Message-ID: <3d7a5f70-7ece-48ba-92bd-8b6473fd8b6c@linux.ibm.com>
Date: Tue, 7 Oct 2025 20:38:38 +0530
From: Venkat Rao Bagalkote <venkat88@...ux.ibm.com>
To: LKML <linux-kernel@...r.kernel.org>,
        linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>,
        Madhavan Srinivasan <maddy@...ux.ibm.com>,
        linux-kselftest@...r.kernel.org, daleksan@...hat.com,
        jstancek@...hat.com, jarkko@...nel.org, pmenzel@...gen.mpg.de
Subject: Re: [bisected][linux-next20251003] tmp2 selftests resulting in Kernel
 OOPs


On 07/10/25 10:59 am, Venkat Rao Bagalkote wrote:
> Greetings!!!
>
>
> IBM CI has reported a kernel OOPs while running TPM2selftests on IBM 
> Power11 system with linux-next20251002 kernel.
>
>
> Test Case:
>
> make run_tests
> TAP version 13
> 1..3
> # timeout set to 600
> # selftests: tpm2: test_smoke.sh
> # test_read_partial_overwrite (tpm2_tests.SmokeTest) ... ok
> # test_read_partial_resp (tpm2_tests.SmokeTest) ... ok
> # test_seal_with_auth (tpm2_tests.SmokeTest) ... ok
> # test_seal_with_policy (tpm2_tests.SmokeTest) ... ok
> # test_seal_with_too_long_auth (tpm2_tests.SmokeTest) ... ok
> # test_send_two_cmds (tpm2_tests.SmokeTest) ... ok
> # test_too_short_cmd (tpm2_tests.SmokeTest) ... ok
> # test_unseal_with_wrong_auth (tpm2_tests.SmokeTest) ... ok
> # test_unseal_with_wrong_policy (tpm2_tests.SmokeTest) ... ERROR
> #
> # ======================================================================
> # ERROR: test_unseal_with_wrong_policy (tpm2_tests.SmokeTest)
> # -----------------------------------------------------
>
>
> Traces:
>
>
> [  452.604333] BUG: KASAN: slab-use-after-free in tpmrm_release+0x78/0xa8
> [  452.604345] Read of size 8 at addr c00000001c650000 by task 
> python3/1856
> [  452.604353]
> [  452.604358] CPU: 24 UID: 0 PID: 1856 Comm: python3 Kdump: loaded 
> Not tainted 6.17.0-next-20251003 #1 VOLUNTARY
> [  452.604364] Hardware name: IBM,9080-HEX Power11 (architected) 
> 0x820200 0xf000007 of:IBM,FW1110.01 (NH1110_069) hv:phyp pSeries
> [  452.604368] Call Trace:
> [  452.604370] [c0000000c1867840] [c00000000187ea4c] 
> dump_stack_lvl+0x84/0xe8 (unreliable)
> [  452.604380] [c0000000c1867870] [c000000000803754] 
> print_address_description.constprop.0+0x11c/0x56c
> [  452.604388] [c0000000c1867910] [c000000000803c84] 
> print_report+0xe0/0x358
> [  452.604394] [c0000000c18679e0] [c000000000804124] 
> kasan_report+0x128/0x1f4
> [  452.604400] [c0000000c1867af0] [c0000000008062b4] 
> __asan_load8+0xa8/0xe0
> [  452.604406] [c0000000c1867b10] [c000000000f2ec18] 
> tpmrm_release+0x78/0xa8
> [  452.604412] [c0000000c1867b40] [c0000000008b6a2c] __fput+0x21c/0x60c
> [  452.604417] [c0000000c1867bc0] [c0000000008ada70] sys_close+0x74/0xd0
> [  452.604424] [c0000000c1867bf0] [c000000000039270] 
> system_call_exception+0x1e0/0x460
> [  452.604431] [c0000000c1867e50] [c00000000000d05c] 
> system_call_vectored_common+0x15c/0x2ec
> [  452.604438] ---- interrupt: 3000 at 0x7fffb7534ab4
