lkml.org 
[lkml]   [2020]   [Dec]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: memory leak in tls_init
On Thu, 26 Sep 2019 18:19:09 -0700 syzbot wrote:
> 2019/09/26 13:11:21 executed programs: 23
> BUG: memory leak
> unreferenced object 0xffff88810e482a00 (size 512):
> comm "syz-executor.4", pid 6874, jiffies 4295090041 (age 14.090s)
> hex dump (first 32 bytes):
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
> 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
> backtrace:
> [<00000000e93f019a>] kmemleak_alloc_recursive
> include/linux/kmemleak.h:43 [inline]
> [<00000000e93f019a>] slab_post_alloc_hook mm/slab.h:586 [inline]
> [<00000000e93f019a>] slab_alloc mm/slab.c:3319 [inline]
> [<00000000e93f019a>] kmem_cache_alloc_trace+0x145/0x2c0 mm/slab.c:3548
> [<00000000268637bd>] kmalloc include/linux/slab.h:552 [inline]
> [<00000000268637bd>] kzalloc include/linux/slab.h:686 [inline]
> [<00000000268637bd>] create_ctx net/tls/tls_main.c:611 [inline]
> [<00000000268637bd>] tls_init net/tls/tls_main.c:794 [inline]
> [<00000000268637bd>] tls_init+0xbc/0x200 net/tls/tls_main.c:773
> [<00000000f52c33c5>] __tcp_set_ulp net/ipv4/tcp_ulp.c:139 [inline]
> [<00000000f52c33c5>] tcp_set_ulp+0xe2/0x190 net/ipv4/tcp_ulp.c:160
> [<0000000009cb49a0>] do_tcp_setsockopt.isra.0+0x1c1/0xe10
> net/ipv4/tcp.c:2825
> [<00000000b9d96429>] tcp_setsockopt+0x71/0x80 net/ipv4/tcp.c:3152
> [<0000000038a5546c>] sock_common_setsockopt+0x38/0x50
> net/core/sock.c:3142
> [<00000000d945b2a0>] __sys_setsockopt+0x10f/0x220 net/socket.c:2084
> [<000000003c3afaa0>] __do_sys_setsockopt net/socket.c:2100 [inline]
> [<000000003c3afaa0>] __se_sys_setsockopt net/socket.c:2097 [inline]
> [<000000003c3afaa0>] __x64_sys_setsockopt+0x26/0x30 net/socket.c:2097
> [<00000000f7f21cbd>] do_syscall_64+0x73/0x1f0
> arch/x86/entry/common.c:290
> [<00000000d4c003b9>] entry_SYSCALL_64_after_hwframe+0x44/0xa9

#syz invalid

This hasn't happened for over a year, so perhaps the TOE-related
restructuring accidentally fixed it, somehow.

\
 
 \ /
  Last update: 2020-12-29 20:03    [W:0.041 / U:1.752 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site