[pkgsrc/trunk]: pkgsrc/net/quagga Update to 0.99.24.1.

To: pkgsrc-changes-hg%NetBSD.org@localhost
Subject: [pkgsrc/trunk]: pkgsrc/net/quagga Update to 0.99.24.1.
From: gdt <gdt%pkgsrc.org@localhost>
Date: Thu, 14 Oct 2021 17:30:26 +0000
details:   https://anonhg.NetBSD.org/pkgsrc/rev/11b6175f255e
branches:  trunk
changeset: 648071:11b6175f255e
user:      gdt <gdt%pkgsrc.org@localhost>
date:      Tue Mar 10 16:46:51 2015 +0000

description:
Update to 0.99.24.1.

Remove patches that were applied upstream.
isisd is enabled, but pimd isn't yet (only because those are upstream defaults).

Upstream changes since 0.99.23:

User-visible changes:
- [pimd] New daemon: pimd provides IPv4 PIM-SSM multicast routing.
- [bgpd] New feature: "next-hop-self all" to override nexthop on iBGP route
    reflector setups.
- [bgpd] route-maps have a new action "set ipv6 next-hop peer-address"
- [bgpd] route-maps have a new action "set as-path prepend last-as"
- [bgpd] Update validity checking (particularly MP-BGP / IPv6 routes) was
    touched up significantly.  Please report possible bugs.
- [ripd] New feature: RIP for IPv4 now supports equal-cost multipath (ECMP)
- [zebra] Multicast RIB support has been extended.  It still is IPv4 only.
- [zebra] "no link-detect" is now printed in configurations since it won't
    be the default anymore soon.  To retain current behaviour, re-save your
    configuration after updating to 0.99.24.

Distributor-visible changes:
- --enable-pimd is added to enable pimd.  It is considered experimental, though
    unless the distribution target is embedded systems with little flash, there
    is no reason to not include it in packages.
- --disable-ipv6 no longer exists as an option.  It's 2015, your C library
    really needs to have IPv6 support by now.
- --disable-netlink no longer exists as an option.  It didn't work anyway.
- --disable-solaris no longer exists as an option.  It only controlled some
    init scripts.
- --enable-isisd is now the default.
- mrlg.cgi is no longer included (it was severely outdated).  It can be found
    independently at http://mrlg.op-sec.us/
- build on Linux with the musl C library should now work

diffstat:

 net/quagga/Makefile                             |   5 +-
 net/quagga/PLIST                                |   5 +-
 net/quagga/distinfo                             |  11 +-
 net/quagga/patches/patch-configure              |  16 ----
 net/quagga/patches/patch-solaris_quagga.init.in |  38 ----------
 net/quagga/patches/patch-zebra-kernel_socket.c  |  90 -------------------------
 6 files changed, 10 insertions(+), 155 deletions(-)

diffs (219 lines):

diff -r ef6ef80487aa -r 11b6175f255e net/quagga/Makefile
--- a/net/quagga/Makefile       Tue Mar 10 16:44:01 2015 +0000
+++ b/net/quagga/Makefile       Tue Mar 10 16:46:51 2015 +0000
@@ -1,8 +1,7 @@
-# $NetBSD: Makefile,v 1.50 2015/02/26 15:32:47 gdt Exp $
+# $NetBSD: Makefile,v 1.51 2015/03/10 16:46:51 gdt Exp $
 #
 
-DISTNAME=      quagga-0.99.23.1
-PKGREVISION=   1
+DISTNAME=      quagga-0.99.24.1
 CATEGORIES=    net
 MASTER_SITES=  http://download.savannah.gnu.org/releases/quagga/
 