> [  452.604443] NIP:  00007fffb7534ab4 LR: 00007fffb7534ab4 CTR: 
> 0000000000000000
> [  452.604446] REGS: c0000000c1867e80 TRAP: 3000   Not tainted 
> (6.17.0-next-20251003)
> [  452.604449] MSR:  800000000280f033 
> <SF,VEC,VSX,EE,PR,FP,ME,IR,DR,RI,LE>  CR: 44284422  XER: 00000000
> [  452.604466] IRQMASK: 0
> [  452.604466] GPR00: 0000000000000006 00007ffff65d76b0 
> 00007fffb7c17700 0000000000000006
> [  452.604466] GPR04: 0000000000000000 0000000000000000 
> 0000000000000000 0000000000000004
> [  452.604466] GPR08: 0000000000000000 0000000000000000 
> 0000000000000000 0000000000000000
> [  452.604466] GPR12: 0000000000000000 00007fffb7e6b8e0 
> 00000000000000a1 00007fffb67acec0
> [  452.604466] GPR16: 0000000164032ad0 00007fffb67aceb0 
> 00007fffb76f6a90 0000000000000000
> [  452.604466] GPR20: 00007fffb6f21850 0000000000000000 
> 00007fffb71062c0 0000000164034490
> [  452.604466] GPR24: 00007fffb6f2fea0 00007fffb67acea8 
> 0000000164032b18 00007fffb7c45b32
> [  452.604466] GPR28: 00007fffb7c678e0 00007fffb67aceb8 
> 0000000000000006 0000000164034490
> [  452.604510] NIP [00007fffb7534ab4] 0x7fffb7534ab4
> [  452.604513] LR [00007fffb7534ab4] 0x7fffb7534ab4
> [  452.604516] ---- interrupt: 3000
> [  452.604518]
> [  452.604601] Allocated by task 1856:
> [  452.604607]  kasan_save_stack+0x34/0x64
> [  452.604614]  kasan_save_track+0x2c/0x50
> [  452.604621]  kasan_save_alloc_info+0x58/0x74
> [  452.604628]  __kasan_kmalloc+0x12c/0x168
> [  452.604635]  __kmalloc_cache_noprof+0x1d8/0x71c
> [  452.604643]  tpmrm_open+0x88/0x168
> [  452.604649]  chrdev_open+0x1f4/0x484
> [  452.604656]  do_dentry_open+0x578/0x9cc
> [  452.604663]  vfs_open+0x68/0x23c
> [  452.604670]  do_open+0x514/0x74c
> [  452.604676]  path_openat+0x16c/0x380
> [  452.604682]  do_filp_open+0x104/0x230
> [  452.604689]  do_sys_openat2+0xb8/0x154
> [  452.604696]  sys_openat+0xcc/0x130
> [  452.604703]  system_call_exception+0x1e0/0x460
> [  452.604710]  system_call_vectored_common+0x15c/0x2ec
> [  452.604718]
> [  452.604722] Freed by task 1856:
> [  452.604726]  kasan_save_stack+0x34/0x64
> [  452.604733]  kasan_save_track+0x2c/0x50
> [  452.604739]  __kasan_save_free_info+0x64/0x110
> [  452.604747]  __kasan_slab_free+0xb0/0x10c
> [  452.604753]  kfree+0x220/0x624
> [  452.604760]  tpmrm_release+0x6c/0xa8
> [  452.604766]  __fput+0x21c/0x60c
> [  452.604772]  sys_close+0x74/0xd0
> [  452.604779]  system_call_exception+0x1e0/0x460
> [  452.604786]  system_call_vectored_common+0x15c/0x2ec
> [  452.604794]
> [  452.604797] The buggy address belongs to the object at 
> c00000001c650000
> [  452.604797]  which belongs to the cache kmalloc-8k of size 8192
> [  452.604806] The buggy address is located 0 bytes inside of
> [  452.604806]  freed 8192-byte region [c00000001c650000, 
> c00000001c652000)
> [  452.604815]
> [  452.604818] The buggy address belongs to the physical page:
> [  452.604824] page: refcount:0 mapcount:0 mapping:0000000000000000 
> index:0xc00000001c644000 pfn:0x1c60
