Message-ID: <e632e378-d019-4de7-8f13-07c572ab37a9@samsung.com>
Date: Mon, 12 Aug 2024 18:55:26 +0200
From: Marek Szyprowski <m.szyprowski@...sung.com>
To: Jiri Pirko <jiri@...nulli.us>
Cc: netdev@...r.kernel.org, davem@...emloft.net, edumazet@...gle.com,
	kuba@...nel.org, pabeni@...hat.com, mst@...hat.com, jasowang@...hat.com,
	xuanzhuo@...ux.alibaba.com, virtualization@...ts.linux.dev, ast@...nel.org,
	daniel@...earbox.net, hawk@...nel.org, john.fastabend@...il.com,
	dave.taht@...il.com, kerneljasonxing@...il.com, hengqi@...ux.alibaba.com
Subject: Re: [PATCH net-next v3] virtio_net: add support for Byte Queue
 Limits

On 12.08.2024 18:47, Jiri Pirko wrote:
> Mon, Aug 12, 2024 at 04:57:24PM CEST, m.szyprowski@...sung.com wrote:
>> On 18.06.2024 16:44, Jiri Pirko wrote:
>>> From: Jiri Pirko <jiri@...dia.com>
>>>
>>> Add support for Byte Queue Limits (BQL).
>>>
>>> Tested on qemu emulated virtio_net device with 1, 2 and 4 queues.
>>> Tested with fq_codel and pfifo_fast. Super netperf with 50 threads is
>>> running in background. Netperf TCP_RR results:
>>>
>>> NOBQL FQC 1q:  159.56  159.33  158.50  154.31    agv: 157.925
>>> NOBQL FQC 2q:  184.64  184.96  174.73  174.15    agv: 179.62
>>> NOBQL FQC 4q:  994.46  441.96  416.50  499.56    agv: 588.12
>>> NOBQL PFF 1q:  148.68  148.92  145.95  149.48    agv: 148.2575
>>> NOBQL PFF 2q:  171.86  171.20  170.42  169.42    agv: 170.725
>>> NOBQL PFF 4q: 1505.23 1137.23 2488.70 3507.99    agv: 2159.7875
>>>     BQL FQC 1q: 1332.80 1297.97 1351.41 1147.57    agv: 1282.4375
>>>     BQL FQC 2q:  768.30  817.72  864.43  974.40    agv: 856.2125
>>>     BQL FQC 4q:  945.66  942.68  878.51  822.82    agv: 897.4175
>>>     BQL PFF 1q:  149.69  151.49  149.40  147.47    agv: 149.5125
>>>     BQL PFF 2q: 2059.32  798.74 1844.12  381.80    agv: 1270.995
>>>     BQL PFF 4q: 1871.98 4420.02 4916.59 13268.16   agv: 6119.1875
>>>
>>> Signed-off-by: Jiri Pirko <jiri@...dia.com>
>>> ---
>>> v2->v3:
>>> - fixed the switch from/to orphan mode while skbs are yet to be
>>>     completed by using the second least significant bit in virtqueue
>>>     token pointer to indicate skb is orphan. Don't account orphan
>>>     skbs in completion.
>>> - reorganized parallel skb/xdp free stats accounting to napi/others.
>>> - fixed kick condition check in orphan mode
>>> v1->v2:
>>> - moved netdev_tx_completed_queue() call into __free_old_xmit(),
>>>     propagate use_napi flag to __free_old_xmit() and only call
>>>     netdev_tx_completed_queue() in case it is true
>>> - added forgotten call to netdev_tx_reset_queue()
>>> - fixed stats for xdp packets
>>> - fixed bql accounting when __free_old_xmit() is called from xdp path
>>> - handle the !use_napi case in start_xmit() kick section
>>> ---
>>>    drivers/net/virtio_net.c | 81 ++++++++++++++++++++++++++++------------
>>>    1 file changed, 57 insertions(+), 24 deletions(-)
>> I've recently found an issue with virtio-net driver and system
>> suspend/resume. Bisecting pointed to the c8bd1f7f3e61 ("virtio_net: add
>> support for Byte Queue Limits") commit and this patch. Once it got
>> merged to linux-next and then Linus trees, the driver occasionally
>> crashes with the following log (captured on QEMU's ARM 32bit 'virt'