diff -r ef6ef80487aa -r 11b6175f255e net/quagga/PLIST
--- a/net/quagga/PLIST  Tue Mar 10 16:44:01 2015 +0000
+++ b/net/quagga/PLIST  Tue Mar 10 16:46:51 2015 +0000
@@ -1,4 +1,4 @@
-@comment $NetBSD: PLIST,v 1.15 2014/12/02 20:22:14 gdt Exp $
+@comment $NetBSD: PLIST,v 1.16 2015/03/10 16:46:51 gdt Exp $
 include/quagga/buffer.h
 include/quagga/checksum.h
 include/quagga/command.h
@@ -52,12 +52,14 @@
 lib/libzebra.la
 man/man1/vtysh.1
 man/man8/bgpd.8
+man/man8/isisd.8
 man/man8/ospfd.8
 man/man8/ripd.8
 man/man8/watchquagga.8
 man/man8/zebra.8
 sbin/babeld
 sbin/bgpd
+sbin/isisd
 sbin/ospfd
 sbin/ripd
 sbin/watchquagga
@@ -71,6 +73,7 @@
 share/examples/quagga/babeld.conf.sample
 share/examples/quagga/bgpd.conf.sample
 share/examples/quagga/bgpd.conf.sample2
+share/examples/quagga/isisd.conf.sample
 share/examples/quagga/log_syslog.conf
 share/examples/quagga/ospfd.conf.sample
 share/examples/quagga/ripd.conf.sample
diff -r ef6ef80487aa -r 11b6175f255e net/quagga/distinfo
--- a/net/quagga/distinfo       Tue Mar 10 16:44:01 2015 +0000
+++ b/net/quagga/distinfo       Tue Mar 10 16:46:51 2015 +0000
@@ -1,9 +1,6 @@
-$NetBSD: distinfo,v 1.22 2015/01/26 00:06:49 gdt Exp $
+$NetBSD: distinfo,v 1.23 2015/03/10 16:46:51 gdt Exp $
 
