BUG_ON hit in "reimplement path_mountpoint() with less magic" [Was: Re: linux-next: Tree for Jan 13]

From: Jason A. Donenfeld
Date: Mon Jan 13 2020 - 13:27:40 EST

Hi Al,

Your "reimplement path_mountpoint() with less magic" commit is causing oopses on the linux-next tests running on https://build.wireguard.com/ . Here's one such crash with debug symbols:

[ 81.310222] ------------[ cut here ]------------
[ 81.310684] kernel BUG at ../../../../../../../../fs/namei.c:684!
[ 81.311261] invalid opcode: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC KASAN
[ 81.311869] CPU: 2 PID: 554 Comm: ip Not tainted 5.5.0-rc5-next-20200113-debug #1
[ 81.312575] RIP: 0010:unlazy_walk+0x20a/0x2a0
[ 81.312987] Code: 43 01 00 83 e3 01 74 3e e8 83 62 dd ff 48 c7 c6 48 b2 47 81 48 c7 c7 40 d5 56 82 e8 f0 c3 d6 ff b8 f6 ff ff ff 5b 5d 41 5c c3 <0f> 0b 48 c7 c7 c0 1e 65 82 e8 53 2c 14 00 0f 0b 48 c7 c7 80 1e 65
[ 81.314668] RSP: 0018:ffff888032b8fc38 EFLAGS: 00010246
[ 81.315144] RAX: 0000000000001009 RBX: ffff888032b8fcd0 RCX: 0000000032b8fcd8
[ 81.315792] RDX: 1ffff11006571fa1 RSI: 1ffff11006571fa0 RDI: ffff888032b8fd08
[ 81.316441] RBP: ffff888034c88008 R08: 0000000000000001 R09: ffff88803222a738
[ 81.317089] R10: fffffbfff04fd048 R11: ffffffff827e8247 R12: ffff888032b8fcd8
[ 81.317738] R13: ffff888032b8fe88 R14: ffff888032b8fd18 R15: ffff888032b8fcd0
[ 81.318383] FS: 00007fcd43cb0d48(0000) GS:ffff888036500000(0000) knlGS:0000000000000000
[ 81.319113] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 81.319638] CR2: 00007fcd43c5b8f0 CR3: 00000000164f1006 CR4: 00000000001606a0
[ 81.320283] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 81.320932] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 81.321580] Call Trace:
[ 81.321813] path_mountpoint+0x148/0x2b0
[ 81.322174] filename_mountpoint+0x149/0x200
[ 81.322571] ? path_mountpoint+0x2b0/0x2b0
[ 81.322954] ? trace_event_raw_event_rcu_quiescent_state_report+0x340/0x340
[ 81.323592] ? strncpy_from_user+0x6f/0x300
[ 81.323977] ? kmem_cache_alloc+0x2d3/0x300
[ 81.324362] ? getname_flags+0x89/0x3c0
[ 81.324724] ksys_umount+0x115/0xba0
[ 81.325057] ? __up_read+0x195/0x720
[ 81.325388] ? __detach_mounts+0x220/0x220
[ 81.325769] ? __down_read+0x3e0/0x3e0
[ 81.326120] ? trace_hardirqs_on_thunk+0x1a/0x1c
[ 81.326547] __x64_sys_umount+0x4b/0x70
[ 81.326900] do_syscall_64+0x90/0xd9a
[ 81.327245] entry_SYSCALL_64_after_hwframe+0x49/0xbe
[ 81.327708] RIP: 0033:0x7fcd43c16d1e
[ 81.328039] Code: b8 a6 00 00 00 31 f6 0f 05 48 89 c7 e8 7b e5 ff ff 48 83 c4 08 c3 66 0f 1f 44 00 00 48 83 ec 08 48 63 f6 b8 a6 00 00 00 0f 05 <48> 89 c7 e8 5a e5 ff ff 48 83 c4 08 c3 0f 1f 44 00 00 48 83 ec 08
[ 81.329716] RSP: 002b:00007fff06d4d850 EFLAGS: 00000206 ORIG_RAX: 00000000000000a6
[ 81.330398] RAX: ffffffffffffffda RBX: 00007fff06d4ea98 RCX: 00007fcd43c16d1e
[ 81.331044] RDX: 00000000ffffffff RSI: 0000000000000002 RDI: 00007fff06d4d860
[ 81.331693] RBP: 0000000000000002 R08: 0000000000000000 R09: 00007fff06d4d680
[ 81.332335] R10: 000055559d446fcf R11: 0000000000000206 R12: 00007fff06d4d860
[ 81.332982] R13: 00007fff06d4ea90 R14: 00007fff06d4ef99 R15: 0000000000000000
[ 81.335146] ---[ end trace 35c7405bba7e573a ]---
[ 81.335583] RIP: 0010:unlazy_walk+0x20a/0x2a0
[ 81.335984] Code: 43 01 00 83 e3 01 74 3e e8 83 62 dd ff 48 c7 c6 48 b2 47 81 48 c7 c7 40 d5 56 82 e8 f0 c3 d6 ff b8 f6 ff ff ff 5b 5d 41 5c c3 <0f> 0b 48 c7 c7 c0 1e 65 82 e8 53 2c 14 00 0f 0b 48 c7 c7 80 1e 65
[ 81.337702] RSP: 0018:ffff888032b8fc38 EFLAGS: 00010246
[ 81.338190] RAX: 0000000000001009 RBX: ffff888032b8fcd0 RCX: 0000000032b8fcd8
[ 81.338842] RDX: 1ffff11006571fa1 RSI: 1ffff11006571fa0 RDI: ffff888032b8fd08
[ 81.339492] RBP: ffff888034c88008 R08: 0000000000000001 R09: ffff88803222a738
[ 81.340137] R10: fffffbfff04fd048 R11: ffffffff827e8247 R12: ffff888032b8fcd8
[ 81.340784] R13: ffff888032b8fe88 R14: ffff888032b8fd18 R15: ffff888032b8fcd0
[ 81.341435] FS: 00007fcd43cb0d48(0000) GS:ffff888036500000(0000) knlGS:0000000000000000
[ 81.343468] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 81.344000] CR2: 00007fcd43c5b8f0 CR3: 00000000164f1006 CR4: 00000000001606a0
[ 81.344652] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 81.345298] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 81.346026] Kernel panic - not syncing: Fatal exception
[ 81.346671] Kernel Offset: disabled