>> machine):
>>
>> root@...get:~# time rtcwake -s10 -mmem
>> rtcwake: wakeup from "mem" using /dev/rtc0 at Sat Aug 10 12:40:26 2024
>> PM: suspend entry (s2idle)
>> Filesystems sync: 0.000 seconds
>> Freezing user space processes
>> Freezing user space processes completed (elapsed 0.006 seconds)
>> OOM killer disabled.
>> Freezing remaining freezable tasks
>> Freezing remaining freezable tasks completed (elapsed 0.001 seconds)
>> ------------[ cut here ]------------
>> kernel BUG at lib/dynamic_queue_limits.c:99!
>> Internal error: Oops - BUG: 0 [#1] SMP ARM
>> Modules linked in: bluetooth ecdh_generic ecc libaes
>> CPU: 1 PID: 1282 Comm: rtcwake Not tainted
>> 6.10.0-rc3-00732-gc8bd1f7f3e61 #15240
>> Hardware name: Generic DT based system
>> PC is at dql_completed+0x270/0x2cc
>> LR is at __free_old_xmit+0x120/0x198
>> pc : [<c07ffa54>]    lr : [<c0c42bf4>]    psr: 80000013
>> ...
>> Flags: Nzcv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
>> Control: 10c5387d  Table: 43a4406a  DAC: 00000051
>> ...
>> Process rtcwake (pid: 1282, stack limit = 0xfbc21278)
>> Stack: (0xe0805e80 to 0xe0806000)
>> ...
>> Call trace:
>>   dql_completed from __free_old_xmit+0x120/0x198
>>   __free_old_xmit from free_old_xmit+0x44/0xe4
>>   free_old_xmit from virtnet_poll_tx+0x88/0x1b4
>>   virtnet_poll_tx from __napi_poll+0x2c/0x1d4
>>   __napi_poll from net_rx_action+0x140/0x2b4
>>   net_rx_action from handle_softirqs+0x11c/0x350
>>   handle_softirqs from call_with_stack+0x18/0x20
>>   call_with_stack from do_softirq+0x48/0x50
>>   do_softirq from __local_bh_enable_ip+0xa0/0xa4
>>   __local_bh_enable_ip from virtnet_open+0xd4/0x21c
>>   virtnet_open from virtnet_restore+0x94/0x120
>>   virtnet_restore from virtio_device_restore+0x110/0x1f4
>>   virtio_device_restore from dpm_run_callback+0x3c/0x100
>>   dpm_run_callback from device_resume+0x12c/0x2a8
>>   device_resume from dpm_resume+0x12c/0x1e0
>>   dpm_resume from dpm_resume_end+0xc/0x18
>>   dpm_resume_end from suspend_devices_and_enter+0x1f0/0x72c
>>   suspend_devices_and_enter from pm_suspend+0x270/0x2a0
>>   pm_suspend from state_store+0x68/0xc8
>>   state_store from kernfs_fop_write_iter+0x10c/0x1cc
>>   kernfs_fop_write_iter from vfs_write+0x2b0/0x3dc
>>   vfs_write from ksys_write+0x5c/0xd4
>>   ksys_write from ret_fast_syscall+0x0/0x54
>> Exception stack(0xe8bf1fa8 to 0xe8bf1ff0)
>> ...
>> ---[ end trace 0000000000000000 ]---
>> Kernel panic - not syncing: Fatal exception in interrupt
>> ---[ end Kernel panic - not syncing: Fatal exception in interrupt ]---
>>
>> I have fully reproducible setup for this issue. Reverting it together
>> with f8321fa75102 ("virtio_net: Fix napi_skb_cache_put warning") (due to
>> some code dependencies) fixes this issue on top of Linux v6.11-rc1 and
>> recent linux-next releases. Let me know if I can help debugging this
>> issue further and help fixing.
> Will fix this tomorrow. In the meantime, could you provide full
> reproduce steps?

Well, it is easy to reproduce it simply by calling

# time rtcwake -s10 -mmem

a few times and sooner or later it will cause a kernel panic.

Best regards
-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland

Powered by blists - more mailing lists