[src/netbsd-2]: src/doc + [bouyer, ticket #5965]

[riz <riz%NetBSD.org@localhost>]

details:   https://anonhg.NetBSD.org/src/rev/04e95f2119bb
branches:  netbsd-2
changeset: 564334:04e95f2119bb
user:      riz <riz%NetBSD.org@localhost>
date:      Sun Nov 06 00:51:13 2005 +0000

description:
+       [bouyer, ticket #5965]
+       [bouyer, ticket #5967]
+       [bouyer, ticket #5968]
+       [chs, ticket #5969]
+       [chs, ticket #5971]
+       [christos, ticket #5972]
Tickets 5965, 5967-5969, 5971-5972.

diffstat:

 doc/CHANGES-2.2 |  48 +++++++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 47 insertions(+), 1 deletions(-)

diffs (59 lines):

diff -r 3f47db2461e3 -r 04e95f2119bb doc/CHANGES-2.2
--- a/doc/CHANGES-2.2   Sun Nov 06 00:46:50 2005 +0000
+++ b/doc/CHANGES-2.2   Sun Nov 06 00:51:13 2005 +0000
@@ -1,4 +1,4 @@
-#      $NetBSD: CHANGES-2.2,v 1.1.2.11 2005/11/03 21:04:07 riz Exp $
+#      $NetBSD: CHANGES-2.2,v 1.1.2.12 2005/11/06 00:51:13 riz Exp $
 
 A complete list of changes from the NetBSD 2.1 release to the NetBSD 2.2 
 release
@@ -154,3 +154,49 @@
        to be redrawn overwriting most of the main set selection window.
        [dsl, ticket #5964]
 
+sys/uvm/uvm_glue.c                             1.90
+
+       Remove the assertion in uvm_swapout_threads() about LSONPROC lwps
+       not running on the same CPU as the swapper.  l_stat is protected by
+       sched_lock, which isn't held here, so we can race with that lwp
+       starting to run and see its l_cpu not updated yet, as in PR 31870.
+       we check l_stat again in uvm_swapout() while holding sched_lock,
+       so the race itself is harmless.
+       [bouyer, ticket #5965]
+
+usr.sbin/rpc.statd/stat_proc.c                 1.7
+
+       When forking, we keep a reference to the status db file, and if the
+       memory copy of the file is dirty, is will be synced when the child
+       exists.  But the parent may have done changes to the database in
+       the meantime, resulting in a db corruption.
+       Fix this by calling sync_file() just before the fork. The child
+       will still have a reference to the db, but won't try to write it
+       on exit.
+       [bouyer, ticket #5967]
+
+sys/dev/sbus/be.c                              1.46
+
+       ETHERCAP_VLAN_MTU is a flag for ec_capabilities, the size of the 802.1q
+       encaptulation header is ETHER_VLAN_ENCAP_LEN. Pointed out by der Mouse
+       on tech-kern.
+       [bouyer, ticket #5968]
+
+lib/libkvm/kvm_x86_64.c                                1.5
+
+       Fix parsing the page tables.
+       [chs, ticket #5969]
+
+sbin/newfs/mkfs.c                              1.92
+
+       don't set fs_maxsymlinklen and fs_old_inodefmt twice,
+       we set them correctly the first time.  fixes PR 26995.
+       [chs, ticket #5971]
+
+sys/fs/msdosfs/denode.h                                1.10
+sys/fs/msdosfs/msdosfs_vnops.c                 1.22
+
+       PR/32003: Brian Buhrow: msdosfs doesn't properly zero out high
+       cluster data on non-FAT32 msdos filesystems.
+       [christos, ticket #5972]
+