-SHA1 (quagga-0.99.23.1.tar.gz) = 0501f527383cfa548a800de9816cf1423f6b2336
-RMD160 (quagga-0.99.23.1.tar.gz) = 6102fdf118cf31de8155dfcaffee91207ff56164
-Size (quagga-0.99.23.1.tar.gz) = 2526612 bytes
-SHA1 (patch-configure) = 1b994eaa57c75ac0703e2383c846734f38587224
+SHA1 (quagga-0.99.24.1.tar.gz) = da5d4b1b6f524becb217355f95f99076b7a969d5
+RMD160 (quagga-0.99.24.1.tar.gz) = a6b5c6d93b2d129aa4f627ae0359fa6539de6645
+Size (quagga-0.99.24.1.tar.gz) = 2683800 bytes
 SHA1 (patch-lib_privs.c) = d86ec56c9f57f38af6cce5285fb95759efc7c8e9
-SHA1 (patch-solaris_quagga.init.in) = 85c63089ecc0852f737f3be347d0ad90093d844f
-SHA1 (patch-zebra-kernel_socket.c) = d4fdc39de43c6946a26bd459326f127f36211c8e
diff -r ef6ef80487aa -r 11b6175f255e net/quagga/patches/patch-configure
--- a/net/quagga/patches/patch-configure        Tue Mar 10 16:44:01 2015 +0000
+++ /dev/null   Thu Jan 01 00:00:00 1970 +0000
@@ -1,16 +0,0 @@
-$NetBSD: patch-configure,v 1.4 2015/01/26 00:06:50 gdt Exp $
-
-This patch has not yet been sent upstream.
-
-Extend ip_mreq hack to DragonFlyBSD and SunOS.
---- configure.orig     2013-08-07 13:54:57.000000000 +0000
-+++ configure
-@@ -16637,6 +16637,8 @@ main ()
- {
- #if (defined(__FreeBSD__) && ((__FreeBSD_version >= 500022 && __FreeBSD_version < 700000) || (__FreeBSD_version < 500000 && __FreeBSD_version >= 440000))) || (defined(__NetBSD__) && 
defined(__NetBSD_Version__) && __NetBSD_Version__ >= 106010000) || defined(__OpenBSD__) || defined(__APPLE__)
-   return (0);
-+#elif defined(__DragonFly__) || defined(__sun)
-+  return (0);
- #else
-   #error No support for BSD struct ip_mreq hack detected
- #endif
diff -r ef6ef80487aa -r 11b6175f255e net/quagga/patches/patch-solaris_quagga.init.in
--- a/net/quagga/patches/patch-solaris_quagga.init.in   Tue Mar 10 16:44:01 2015 +0000
+++ /dev/null   Thu Jan 01 00:00:00 1970 +0000
@@ -1,38 +0,0 @@
-$NetBSD: patch-solaris_quagga.init.in,v 1.1 2015/01/25 22:14:05 fhajny Exp $
-
-Fix SMF method based on the manifest we provide.
---- solaris/quagga.init.in.orig        2014-08-25 16:56:53.000000000 +0000
-+++ solaris/quagga.init.in
-@@ -156,16 +156,16 @@ routeadm_zebra_enable () {
-       enable_zebra=`/usr/bin/svcprop -p \
-               routing/enable_zebra $SMF_FMRI 2> /dev/null`
-       if [ "$enable_zebra" != "false" ]; then
--              zenabled=`/usr/bin/svcprop -p general/enabled zebra:quagga`
--              zenabledt=`/usr/bin/svcprop -p general_ovr/enabled zebra:quagga`
-+              zenabled=`/usr/bin/svcprop -p general/enabled quagga:zebra`
-+              zenabledt=`/usr/bin/svcprop -p general_ovr/enabled quagga:zebra`
-               if [ "$zenabled" = "true" -o "$zenabledt" = "true" ]; then
--                      /usr/sbin/svcadm disable zebra:quagga
--                      /usr/sbin/svcadm enable -st zebra:quagga
-+                      /usr/sbin/svcadm disable quagga:zebra
-+                      /usr/sbin/svcadm enable -st quagga:zebra
-               else
--                      /usr/sbin/svcadm enable -st zebra:quagga 
-+                      /usr/sbin/svcadm enable -st quagga:zebra 
-               fi
-               if [ "$?" != "0" ]; then
--                      echo "Could not enable zebra:quagga"
-+                      echo "Could not enable quagga:zebra"
-                       exit $SMF_EXIT_ERR_FATAL
-               fi
-       fi
-@@ -255,7 +255,8 @@ else
-       fi
- fi
- 
--upgrade_config "$DAEMON"
-+# Do not try to upgrade, keep pkgsrc separate
-+# upgrade_config "$DAEMON"
- 
- if [ ! -f "@sysconfdir@/${DAEMON}.conf" ] ; then
-       echo "Could not find config file, @sysconfdir@/${DAEMON}.conf"
diff -r ef6ef80487aa -r 11b6175f255e net/quagga/patches/patch-zebra-kernel_socket.c
--- a/net/quagga/patches/patch-zebra-kernel_socket.c    Tue Mar 10 16:44:01 2015 +0000
+++ /dev/null   Thu Jan 01 00:00:00 1970 +0000
