Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
From:   Zhouyi Zhou <zhouzhouyi@gmail.com>
Date:   Sat, 22 Apr 2023 20:46:44 +0800
Message-ID: <CAABZP2xJRGhPmfB-PrfesQKzP7fsuZsj+3TewAiLLW8u=YK4dg@mail.gmail.com>
Subject: BUG : PowerPC RCU: torture test failed with __stack_chk_fail
To:     linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
        rcu <rcu@vger.kernel.org>,
        linux-kernel <linux-kernel@vger.kernel.org>, lance@osuosl.org,
        "Paul E. McKenney" <paulmck@kernel.org>,
        Michael Ellerman <mpe@ellerman.id.au>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Dear PowerPC and RCU developers:
During the RCU torture test on mainline (on the VM of Opensource Lab
of Oregon State University), SRCU-P failed with __stack_chk_fail:
[  264.381952][   T99] [c000000006c7bab0] [c0000000010c67c0]
dump_stack_lvl+0x94/0xd8 (unreliable)
[  264.383786][   T99] [c000000006c7bae0] [c00000000014fc94] panic+0x19c/0x468
[  264.385128][   T99] [c000000006c7bb80] [c0000000010fca24]
__stack_chk_fail+0x24/0x30
[  264.386610][   T99] [c000000006c7bbe0] [c0000000002293b4]
srcu_gp_start_if_needed+0x5c4/0x5d0
[  264.388188][   T99] [c000000006c7bc70] [c00000000022f7f4]
srcu_torture_call+0x34/0x50
[  264.389611][   T99] [c000000006c7bc90] [c00000000022b5e8]
rcu_torture_fwd_prog+0x8c8/0xa60
[  264.391439][   T99] [c000000006c7be00] [c00000000018e37c] kthread+0x15c/0x170
[  264.392792][   T99] [c000000006c7be50] [c00000000000df94]
ret_from_kernel_thread+0x5c/0x64
The kernel config file can be found in [1].
And I write a bash script to accelerate the bug reproducing [2].
After a week's debugging, I found the cause of the bug is because the
register r10 used to judge for stack overflow is not constant between
context switches.
The assembly code for srcu_gp_start_if_needed is located at [3]:
c000000000226eb4:   78 6b aa 7d     mr      r10,r13
c000000000226eb8:   14 42 29 7d     add     r9,r9,r8
c000000000226ebc:   ac 04 00 7c     hwsync
c000000000226ec0:   10 00 7b 3b     addi    r27,r27,16
c000000000226ec4:   14 da 29 7d     add     r9,r9,r27
c000000000226ec8:   a8 48 00 7d     ldarx   r8,0,r9
c000000000226ecc:   01 00 08 31     addic   r8,r8,1
c000000000226ed0:   ad 49 00 7d     stdcx.  r8,0,r9
c000000000226ed4:   f4 ff c2 40     bne-    c000000000226ec8
<srcu_gp_start_if_needed+0x1c8>
c000000000226ed8:   28 00 21 e9     ld      r9,40(r1)
c000000000226edc:   78 0c 4a e9     ld      r10,3192(r10)
c000000000226ee0:   79 52 29 7d     xor.    r9,r9,r10
c000000000226ee4:   00 00 40 39     li      r10,0
c000000000226ee8:   b8 03 82 40     bne     c0000000002272a0
<srcu_gp_start_if_needed+0x5a0>
by debugging, I see the r10 is assigned with r13 on c000000000226eb4,
but if there is a context-switch before c000000000226edc, a false
positive will be reported.

[1] http://154.220.3.115/logs/0422/configformainline.txt
[2] 154.220.3.115/logs/0422/whilebash.sh
[3] http://154.220.3.115/logs/0422/srcu_gp_start_if_needed.txt

My analysis and debugging may not be correct, but the bug is easily
reproducible.

Thanks
Zhouyi