home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
SGI Origin & Onyx2 Patches 1998 May
/
Origin and Onyx2 System Disk Patches May 1998.img
/
dist
/
patchSG0002923.idb
/
usr
/
relnotes
/
patchSG0002923
/
ch1.z
/
ch1
Wrap
Text File
|
1998-04-01
|
11KB
|
397 lines
- 1 -
1. _P_a_t_c_h__S_G_0_0_0_2_9_2_3__R_e_l_e_a_s_e__N_o_t_e
This release note describes patch SG0002923 to release 6.4
of the IRIX operating system.
1.1 _S_u_p_p_o_r_t_e_d__H_a_r_d_w_a_r_e__P_l_a_t_f_o_r_m_s
This patch contains bug fixes for the IP27 and IP30
platforms running release 6.4 of the IRIX operating system.
The software cannot be installed on other configurations.
This patch replaces patches SG0001904, SG0001994, SG0002019,
SG0002308, SG0002453, and SG0002705.
1.2 _S_u_p_p_o_r_t_e_d__S_o_f_t_w_a_r_e__P_l_a_t_f_o_r_m_s
This patch contains bug fixes for the IRIX Execution
Environment (version 1263561140) on a system running release
6.4 of the IRIX operating system. The software cannot be
installed on other configurations.
1.3 _B_u_g_s__F_i_x_e_d__b_y__P_a_t_c_h__S_G_0_0_0_2_9_2_3
This patch contains fixes and performance enhancements for
the following bugs. Bug numbers from Silicon Graphics bug
tracking system are included for reference.
1.3.1 _N_e_w__F_i_x_e_s__i_n__P_a_t_c_h__S_G_0_0_0_2_9_2_3
+o Multicast packets can have bad checksum (Bug #549465).
+o Data corruption seen with patch 2121 installed (Bug
#552691).
+o SIGIO not sent when data drained from TCP (Bug
#568129).
1.3.2 _B_u_g_s__F_i_x_e_d__i_n__P_a_t_c_h__S_G_0_0_0_2_7_0_5
+o Add the necessary stubs to the bsd.a library to support
adding SPANS signaling to the ATM 2.2 software. SPANS
implementation for ATM-XIO card needed in IRIX (Bug
#545698).
+o Support for Gauntlet Patch 2425.
+o The system could crash in in_pcbassign (Bug #550117).
1.3.3 _B_u_g_s__f_i_x_e_d__b_y__P_a_t_c_h__S_G_0_0_0_2_4_5_3
- 2 -
+o GEMS TLBmiss crash (Bug #526256).
This fixes a host of kernel crashes first found on a
GEMS BIT3 VME setup for an OCTANE. Usually the crashes
manifest themselves as tlbmiss or illegal instruction
or a bogus back trace which all point to sthread stack
overflow. This has been fixed.
1.3.4 _B_u_g_s__F_i_x_e_d__b_y__P_a_t_c_h__S_G_0_0_0_2_3_0_8
+o The system could crash in tpisocket_scann_conn_q().
(Bug #489360)
+o rcp I/O error after the process is suspended. (Bug
#497875)
+o Processes could livelock when using UNIX-domain
sockets. (Bug #500939)
+o The system could panic in_pcbassign(). (Bug #513225)
+o The system could panic in mbscavenge(). (Bug #514382)
+o Processes could hang waiting for node-specific memory.
(Bug #516917)
+o The system could appear to be out of mbufs due to zone
memory fragmentation. (Bug #517472)
+o It was not possible to boot certain devices via BOOTP.
(Bug #522016)
+o Stack overflow in 'sockd'. (Bug #526256)
+o netstat could core dump if killed while being run with
-C. (Bug #536542).
+o Crash when using 'fuser <port>/tcp'. (Bug #542760).
1.3.5 _B_u_g_s__F_i_x_e_d__b_y__P_a_t_c_h__S_G_0_0_0_2_0_1_9
+o Transport state was being reported as TS_IDLE instead
of TS_WRES_CIND. (Bug #390346)
+o TCP retransmission granularity is too coarse for very
lossy networks. (Bug #450482)
+o Wrong length in kernel-generated routing socket
messages. (Bug #456853)
- 3 -
+o TCP connections hang in FIN-WAIT-2. (Bug #458244)
+o tpisocket used M_PROTO messages for acknowledgments
that should have been M_PCPROTO. (Bug #459724)
+o Problem with user-level appletalk on 6.2. (Bug
#462312)
+o Potential structure leak in tpisocket. (Bug #467421)
+o Under some circumstances, a bookkeeping error in IP
could cause a crash in ip_freef(). (Bug #468607)
+o The kernel mbuf allocator needs to be improved for
better performance on multi-node systems. (Bug
#469471)
+o Network input performance needs to be improved for
multi-node systems. (Bug #470993)
+o Under some circumstances, the system could crash in
vsocket_release(). (Bug #472592)
+o Invalid socket pointer dereference in tpisocket. (Bug
#473007)
+o RSTs could be ignored in state SYN-RCVD. (Bug #473346)
+o Processing of large numbers of connections in TIME-WAIT
is more expensive than necessary. (Bug #473347)
+o Busy web servers spend too much time copying data from
user-space to kernel-space. (Bug #473348)
+o TCP wakes up applications more often than necessary.
