From: Finn Thain <fthain@linux-m68k.org>
To: Michael Schmitz <schmitzmic@gmail.com>
Cc: linux-m68k@vger.kernel.org
Subject: Re: Mainline kernel crashes, was Re: RFC: remove set_fs for m68k
Date: Sun, 12 Sep 2021 10:51:21 +1000 (AEST) [thread overview]
Message-ID: <2c624213-6a4-799c-45e-a1be578dd5f@linux-m68k.org> (raw)
In-Reply-To: <0477f373-86c9-dacb-a7b1-25fe4b3befd3@gmail.com>
On Fri, 10 Sep 2021, Michael Schmitz wrote:
> On 10/09/21 10:51, Finn Thain wrote:
> > On Wed, 8 Sep 2021, Michael Schmitz wrote:
> >
> > >
> > > In a related case, I've managed to swap my 'resume_userspace' format
> > > error for a nice 'illegal instruction' format error apparently
> > > caused by an invalid function pointer in
> > > __handle_irq_event_percpu(), just by disabling all interrupts upon
> > > entering the auto_inthandler and user_inthandler exception handlers.
> > > This bug is quite readily reproduced by running your
> > > kernel_coverage.sh script in a loop (panics on the first stress test
> > > on the second pass):
> > >
> > > Stress run 2
> > > Logging to stress-ng-20210908-0838.log
> > > ./kernel-coverage.sh: line 272: lcov: command not found
> > > running --fork 1 --fork-vm -t 60 --timestamp --no-rand-seed --times
> > > stress-ng: 08:40:08.70 info: [1914] setting to a 60 second run per
> > > stressor
> > > stress-ng: 08:40:08.82 info: [1914] dispatching hogs: 1 fork
> > > packet_write_wait: Connection to 10.1.1.4 port 22: Broken pipe
> > >
> > > Why disabling interrupts during interrupt processing would make
> > > matters worse doesn't make any sense to me...
> > >
> >
> > Are you able to reproduce that with a stock mainline build?
>
> I still need to try that. The kernel panic went away when adding
> additional local_irq_save()/local_irq_restore() in do_IRQ(),
I eventually realized that the code using percpu variables inside
add_interrupt_randomness() outside of the spinlock may also need irqs
disabled. So I've now done as you did, that is,
diff --git a/arch/m68k/kernel/irq.c b/arch/m68k/kernel/irq.c
index 9ab4f550342e..b46d8a57f4da 100644
--- a/arch/m68k/kernel/irq.c
+++ b/arch/m68k/kernel/irq.c
@@ -20,10 +20,13 @@
asmlinkage void do_IRQ(int irq, struct pt_regs *regs)
{
struct pt_regs *oldregs = set_irq_regs(regs);
+ unsigned long flags;
+ local_irq_save(flags);
irq_enter();
generic_handle_irq(irq);
irq_exit();
+ local_irq_restore(flags);
set_irq_regs(oldregs);
}
There may be a better way to achieve that. If the final IPL can be found
in regs then it doesn't need to be saved again.
I haven't looked for a possible entropy pool improvement from correct
locking in random.c -- it would not surprise me if there was one.
But none of this explains the panics I saw so I went looking for potential
race conditions in the irq_enter_rcu() and irq_exit_rcu() code. I haven't
found the bug yet.
However, I did notice that preempt_count_add() and preempt_count_sub() and
the associated macros in linux/preempt.h produce very inefficient code in
the interrupt fast path. Here's one example (there are others).
000323de <irq_enter_rcu>:
323de: 4e56 0000 linkw %fp,#0
323e2: 200f movel %sp,%d0
323e4: 0280 ffff e000 andil #-8192,%d0
323ea: 2040 moveal %d0,%a0
323ec: 2028 000c movel %a0@(12),%d0
323f0: 0680 0001 0000 addil #65536,%d0
323f6: 2140 000c movel %d0,%a0@(12)
323fa: 082a 0001 000f btst #1,%a2@(15)
32400: 670c beqs 3240e <irq_enter_rcu+0x30>
32402: 2028 000c movel %a0@(12),%d0
32406: 2028 000c movel %a0@(12),%d0
3240a: 2028 000c movel %a0@(12),%d0
3240e: 4e5e unlk %fp
32410: 4e75 rts
next prev parent reply other threads:[~2021-09-12 0:51 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-09 7:01 RFC: remove set_fs for m68k Christoph Hellwig
2021-07-09 7:01 ` [PATCH 1/7] m68k: document that access_ok is broken for !CONFIG_CPU_HAS_ADDRESS_SPACES Christoph Hellwig
