OBSOLETE Patch-ID# 147440-15

	***********************************************************
	NOTE: This patch may contain one or more OEM-specific platform ports.
	      See the appropriate OEM_NOTES file within the patch for
	      information specific to these platforms.
	      DO NOT INSTALL this patch on an OEM system if a corresponding
	      OEM_NOTES file is not present (or is present, but instructs not
	      to install the patch), unless the OEM vendor directs otherwise.
	***********************************************************

6986796 SCSI code using _DMA_USES_PHYSADDR has problems
6994014 suspended pools should be reported as SUSPENDED
7001055 scsi_hba_pkt interfaces/support-functions need cleanup
7001057 SCSA tran_setup_pkt does not align extensions
7032667 if MPXIO is enabled on Solaris, the bridge drive will disappear from Netbackup
7047502 assertion failed: pkt->pkt_numcookies <= tran->tran_dma_attr.dma_attr_sgllen
7061583 obsolete function scsi_dmaget() missing from scsi_obsolete.h
7068460 problem with SCTP
7079785 problem with TCP/IP
7103545 SCSA scsi_pkt allocation Method-2 plus
7113528 CONN_DEC_REF panic
 
(from 147440-14)
 
6742788 assertion panic at: zfs:zap_deref_leaf
6964945 inbound IPMP echo replies are delayed in the soft_rings layer of the network stack
6976220 unused bits in dls_soft_ring.h
6997510 observability of dls soft rings could be improved
6997512 dls soft_ring_workers are unbound when a CPU is offlined, and never rebound again
7067640 soft_ring_workerwait_tick and soft_ring_workerwait_ms are unused
7068363 various dls functions are headed with ARGSUSED lint directive, but args are used
7077961 dtrace repeatedly fails to allocate a thread-local variable and finally panics due to stack overflow
7089512 mutex_enter: bad mutex in vdev_dtl_sync() during zpool split
7097870 spill block can be dropped in some situations during incremental receive
7099310 links within an aggregation do not all come back up until all links are up within the aggr
7107814 zfs receive error message always displays feature flags as 0
7150560 S10U10B09 KU patches' log show checkinstall errors after fix for 7145394
 
(from 147440-13)
 
6938116 Solaris regression: not respecting SA_RESTART when syscalls use NFS subsystem
7028550 after fix for CR#6998596, ls command displays ACLs for subdirs, if no ACLs are present
7063556 system hang due to race condition of threads accessing kern_preprom
7096946 Cluster node paniced while being closed down
7108029 with 147440-04 installed on vxvm root, system panics in swapify()
7145394 147440-12/147441-12 and newer should backout the dir ownership changes on patch backout
 
(from 147440-12)
 
6673627 munmap slow when process has many shared mutexes with contention
6811414 problem with NFS
7073951 mac_pkt_hash() causes IP fragments to go out of order
 
(from 147440-11)
 
7076775 system panic occurred while lucreate(1M) on a pool with lesser space
7108689 cpupart_move_cpu() is missing a DISP_BOUND_SKIP_AFF option to disp_bound_check()
7110984 attempt to receive a ZFS stream created on Solaris 11 on to a S10U10 system aborts
7117823 "RPC: Can't decode result" after READ of zero bytes
7126941 copyright year needs updating to 2012 for S10U11
 
(from 147440-10)
 
6199185 netname2user() code has a limit for the number of groups
6373420 panic during network I/O
6940432 false positive TOD STALLED message from tod_validate()
6978340 date(1) causes the message "Time of Day clock error: reason [Stalled]. ..."
7029595 panic in ip:ip_rput_fragment
7044547 kernel rpc should call KEY_GETCRED_3 and get all available gids
7044600 keyserv dumps core when the remote procedure KEY_GETCRED_3 is called
7044891 groups aren't always sorted in the credential
7047829 AUTH_LOOPBACK corrupts data when > 32 groups are available
7068054 Code ordering issue introduced by 6978340
7085977 Solaris 10 system not responding to IGMP membership queries after some time
7107093 fix for 6486764 was only partially backported to Solaris 10
7107096 fix for 7016532 failed - fix that *panic at tcp_xmit_early_reset* problem correctly via this CR
 