(Bug #473349)
This behavior can be controlled by adjusting the
tunable parameter compatible with release 6.4 as
shipped. For servers which transmit large amounts of
data (e.g. web), setting tcp_wakethresh to the
expression 'NBPP/2' may improve performance.
+o IP_HDRINCL does not work on IP options and ip_id is
bad. (Bug #473385)
+o NET_RT_IFLIST sysctl does not return unconfigured
interfaces. (Bug #477568)
+o socket() could return an incorrect error code. (Bug
#477832)
- 4 -
+o T_DISCON_REQ not processed correctly when in state
T_WRES_CIND. (Bug #480545)
+o IRIX 6.2 and newer have different interface ordering
than 5.3. (Bug #480831)
+o It was possible to panic the system by specifying
illegal ifconfig commands for adding aliases (Bug
#483646).
+o Fix for occasional IP fragmentation checksum errors
under heavy load because ip_id was not MP protected.
(Bug #486283)
+o Processes could hang in sosend(). (Bug #489317)
+o uipc_accept() could sleep in kmem_zone_alloc() while
locked. (Bug #499837)
+o Ensure that fstat() returns correct st_mode for sockets
(Bug #705918).
1.4 _S_u_b_s_y_s_t_e_m_s__I_n_c_l_u_d_e_d__i_n__P_a_t_c_h__S_G_0_0_0_2_9_2_3
This patch release includes these subsystems:
+o patchSG0002923.eoe_sw.svr4net
+o patchSG0002923.eoe_sw.unix
1.5 _I_n_s_t_a_l_l_a_t_i_o_n__I_n_s_t_r_u_c_t_i_o_n_s
Because you want to install only the patches for problems
you have encountered, patch software is not installed by
default. After reading the descriptions of the bugs fixed
in this patch (see Section 1.3), determine the patches that
meet your specific needs.
If, after reading Sections 1.1 and 1.2 of these release
notes, you are unsure whether your hardware and software
meet the requirements for installing a particular patch, run
_i_n_s_t. The _i_n_s_t program does not allow you to install
patches that are incompatible with your hardware or
software.
Patch software is installed like any other Silicon Graphics
software product. Follow the instructions in your _S_o_f_t_w_a_r_e
_I_n_s_t_a_l_l_a_t_i_o_n _A_d_m_i_n_i_s_t_r_a_t_o_r'_s _G_u_i_d_e to bring up the miniroot
form of the software installation tools.
- 5 -
Follow these steps to select a patch for installation:
1. At the Inst> prompt, type
iiiinnnnssssttttaaaallllllll ppppaaaattttcccchhhhSSSSGGGG_x_x_x_x_x_x_x
where _x_x_x_x_x_x_x is the patch number.
2. Initiate the installation sequence. Type
IIIInnnnsssstttt>>>> ggggoooo
3. You may find that two patches have been marked as
incompatible. (The installation tools reject an
installation request if an incompatibility is
detected.) If this occurs, you must deselect one of
the patches.
IIIInnnnsssstttt>>>> kkkkeeeeeeeepppp ppppaaaattttcccchhhhSSSSGGGG_x_x_x_x_x_x_x
where _x_x_x_x_x_x_x is the patch number.
4. After completing the installation process, exit the
_i_n_s_t program by typing
IIIInnnnsssstttt>>>> qqqquuuuiiiitttt
1.6 _P_a_t_c_h__R_e_m_o_v_a_l__I_n_s_t_r_u_c_t_i_o_n_s
To remove a patch, use the _v_e_r_s_i_o_n_s _r_e_m_o_v_e command as you
would for any other software subsystem. The removal process
reinstates the original version of software unless you have
specifically removed the patch history from your system.
vvvveeeerrrrssssiiiioooonnnnssss rrrreeeemmmmoooovvvveeee ppppaaaattttcccchhhhSSSSGGGG_x_x_x_x_x_x_x
where _x_x_x_x_x_x_x is the patch number.
To keep a patch but increase your disk space, use the
_v_e_r_s_i_o_n_s _r_e_m_o_v_e_h_i_s_t command to remove the patch history.
vvvveeeerrrrssssiiiioooonnnnssss rrrreeeemmmmoooovvvveeeehhhhiiiisssstttt ppppaaaattttcccchhhhSSSSGGGG_x_x_x_x_x_x_x
where _x_x_x_x_x_x_x is the patch number.
- 6 -
1.7 _K_n_o_w_n__P_r_o_b_l_e_m_s