[src/netbsd-9]: src/doc Tickets #757 - #768

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

details:   https://anonhg.NetBSD.org/src/rev/0c8b855faa0a
branches:  netbsd-9
changeset: 1001562:0c8b855faa0a
user:      martin <martin%NetBSD.org@localhost>
date:      Sun Mar 08 11:22:49 2020 +0000

description:
Tickets #757 - #768

diffstat:

 doc/CHANGES-9.1 |  91 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 files changed, 90 insertions(+), 1 deletions(-)

diffs (102 lines):

diff -r 26a2b12268cf -r 0c8b855faa0a doc/CHANGES-9.1
--- a/doc/CHANGES-9.1   Sun Mar 08 11:21:29 2020 +0000
+++ b/doc/CHANGES-9.1   Sun Mar 08 11:22:49 2020 +0000
@@ -1,4 +1,4 @@
-# $NetBSD: CHANGES-9.1,v 1.1.2.9 2020/03/03 18:56:34 martin Exp $
+# $NetBSD: CHANGES-9.1,v 1.1.2.10 2020/03/08 11:22:49 martin Exp $
 
 A complete list of changes from the NetBSD 9.0 release to the NetBSD 9.1
 release:
@@ -347,3 +347,92 @@
        Avoid running useless makes in every subdirectory.
        [riastradh, ticket #756]
 
+sys/conf/files                                 1.1257-1.1259
+
+       Fix COMPAT_NETBSD32 to properly handle EXEC_AOUT binaries.
+       [pgoyette, ticket #757]
+
+libexec/ld.elf_so/headers.c                    1.68
+libexec/ld.elf_so/map_object.c                 1.61
+libexec/ld.elf_so/rtld.c                       1.203
+
+       PR 55043: fix protection update for the last page of a PT_GNU_RELRO
+       segment.
+       [thorpej, ticket #758]
+
+sys/dev/pci/if_ena.c                           1.24
+
+       PR 55044: replace #if condition for irq_slot with __diagused on the
+       cleanup path.
+       [jdolecek, ticket #759]
+
+sys/arch/amiga/dev/gayle_pcmcia.c              1.32
+sys/arch/amigappc/include/bus_defs.h           1.8
+
+       The Gayle interface uses swapped (little-endian) word accesses, so we
+       need to use the amiga_bus_stride_1swap methods for the word accesses.
+       [is, ticket #760]
+
+sys/compat/netbsd32/netbsd32_compat_43.c       1.60
+sys/compat/netbsd32/syscalls.master            1.132
+sys/compat/netbsd32/netbsd32_syscallargs.h     (regen)
+sys/compat/netbsd32/netbsd32_syscall.h         (regen)
+sys/compat/netbsd32/netbsd32_sysent.c          (regen)
+sys/compat/netbsd32/netbsd32_syscalls.c                (regen)
+sys/compat/netbsd32/netbsd32_syscalls_autoload.c (regen)
+sys/compat/netbsd32/netbsd32_systrace_args.c   (regen)
+
+       PR 55047: properly mark netbsd32_compat_43 syscalls that are
+       directly implemented by calling their compat_43 equivalents.
+       [pgoyette, ticket #761]
+
+sys/arch/amd64/amd64/trap.c                    1.126
+sys/compat/netbsd32/netbsd32_mod.c             1.17,1.18
+sys/sys/compat_stub.h                          1.23
+
+       PR 55038: fix handling of "oosyscall" under compat_netbsd32.
+       [pgoyette, ticket #762]
+
+lib/libkvm/kvm.c                               1.105
+lib/libkvm/kvm_x86_64.c                                1.11
+
+       Only do bounce buffering for character devices.
+       Uuse a binary search in _kvm_pa2off().
+       [chs, ticket #763]
+
+sys/uvm/uvm_fault.c                            1.207
+
+        - in uvm_fault_check(), if the map entry is wired, handle the fault
+          the same way that we would handle UVM_FAULT_WIRE.
+          Faulting on wired mappings is valid if the mapped object was
+          truncated and then later grown again.
+
+        - in uvm_fault_unwire_locked(), we must hold the locks for the
+          vm_map_entry while calling pmap_extract() in order to avoid
+          races with the mapped object being truncated while we are
+          unwiring it.
+       [chs, ticket #764]
+
+sys/dev/acpi/acpi_pci_link.c                   1.25
+
+       Fix a use-after-free bug for extended IRQ resource.
+       [chs, ticket #765]
+
+sys/kern/subr_pool.c                           1.265
+
+       Fix assertions about when it is ok for pool_get() to return NULL.
+       [chs, ticket #766]
+
+sys/ufs/ufs/ufs_dirhash.c                      1.38
+
+       In ufsdirhash_free(), only examine dh->dh_onlist after taking the
+       dirhashlist lock.
+       [chs, ticket #767]
+
+sys/kern/kern_mutex.c                          1.90
+
+       Split an "a && b" assertion into two so it's clear in the dump
+       which condition was not true even if both are true by the time
+       the dump is written.
+       [chs, ticket #768]
+