* [MPTCP] Re: [MPTCP][PATCH v2 mptcp-next 2/3] mptcp: send ack for rm_addr
@ 2021-03-17 4:09 Geliang Tang
0 siblings, 0 replies; 3+ messages in thread
From: Geliang Tang @ 2021-03-17 4:09 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 5515 bytes --]
Hi Mat,
Mat Martineau <mathew.j.martineau(a)linux.intel.com> 于2021年3月17日周三 上午8:31写道:
>
> On Tue, 16 Mar 2021, Geliang Tang wrote:
>
> > This patch changes the sending ACK conditions for the ADD_ADDR, send an
> > ACK packet for RM_ADDR too.
> >
> > In mptcp_pm_remove_addr, invoke mptcp_pm_nl_add_addr_send_ack to send
> > the ACK packet.
> >
> > Signed-off-by: Geliang Tang <geliangtang(a)gmail.com>
> > ---
> > net/mptcp/pm.c | 1 +
> > net/mptcp/pm_netlink.c | 10 +++++-----
> > net/mptcp/protocol.h | 1 +
> > 3 files changed, 7 insertions(+), 5 deletions(-)
> >
>
> Thanks for the v2 update!
>
> The code looked ok on visual inspection, but I ran in to problems with
> mptcp_join.sh tests (looks like the "remove id 0 address" test that
> triggers it):
Did you apply the patch (Squash to "mptcp: remove id 0 address") before
applying this patch set? We need to apply the squash-to patch first.
I didn't get this "bad unlock balance" warning on my test, could you please
share your kconfig file to me?
Thanks.
-Geliang
>
> [ 286.809888] =====================================
> [ 286.810818] WARNING: bad unlock balance detected!
> [ 286.811921] 5.12.0-rc2+ #5 Tainted: G W
> [ 286.813119] -------------------------------------
> [ 286.814218] pm_nl_ctl/3445 is trying to release lock (&msk->pm.lock)
> at:
> [ 286.815822] [<ffffffffa5ede78b>] mptcp_pm_nl_addr_send_ack+0xab/0x1a0
> [ 286.816979] but there are no more locks to release!
> [ 286.817795]
> [ 286.817795] other info that might help us debug this:
> [ 286.818874] 3 locks held by pm_nl_ctl/3445:
> [ 286.819598] #0: ffffffffa7514770 (cb_lock){++++}-{3:3}, at:
> genl_rcv+0x10/0x30
> [ 286.820982] #1: ffffffffa7514828 (genl_mutex){+.+.}-{3:3}, at:
> genl_rcv_msg+0x1f5/0x2c0
> [ 286.822520] #2: ffff888104840120 (sk_lock-AF_INET){+.+.}-{0:0}, at:
> mptcp_nl_remove_id_zero_address+0x10f/0x280
> [ 286.824648]
> [ 286.824648] stack backtrace:
> [ 286.825355] CPU: 1 PID: 3445 Comm: pm_nl_ctl Tainted: G W
> 5.12.0-rc2+ #5
> [ 286.826666] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> 1.14.0-1.fc33 04/01/2014
> [ 286.828234] Call Trace:
> [ 286.828688] dump_stack+0x93/0xc2
> [ 286.829296] lock_release.cold+0x10/0x4e
> [ 286.830000] ? mptcp_pm_nl_addr_send_ack+0xab/0x1a0
> [ 286.830833] ? lock_downgrade+0x390/0x390
> [ 286.831553] ? __mptcp_flush_join_list+0x20/0x180
> [ 286.832480] _raw_spin_unlock_bh+0x12/0x30
> [ 286.833409] mptcp_pm_nl_addr_send_ack+0xab/0x1a0
> [ 286.834453] mptcp_pm_remove_addr+0x7a/0xb0
> [ 286.835203] mptcp_nl_remove_id_zero_address+0x11f/0x280
> [ 286.836204] ? lookup_subflow_by_saddr+0x1a0/0x1a0
> [ 286.837116] ? mptcp_nl_fill_addr+0x320/0x320
> [ 286.837848] ? lock_release+0x1d0/0x400
> [ 286.838565] ? lock_release+0x1d0/0x400