Regards,
Jason

On 1/13/20 2:14 AM, Stephen Rothwell wrote:
Hi all,

Changes since 20200110:

The vfs tree gained a conflict against the m68k tree.

The keys tree still had build failures so I used the version from
next-20191211. It also gained a conflict against the m68k tree.

The usb-gadget tree still had its conflict against Linus' tree, so I dropped
it for today.

The pidfd tree gained conflicts against the m68k, vfs and keys trees.

The generic-ioremap tree gained a conflict against the tegra tree.

The kunit-next tree lost its failures.

Non-merge commits (relative to Linus' tree): 7258
7462 files changed, 345876 insertions(+), 165190 deletions(-)

----------------------------------------------------------------------------

I have created today's linux-next tree at
git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
(patches at http://www.kernel.org/pub/linux/kernel/next/ ). If you
are tracking the linux-next tree using git, you should not use "git pull"
to do so as that will try to merge the new linux-next release with the
old one. You should use "git fetch" and checkout or reset to the new
master.

You can see which trees have been included by looking in the Next/Trees
file in the source. There are also quilt-import.log and merge.log
files in the Next directory. Between each merge, the tree was built
with a ppc64_defconfig for powerpc, an allmodconfig for x86_64, a
multi_v7_defconfig for arm and a native build of tools/perf. After
the final fixups (if any), I do an x86_64 modules_install followed by
builds for x86_64 allnoconfig, powerpc allnoconfig (32 and 64 bit),
ppc44x_defconfig, allyesconfig and pseries_le_defconfig and i386, sparc
and sparc64 defconfig. And finally, a simple boot test of the powerpc
pseries_le_defconfig kernel in qemu (with and without kvm enabled).

Below is a summary of the state of the merge.

I am currently merging 314 trees (counting Linus' and 78 trees of bug
fix patches pending for the current merge release).

Stats about the size of the tree over time can be seen at
http://neuling.org/linux-next-size.html .

Status of my local build tests will be at
http://kisskb.ellerman.id.au/linux-next . If maintainers want to give
advice about cross compilers/configs that work, we are always open to add
more builds.

Thanks to Randy Dunlap for doing many randconfig builds. And to Paul
Gortmaker for triage and bug fixes.