Date:   Thu, 3 Dec 2020 16:46:04 +0100
From:   Marek Szyprowski <m.szyprowski@...sung.com>
To:     Matthew Wilcox <willy@...radead.org>,
        Hugh Dickins <hughd@...gle.com>
Cc:     Andrew Morton <akpm@...ux-foundation.org>, Jan Kara <jack@...e.cz>,
        William Kucharski <william.kucharski@...cle.com>,
        linux-fsdevel@...r.kernel.org, linux-mm@...ck.org, hch@....de,
        hannes@...xchg.org, yang.shi@...ux.alibaba.com,
        dchinner@...hat.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 00/16] Overhaul multi-page lookups for THP

Hi

On 25.11.2020 03:32, Matthew Wilcox wrote:
> On Tue, Nov 17, 2020 at 11:43:02PM +0000, Matthew Wilcox wrote:
>> On Tue, Nov 17, 2020 at 07:15:13PM +0000, Matthew Wilcox wrote:
>>> I find both of these functions exceptionally confusing.  Does this
>>> make it easier to understand?
>> Never mind, this is buggy.  I'll send something better tomorrow.
> That took a week, not a day.  *sigh*.  At least this is shorter.
>
> commit 1a02863ce04fd325922d6c3db6d01e18d55f966b
> Author: Matthew Wilcox (Oracle) <willy@...radead.org>
> Date:   Tue Nov 17 10:45:18 2020 -0500
>
>      fix mm-truncateshmem-handle-truncates-that-split-thps.patch

This patch landed in todays linux-next (20201203) as commit 8678b27f4b8b 
("8678b27f4b8bfc130a13eb9e9f27171bcd8c0b3b"). Sadly it breaks booting of 
ANY of my ARM 32bit test systems, which use initrd. ARM64bit based 
systems boot fine. Here is example of the crash:

Waiting 2 sec before mounting root device...
RAMDISK: squashfs filesystem found at block 0
RAMDISK: Loading 37861KiB [1 disk] into ram disk... /
/
/
/
done.
using deprecated initrd support, will be removed in 2021.
------------[ cut here ]------------
kernel BUG at fs/inode.c:531!
Internal error: Oops - BUG: 0 [#1] PREEMPT SMP ARM
Modules linked in:
CPU: 0 PID: 12 Comm: kworker/0:1 Not tainted 5.10.0-rc6-next-20201203 #2131
Hardware name: Samsung Exynos (Flattened Device Tree)
Workqueue: events delayed_fput
PC is at clear_inode+0x74/0x88
LR is at clear_inode+0x14/0x88
pc : [<c02fb334>]    lr : [<c02fb2d4>]    psr: 200001d3
sp : c1d2be68  ip : c1736ff4  fp : c1208f14
r10: c1208ec8  r9 : c20020c0  r8 : c209b0d8
r7 : c02f759c  r6 : c0c13940  r5 : c209b244  r4 : c209b0d8
r3 : 000024f9  r2 : 00000000  r1 : 00000000  r0 : c209b244
Flags: nzCv  IRQs off  FIQs off  Mode SVC_32  ISA ARM  Segment none
Control: 10c5387d  Table: 4000404a  DAC: 00000051
Process kworker/0:1 (pid: 12, stack limit = 0x(ptrval))
Stack: (0xc1d2be68 to 0xc1d2c000)
...
[<c02fb334>] (clear_inode) from [<c02fc8a0>] (evict+0x12c/0x13c)
[<c02fc8a0>] (evict) from [<c02f648c>] (__dentry_kill+0xb0/0x188)
[<c02f648c>] (__dentry_kill) from [<c02f7714>] (dput+0x2d8/0x67c)
[<c02f7714>] (dput) from [<c02dd300>] (__fput+0xd4/0x24c)
[<c02dd300>] (__fput) from [<c02dd4b4>] (delayed_fput+0x3c/0x48)
[<c02dd4b4>] (delayed_fput) from [<c0149660>] (process_one_work+0x234/0x7e4)
[<c0149660>] (process_one_work) from [<c0149c54>] (worker_thread+0x44/0x51c)
[<c0149c54>] (worker_thread) from [<c0150a88>] (kthread+0x158/0x1a0)
[<c0150a88>] (kthread) from [<c010011c>] (ret_from_fork+0x14/0x38)
Exception stack(0xc1d2bfb0 to 0xc1d2bff8)
...
---[ end trace b3c68905048e7f9b ]---
note: kworker/0:1[12] exited with preempt_count 1
BUG: sleeping function called from invalid context at 
./include/linux/percpu-rwsem.h:49
in_atomic(): 0, irqs_disabled(): 128, non_block: 0, pid: 12, name: 
kworker/0:1
INFO: lockdep is turned off.
irq event stamp: 7498
hardirqs last  enabled at (7497): [<c02b7fcc>] free_unref_page+0x80/0x88
hardirqs last disabled at (7498): [<c0b40b18>] _raw_spin_lock_irq+0x24/0x5c
softirqs last  enabled at (6234): [<c0966af4>] linkwatch_do_dev+0x20/0x80
softirqs last disabled at (6232): [<c0966a60>] rfc2863_policy+0x30/0xa4
CPU: 0 PID: 12 Comm: kworker/0:1 Tainted: G      D 
5.10.0-rc6-next-20201203 #2131
Hardware name: Samsung Exynos (Flattened Device Tree)
Workqueue: events delayed_fput
[<c0111718>] (unwind_backtrace) from [<c010d050>] (show_stack+0x10/0x14)
[<c010d050>] (show_stack) from [<c0b34310>] (dump_stack+0xb4/0xd4)
[<c0b34310>] (dump_stack) from [<c015a9d4>] (___might_sleep+0x288/0x2d8)
[<c015a9d4>] (___might_sleep) from [<c013c744>] (exit_signals+0x38/0x428)
[<c013c744>] (exit_signals) from [<c012ce18>] (do_exit+0xe4/0xc88)
[<c012ce18>] (do_exit) from [<c010d28c>] (die+0x238/0x30c)
[<c010d28c>] (die) from [<c010d560>] (do_undefinstr+0xbc/0x26c)
[<c010d560>] (do_undefinstr) from [<c0100c1c>] (__und_svc_finish+0x0/0x44)
Exception stack(0xc1d2be18 to 0xc1d2be60)
VFS: Mounted root (squashfs filesystem) readonly on device 1:0.
be00: c209b244 00000000
be20: 00000000 000024f9 c209b0d8 c209b244 c0c13940 c02f759c c209b0d8 
c20020c0
be40: c1208ec8 c1208f14 c1736ff4 c1d2be68 c02fb2d4 c02fb334 200001d3 
ffffffff
[<c0100c1c>] (__und_svc_finish) from [<c02fb334>] (clear_inode+0x74/0x88)
[<c02fb334>] (clear_inode) from [<c02fc8a0>] (evict+0x12c/0x13c)
[<c02fc8a0>] (evict) from [<c02f648c>] (__dentry_kill+0xb0/0x188)
[<c02f648c>] (__dentry_kill) from [<c02f7714>] (dput+0x2d8/0x67c)
[<c02f7714>] (dput) from [<c02dd300>] (__fput+0xd4/0x24c)
[<c02dd300>] (__fput) from [<c02dd4b4>] (delayed_fput+0x3c/0x48)
[<c02dd4b4>] (delayed_fput) from [<c0149660>] (process_one_work+0x234/0x7e4)
[<c0149660>] (process_one_work) from [<c0149c54>] (worker_thread+0x44/0x51c)
[<c0149c54>] (worker_thread) from [<c0150a88>] (kthread+0x158/0x1a0)
[<c0150a88>] (kthread) from [<c010011c>] (ret_from_fork+0x14/0x38)
Exception stack(0xc1d2bfb0 to 0xc1d2bff8)
bfa0:                                     00000000 00000000 00000000 
00000000
bfc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 
00000000
bfe0: 00000000 00000000 00000000 00000000 00000013 00000000
EXT4-fs (mmcblk0p6): INFO: recovery required on readonly filesystem
EXT4-fs (mmcblk0p6): write access will be enabled during recovery
EXT4-fs (mmcblk0p6): recovery complete
EXT4-fs (mmcblk0p6): mounted filesystem with ordered data mode. Opts: (null)
VFS: Mounted root (ext4 filesystem) readonly on device 179:6.
Trying to move old root to /initrd ...

I suppose this issue can be also reproduced with qemu.

Best regards

-- 
Marek Szyprowski, PhD
Samsung R&D Institute Poland

Powered by blists - more mailing lists