> [ 286.839302] mptcp_nl_cmd_del_addr+0x497/0x5e0
> [ 286.840130] ? mptcp_pm_remove_anno_addr.isra.0+0x180/0x180
> [ 286.841165] ? nla_get_range_signed+0x1c0/0x1c0
> [ 286.842053] ? lock_is_held_type+0x98/0x110
> [ 286.842820] ? __nla_parse+0x22/0x30
> [ 286.843503] ? genl_family_rcv_msg_attrs_parse.constprop.0+0xd1/0x130
> [ 286.844739] genl_family_rcv_msg_doit+0x17a/0x200
> [ 286.845641] ? genl_family_rcv_msg_attrs_parse.constprop.0+0x130/0x130
> [ 286.847167] ? security_capable+0x48/0x60
> [ 286.848086] genl_rcv_msg+0x18d/0x2c0
> [ 286.848873] ? genl_get_cmd+0x1b0/0x1b0
> [ 286.849780] ? genl_rcv+0x10/0x30
> [ 286.850730] ? mptcp_pm_remove_anno_addr.isra.0+0x180/0x180
> [ 286.852074] ? netlink_deliver_tap+0xa6/0x570
> [ 286.852801] netlink_rcv_skb+0xc4/0x1f0
> [ 286.853434] ? find_held_lock+0x85/0xa0
> [ 286.854074] ? genl_get_cmd+0x1b0/0x1b0
> [ 286.854713] ? netlink_ack+0x4f0/0x4f0
> [ 286.855336] genl_rcv+0x1f/0x30
> [ 286.855863] netlink_unicast+0x2d3/0x410
> [ 286.856516] ? netlink_attachskb+0x400/0x400
> [ 286.857219] ? _copy_from_iter_full+0xd8/0x360
> [ 286.857986] netlink_sendmsg+0x394/0x670
> [ 286.858644] ? netlink_unicast+0x410/0x410
> [ 286.859343] ? netlink_unicast+0x410/0x410
> [ 286.860256] sock_sendmsg+0x91/0xa0
> [ 286.861032] __sys_sendto+0x16e/0x210
> [ 286.861881] ? __ia32_sys_getpeername+0x40/0x40
> [ 286.862902] ? handle_mm_fault+0x103/0x270
> [ 286.863807] ? do_user_addr_fault+0x3f0/0x950
> [ 286.864781] ? trace_page_fault_kernel+0x110/0x110
> [ 286.865862] ? mark_held_locks+0x24/0x90
> [ 286.866777] __x64_sys_sendto+0x6f/0x80
> [ 286.867621] do_syscall_64+0x33/0x40
> [ 286.868504] entry_SYSCALL_64_after_hwframe+0x44/0xae
> [ 286.869788] RIP: 0033:0x7f1279a1014a
> [ 286.870857] Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f
> 1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05
> <48> 3d 00 f0 ff ff 77 76 c3 0f 1f 44 00 00 55 48 83 ec 30 44 89 4c
> [ 286.875110] RSP: 002b:00007ffe1a791af8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
> [ 286.876758] RAX: ffffffffffffffda RBX: 00007ffe1a791b40 RCX: 00007f1279a1014a
> [ 286.878317] RDX: 0000000000000030 RSI: 00007ffe1a791b40 RDI: 0000000000000003
> [ 286.879885] RBP: 0000000000000030 R08: 00007ffe1a791b04 R09: 000000000000000c
> [ 286.881537] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
> [ 286.883189] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000
>
>
> --
> Mat Martineau
> Intel
^ permalink raw reply [flat|nested] 3+ messages in thread
* [MPTCP] Re: [MPTCP][PATCH v2 mptcp-next 2/3] mptcp: send ack for rm_addr
@ 2021-03-18 0:08 Mat Martineau
0 siblings, 0 replies; 3+ messages in thread
From: Mat Martineau @ 2021-03-18 0:08 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 5862 bytes --]
On Wed, 17 Mar 2021, Geliang Tang wrote:
> Hi Mat,
>
> Mat Martineau <mathew.j.martineau(a)linux.intel.com> 于2021年3月17日周三 上午8:31写道:
>>
>> On Tue, 16 Mar 2021, Geliang Tang wrote:
>>
>>> This patch changes the sending ACK conditions for the ADD_ADDR, send an
>>> ACK packet for RM_ADDR too.