(from 147440-09)
 
6878961 problem with NFS
6927023 need support for AMD family 15h processor
6932919 need support for AMD's family 15h ISA enhancements
7030516 hpet_acpi_init() should be called only if hpet is really required
7047435 'genunix: WARNING: preconfig failed: disk' when configure hard disk drive for removal
7105132 deadman panic on a T3-1
 
(from 147440-08)
 
6500167 mpt_ioctl() panics if negative unitid passed in argument
6786168 change mpt driver's property initiator-interconnect-type from "SATA" to "SAS"
6898322 sprunlock() leaks SIGKILL
6973749 stdio ftell() misbehaves in some cases
7099298 allow IPMP test addresses to be used if no data addresses are available
7111032 prunlock() may not honor a pending SIGKILL
7113098 deadlock between pause_cpus and cpu_pause hangs system
 
(from 147440-07)
 
6868727 panic BAD TRAP type=34 addr=deadbeefdeadbeff mmu_fsr=0
6906140 BAD TRAP panic with nfs:nfs4_mnt_kstat_update
6951642 problem with kernel tty
6970123 multihomed SCTP connection should stick to primary path
7000292 changes made for 'PSARC/2010/029 -- Improved ACL interoperability' cause CIFS compatibility issues
7001431 NFSv4 client needs to clean up all stale kstats
7009755 client panic at update_recov_kstats
7040332 delete_child being set when it shouldn't
7041110 mdb STREAMS flag parser definitions wrong
7044516 cannot connect to loopback address
7092682 /dev/zvol/[r]dsk links are not created when importing zpool in 144500-19 & 147440-xx
7108367 remove changes introduced to Solaris 10 by 6898318, 6980085, 7055548, and 7055550
 
(from 147440-06)
 
6463285 problem with TCP/IP
6512250 dtrace_getustack_common() could be improved
7021231 cpupart_attach_cpu with force flag can return EBUSY
7086393 assertion panic when system runs out of preferred large page for DISM
 
(from 147440-05)
 
5102294 Bad Trap 0x31 in nfslog_write_logrecords+0x54() during two "share"
6399681 aggr should adjust its capabilities dynamically according to constituent ports
6586968 interrupts dcmd traverses bogus entries, misses others
6845208 IPsec and near-MTU datagrams act a bit funny, and cause some IPv6 datagrams to drop
6881114 DB redo log stream is not prefetched
6889658 zpool status -v aborts
6896447 new R_SPARC_WDISP10 relocation type
6939851 ZFS hang at vmem_xalloc due to segzio exhaustion
6941744 libsharecore fix_notice function incorrectly advise Solaris 10 users to use sharemgr(1m) command
6977755 libshare incorrectly comments out legacy dfstab shares of ZFS filesystems or sub-dirs on shutdown
6978581 t_look() should never return T_ERROR
7000893 SCTP send() larger data which needs to be fragmented is delayed ~5 seconds
7012256 pools on iSCSI devices unavailable upon boot
7012939 since 6621614 no data will be sent to wire when sendmsg() is called with mss and unsent data exists
7029430 ::ire -v causes mdb to core dump
7034287 sctp_congest_reset() needs also to cleanup the unsent list as it does the xmit list
7038404 outstanding zone references should not cause zone_destroy(2) to hang
7054758 tmp_statvfs() is confused about zone_id vs zone_uniqid
7068461 system panics when the snoop command and the rcp command are executed
7076483 deadlock between aggr and callout threads
7083001 watchdog timeouts seen on max config (64x16GB DIMMs) T4-4 systems, even with fix for 6994535
7084866 aggr: mac_close should be part of aggr_port_free
 
(from 147440-04)
 
