linux-kernel - [powerpc] Kernel crash with THP tests (next-20220920)

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-next>] [day] [month] [year] [list]

Message-Id: <C2C8DA4F-F00F-43E9-ACD8-2A8BACA55893@linux.ibm.com>
Date:   Wed, 21 Sep 2022 12:00:57 +0530
From:   Sachin Sant <sachinp@...ux.ibm.com>
To:     linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>, linux-mm@...ck.org
Cc:     mike.kravetz@...cle.com, open list <linux-kernel@...r.kernel.org>
Subject: [powerpc] Kernel crash with THP tests (next-20220920)

While running transparent huge page tests [1] against 6.0.0-rc6-next-20220920
following crash is seen on IBM Power server.

Kernel attempted to read user page (34) - exploit attempt? (uid: 0)
BUG: Kernel NULL pointer dereference on read at 0x00000034
Faulting instruction address: 0xc0000000004d2744
Oops: Kernel access of bad area, sig: 11 [#1]
LE PAGE_SIZE=64K MMU=Radix SMP NR_CPUS=2048 NUMA pSeries
Modules linked in: dm_mod(E) bonding(E) rfkill(E) tls(E) sunrpc(E) nd_pmem(E) nd_btt(E) dax_pmem(E) papr_scm(E) libnvdimm(E) pseries_rng(E) vmx_crypto(E) ext4(E) mbcache(E) jbd2(E) sd_mod(E) t10_pi(E) crc64_rocksoft(E) crc64(E) sg(E) ibmvscsi(E) scsi_transport_srp(E) ibmveth(E) fuse(E)
CPU: 37 PID: 2219255 Comm: sysctl Tainted: G            E      6.0.0-rc6-next-20220920 #1
NIP:  c0000000004d2744 LR: c0000000004d2734 CTR: 0000000000000000
REGS: c0000012801bf660 TRAP: 0300   Tainted: G            E       (6.0.0-rc6-next-20220920)
MSR:  8000000000009033 <SF,EE,ME,IR,DR,RI,LE>  CR: 24048222  XER: 20040000
CFAR: c0000000004b0eac DAR: 0000000000000034 DSISR: 40000000 IRQMASK: 0 
GPR00: c0000000004d2734 c0000012801bf900 c000000002a92300 0000000000000000 
GPR04: c000000002ac8ac0 c000000001209340 0000000000000005 c000001286714b80 
GPR08: 0000000000000034 0000000000000000 0000000000000000 0000000000000000 
GPR12: 0000000028048242 c00000167fff6b00 0000000000000000 0000000000000000 
GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 
GPR20: c0000012801bfae8 0000000000000001 0000000000000100 0000000000000001 
GPR24: c0000012801bfae8 c000000002ac8ac0 0000000000000002 0000000000000005 
GPR28: 0000000000000000 0000000000000001 0000000000000000 0000000000346cca 
NIP [c0000000004d2744] alloc_buddy_huge_page+0xd4/0x240
LR [c0000000004d2734] alloc_buddy_huge_page+0xc4/0x240
Call Trace:
[c0000012801bf900] [c0000000004d2734] alloc_buddy_huge_page+0xc4/0x240 (unreliable)
[c0000012801bf9b0] [c0000000004d46a4] alloc_fresh_huge_page.part.72+0x214/0x2a0
[c0000012801bfa40] [c0000000004d7f88] alloc_pool_huge_page+0x118/0x190
[c0000012801bfa90] [c0000000004d84dc] __nr_hugepages_store_common+0x4dc/0x610
[c0000012801bfb70] [c0000000004d88bc] hugetlb_sysctl_handler_common+0x13c/0x180
[c0000012801bfc10] [c0000000006380e0] proc_sys_call_handler+0x210/0x350
[c0000012801bfc90] [c000000000551c00] vfs_write+0x2e0/0x460
[c0000012801bfd50] [c000000000551f5c] ksys_write+0x7c/0x140
[c0000012801bfda0] [c000000000033f58] system_call_exception+0x188/0x3f0
[c0000012801bfe10] [c00000000000c53c] system_call_common+0xec/0x270
--- interrupt: c00 at 0x7fffa9520c34
NIP:  00007fffa9520c34 LR: 00000001024754bc CTR: 0000000000000000
REGS: c0000012801bfe80 TRAP: 0c00   Tainted: G            E       (6.0.0-rc6-next-20220920)
MSR:  800000000280f033 <SF,VEC,VSX,EE,PR,FP,ME,IR,DR,RI,LE>  CR: 28002202  XER: 00000000
IRQMASK: 0 
GPR00: 0000000000000004 00007fffccd76cd0 00007fffa9607300 0000000000000003 
GPR04: 0000000138da6970 0000000000000006 fffffffffffffff6 0000000000000000 
GPR08: 0000000138da6970 0000000000000000 0000000000000000 0000000000000000 
GPR12: 0000000000000000 00007fffa9a40940 0000000000000000 0000000000000000 
GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 
GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 
GPR24: 0000000000000001 0000000000000010 0000000000000006 0000000138da8aa0 
GPR28: 00007fffa95fc2c8 0000000138da8aa0 0000000000000006 0000000138da6930 
NIP [00007fffa9520c34] 0x7fffa9520c34
LR [00000001024754bc] 0x1024754bc
--- interrupt: c00
Instruction dump:
3b400002 3ba00001 3b800000 7f26cb78 7fc5f378 7f64db78 7fe3fb78 4bfde5b9 
60000000 7c691b78 39030034 7c0004ac <7d404028> 7c0ae800 40c20010 7f80412d 
---[ end trace 0000000000000000 ]---

Kernel panic - not syncing: Fatal exception

Bisect points to following patch:
commit f2f3c25dea3acfb17aecb7273541e7266dfc8842
    hugetlb: freeze allocated pages before creating hugetlb pages

Reverting the patch allows the test to run successfully.

Thanks
- Sachin

[1] https://github.com/avocado-framework-tests/avocado-misc-tests/blob/master/memory/transparent_hugepages_defrag.py