>>>
>>> In mptcp_pm_remove_addr, invoke mptcp_pm_nl_add_addr_send_ack to send
>>> the ACK packet.
>>>
>>> Signed-off-by: Geliang Tang <geliangtang(a)gmail.com>
>>> ---
>>> net/mptcp/pm.c | 1 +
>>> net/mptcp/pm_netlink.c | 10 +++++-----
>>> net/mptcp/protocol.h | 1 +
>>> 3 files changed, 7 insertions(+), 5 deletions(-)
>>>
>>
>> Thanks for the v2 update!
>>
>> The code looked ok on visual inspection, but I ran in to problems with
>> mptcp_join.sh tests (looks like the "remove id 0 address" test that
>> triggers it):
>
> Did you apply the patch (Squash to "mptcp: remove id 0 address") before
> applying this patch set? We need to apply the squash-to patch first.
>
> I didn't get this "bad unlock balance" warning on my test, could you please
> share your kconfig file to me?
>
You're right, I was testing the patch sets in separate branches. I rebased
on the current export branch with the locking fix squashed and I'm not
seeing the crash.
Mat
>>
>> [ 286.809888] =====================================
>> [ 286.810818] WARNING: bad unlock balance detected!
>> [ 286.811921] 5.12.0-rc2+ #5 Tainted: G W
>> [ 286.813119] -------------------------------------
>> [ 286.814218] pm_nl_ctl/3445 is trying to release lock (&msk->pm.lock)
>> at:
>> [ 286.815822] [<ffffffffa5ede78b>] mptcp_pm_nl_addr_send_ack+0xab/0x1a0
>> [ 286.816979] but there are no more locks to release!
>> [ 286.817795]
>> [ 286.817795] other info that might help us debug this:
>> [ 286.818874] 3 locks held by pm_nl_ctl/3445:
>> [ 286.819598] #0: ffffffffa7514770 (cb_lock){++++}-{3:3}, at:
>> genl_rcv+0x10/0x30
>> [ 286.820982] #1: ffffffffa7514828 (genl_mutex){+.+.}-{3:3}, at:
>> genl_rcv_msg+0x1f5/0x2c0
>> [ 286.822520] #2: ffff888104840120 (sk_lock-AF_INET){+.+.}-{0:0}, at:
>> mptcp_nl_remove_id_zero_address+0x10f/0x280
>> [ 286.824648]
>> [ 286.824648] stack backtrace:
>> [ 286.825355] CPU: 1 PID: 3445 Comm: pm_nl_ctl Tainted: G W
>> 5.12.0-rc2+ #5
>> [ 286.826666] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
>> 1.14.0-1.fc33 04/01/2014
>> [ 286.828234] Call Trace:
>> [ 286.828688] dump_stack+0x93/0xc2
>> [ 286.829296] lock_release.cold+0x10/0x4e
>> [ 286.830000] ? mptcp_pm_nl_addr_send_ack+0xab/0x1a0
>> [ 286.830833] ? lock_downgrade+0x390/0x390
>> [ 286.831553] ? __mptcp_flush_join_list+0x20/0x180
>> [ 286.832480] _raw_spin_unlock_bh+0x12/0x30
>> [ 286.833409] mptcp_pm_nl_addr_send_ack+0xab/0x1a0
>> [ 286.834453] mptcp_pm_remove_addr+0x7a/0xb0
>> [ 286.835203] mptcp_nl_remove_id_zero_address+0x11f/0x280
>> [ 286.836204] ? lookup_subflow_by_saddr+0x1a0/0x1a0
>> [ 286.837116] ? mptcp_nl_fill_addr+0x320/0x320
>> [ 286.837848] ? lock_release+0x1d0/0x400
>> [ 286.838565] ? lock_release+0x1d0/0x400
>> [ 286.839302] mptcp_nl_cmd_del_addr+0x497/0x5e0
>> [ 286.840130] ? mptcp_pm_remove_anno_addr.isra.0+0x180/0x180
