lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1ecd64a4eb3d0206de49efd753e3c59aa6ec3348.camel@declera.com>
Date:   Mon, 07 Jan 2019 12:22:23 +0200
From:   Yanko Kaneti <yaneti@...lera.com>
To:     Ido Schimmel <idosch@...sch.org>,
        Jeff Kirsher <jeffrey.t.kirsher@...el.com>
Cc:     davem@...emloft.net, netdev@...r.kernel.org, nhorman@...hat.com,
        sassmann@...hat.com,
        Bernhard Kaindl <bernhard.kaindl@...lesgroup.com>,
        Jan Jablonsky <jan.jablonsky@...lesgroup.com>
Subject: Re: [net-next 01/10] igb: reduce CPU0 latency when updating
 statistics

On Tue, 2018-12-25 at 17:47 +0200, Ido Schimmel wrote:
> On Thu, Dec 20, 2018 at 12:56:25PM -0800, Jeff Kirsher wrote:
> > This change is based off of the work and suggestion of Jan Jablonsky
> > <jan.jablonsky@...lesgroup.com>.
> > 
> > The Watchdog workqueue in igb driver is scheduled every 2s for each
> > network interface. That includes updating a statistics protected by
> > spinlock. Function igb_update_stats in this case will be protected
> > against preemption. According to number of a statistics registers
> > (cca 60), processing this function might cause additional cpu load
> >  on CPU0.
> > 
> > In case of statistics spinlock may be replaced with mutex, which
> > reduce latency on CPU0.
> > 
> > CC: Bernhard Kaindl  <bernhard.kaindl@...lesgroup.com>
> > CC: Jan Jablonsky <jan.jablonsky@...lesgroup.com>
> > Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@...el.com>
> > Tested-by: Aaron Brown <aaron.f.brown@...el.com>
> > Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@...el.com>
> 
> Hi,
> 
> I'm hitting following issue with this commit:
> https://lists.01.org/pipermail/lkp/2018-September/009139.html

Hitting what looks like the same thing with linus tip from today and fedora rawhide config. 
There is some rcu splat that might help.. (ignore the drm,it87 bits)

.....
[   17.150603] WARNING: suspicious RCU usage
[   17.150605] 5.0.0-rc1+ #3 Not tainted
[   17.150606] -----------------------------
[   17.150607] include/linux/rcupdate.h:281 Illegal context switch in RCU read-side critical section!
[   17.150608] 
               other info that might help us debug this:

[   17.150609] 
               rcu_scheduler_active = 2, debug_locks = 1
[   17.150610] 2 locks held by sadc/819:
[   17.150610]  #0: 000000006764297e (&p->lock){+.+.}, at: seq_read+0x41/0x430
[   17.150616]  #1: 00000000488cf3bc (rcu_read_lock){....}, at: dev_seq_start+0x5/0x120
[   17.150621] 
               stack backtrace:
[   17.150623] CPU: 3 PID: 819 Comm: sadc Not tainted 5.0.0-rc1+ #3
[   17.150624] Hardware name: Gigabyte Technology Co., Ltd. X470 AORUS ULTRA GAMING/X470 AORUS ULTRA GAMING-CF, BIOS F4 12/24/2018
[   17.150624] Call Trace:
[   17.150628]  dump_stack+0x85/0xc0
[   17.150632]  ___might_sleep+0x100/0x180
[   17.150635]  __mutex_lock+0x55/0x9a0
[   17.241240] [drm] VCE initialized successfully.
[   17.244179]  ? seq_vprintf+0x30/0x50
[   17.244181]  ? seq_printf+0x53/0x70
[   17.244185]  ? __lock_acquire+0x279/0x1650
[   17.244193]  ? igb_get_stats64+0x29/0x80 [igb]
[   17.287914]  igb_get_stats64+0x29/0x80 [igb]
[   17.287920]  dev_get_stats+0x5b/0xc0
[   17.298714]  dev_seq_printf_stats+0x32/0xe0
[   17.298721]  dev_seq_show+0x10/0x30
[   17.307934]  seq_read+0x2fd/0x430
[   17.307940]  proc_reg_read+0x39/0x60
[   17.317677]  __vfs_read+0x36/0x1a0
[   17.317683]  vfs_read+0x9f/0x160
[   17.317686]  ksys_read+0x52/0xc0
[   17.330601]  do_syscall_64+0x60/0x1f0
[   17.330604]  entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   17.330606] RIP: 0033:0x7fcce3c5e1b5
[   17.330608] Code: fe ff ff 50 48 8d 3d 9a d3 09 00 e8 75 00 02 00 0f 1f 44 00 00 f3 0f 1e fa 48 8d 05 65 55 0d 00 8b 00 85 c0 75 0f 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 53 c3 66 90 41 54 49 89 d4 55 48 89 f5 53 89
[   17.330609] RSP: 002b:00007ffdd6d957e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
[   17.375503] RAX: ffffffffffffffda RBX: 00000000012dd2a0 RCX: 00007fcce3c5e1b5
[   17.375504] RDX: 0000000000000400 RSI: 00000000012d89e0 RDI: 0000000000000003
[   17.375505] RBP: 0000000000000d68 R08: 0000000000000001 R09: 0000000000000000
[   17.375506] R10: 00007fcce3a29740 R11: 0000000000000246 R12: 00007fcce3d2a740
[   17.375507] R13: 00007fcce3d2b340 R14: 000000000000007f R15: 00000000012dd2a0
[   17.375577] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:908
[   17.428833] in_atomic(): 1, irqs_disabled(): 0, pid: 819, name: sadc
[   17.428834] 2 locks held by sadc/819:
[   17.428835]  #0: 000000006764297e (&p->lock){+.+.}, at: seq_read+0x41/0x430
[   17.428841]  #1: 00000000488cf3bc (rcu_read_lock){....}, at: dev_seq_start+0x5/0x120
[   17.428847] CPU: 3 PID: 819 Comm: sadc Not tainted 5.0.0-rc1+ #3
[   17.428848] Hardware name: Gigabyte Technology Co., Ltd. X470 AORUS ULTRA GAMING/X470 AORUS ULTRA GAMING-CF, BIOS F4 12/24/2018
[   17.428849] Call Trace:
[   17.428853]  dump_stack+0x85/0xc0
[   17.428857]  ___might_sleep.cold.73+0xac/0xbc
[   17.429616] it87: Found IT8733E chip at 0xa60, revision 3
[   17.429704] it87: Beeping is supported
[   17.506837]  __mutex_lock+0x55/0x9a0
[   17.506842]  ? seq_vprintf+0x30/0x50
[   17.516925]  ? seq_printf+0x53/0x70
[   17.516929]  ? __lock_acquire+0x279/0x1650
[   17.516937]  ? igb_get_stats64+0x29/0x80 [igb]
[   17.533393]  igb_get_stats64+0x29/0x80 [igb]
[   17.533397]  dev_get_stats+0x5b/0xc0
[   17.533401]  dev_seq_printf_stats+0x32/0xe0
[   17.549848]  dev_seq_show+0x10/0x30
[   17.549850]  seq_read+0x2fd/0x430
[   17.559502]  proc_reg_read+0x39/0x60
[   17.559505]  __vfs_read+0x36/0x1a0
[   17.559510]  vfs_read+0x9f/0x160
[   17.559513]  ksys_read+0x52/0xc0
[   17.559517]  do_syscall_64+0x60/0x1f0
[   17.559519]  entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   17.559521] RIP: 0033:0x7fcce3c5e1b5
[   17.559523] Code: fe ff ff 50 48 8d 3d 9a d3 09 00 e8 75 00 02 00 0f 1f 44 00 00 f3 0f 1e fa 48 8d 05 65 55 0d 00 8b 00 85 c0 75 0f 31 c0 0f 05 <48> 3d 00 f0 ff ff 77 53 c3 66 90 41 54 49 89 d4 55 48 89 f5 53 89
[   17.559524] RSP: 002b:00007ffdd6d957e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
[   17.559526] RAX: ffffffffffffffda RBX: 00000000012dd2a0 RCX: 00007fcce3c5e1b5
[   17.559527] RDX: 0000000000000400 RSI: 00000000012d89e0 RDI: 0000000000000003
[   17.559528] RBP: 0000000000000d68 R08: 0000000000000001 R09: 0000000000000000
[   17.559528] R10: 00007fcce3a29740 R11: 0000000000000246 R12: 00007fcce3d2a740
[   17.559529] R13: 00007fcce3d2b340 R14: 000000000000007f R15: 00000000012dd2a0
.....

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