6442609 TCP_EXCLBIND handling has problem with server restarting
6828121 sun4u unix ambushed by alloc-mem
6876962 degraded write performance with threads held up by space_map_load_wait()
(removed) 6898318 ZFS root system can hang swapping to zvol
6898885 pool import with cachefile does not check if pool is renamed or in use
6950017 problem with streams
6962304 metaslab_min_alloc_size may be too big
6967597 comment in putback() doesn't make sense
(removed) 6980085 setting invalid volsize on a dump device is possible
6993637 metaslab_group_alloc() should passivate a slab that fails to activate
7000208 space map thrashing affects NFS write throughput
7001813 upgrading pools 2010.Q3 -> 2011.Q1 causes all fmods to become faulted on import
7015535 pool import with cachefile does not check if pool is destroyed
7019636 upgrade from 2010.Q3.3 -> fw_127 faults mirrored fmods
7024586 shmat fails if address is not aligned to largest pagesize
7030684 upgrade from 2010.Q3.3 -> fw_133 followed by pool scrub faults fmods
7033559 metaslab group allocation biasing is broken
7036242 ds_lock should not be held over I/O calls
7037717 page_retire_thread should take into account lbolt rollover on 32-bit machines
7039566 all Interfaces in IPMP group fail if there is not at least one non-deprecated interface
7039969 mpt driver returns EBUSY for passthru commands
7039973 need fix for 6914652 in mpt driver
7039974 need fix for 6878430 in mpt driver
7040974 zdb_002_neg fails
7042770 remove metaslab_smo_bonus_pct code
7043235 callouts not firing due to missing interrupts
7046033 'cfgadm -alv' shows a WHEN date of "Jan  1  1970" for the IOUs
7053560 panic in nfssrv:srv_secinfo_remove with KUP 144488-11
(removed) 7055548 preallocated zvol ends up with small blocksize after certain init failures
(removed) 7055550 preallocating a zvol can fail due to cached dbufs
7060695 zfs rename can result in directory is not empty and out-of-order mounting problems
7062553 need to enable 2GB pagesize support by default
7072532 general protection fault in nfslog_share_record
7074912 NFSv4: delays in getting lock-requests granted, nfs4_block_and_wait doubles interval for polling
7079281 anon_get_next_ptr() does not honor end argument in some cases
7087601 patches for CR 7058265 did not deliver header files
 
(from 147440-03)
 
6437646 pm_name_to_dip() can be simplified
6993538 Solana RQT CSMX16-C05 (SN:1039BDR057): panic page_unlock_noretire: page 70003b80c00 is not locked
7018576 memory dr must be able to split up pages larger than 256M to maintain 256M granularity for dr delete
7031109 16.6% osol_1002-161 hatbench_16G_256M performance regression SPARC
7033859 map_addr_proc() does not support 2G pages properly on sun4v platforms
7075703 coredump due to "Illegal Instruction" in multi-threaded tests on T4
 
(from 147440-02)
 
6491230 problem with procfs
6533904 SCTP: doesn't support traffic class for IPv6
6593014 pools and zones do not correctly deal with offline CPUs when a pset is destroyed
6805468 stack instances introduced a header format error in usr/src/uts/common/inet/sctp/sctp_impl.h
6885065 problem with devfs
6921225 problem with kernel/fs
6994468 recv() with MSG_PEEK causes death by a thousand tiny msgbs after 6666472
6994535 watchdog timeouts seen on small subset of T3/T4 systems
6995260 killing 10's of 1000's of processes does not scale
7002817 IPv6 Traffic Class (TCLASS) header option cannot be set on Solaris 10 running on x86
7013680 need dis support for new instrs
7014100 Solaris cpu-mondo queue filling up on T3/T4 systems
7040407 T4 should not advertise cspare hwcap
7042526 poolbind(1m) can cause processes to be unkillably stuck
7045829 libc_psr memmove needs to avoid use of block init store for T4
7045836 memcpy on T4 could run substantially faster
7047568 perf counter changes for T4 1.2
7049226 hot lock in ire_cache_lookup
7058265 fix for 6919646 can be optimized to use demap-all only when shared context is enabled
7079983 T4 memcpy triggers compiler CR 7076485
 
(from 147440-01)
 
7054149 KU rejuvenation post Solaris 10 Update Release 10
7075118 drivers with null probe entry point fail attach
 
(from 146322-01)
 
        This revision accumulates generic Sustaining patch 146283-01
        into Solaris S10U10 update.
 