>> [ 286.841165] ? nla_get_range_signed+0x1c0/0x1c0
>> [ 286.842053] ? lock_is_held_type+0x98/0x110
>> [ 286.842820] ? __nla_parse+0x22/0x30
>> [ 286.843503] ? genl_family_rcv_msg_attrs_parse.constprop.0+0xd1/0x130
>> [ 286.844739] genl_family_rcv_msg_doit+0x17a/0x200
>> [ 286.845641] ? genl_family_rcv_msg_attrs_parse.constprop.0+0x130/0x130
>> [ 286.847167] ? security_capable+0x48/0x60
>> [ 286.848086] genl_rcv_msg+0x18d/0x2c0
>> [ 286.848873] ? genl_get_cmd+0x1b0/0x1b0
>> [ 286.849780] ? genl_rcv+0x10/0x30
>> [ 286.850730] ? mptcp_pm_remove_anno_addr.isra.0+0x180/0x180
>> [ 286.852074] ? netlink_deliver_tap+0xa6/0x570
>> [ 286.852801] netlink_rcv_skb+0xc4/0x1f0
>> [ 286.853434] ? find_held_lock+0x85/0xa0
>> [ 286.854074] ? genl_get_cmd+0x1b0/0x1b0
>> [ 286.854713] ? netlink_ack+0x4f0/0x4f0
>> [ 286.855336] genl_rcv+0x1f/0x30
>> [ 286.855863] netlink_unicast+0x2d3/0x410
>> [ 286.856516] ? netlink_attachskb+0x400/0x400
>> [ 286.857219] ? _copy_from_iter_full+0xd8/0x360
>> [ 286.857986] netlink_sendmsg+0x394/0x670
>> [ 286.858644] ? netlink_unicast+0x410/0x410
>> [ 286.859343] ? netlink_unicast+0x410/0x410
>> [ 286.860256] sock_sendmsg+0x91/0xa0
>> [ 286.861032] __sys_sendto+0x16e/0x210
>> [ 286.861881] ? __ia32_sys_getpeername+0x40/0x40
>> [ 286.862902] ? handle_mm_fault+0x103/0x270
>> [ 286.863807] ? do_user_addr_fault+0x3f0/0x950
>> [ 286.864781] ? trace_page_fault_kernel+0x110/0x110
>> [ 286.865862] ? mark_held_locks+0x24/0x90
>> [ 286.866777] __x64_sys_sendto+0x6f/0x80
>> [ 286.867621] do_syscall_64+0x33/0x40
>> [ 286.868504] entry_SYSCALL_64_after_hwframe+0x44/0xae
>> [ 286.869788] RIP: 0033:0x7f1279a1014a
>> [ 286.870857] Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f
>> 1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05
>> <48> 3d 00 f0 ff ff 77 76 c3 0f 1f 44 00 00 55 48 83 ec 30 44 89 4c
>> [ 286.875110] RSP: 002b:00007ffe1a791af8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
>> [ 286.876758] RAX: ffffffffffffffda RBX: 00007ffe1a791b40 RCX: 00007f1279a1014a
>> [ 286.878317] RDX: 0000000000000030 RSI: 00007ffe1a791b40 RDI: 0000000000000003
>> [ 286.879885] RBP: 0000000000000030 R08: 00007ffe1a791b04 R09: 000000000000000c
>> [ 286.881537] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
>> [ 286.883189] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000
>>
>>
>> --
>> Mat Martineau
>> Intel
>
--
Mat Martineau
Intel
^ permalink raw reply [flat|nested] 3+ messages in thread
* [MPTCP] Re: [MPTCP][PATCH v2 mptcp-next 2/3] mptcp: send ack for rm_addr
@ 2021-03-17 0:31 Mat Martineau
0 siblings, 0 replies; 3+ messages in thread
From: Mat Martineau @ 2021-03-17 0:31 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 4916 bytes --]
On Tue, 16 Mar 2021, Geliang Tang wrote:
> This patch changes the sending ACK conditions for the ADD_ADDR, send an
> ACK packet for RM_ADDR too.
