RDS: Back out OoO send status fix since it causes the regression
With the DB build, the crash was observed which was boiled down to
this change on UEK2. Proactively we back this out on UEK4 as well
till the issue gets addresssed.
<0>------------[ cut here ]------------
<2>kernel BUG at net/rds/send.c:511!
<0>invalid opcode: 0000 [#1] SMP
<4>CPU 197
<4>Modules linked in: oracleacfs(P)(U) oracleadvm(P)(U) oracleoks(P)(U)
ipmi_poweroff ipmi_devintf ipmi_si ipmi_msghandler bonding acpi_cpufreq
freq_table mperf rds_rdma rds ib_sdp ib_ipoib rdma_ucm ib_ucm ib_uverbs
ib_umad rdma_cm ib_cm iw_cm ib_addr ipv6 mlx4_ib ib_sa ib_mad ib_core
mlx4_core ext3 jbd fuse ghes hed wmi i2c_i801 iTCO_wdt
iTCO_vendor_support
igb i2c_algo_bit i2c_core ixgbe hwmon dca sg ext4 mbcache jbd2 sd_mod
crc_t10dif megaraid_sas dm_mirror dm_region_hash dm_log dm_mod [last
unloaded: scsi_wait_scan]
<4>
<4>Pid: 85252, comm: oracle_85252_tp Tainted: P
[...]
<0>Call Trace:
<4> [<
ffffffffa0340cab>] rds_send_drop_to+0xcb/0x470 [rds]
<4> [<
ffffffffa033a19e>] rds_release+0x8e/0x110 [rds]
<4> [<
ffffffff8142a369>] sock_release+0x29/0x90
<4> [<
ffffffff8142a3e7>] sock_close+0x17/0x30
<4> [<
ffffffff8116f2fe>] __fput+0xbe/0x240
<4> [<
ffffffff8116f4a5>] fput+0x25/0x30
<4> [<
ffffffff8116b3e3>] filp_close+0x63/0x90
<4> [<
ffffffff8116b4c7>] sys_close+0xb7/0x120
<4> [<
ffffffff81516762>] system_call_fastpath+0x16/0x1b
<0>Code: 00 75 1f 65 8b 14 25 58 c3 00 00 48 63 d2 48 c7 c0 80 26 01 00
48 03
04 d5 40 0e 99 81 48 83 40 68 01 5b 41 5c c9 c3 0f 0b eb fe <0f> 0b eb
fe 0f
1f 40 00 55 48 89 e5 48 83 ec 40 48 89 5d d8 4c
<1>RIP [<
ffffffffa033e5e8>] rds_send_sndbuf_remove+0x68/0x70 [rds]
<4> RSP <
ffff8825b0577dd8>
Revert "RDS: Fix out-of-order RDS_CMSG_RDMA_SEND_STATUS"
This reverts commit
5631f1a303104d41f6ded0d603011d6c172b8644.
Orabug:
21894138
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>