pkgsrc-WIP-changes archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

dmake: Update to 1.1nb2



Module Name:	pkgsrc-wip
Committed By:	Michael Baeuerle <micha%NetBSD.org@localhost>
Pushed By:	micha
Date:		Mon Aug 30 14:25:45 2021 +0200
Changeset:	2a795875dd64bc23ebd4c2073ba7ea67a305bc66

Modified Files:
	dmake/Makefile
	dmake/distinfo

Log Message:
dmake: Update to 1.1nb2

- SunPro Make: Some static wide character string variables introduced
  by our development activities in March 2018 did not have the needed
  room for the final null character at the end of the string.

  This caused a coredump on z/OS. We added the needed room to these
  variables and finally got a working SunPro Make for z/OS.

  The missing coredump on intel CPUs was most likely a result of the
  byte-order used on these machines. z/OS is using network byte order
  like Sparc. But Sparc users seem to have become rare these days and
  this is the reason, why this bug could stay in the code for more than
  3 years wihout being detected.

  Thanks to a report from Matthew R. Wilson

- SunPro Make: The operators :::= and +:= no longer expand $$ to
  $ while expanding the right side of the assignment. $$ is left
  unchanged in such a case.

  This change is a result of recent discussions for the current POSIX
  standardization and based on a concept change for := from BSD make,
  that was introduced in 2016, see smake below.

- SunPro Make: Calling "dmake -x SUN_MAKE_COMPAT_MODE=value" now works
  as documented. Sun introduced that in the docmentation long ago, but
  did not add working related code. The state since long ago was that
  Sun recognised that the code did not work because the options have
  been scanned too late fo being able to let
  -x SUN_MAKE_COMPAT_MODE=value become effective before parsing the
  makefiles. Sun silently deactivated the related code without changing
  the documentation.

  We now added a workaround for the problem and have been able to
  evaluate -x SUN_MAKE_COMPAT_MODE=value earlier (before compatibility
  mode related features will be used in make) and thus before it is no
  longer possible to change the mode related features anymore.

- SunPro Make: The pseudo targets .POSIX: and .SVR4: now set all
  variables that are used to control the operating mode in dmake.
  This now includes sunpro_compat and gnu_style.

- SunPro Make: Using the pseudo target .POSIX:, calling make under
  the name /usr/xpg4/bin/make or calling

    make -x SUN_MAKE_COMPAT_MODE=posix ...

  now also results in setting DMAKE_ADJUST_MAX_JOBS=M2, as the
  recent changes in POSIX decided to require a pool of job slots to
  be maintained for a make program group in case that the parallel
  mode is enabled via -j N with an N > 0.

  Before, SunPro Make by default did always use an algorithm for
  handling the number of parallel jobs that is only based on the
  system load retrieved via getloadavg().

- SunPro Make: The man page now documents that .POSIX: is silently
  ignored if make was in SVR4 mode before.

- SunPro Make: The man page now documents the .SVR4: pseudo target.

- SunPro Make: The man page added better documentation for the
  .WAIT: pseudo target.

- SunPro Make: The man page now documents the .BUILT_LAST_MAKE_RUN:
  pseudo target. .BUILT_LAST_MAKE_RUN: is automatically inserted
  as a separator into the auto-created file .make.state that is
  maintained by make if .KEEP_STATE: or .KEEP_STATE_FILE: is in
  effect.

- SunPro Make: The man page now mentions that the POSIX operator
  :::= (based on the := operator form smake and BSD make) should be
  preferred over the operator ::= (based on the := operator from
  GNU make) even tough ::= is also a POSIX operator as well.

- SunPro Make: Is now using a new versions date to mark the new
  milestone of features in SunPro Make.

To see a diff of this commit:
https://wip.pkgsrc.org/cgi-bin/gitweb.cgi?p=pkgsrc-wip.git;a=commitdiff;h=2a795875dd64bc23ebd4c2073ba7ea67a305bc66

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.

diffstat:
 dmake/Makefile | 4 ++--
 dmake/distinfo | 8 ++++----
 2 files changed, 6 insertions(+), 6 deletions(-)

diffs:
diff --git a/dmake/Makefile b/dmake/Makefile
index ea2b6f7751..04ad919a92 100644
--- a/dmake/Makefile
+++ b/dmake/Makefile
@@ -1,8 +1,8 @@
 # $NetBSD$
 
-DISTNAME=	schily-2021-07-29
+DISTNAME=	schily-2021-08-14
 PKGNAME=	dmake-1.1
-PKGREVISION=	1
+PKGREVISION=	2
 CATEGORIES=	devel
 MASTER_SITES=	${MASTER_SITE_SOURCEFORGE:=schilytools/}
 EXTRACT_SUFX=	.tar.bz2
diff --git a/dmake/distinfo b/dmake/distinfo
index 8fe040da76..7982521422 100644
--- a/dmake/distinfo
+++ b/dmake/distinfo
@@ -1,8 +1,8 @@
 $NetBSD$
 
-SHA1 (schily-2021-07-29.tar.bz2) = 6de3773faf2aa0418ac761c011736ea05c4a66a8
-RMD160 (schily-2021-07-29.tar.bz2) = cde4b154b66cbb590faa32aae3448c62979e6c18
-SHA512 (schily-2021-07-29.tar.bz2) = 095740bf1190440d386145653763d74ad93abf996e6a1e56279b5441990898515356eda0f7b5d07386d2aac1ba1adf0e9dac423ce49ced4fdc13914857b6afc1
-Size (schily-2021-07-29.tar.bz2) = 4927143 bytes
+SHA1 (schily-2021-08-14.tar.bz2) = 07ca32bfbdac2be868d43729a7a358161f97fd30
+RMD160 (schily-2021-08-14.tar.bz2) = 543946479074743aa81b9d8f8ebbbec235c48002
+SHA512 (schily-2021-08-14.tar.bz2) = 48c015041edf4dd28ce82481cf6cb405ba44de0d695ff5b718d2d1a0c38ad7fee49a023064e123e9dfaa0a04fc4c8a0ba405b41ae16f0946436b7d7debf436a5
+Size (schily-2021-08-14.tar.bz2) = 4942864 bytes
 SHA1 (patch-sunpro_Make_bin_make_common_Makefile.def) = 83f2f7ef8d8aa6fc61566fe8a0301e0038a16c88
 SHA1 (patch-sunpro_Make_bin_make_common_read.cc) = 1bdfe9b623f83239989288a585dc9804e8dbef73


Home | Main Index | Thread Index | Old Index