>
> In mptcp_pm_remove_addr, invoke mptcp_pm_nl_add_addr_send_ack to send
> the ACK packet.
>
> Signed-off-by: Geliang Tang <geliangtang(a)gmail.com>
> ---
> net/mptcp/pm.c | 1 +
> net/mptcp/pm_netlink.c | 10 +++++-----
> net/mptcp/protocol.h | 1 +
> 3 files changed, 7 insertions(+), 5 deletions(-)
>
Thanks for the v2 update!
The code looked ok on visual inspection, but I ran in to problems with
mptcp_join.sh tests (looks like the "remove id 0 address" test that
triggers it):
[ 286.809888] =====================================
[ 286.810818] WARNING: bad unlock balance detected!
[ 286.811921] 5.12.0-rc2+ #5 Tainted: G W
[ 286.813119] -------------------------------------
[ 286.814218] pm_nl_ctl/3445 is trying to release lock (&msk->pm.lock)
at:
[ 286.815822] [<ffffffffa5ede78b>] mptcp_pm_nl_addr_send_ack+0xab/0x1a0
[ 286.816979] but there are no more locks to release!
[ 286.817795]
[ 286.817795] other info that might help us debug this:
[ 286.818874] 3 locks held by pm_nl_ctl/3445:
[ 286.819598] #0: ffffffffa7514770 (cb_lock){++++}-{3:3}, at:
genl_rcv+0x10/0x30
[ 286.820982] #1: ffffffffa7514828 (genl_mutex){+.+.}-{3:3}, at:
genl_rcv_msg+0x1f5/0x2c0
[ 286.822520] #2: ffff888104840120 (sk_lock-AF_INET){+.+.}-{0:0}, at:
mptcp_nl_remove_id_zero_address+0x10f/0x280
[ 286.824648]
[ 286.824648] stack backtrace:
[ 286.825355] CPU: 1 PID: 3445 Comm: pm_nl_ctl Tainted: G W
5.12.0-rc2+ #5
[ 286.826666] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.14.0-1.fc33 04/01/2014
[ 286.828234] Call Trace:
[ 286.828688] dump_stack+0x93/0xc2
[ 286.829296] lock_release.cold+0x10/0x4e
[ 286.830000] ? mptcp_pm_nl_addr_send_ack+0xab/0x1a0
[ 286.830833] ? lock_downgrade+0x390/0x390
[ 286.831553] ? __mptcp_flush_join_list+0x20/0x180
[ 286.832480] _raw_spin_unlock_bh+0x12/0x30
[ 286.833409] mptcp_pm_nl_addr_send_ack+0xab/0x1a0
[ 286.834453] mptcp_pm_remove_addr+0x7a/0xb0
[ 286.835203] mptcp_nl_remove_id_zero_address+0x11f/0x280
[ 286.836204] ? lookup_subflow_by_saddr+0x1a0/0x1a0
[ 286.837116] ? mptcp_nl_fill_addr+0x320/0x320
[ 286.837848] ? lock_release+0x1d0/0x400
[ 286.838565] ? lock_release+0x1d0/0x400
[ 286.839302] mptcp_nl_cmd_del_addr+0x497/0x5e0
[ 286.840130] ? mptcp_pm_remove_anno_addr.isra.0+0x180/0x180
[ 286.841165] ? nla_get_range_signed+0x1c0/0x1c0
[ 286.842053] ? lock_is_held_type+0x98/0x110
[ 286.842820] ? __nla_parse+0x22/0x30
[ 286.843503] ? genl_family_rcv_msg_attrs_parse.constprop.0+0xd1/0x130