@@ -1,90 +0,0 @@
-$NetBSD: patch-zebra-kernel_socket.c,v 1.2 2015/01/26 00:06:50 gdt Exp $
-
-This patch has already been applied upstream.
-
-From 2fa9994ed80f1b5620cbebf040e85127992ec20d Mon Sep 17 00:00:00 2001
-From: Greg Troxel <gdt%ir.bbn.com@localhost>
-Date: Tue, 2 Dec 2014 14:51:49 -0500
-Subject: [PATCH] zebra/kernel_socket.c: Use platform alignment.
-
-Use the platform-provided RT_ROUNDUP macro to align sockaddrs on the
-routing socket, rather than using hard-coded assumptions about
-alignment.  Emit a warning if the OS doesn't define alignment macros.
-
-Resolves failure of ripngd on NetBSD 6 i386, which changed alignment
-to uint64_t from long.
----
- zebra/kernel_socket.c | 51 +++++++++++++++++++++++++++++++++++++++++----------
- 1 file changed, 41 insertions(+), 10 deletions(-)
-
-diff --git a/zebra/kernel_socket.c b/zebra/kernel_socket.c
-index 3dbeb98..c2ede09 100644
---- zebra/kernel_socket.c
-+++ zebra/kernel_socket.c
-@@ -42,22 +42,53 @@ extern struct zebra_privs_t zserv_privs;
- extern struct zebra_t zebrad;
- 
- /*
-- * Given a sockaddr length, round it up to include pad bytes following
-- * it.  Assumes the kernel pads to sizeof(long).
-+ * Historically, the BSD routing socket has aligned data following a
-+ * struct sockaddr to sizeof(long), which was 4 bytes on some
-+ * platforms, and 8 bytes on others.  bytes.  NetBSD 6 changed the
-+ * routing socket to align to sizeof(uint64_t), which is 8 bytes.  OS
-+ * X appears to align to sizeof(int), which is 4 bytes.
-  *
-- * XXX: why is ROUNDUP(0) sizeof(long)?  0 is an illegal sockaddr
-- * length anyway (< sizeof (struct sockaddr)), so this shouldn't
-- * matter.
-- * On OS X, both 32, 64bit syatems align on 4 byte boundary
-+ * Alignment of zero-sized sockaddrs is nonsensical, but historically
-+ * BSD defines RT_ROUNDUP(0) to be the alignment interval (rather than
-+ * 0).  We follow this practice without questioning it, but it is a
-+ * bug if quagga calls ROUNDUP with 0.
-  */
-+
-+/*
-+ * Because of these varying conventions, the only sane approach is for
-+ * the <net/route.h> header to define a ROUNDUP macro.  We use the
-+ * provided ROUNDUP macro when possible, and provide a workaround with
-+ * a warning if not.
-+ */
-+#if defined(RT_ROUNDUP)
-+#define ROUNDUP(a)    RT_ROUNDUP(a)
-+#endif /* defined(RT_ROUNDUP) */
-+
-+/*
-+ * If ROUNDUP is not defined in terms of platform-provided defines,
-+ * attempt to cope with heuristics.
-+ */
-+#if !defined(ROUNDUP)
-+
-+/*
-+ * It's a bug for a platform not to define rounding/alignment for
-+ * sockaddrs on the routing socket.  This warning really is
-+ * intentional, to provoke filing bug reports with operating systems
-+ * that don't define RT_ROUNDUP or equivalent.
-+ */
-+#warning "net/route.h does not define RT_ROUNDUP; makeing unwarranted assumptions!"
-+
- #ifdef __APPLE__
--#define ROUNDUP(a) \
--  ((a) > 0 ? (1 + (((a) - 1) | (sizeof(int) - 1))) : sizeof(int))
-+#define ROUNDUP_TYPE  long
- #else
--#define ROUNDUP(a) \
--  ((a) > 0 ? (1 + (((a) - 1) | (sizeof(long) - 1))) : sizeof(long))
-+#define ROUNDUP_TYPE  int
- #endif
- 
-+#define ROUNDUP(a) \
-+  ((a) > 0 ? (1 + (((a) - 1) | (sizeof(ROUNDUP_TYPE) - 1))) : sizeof(ROUNDUP_TYPE))
-+
-+#endif /* defined(ROUNDUP) */
-+
- /*
-  * Given a pointer (sockaddr or void *), return the number of bytes
-  * taken up by the sockaddr and any padding needed for alignment.
--- 
-2.1.0
-
Prev by Date: [pkgsrc/trunk]: pkgsrc/doc Updated devel/p5-Class-Base to 0.06
Next by Date: [pkgsrc/trunk]: pkgsrc/net/openntpd Create ${VARBASE}/db/openntpd for the dri...
Previous by Thread: [pkgsrc/trunk]: pkgsrc/doc Updated devel/p5-Class-Base to 0.06
Next by Thread: [pkgsrc/trunk]: pkgsrc/net/openntpd Create ${VARBASE}/db/openntpd for the dri...
Indexes:
Home | Main Index | Thread Index | Old Index