(from 146283-01)
 
6382683 lofi is confused about sync/async I/O
6603856 lofi(7D) can thrash the page cache
6743207 lofi page cache fix does not work for non-8K pagesizes
 
(from 147268-01)
 
6216670 NFS server needs a bigger transmit buffer
6261858 ls(1) -l, getfacl(1), and setfacl(1) can return "Permission denied" due to "nobody" and ACLs
6712803 4way NFSv4 deadlock between pvn_vplist_dirty, nfs4_attr_cache and nfs4_putpages
6918232 /etc/ftpd/ftpusers needs to be updated to include unknown, upnp and svctag users
6942081 Solaris doesn't deliver SIGLOST for lost lock when Linux NFS server returns NFS4ERR_BAD_STATEID
6977020 patch 143131-04 or higher causes performance issues with forcedirectio in place
7036028 nfs{3,4}_close() has O(n*n) time complexity after forced umount where n is number of pages
 
(from 147699-01)
 
6976827 unable to remove files once ZFS filesystem is totally filled
7003251 'zfs recv' fails with EINVAL in restore_free()
 
(from 147701-01)
 
7026264 T5120 panicked with MTLP error from LSI 1068E B1 controller
 
(from 146294-03)
 
7011214 UNIX03: /tset/ANSI.hdr/misc/stdlib/T.stdlib{48} /tset/POSIX.hdr/misc/wait/T.wait{10} failed
7052175 UNIX03: VSC /tset/XOPEN.cmd/uucp/uucp.sh {1039} still fails
 
(from 146294-02)
 
6235135 UNIX03/UNIX98: *vsc* uux 1008, 1009, 1010, 1018, 1032, uuto 1029 and uucp 1021 fail
7020236 uucp/uux cannot recognize host names greather than 14 characters
 
(from 146294-01)
 
        This revision accumulates generic Sustaining patch 146279-01
        into Solaris S10U10 update.
 
(from 146279-01)
 
6177817 uucp uses unsafe strcat() routines for processing "-s" & "-m" options
 
(from 137276-01)
 
        This revision accumulates generic Sustaining patch 125902-02
        into Solaris S10U5 update.
 
(from 125902-02)
 
6542991 UUCICO reports ASSERT error after installing su driver patch 118557-07
 
(from 125902-01)
 
        This revision accumulates generic Sustaining patch 125804-01
        into Solaris S10U4 update.
 
(from 125804-01)
 
6500236 several times a day uucp stops working and creates a CPU workload up to 100%
 
(from 147012-02)
 
        This revision accumulates generic Sustaining patch 146767-02
        into Solaris S10U10 update.
 
(from 147012-01)
 
        This revision accumulates generic Sustaining patch 146767-01
        into Solaris S10U10 update.
 
(from 146767-02)
 
6993217 zoneadm install -b aborts instead of removing patches in the correct order before installing
 
(from 146767-01)
 
6757506 race can cause zoneadm to emit 'zoneadm: failed to get zone name: Invalid argument' messages
 
(from 147258-01)
 
        This revision accumulates generic Sustaining patch 144534-02
        into Solaris S10U10 update.
 
(from 144534-02)
 
6418178 system can panic if MAC notifications come immediately after a port is added into an aggregation
 
(from 144534-01)
 
6946534 aggr interface not multicast capable wrt all hosts group on reboot
6959661 possible memory leaks in aggr_m_multicst
 
(from 147436-01)
 
7083331 copy relocation against partially initialized bss does not do the copy
 
(from 147713-01)
 
6987395 fgetgrent_r() only handles input lines up to 2K regardless of the supplied buffer length
7019305 problem with library libc
 
(from 146575-03)
 
6642281 substr() doesn't properly null-terminate
6642734 substr() can do the wrong thing for invalid index values
6642972 substr() is broken with a negative third argument
 
(from 146575-02)
 
6981293 potential memory leak in fasttrap_ioctl()
6981301 potential memory leak in pm_ioctl()
6986248 missing unlock in dr_attach()
 
(from 146575-01)
 
        This revision accumulates generic Sustaining patch 143941-02
        into Solaris S10U10 update.
 
