Re: [PATCH] kgdboc: Be a bit more robust about handling earlycon leaving
From: Daniel Thompson <hidden>
Date: 2020-05-01 11:49:51
Also in:
lkml
On Thu, Apr 30, 2020 at 09:59:09AM -0700, Douglas Anderson wrote:
The original implementation of kgdboc_earlycon included a comment about how it was impossible to get notified about the boot console going away without making changes to the Linux core. Since folks often don't want to change the Linux core for kgdb's purposes, the kgdboc_earlycon implementation did a bit of polling to figure out when the boot console went away. It turns out, though, that it is possible to get notified about the boot console going away. The solution is either clever or a hack depending on your viewpoint. ...or, perhaps, a clever hack. All we need to do is head-patch the "exit" routine of the boot console. We know that "struct console" must be writable because it has a "next" pointer in it, so we can just put our own exit routine in, do our stuff, and then call back to the original.
I think I'm in the hack camp on this one!
This works great to get notified about the boot console going away. The (slight) problem is that in the context of the boot console's exit routine we can't call tty_find_polling_driver().
I presume this is something to do with the tty_mutex?
We solve this by kicking off some work on the system_wq when we get notified and this works pretty well.
There are some problems with the workqueue approach. Firstly, its runs too early on many systems (esp. those that register the console from a console initcall. kgdboc cannot find the tty, I think because the console is registered a long time before the corresponding tty comes up. Secondly I am seeing warnings related to the tty_mutex where the might_sleep() machinery ends up flushing the active workqueue. [ 39.298332] ------------[ cut here ]------------ [ 39.298332] WARNING: CPU: 0 PID: 5 at kernel/workqueue.c:3033 __flush_work+00 [ 39.298332] Modules linked in: [ 39.298332] CPU: 0 PID: 5 Comm: kworker/0:0 Not tainted 5.7.0-rc3+ #47 [ 39.298332] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-204 [ 39.298332] Workqueue: events kgdboc_earlycon_exit_work_fn [ 39.298332] RIP: 0010:__flush_work+0x19c/0x1c0 [ 39.298332] Code: 4c 8b 6d 20 e9 06 ff ff ff 41 c6 04 24 00 fb 45 31 f6 eb 8f [ 39.298332] RSP: 0018:ffff993500033dd0 EFLAGS: 00010246 [ 39.298332] RAX: 0000000000000000 RBX: ffffffffadcd0720 RCX: 0000000000000001 [ 39.298332] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffffffadcd0820 [ 39.298332] RBP: ffff8a633ec299c0 R08: 0000000000000000 R09: 0000000000000001 [ 39.298332] R10: 000000000000000a R11: f000000000000000 R12: 00000000ffffffed [ 39.298332] R13: ffff8a633e408840 R14: 0000000000000000 R15: ffff8a633e408840 [ 39.298332] FS: 0000000000000000(0000) GS:ffff8a633ec00000(0000) knlGS:00000 [ 39.298332] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 39.298332] CR2: ffff8a6333201000 CR3: 0000000032a0a000 CR4: 00000000000006f0 [ 39.298332] Call Trace: [ 39.298332] ? _cond_resched+0x10/0x20 [ 39.298332] ? mutex_lock+0x9/0x30 [ 39.298332] ? tty_find_polling_driver+0x134/0x1a0 [ 39.298332] configure_kgdboc+0x12d/0x1c0 [ 39.298332] kgdboc_earlycon_exit_work_fn+0x1a/0x40 [ 39.298332] process_one_work+0x1d3/0x380 [ 39.298332] worker_thread+0x45/0x3c0 [ 39.298332] kthread+0xf6/0x130 [ 39.298332] ? process_one_work+0x380/0x380 [ 39.298332] ? kthread_park+0x80/0x80 [ 39.298332] ret_from_fork+0x22/0x40 [ 39.298332] ---[ end trace 1190f578d6e11204 ]--- [ 39.298338] KGDB: Unregistered I/O driver kgdboc_earlycon, debugger disabled Daniel.