> [  452.604833] head: order:3 mapcount:0 entire_mapcount:0 
> nr_pages_mapped:0 pincount:0
> [  452.604840] flags: 
> 0x3ffffe00000040(head|node=0|zone=0|lastcpupid=0x1fffff)
> [  452.604849] page_type: f5(slab)
> [  452.604856] raw: 003ffffe00000040 c000000007012300 5deadbeef0000122 
> 0000000000000000
> [  452.604864] raw: c00000001c644000 000000008020001e 00000000f5000000 
> 0000000000000000
> [  452.604872] head: 003ffffe00000040 c000000007012300 
> 5deadbeef0000122 0000000000000000
> [  452.604879] head: c00000001c644000 000000008020001e 
> 00000000f5000000 0000000000000000
> [  452.604887] head: 003ffffe00000003 c00c000000071801 
> 00000000ffffffff 00000000ffffffff
> [  452.604894] head: ffffffffffffffff 0000000000000000 
> 00000000ffffffff 0000000000000008
> [  452.604900] page dumped because: kasan: bad access detected
> [  452.604905]
> [  452.604908] Memory state around the buggy address:
> [  452.604914]  c00000001c64ff00: fc fc fc fc fc fc fc fc fc fc fc fc 
> fc fc fc fc
> [  452.604920]  c00000001c64ff80: fc fc fc fc fc fc fc fc fc fc fc fc 
> fc fc fc fc
> [  452.604927] >c00000001c650000: fa fb fb fb fb fb fb fb fb fb fb fb 
> fb fb fb fb
> [  452.604933]                    ^
> [  452.604937]  c00000001c650080: fb fb fb fb fb fb fb fb fb fb fb fb 
> fb fb fb fb
> [  452.604944]  c00000001c650100: fb fb fb fb fb fb fb fb fb fb fb fb 
> fb fb fb fb
> [  452.604950] 
> ==================================================================
> [  452.604955] Disabling lock debugging due to kernel taint
> [  452.604961] Kernel attempted to read user page (770) - exploit 
> attempt? (uid: 0)
> [  452.604969] BUG: Kernel NULL pointer dereference on read at 0x00000770
> [  452.604975] Faulting instruction address: 0xc0000000002b2e0c
> [  452.604982] Oops: Kernel access of bad area, sig: 11 [#1]
> [  452.604987] LE PAGE_SIZE=64K MMU=Radix  SMP NR_CPUS=8192 NUMA pSeries
> [  452.604996] Modules linked in: nft_fib_inet nft_fib_ipv4 
> nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 
> nft_reject nft_ct nft_chain_nat nf_nat bonding nf_conntrack tls 
> nf_defrag_ipv6 nf_defrag_ipv4 rfkill ip_set nf_tables nfnetlink sunrpc 
> pseries_rng vmx_crypto fuse ext4 crc16 mbcache jbd2 sd_mod sg ibmvscsi 
> ibmveth scsi_transport_srp pseries_wdt
> [  452.605073] CPU: 24 UID: 0 PID: 1856 Comm: python3 Kdump: loaded 
> Tainted: G    B               6.17.0-next-20251003 #1 VOLUNTARY
> [  452.605084] Tainted: [B]=BAD_PAGE
> [  452.605089] Hardware name: IBM,9080-HEX Power11 (architected) 
> 0x820200 0xf000007 of:IBM,FW1110.01 (NH1110_069) hv:phyp pSeries
> [  452.605096] NIP:  c0000000002b2e0c LR: c0000000002b2e08 CTR: 
> 0000000000000000
> [  452.605103] REGS: c0000000c1867820 TRAP: 0300   Tainted: G B       
>       (6.17.0-next-20251003)
> [  452.605110] MSR:  8000000000009033 <SF,EE,ME,IR,DR,RI,LE>  CR: 
> 28284420  XER: 0000000d
> [  452.605132] CFAR: c000000000807920 DAR: 0000000000000770 DSISR: 
> 40000000 IRQMASK: 0