(from 143941-02)
 
6712247 dtrace -c runs the program despite errors
 
(from 143941-01)
 
        This revision accumulates generic Sustaining patch 141590-02
        into Solaris S10U9 update.
 
(from 141590-02)
 
6677812 race between dtrace activities in non-local zones
6718877 crgetzoneid() use can cause problems when forking processes with USDT providers in a non global zone
6735480 race between probe enabling and provider registration
6886953 large symbols lead to stack exhaustion
 
(from 141590-01)
 
	This revision accumulates generic Sustaining patch 141765-01
	into Solaris S10U8 update.
 
(from 141765-01)
 
6823388 DTrace ioctl handlers must validate all structure members
 
(from 146320-03)
 
        This revision accumulates generic Sustaining patch 140563-03
        into Solaris S10U10 update.
 
(from 146320-02)
 
6985890 UNIX03 VSU CAPI.os/streams/ioctl/T.ioctl{19,99} FAILS intermittently
 
(from 146320-01)
 
        This revision accumulates generic Sustaining patch 146281-01
        into Solaris S10U10 update.
 
(from 146281-01)
 
6919748 panic caused by bad mblk_t after a call to ldtermrput
 
(from 140563-03)
 
7047554 backout 6651616 - fix requires more work
 
(from 140563-02)
 
4491376 extraneous statement in ptem.c
4726809 duplicate M_IOCNAK from 'pts'
5012244 ptslwput() holds lock across putnext()
5021554 ldterm doesn't honor flow control when handling upstream M_BREAK messages
(removed) 6651616 ptm streams driver misbehaving in some corner cases
6955019 oplmsu maintainability
6966955 ptem.c cstyle failures
 
(from 140563-01)
 
        This revision accumulates generic Sustaining patch 140383-01
        into Solaris S10U7 update.
 
(from 140383-01)
 
6433954 ptc and ptsl needs to have properly sequenced code
 
(from 147770-01)
 
7034726 multi-threaded service calls acl_fromtext() from multiple threads at the same time and dumps core
 
(from 147157-02)
 
7020180 postinstall script of SUNWn2cp.v in 142909-17 doesn't properly add n2cp driver aliases
 
(from 147157-01)
 
7041435 new assertion panic occurring in common/os/cpu.c on sun4v T3 machines.
7048323 panic seen in n2cp_cwq_q_unconfigure() while running numa test suite from STC
 
(from 148229-01)
 
7080247 automount home directory access can trigger gssd circular dependency
 
(from 147149-01)
 
6989192 I/Os are stuck in mptsas
7052663 panic BAD TRAP: type=31 occurred in module "unix" due to a NULL pointer dereference
7065412 t4-2 panics during disk/DVD testing
 
(from 148098-03)
 
6228435 undecoded command in var/adm/messages - Error for Command: undecoded cmd 0x5a
6241086 format should allow label adjustment when disk/LUN size changes
6430818 Solaris needs mechanism of dynamically increasing LUN size
6925496 kernel panic in ata_wait when opening sound-juicer application (intermittent)
7026372 x86 is reporting "Error setting up next portion of DMA transfer" after each exception
7068614 add DELL MD32xx/MD32xxi to lsi_asymmetric_fops.c and sd.c/ssd.c
 
(from 148098-02)
 
6981452 scsi_watch_thread is not well-serialized
 
(from 148098-01)
 
6938643 Erie PCI-EM hotplug failed to detach driver
 
(from 144524-03)
 
6625553 BAD TRAP: type=31 rp=2a100297730 addr=0 mmu_fsr=0 occurred in module "genunix"
7068412 scsi_vhci can hold LUN with VHCI_HOLD_LUN() indefinitely
7085968 VHCI_PKT_IN_FAILOVER flag is never cleared
 
(from 144524-02)
 
7079724 system panics in scsi_vhci:hds_sym_path_get_opinfo after installing patch 144501-19
 
(from 144524-01)
 
7010193 Reservation Conflict on path failback with IBM SVC
7068425 I/O is not always quiesced when SCSI-2 reservation is sent
7075285 panic while sending reset target to a FC device on S10U10
 
(from 145937-01)
 
6937968 request addition to conversion_array in rdwr_efi.c to support CDS layout on disks greater than 1TB
 
(from 147784-01)
 
7017186 "WARNING:Command failed to complete...Device is gone" generated when resetting SP
7068505 disk unavailable after re-insertion

--------------------------------
 
Please refer to the man pages for instructions on using 'patchadd'
and 'patchrm' commands provided with Solaris.
 
