "foo") that you can use if you like. HTH... -- --- Paul D. Smith <[EMAIL PROTECTED]> Find some GNU make tips at: http://www.gnu.org http://www.ultranet.com/~pauld/gmake/ "Please remain calm...I may be mad, but I am a professional." --Mad Scientist

<    3   4   5   6   7   8   9   10   11   12   >