> [  452.605132] GPR00: c0000000002b2e08 c0000000c1867ac0 
> c00000000234a500 0000000000000001
> [  452.605132] GPR04: 0000000000000008 0000000000000000 
> c0000000002b2e08 0000000000000001
> [  452.605132] GPR08: 0000000000000020 0000000000000001 
> 0000000000000001 a80e000000000000
> [  452.605132] GPR12: c00e0000009b1c8c c000000d0ddeb700 
> 0000000000000000 0000000000000000
> [  452.605132] GPR16: 0000000000000000 0000000000000000 
> 0000000000000000 0000000000000000
> [  452.605132] GPR20: 0000000000000008 0000000000000000 
> c000000008202f00 c00000007b9ff620
> [  452.605132] GPR24: c00000008a76cb20 c00000008a76cb40 
> c00000008a76cb08 c000000002201e80
> [  452.605132] GPR28: c000000061569248 0000000000000770 
> c00000008a76cb00 0000000000000768
> [  452.605227] NIP [c0000000002b2e0c] up_read+0x50/0x17c
> [  452.605237] LR [c0000000002b2e08] up_read+0x4c/0x17c
> [  452.605245] Call Trace:
> [  452.605249] [c0000000c1867ac0] [c0000000002b2e08] 
> up_read+0x4c/0x17c (unreliable)
> [  452.605261] [c0000000c1867b10] [c000000000f2ec28] 
> tpmrm_release+0x88/0xa8
> [  452.605271] [c0000000c1867b40] [c0000000008b6a2c] __fput+0x21c/0x60c
> [  452.605280] [c0000000c1867bc0] [c0000000008ada70] sys_close+0x74/0xd0
> [  452.605291] [c0000000c1867bf0] [c000000000039270] 
> system_call_exception+0x1e0/0x460
> [  452.605301] [c0000000c1867e50] [c00000000000d05c] 
> system_call_vectored_common+0x15c/0x2ec
> [  452.605312] ---- interrupt: 3000 at 0x7fffb7534ab4
> [  452.605319] NIP:  00007fffb7534ab4 LR: 00007fffb7534ab4 CTR: 
> 0000000000000000
> [  452.605326] REGS: c0000000c1867e80 TRAP: 3000   Tainted: G B       
>       (6.17.0-next-20251003)
> [  452.605333] MSR:  800000000280f033 
> <SF,VEC,VSX,EE,PR,FP,ME,IR,DR,RI,LE>  CR: 44284422  XER: 00000000
> [  452.605362] IRQMASK: 0
> [  452.605362] GPR00: 0000000000000006 00007ffff65d76b0 
> 00007fffb7c17700 0000000000000006
> [  452.605362] GPR04: 0000000000000000 0000000000000000 
> 0000000000000000 0000000000000004
> [  452.605362] GPR08: 0000000000000000 0000000000000000 
> 0000000000000000 0000000000000000
> [  452.605362] GPR12: 0000000000000000 00007fffb7e6b8e0 
> 00000000000000a1 00007fffb67acec0
> [  452.605362] GPR16: 0000000164032ad0 00007fffb67aceb0 
> 00007fffb76f6a90 0000000000000000
> [  452.605362] GPR20: 00007fffb6f21850 0000000000000000 
> 00007fffb71062c0 0000000164034490
> [  452.605362] GPR24: 00007fffb6f2fea0 00007fffb67acea8 
> 0000000164032b18 00007fffb7c45b32
> [  452.605362] GPR28: 00007fffb7c678e0 00007fffb67aceb8 
> 0000000000000006 0000000164034490
> [  452.605450] NIP [00007fffb7534ab4] 0x7fffb7534ab4
> [  452.605456] LR [00007fffb7534ab4] 0x7fffb7534ab4
> [  452.605462] ---- interrupt: 3000
> [  452.605467] Code: fbc1fff0 7c7f1b78 f8010010 f821ffb1 e92d0c78 
> f9210028 39200000 3ba30008 38800008 7fa3eb78 48554af5 60000000 
> <ebdf0008> eb8d0908 7bc90764 fbc10020
> [  452.605501] ---[ end trace 0000000000000000 ]---
> [  452.613685] pstore: backend (nvram) writing error (-1)
> [  452.613691]
>
>