[ 286.844739] genl_family_rcv_msg_doit+0x17a/0x200
[ 286.845641] ? genl_family_rcv_msg_attrs_parse.constprop.0+0x130/0x130
[ 286.847167] ? security_capable+0x48/0x60
[ 286.848086] genl_rcv_msg+0x18d/0x2c0
[ 286.848873] ? genl_get_cmd+0x1b0/0x1b0
[ 286.849780] ? genl_rcv+0x10/0x30
[ 286.850730] ? mptcp_pm_remove_anno_addr.isra.0+0x180/0x180
[ 286.852074] ? netlink_deliver_tap+0xa6/0x570
[ 286.852801] netlink_rcv_skb+0xc4/0x1f0
[ 286.853434] ? find_held_lock+0x85/0xa0
[ 286.854074] ? genl_get_cmd+0x1b0/0x1b0
[ 286.854713] ? netlink_ack+0x4f0/0x4f0
[ 286.855336] genl_rcv+0x1f/0x30
[ 286.855863] netlink_unicast+0x2d3/0x410
[ 286.856516] ? netlink_attachskb+0x400/0x400
[ 286.857219] ? _copy_from_iter_full+0xd8/0x360
[ 286.857986] netlink_sendmsg+0x394/0x670
[ 286.858644] ? netlink_unicast+0x410/0x410
[ 286.859343] ? netlink_unicast+0x410/0x410
[ 286.860256] sock_sendmsg+0x91/0xa0
[ 286.861032] __sys_sendto+0x16e/0x210
[ 286.861881] ? __ia32_sys_getpeername+0x40/0x40
[ 286.862902] ? handle_mm_fault+0x103/0x270
[ 286.863807] ? do_user_addr_fault+0x3f0/0x950
[ 286.864781] ? trace_page_fault_kernel+0x110/0x110
[ 286.865862] ? mark_held_locks+0x24/0x90
[ 286.866777] __x64_sys_sendto+0x6f/0x80
[ 286.867621] do_syscall_64+0x33/0x40
[ 286.868504] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 286.869788] RIP: 0033:0x7f1279a1014a
[ 286.870857] Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f
1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05
<48> 3d 00 f0 ff ff 77 76 c3 0f 1f 44 00 00 55 48 83 ec 30 44 89 4c
[ 286.875110] RSP: 002b:00007ffe1a791af8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
[ 286.876758] RAX: ffffffffffffffda RBX: 00007ffe1a791b40 RCX: 00007f1279a1014a
[ 286.878317] RDX: 0000000000000030 RSI: 00007ffe1a791b40 RDI: 0000000000000003
[ 286.879885] RBP: 0000000000000030 R08: 00007ffe1a791b04 R09: 000000000000000c
[ 286.881537] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
[ 286.883189] R13: 0000000000000003 R14: 0000000000000000 R15: 0000000000000000
--
Mat Martineau
Intel
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-03-18 0:08 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-17 4:09 [MPTCP] Re: [MPTCP][PATCH v2 mptcp-next 2/3] mptcp: send ack for rm_addr Geliang Tang
-- strict thread matches above, loose matches on Subject: below --
2021-03-18 0:08 Mat Martineau
2021-03-17 0:31 Mat Martineau
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).