Re: spidev regression in 6.2-rc kernel

From: Bartosz Golaszewski
Date: Mon Jan 16 2023 - 07:47:52 EST


On Mon, 16 Jan 2023 at 13:19, Bartosz Golaszewski <brgl@xxxxxxxx> wrote:
>
> On Mon, Jan 16, 2023 at 1:06 PM Francesco Dolcini <francesco@xxxxxxxxxx> wrote:
> >
> > Hello,
> > we spotted a regression on spidev on latest 6.2-rc kernel.
> >
> > [ 214.047619]
> > [ 214.049198] ============================================
> > [ 214.054533] WARNING: possible recursive locking detected
> > [ 214.059858] 6.2.0-rc3-0.0.0-devel+git.97ec4d559d93 #1 Not tainted
> > [ 214.065969] --------------------------------------------
> > [ 214.071290] spidev_test/1454 is trying to acquire lock:
> > [ 214.076530] c4925dbc (&spidev->spi_lock){+.+.}-{3:3}, at: spidev_ioctl+0x8e0/0xab8
> > [ 214.084164]
> > [ 214.084164] but task is already holding lock:
> > [ 214.090007] c4925dbc (&spidev->spi_lock){+.+.}-{3:3}, at: spidev_ioctl+0x44/0xab8
> > [ 214.097537]
> > [ 214.097537] other info that might help us debug this:
> > [ 214.104075] Possible unsafe locking scenario:
> > [ 214.104075]
> > [ 214.110004] CPU0
> > [ 214.112461] ----
> > [ 214.114916] lock(&spidev->spi_lock);
> > [ 214.118687] lock(&spidev->spi_lock);
> > [ 214.122457]
> > [ 214.122457] *** DEADLOCK ***
> > [ 214.122457]
> > [ 214.128386] May be due to missing lock nesting notation
> > [ 214.128386]
> > [ 214.135183] 2 locks held by spidev_test/1454:
> > [ 214.139553] #0: c4925dbc (&spidev->spi_lock){+.+.}-{3:3}, at: spidev_ioctl+0x44/0xab8
> > [ 214.147524] #1: c4925e14 (&spidev->buf_lock){+.+.}-{3:3}, at: spidev_ioctl+0x70/0xab8
> > [ 214.155493]
> > [ 214.155493] stack backtrace:
> > [ 214.159861] CPU: 0 PID: 1454 Comm: spidev_test Not tainted 6.2.0-rc3-0.0.0-devel+git.97ec4d559d93 #1
> > [ 214.169012] Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree)
> > [ 214.175555] unwind_backtrace from show_stack+0x10/0x14
> > [ 214.180819] show_stack from dump_stack_lvl+0x60/0x90
> > [ 214.185900] dump_stack_lvl from __lock_acquire+0x874/0x2858
> > [ 214.191584] __lock_acquire from lock_acquire+0xfc/0x378
> > [ 214.196918] lock_acquire from __mutex_lock+0x9c/0x8a8
> > [ 214.202083] __mutex_lock from mutex_lock_nested+0x1c/0x24
> > [ 214.207597] mutex_lock_nested from spidev_ioctl+0x8e0/0xab8
> > [ 214.213284] spidev_ioctl from sys_ioctl+0x4d0/0xe2c
> > [ 214.218277] sys_ioctl from ret_fast_syscall+0x0/0x1c
> > [ 214.223351] Exception stack(0xe75cdfa8 to 0xe75cdff0)
> > [ 214.228422] dfa0: 00000000 00001000 00000003 40206b00 bee266e8 bee266e0
> > [ 214.236617] dfc0: 00000000 00001000 006a71a0 00000036 004c0040 004bfd18 00000000 00000003
> > [ 214.244809] dfe0: 00000036 bee266c8 b6f16dc5 b6e8e5f6
> >
> >
> > This is not running the latest rc4, but on sha 97ec4d559d93 (this is
> > just what our CI had available when this test was run). I was not able
> > to bisect it, but it seems something that you could have introduced.
> >
> > The log is from an apalis-imx6, but I have the same on other ARM SOC.
> >
> > Can you have a look?
> > Thanks
> >
> > Francesco
> >
>
> Eek! Yes it's commit 1f4d2dd45b6e ("spi: spidev: fix a race condition
> when accessing spidev->spi"): spidev_ioctl() takes the lock and in
> certain instances can end up calling spidev_compat_ioc_message() which
> takes the same lock again. I'll send a fix shortly.
>
> Bart

Seems, like that's not it... Francesco: what is the output of:

./scripts/faddr2line drivers/spi/spidev.o spidev_ioctl+0x44/0xab8

and

./scripts/faddr2line drivers/spi/spidev.o spidev_ioctl+0x44/0xab8

on the spidev.o object for that build?

Bart