Git bisect is pointing to eb28a2adba0654878bcfd909b429bf567b35922b as 
first bad commit.


eb28a2adba0654878bcfd909b429bf567b35922b is the first bad commit
commit eb28a2adba0654878bcfd909b429bf567b35922b
Author: Jonathan McDowell <noodles@...a.com>
Date:   Tue Sep 23 18:10:00 2025 +0100

     tpm: Ensure exclusive userspace access when using /dev/tpm<n>

     There is an is_open lock on /dev/tpm<n> that dates back to at least
     2013, but it only prevents multiple accesses via *this* interface. 
It is
     perfectly possible for userspace to use /dev/tpmrm<n>, or the kernel to
     use the internal interfaces, to access the TPM. For tooling expecting
     exclusive access, such as firmware updates, this can cause issues.

     Close the userspace loophole by changing the simple bit lock to a full
     read/write mutex. Direct /dev/tpm<n> access needs an exclusive write
     lock, the resource broker continues to allow concurrent access *except*
     when /dev/tpm<n> is open.

     Signed-off-by: Jonathan McDowell <noodles@...a.com>
     Reviewed-by: Jarkko Sakkinen <jarkko@...nel.org>
     Signed-off-by: Jarkko Sakkinen <jarkko@...nel.org>

  drivers/char/tpm/tpm-chip.c  |  1 +
  drivers/char/tpm/tpm-dev.c   | 14 ++++++++------
  drivers/char/tpm/tpmrm-dev.c | 20 ++++++++++++++++++--
  include/linux/tpm.h          |  3 ++-
  4 files changed, 29 insertions(+), 9 deletions(-)



Git bisect log:


git bisect log
git bisect start
# status: waiting for both good and bad commits
# good: [e5f0a698b34ed76002dc5cff3804a61c80233a7a] Linux 6.17
git bisect good e5f0a698b34ed76002dc5cff3804a61c80233a7a
# status: waiting for bad commit, 1 good commit known
# bad: [47a8d4b89844f5974f634b4189a39d5ccbacd81c] Add linux-next 
specific files for 20251003
git bisect bad 47a8d4b89844f5974f634b4189a39d5ccbacd81c
# good: [f79e772258df311c2cb21594ca0996318e720d28] Merge tag 
'media/v6.18-1' of 
git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
git bisect good f79e772258df311c2cb21594ca0996318e720d28
# good: [6bda5a67f6a2ae4c0153e693e96bd408d054c732] Merge branch 
'xtensa-for-next' of https://github.com/jcmvbkbc/linux-xtensa.git
git bisect good 6bda5a67f6a2ae4c0153e693e96bd408d054c732
# bad: [09026363ffabf7bb33e0ce000d8bff3e41b0c3de] Merge branch 'next' of 
https://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit.git
git bisect bad 09026363ffabf7bb33e0ce000d8bff3e41b0c3de
# good: [c32d529ea3e39ba4918f3865ef90ef6bddaa498c] Merge branch 
'for-next' of https://git.kernel.org/pub/scm/linux/kernel/git/hid/hid.git
git bisect good c32d529ea3e39ba4918f3865ef90ef6bddaa498c
# good: [3fa51882336f09d1a8a03edf107ef43e3b872fc1] Merge branch 'next' 
of https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/iwlwifi-next.git
git bisect good 3fa51882336f09d1a8a03edf107ef43e3b872fc1
# good: [acc26ceeebaa1880ceb70379abb3fac92d1007d5] Merge branch 
'drm-xe-next' of https://gitlab.freedesktop.org/drm/xe/kernel.git
git bisect good acc26ceeebaa1880ceb70379abb3fac92d1007d5
# good: [3e06c10cef8079782836155d66c2a91798600cfc] Merge branch 
'for-next' of 
https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git
git bisect good 3e06c10cef8079782836155d66c2a91798600cfc
# bad: [cf305bb4070bef42cea469a6c4e751a63f5cacf2] Merge branch 'next' of 
https://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd.git
git bisect bad cf305bb4070bef42cea469a6c4e751a63f5cacf2
# good: [19766d4984c39d75393d22cbdad14974cb7f9366] Merge branch 'next' 
of https://github.com/cschaufler/smack-next
git bisect good 19766d4984c39d75393d22cbdad14974cb7f9366
# good: [4e349e68974e71da13d4b34988f795f4cfe29650] tpm: use a map for 
tpm2_calc_ordinal_duration()
git bisect good 4e349e68974e71da13d4b34988f795f4cfe29650
# bad: [b6889908d493fe03a1db28aa9afdade6bceda158] tpm: Allow for 
exclusive TPM access when using /dev/tpm<n>
git bisect bad b6889908d493fe03a1db28aa9afdade6bceda158
# bad: [eb28a2adba0654878bcfd909b429bf567b35922b] tpm: Ensure exclusive 
userspace access when using /dev/tpm<n>
git bisect bad eb28a2adba0654878bcfd909b429bf567b35922b
# good: [11baa7201b1baefd281e5a7faf7de5b7e407364c] tpm: Prevent local 
DOS via tpm/tpm0/ppi/*operations
git bisect good 11baa7201b1baefd281e5a7faf7de5b7e407364c
# first bad commit: [eb28a2adba0654878bcfd909b429bf567b35922b] tpm: 
Ensure exclusive userspace access when using /dev/tpm<n>

>
> If you happen to fix this, please add below tag.
>
>
> Reported-by: Venkat Rao Bagalkote <venkat88@...ux.ibm.com>
>
>
> Regards,
>
> Venkat.
>

Powered by blists - more mailing lists