APAR status
Closed as program error.
Error description
There is a deadlock kind of situation in the routines rrent_output and ethchan_link_status_callback_func resulting in lock contention problems among some threads making all the LPAR's inaccessible. Call stacks of these CPU's are as follows: --- CPU #0 --- pvthread+00AA00 STACK: ?005983D4?drw_lock_write+000114 (0000000000000001 ????) ?00014F50?.kernel_add_gate_cstack+000030 () ?F1000000C05D5D74?ethchandd:ethchan_link_status_callback_ func+000094 (??, ??, ??) ?00014D70?.hkey_legacy_gate+00004C () ?047E57CC?rrentdd:rrent_event_notify+000B4C (??, ??, ??) ?047E11F8?rrentdd:librrhba_user_errproc+0004F8 (??, ??, ??) ?0020A1F0?procentry+000010 (??, ??, ??, ??) ?kdb_read_mem? no real storage @ FFFFFFFFFFF8900 . --- CPU #8 --- pvthread+00C600 STACK: ?0059B270?slock+000330 (0000000002C70000, 0000000000001000 ????) ?000090C4?.disable_lock+0000C0 () ?047EBA60?rrentdd:rrent_output+0002C0 (??, ??) ?F1000000C05D7A7C?ethchandd:ethchan_output+0002FC (??, ??) ?F1000000C0601BE8?seadd:sea_virt_input_post_q+000728 (F1000A0035AC0000, F1000A0029E1A6E8, F1000E000018B600) ?F1000000C0600AA8?seadd:sea_send_packet+000108 (??, ??, ??) ?F1000000C0600068?seadd:sea_thread+0002A8 (??) ?00014D70?.hkey_legacy_gate+00004C () ?003BC0D4?threadentry+000094 (??, ??, ??, ??) ?kdb_read_mem? no real storage @ FFFFFFFFFFF8900 . --- CPU #12 --- pvthread+00C500 STACK: ?005949F4?drw_lock_read_ppc+000954 (F1000A00356A3448 ????) ?F1000000C05D7848?ethchandd:ethchan_output+0000C8 (??, ??) ?F1000000C0601BE8?seadd:sea_virt_input_post_q+000728 (F1000A0035AC0000, F1000A0029E1A6E8, F1000E00011BD200) ?F1000000C0600AA8?seadd:sea_send_packet+000108 (??, ??, ??) ?F1000000C0600068?seadd:sea_thread+0002A8 (??) ?00014D70?.hkey_legacy_gate+00004C () ?003BC0D4?threadentry+000094 (??, ??, ??, ??) ?kdb_read_mem? no real storage @ FFFFFFFFFFF8900 . CPU 8 is spinning in slock code for lock F1000A002BE5D600. . (8)> slk @r31 F1000A002BE5D600 krlock association: kp:F1000A03F00F5F00 : 1 total threads contending: 00000001 gp:0: F1000A03F00F6F00 : krl owner: 8 lock FE00100000AA005D thread_owner............ 00AA005D @ pvthread+00AA00 (8)> status | grep AA005D 0 AA005D 170 5600B4 86 librrhba_errproc (8)> . CPU 0 is in drw_lock_write code and the lock address passed to drw_lock_write was F1000A00356A3448. In cpu 12,the call stack includes drw_lock_read_ppc and the lock address passed is the same i.e. F1000A00356A3448.
Local fix
Problem summary
Etherchannel over CT3 adapter may result in hang situation due to deadlock between rrent_output function and ethchan_link_status_callback_func function.
Problem conclusion
CT3 driver code is modified to fix the lock contention problem.
Temporary fix
Comments
6100-09 - use AIX APAR IV76243 6100-09 - use AIX APAR IV76243 7100-04 - use AIX APAR IV84544 7200-01 - use AIX APAR IV84650
APAR Information
APAR number
IV84544
Reported component name
AIX V7.1
Reported component ID
5765H4000
Reported release
710
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2016-05-09
Closed date
2016-05-09
Last modified date
2017-01-20
APAR is sysrouted FROM one or more of the following:
IV76243
APAR is sysrouted TO one or more of the following:
Fix information
Fixed component name
AIX V7.1
Fixed component ID
5765H4000
Applicable component levels
R710 PSY U869863
UP16/10/25 I 1000
[{"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SG11R"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"710","Line of Business":{"code":"LOB08","label":"Cognitive Systems"}}]
Document Information
Modified date:
19 April 2022