The following example installs a patch to a standalone machine:
 
       example# patchadd /var/spool/patch/123456-07
 
The following example removes a patch from a standalone system:
 
       example# patchrm 123456-07
 
For additional examples please see the appropriate man pages. Any
other special or non-generic installation instructions should be
described below as special instructions.

-----------------------------
 
NOTE 1:  Following installation of Solaris kernel patch 147440-02 (or greater)
         on sun4v systems, a panic will occur with "xt_sync:timeout" after the
         system is rebooted, if the firmware is not at revision 6.4.6 or later.
 
         Please ensure firmware on sun4v hardware is version 6.4.6 or higher.
 
         The following examples illustrate how to check the firmware version:
 
         Example ALOM:
 
         sc> showhost
         Sun System Firmware 7.1.5.b 2008/08/05 16:58
 
         Example ILOM:
 
         -> show /HOST/sysfw_version
 
         if the show command fails, then issue the following commands:
 
         > cd /HOST
         /HOST
 
         -> ls
 
         /HOST
            Targets:
               bootmode
               console
               diag
               domain
               tpm
 
             Properties:
                ....
                sysfw_version = Sun System Firmware 7.3.0.build_35 2010/09/09 06:07
 
         If firmware is not at the required level then the system will fail
         to boot due to CR 7098393 (xt_sync timeout on sun4v following
         installation of patch 147440-02 or later).
 
NOTE 2:  Due to the fix for CR 6898318 (ZFS root system can hang swapping to
         zvol) having been removed from 147440-04 (it was initially fixed in
         147440-04), users can now hit CR 6898318.  CR 6898318 only affects
         systems with ZFS root; the workaround is to set primarycache=metadata
         for all zvol based swap devices.
 
         For example:
 
         # swap -l
         swapfile             dev  swaplo blocks   free
         /dev/zvol/dsk/rpool/swap 256,1      16 10246128 10246128
         #
 
         shows that we have a zvol swap device rpool/swap, so then we run:
 
         # zfs set primarycache=metadata rpool/swap
 
         If there were other zvol swap devices, then we have to set
         primarycache=metadata for each one.
 
NOTE 3:  Fix to 7097870 (spill block can be dropped in some situations during
         incremental receive) introduces backward incompatibility with zfs
         send/recv.  I.e., ZFS streams created with this patch will not be
         receivable with older ZFS versions and it will fail with below error
         message in destination host:
 
         "cannot receive: stream has unsupported feature, feature flags = 24"
 
         This change is to allow important fix in ZFS to avoid metadata
         corruptions related to ACL.  An upgrade to same or greater version
         of ZFS is required in destination for ZFS streams to work properly.
 
NOTE 4:  To work around delayed processing of inbound network traffic described
         by CR 6964945 (inbound IPMP echo replies are delayed in the soft_rings
         layer of the network stack), add the following lines to /etc/system
         and reboot:
 
         set dls:soft_ring_allow_bind = 0
         set dls:soft_ring_worker_pri = 99
 
         If this does not resolve the issue fully, add the following line:
 
         set ip:ip_squeue_bind = 0
 
         If this still does not resolve the issue, then either CR 6964945 is
         not the cause, or the system is too busy for the workload.

NOTE 5:  When using SCTP with unreliable networks, a system panic in
         sctp_rotate_faddr() may occur.  To fix, please also install
         the following patch:

         150400-10 (or greater)  kernel patch

         For more details, see Service Alert Document 1607306.1:

         https://support.oracle.com/rs?type=doc&id=1607306.1