2021-07-09 7:01 ` [PATCH 2/7] m68k: use BUILD_BUG for passing invalid sizes to get_user/put_user Christoph Hellwig
2021-07-09 7:01 ` [PATCH 3/7] m68k: remove the inline copy_{from,to}_user variants Christoph Hellwig
2021-07-09 7:01 ` [PATCH 4/7] m68k: remove the err argument to the get_user/put_user assembly helpers Christoph Hellwig
2021-07-09 7:01 ` [PATCH 5/7] m68k: factor the 8-byte lowlevel {get,put}_user code into helpers Christoph Hellwig
2021-07-09 7:01 ` [PATCH 6/7] m68k: provide __{get,put}_kernel_nofault Christoph Hellwig
2021-07-09 7:01 ` [PATCH 7/7] m68k: remove set_fs() Christoph Hellwig
2021-07-11 7:20 ` RFC: remove set_fs for m68k Michael Schmitz
2021-07-12 9:50 ` Christoph Hellwig
2021-07-12 10:20 ` Andreas Schwab
2021-07-12 19:12 ` Michael Schmitz
2021-07-13 5:41 ` Christoph Hellwig
2021-07-13 8:16 ` Michael Schmitz
2021-07-13 8:54 ` Christoph Hellwig
2021-07-14 19:26 ` Michael Schmitz
2021-07-14 20:03 ` Andreas Schwab
2021-07-15 5:44 ` Michael Schmitz
2021-07-16 2:03 ` Michael Schmitz
2021-07-17 5:41 ` Michael Schmitz
2021-07-18 1:14 ` Michael Schmitz
2021-07-21 17:05 ` Christoph Hellwig
2021-07-21 19:20 ` Michael Schmitz
2021-07-23 4:00 ` Michael Schmitz
2021-07-23 5:11 ` Christoph Hellwig
2021-07-25 7:36 ` Michael Schmitz
2021-07-31 19:31 ` Michael Schmitz
2021-08-06 3:10 ` Michael Schmitz
2021-08-11 9:12 ` Christoph Hellwig
2021-08-12 3:37 ` Michael Schmitz
2021-08-15 7:42 ` Christoph Hellwig
2021-08-15 19:17 ` Michael Schmitz
2021-08-16 6:58 ` Christoph Hellwig
[not found] ` <23f745f2-9086-81fb-3d9e-40ea08a1923@linux-m68k.org>
[not found] ` <20210816075155.GA29187@lst.de>
[not found] ` <d407a2a1-738b-5cd5-c2ed-b7250c5da8ec@gmail.com>
[not found] ` <83571ae-10ae-2919-cde-b6b4a5769c9@linux-m68k.org>
[not found] ` <dc594142-e459-533e-cac2-c7a213cec464@gmail.com>
[not found] ` <f4ab2dcb-6761-c60b-54ce-35d0d017d371@gmail.com>
[not found] ` <d772d22e-a945-3e35-80a2-f4783893bea@linux-m68k.org>
[not found] ` <b2c55280-657b-51c2-065c-3fc93db050b9@gmail.com>
[not found] ` <d7b8f7eb-fc18-c8d-fe3e-dcdf19d3f4b@linux-m68k.org>
[not found] ` <755e55ba-4ce2-b4e4-a628-5abc183a557a@linux-m68k.org>
[not found] ` <b52a10fe-3e4b-5740-d3f8-52bce3bc988@linux-m68k.org>
[not found] ` <31f27da7-be60-8eb-9834-748b653c2246@linux-m68k.org>
2021-09-07 3:28 ` Mainline kernel crashes, was " Finn Thain
2021-09-07 5:53 ` Michael Schmitz
2021-09-07 23:50 ` Finn Thain
2021-09-08 8:54 ` Michael Schmitz
2021-09-09 9:40 ` Finn Thain
2021-09-09 23:29 ` Michael Schmitz
2021-09-09 22:51 ` Finn Thain
2021-09-10 0:03 ` Michael Schmitz
2021-09-12 0:51 ` Finn Thain [this message]
2021-09-12 3:55 ` Brad Boyer
2021-09-13 1:27 ` Finn Thain
2021-09-13 3:26 ` Michael Schmitz
2021-09-13 5:22 ` Finn Thain
2021-09-13 7:20 ` Michael Schmitz
2021-09-14 3:13 ` Michael Schmitz
2021-09-15 1:38 ` Finn Thain
2021-09-15 8:37 ` Michael Schmitz
2021-09-16 9:04 ` Finn Thain
2021-09-16 22:28 ` Michael Schmitz
2021-09-21 21:14 ` Michael Schcmitz
2021-08-22 19:33 ` Michael Schmitz
2021-08-23 4:04 ` Michael Schmitz
2021-08-23 17:59 ` Linus Torvalds
2021-08-23 21:31 ` Michael Schmitz
2021-08-23 21:49 ` Linus Torvalds
2021-08-24 8:08 ` Andreas Schwab
2021-08-24 8:44 ` Michael Schmitz
2021-08-24 8:59 ` Andreas Schwab
2021-08-25 7:51 ` Michael Schmitz
2021-08-25 8:44 ` Andreas Schwab
2021-08-25 22:59 ` Michael Schmitz
2021-08-25 23:30 ` Brad Boyer
2021-08-26 7:46 ` Michael Schmitz
2021-08-26 7:45 ` Andreas Schwab
2021-09-14 2:43 ` Michael Schmitz
2021-09-14 15:54 ` Linus Torvalds
2021-09-14 16:28 ` Al Viro
2021-09-14 16:38 ` Linus Torvalds
2021-09-15 1:06 ` Al Viro
2021-07-12 19:04 ` Michael Schmitz
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2c624213-6a4-799c-45e-a1be578dd5f@linux-m68k.org \
--to=fthain@linux-m68k.org \
--cc=linux-m68k@vger.kernel.org \
--cc=schmitzmic@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).