All of lore.kernel.org
 help / color / mirror / Atom feed
* RE: Bug Report:bpf, sockmap: On cleanup we additionally need to remove cached skb
       [not found] <f733db94-c75f-2091-4bae-c99daa5a555d@windriver.com>
@ 2021-10-13 13:57 ` John Fastabend
  0 siblings, 0 replies; only message in thread
From: John Fastabend @ 2021-10-13 13:57 UTC (permalink / raw)
  To: Jun Miao, john.fastabend, cong.wang; +Cc: netdev

Jun Miao wrote:
> Hi,
> 
>    From our netperf stress test with your patch, the Call trace still 
> can reproduce, as follow:

Hi Jun,

Is this testing with sockmap and BPF? If not I can't see how the above
patch is the correct bisection. That patch touches internals of skmsg
handling in a specific backlog case which would only be used in BPF
use case.

Thanks,
John


> 
> *Test Log:*
> 
> [ 6405.990626] Writes:  Total: 12095025292  Max/Min: 379145540/237527212   Fail: 0
> [ 6436.970251] ------------[ cut here ]------------
> [ 6436.975016] WARNING: CPU: 0 PID: 42336 at net/core/stream.c:207 sk_stream_kill_queues+0x11a/0x140
> [ 6436.984044] Modules linked in: i10nm_edac iTCO_wdt intel_pmc_bxt iTCO_vendor_support intel_rapl_msr watchdog intel_rapl_common intel_th_gth x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crct10dif_common aesni_intel crypto_simd cryptd qat_c4xxx(O) intel_qat(O) dh_generic uio intel_spi_pci intel_spi intel_th_pci ice i2c_i801 spi_nor intel_th i2c_smbus i2c_ismt wmi acpi_cpufreq sch_fq_codel openvswitch nsh nf_conncount nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 fuse [last unloaded: i10nm_edac]
> [ 6437.029831] CPU: 0 PID: 42336 Comm: netperf Tainted: G        W  O      5.15.0-rc3 #1
> [ 6437.039374] Hardware name: Intel Corporation JACOBSVILLE/JACOBSVILLE, BIOS JBVLCRB2.86B.0014.P67.2103111848 03/11/2021
> [ 6437.050212] RIP: 0010:sk_stream_kill_queues+0x11a/0x140
> [ 6437.055581] Code: 93 90 02 00 00 85 d2 75 21 8b 83 48 02 00 00 85 c0 75 23 5b 41 5c 5d c3 48 89 df e8 10 d1 fe ff 8b 93 90 02 00 00 85 d2 74 df <0f> 0b 8b 83 48 02 00 00 85 c0 74 dd 0f 0b 5b 41 5c 5d c3 0f 0b eb
> [ 6437.074492] RSP: 0018:ffffac32425d7d90 EFLAGS: 00010282
> [ 6437.079861] RAX: ffffffff9931cfc0 RBX: ffff9cd4eda58d40 RCX: 0000000000000000
> [ 6437.087139] RDX: 00000000fffffb80 RSI: 0000000000000480 RDI: ffff9cd4eda58ea8
> [ 6437.094424] RBP: ffffac32425d7da0 R08: 0000000000000000 R09: 0000000000000000
> [ 6437.101705] R10: 0000000000000000 R11: 0000000000000000 R12: ffff9cd4eda58ea8
> [ 6437.108986] R13: ffff9cd4eda58dc8 R14: 0000000000000000 R15: ffff9cd57079d230
> [ 6437.116263] FS:  00007fe132d96740(0000) GS:ffff9ce3ba200000(0000) knlGS:0000000000000000
> [ 6437.124498] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 6437.130385] CR2: 0000556041d81128 CR3: 0000000165aa4000 CR4: 0000000000350ef0
> [ 6437.137668] Call Trace:
> [ 6437.140259]  inet_csk_destroy_sock+0x64/0x150
> [ 6437.144762]  __tcp_close+0x3b2/0x4e0
> [ 6437.148484]  tcp_close+0x25/0x80
> [ 6437.151858]  inet_release+0x4d/0x90
> [ 6437.155494]  __sock_release+0x3f/0xb0
> [ 6437.159305]  sock_close+0x18/0x20
> [ 6437.162765]  __fput+0xb0/0x280
> [ 6437.165967]  ____fput+0xe/0x10
> [ 6437.169167]  task_work_run+0x61/0xb0
> [ 6437.172889]  exit_to_user_mode_loop+0x114/0x120
> [ 6437.177568]  exit_to_user_mode_prepare+0xe9/0x150
> [ 6437.182421]  syscall_exit_to_user_mode+0x1e/0x60
> [ 6437.187183]  do_syscall_64+0x50/0x90
> [ 6437.190904]  entry_SYSCALL_64_after_hwframe+0x44/0xae
> [ 6437.196097] RIP: 0033:0x7fe132e950c7
> [ 6437.199820] Code: b8 ff ff ff ff e9 3e ff ff ff 66 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 41 c3 48 83 ec 18 89 7c 24 0c e8 33 b1 f8 ff
> [ 6437.218732] RSP: 002b:00007fffc46df2c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000003
> [ 6437.226449] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00007fe132e950c7
> [ 6437.233727] RDX: 0000000000000001 RSI: 00007fffc46df2e4 RDI: 0000000000000004
> [ 6437.241007] RBP: 0000000000000004 R08: 0000000000000000 R09: 0000000000000000
> [ 6437.248286] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001
> [ 6437.255565] R13: 0000000000000001 R14: 0000000000000000 R15: 0000559ace3650c0
> [ 6437.262847] irq event stamp: 0
> [ 6437.266049] hardirqs last  enabled at (0): [<0000000000000000>] 0x0
> [ 6437.272463] hardirqs last disabled at (0): [<ffffffff96d134fd>] copy_process+0x74d/0x18a0
> [ 6437.280786] softirqs last  enabled at (0): [<ffffffff96d134fd>] copy_process+0x74d/0x18a0
> [ 6437.289109] softirqs last disabled at (0): [<0000000000000000>] 0x0
> [ 6437.295524] ---[ end trace 42c5ce4ee9dd57c1 ]---

[...]

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2021-10-13 13:57 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <f733db94-c75f-2091-4bae-c99daa5a555d@windriver.com>
2021-10-13 13:57 ` Bug Report:bpf, sockmap: On cleanup we additionally need to remove cached skb John Fastabend

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.