public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: will.deacon@arm•com (Will Deacon)
To: linux-arm-kernel@lists•infradead.org
Subject: [PATCH v8 3/4] arm64: Add do_softirq_own_stack() and enable irq_stacks
Date: Wed, 9 Dec 2015 11:38:53 +0000	[thread overview]
Message-ID: <20151209113852.GC9303@arm.com> (raw)
In-Reply-To: <5667F897.5070005@arm.com>

On Wed, Dec 09, 2015 at 09:47:03AM +0000, James Morse wrote:
> On 08/12/15 17:23, James Morse wrote:
> > My only theory is that there is an off by one, and its reading what was
> > x29 instead. This wouldn't show up in these tests, but might be a
> > problem for aarch32 user-space, as presumably x29==0 when it switches to
> > aarch64 mode for el0_irq(). I will try this tomorrow.
> 
> Yup, this is what is happening. Its an off-by-one due to broken thinking
> about how the stack works. My broken thinking was:
> 
> >   top ------------
> >       | dummy_lr | <- irq_stack_ptr
> >       ------------
> >       |   x29    |
> >       ------------
> >       |   x19    | <- irq_stack_ptr - 0x10
> >       ------------
> >       |   xzr    |
> >       ------------
> 
> But the stack-pointer is decreased before use. So it actually looks like
> this:
> 
> >       ------------
> >       |          |  <- irq_stack_ptr
> >   top ------------
> >       | dummy_lr |
> >       ------------
> >       |   x29    | <- irq_stack_ptr - 0x10
> >       ------------
> >       |   x19    |
> >       ------------
> >       |   xzr    | <- irq_stack_ptr - 0x20
> >       ------------
> 
> The value being used as the original stack is x29, which in all the
> tests is sp but without the current frames data, hence there are no
> missing frames in the output.
> 
> Jungseok Lee picked it up with a 32bit user space because aarch32 can't
> use x29, so it remains 0 forever. The fix he posted is correct.
> 
> Will: do you want to take Jungseok Lee's patch as a 'Fixes:', or is it
> easier if I repost the series?

I'll take it as a fix on top, but I still want to get to the bottom of
why unwind_frame appeared to be exploding. We really shouldn't be relying
on the frame layout to provide us with safe addresses in there.

Will

  reply	other threads:[~2015-12-09 11:38 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-04 11:02 [PATCH v8 0/4] arm64: Add support for IRQ stack James Morse
2015-12-04 11:02 ` [PATCH v8 1/4] arm64: Store struct task_info in sp_el0 James Morse
2015-12-04 13:27   ` Catalin Marinas
2015-12-04 14:55     ` James Morse
2015-12-04 16:18       ` Catalin Marinas
2015-12-06 13:15     ` Jungseok Lee
2015-12-04 11:02 ` [PATCH v8 2/4] arm64: Modify stack trace and dump for use with irq_stack James Morse
2015-12-04 12:21   ` Jungseok Lee
2015-12-04 14:31   ` Catalin Marinas
2015-12-04 11:02 ` [PATCH v8 3/4] arm64: Add do_softirq_own_stack() and enable irq_stacks James Morse
2015-12-04 13:46   ` Catalin Marinas
2015-12-04 13:47     ` Catalin Marinas
2015-12-07 22:48   ` Catalin Marinas
2015-12-08 11:43     ` Will Deacon
2015-12-08 16:02       ` Jungseok Lee
2015-12-08 17:23         ` James Morse
2015-12-08 17:27           ` Will Deacon
2015-12-08 23:13           ` Jungseok Lee
2015-12-09  9:47           ` James Morse
2015-12-09 11:38             ` Will Deacon [this message]
2015-12-09 13:45   ` Will Deacon
2015-12-09 14:36     ` James Morse
2015-12-04 11:02 ` [PATCH v8 4/4] arm64: switch to irq_stack during softirq James Morse
2015-12-04 14:01   ` Catalin Marinas
2015-12-04 14:39     ` James Morse
2015-12-04 18:40       ` Catalin Marinas
2015-12-08 10:29         ` James Morse
2015-12-06 13:51       ` Jungseok Lee
2015-12-04 12:17 ` [PATCH v8 0/4] arm64: Add support for IRQ stack Jungseok Lee
2015-12-06 13:56   ` Jungseok Lee
2015-12-04 13:57 ` Catalin Marinas
2015-12-06 13:33   ` Jungseok Lee
2015-12-10 10:22 ` [PATCH v8 5/4] arm64: Fix off-by-one in stack tracing when stepping off irq stack James Morse
2015-12-10 10:22   ` [PATCH v8 6/4] arm64: Add this_cpu_ptr() assembler macro for use in entry.S James Morse
2015-12-10 10:22   ` [PATCH v8 7/4] arm64: when walking onto the task stack, check sp & fp are in current->stack James Morse
2015-12-10 10:22   ` [PATCH v8 8/4] arm64: don't call C code with el0's fp register James Morse
2015-12-10 14:03   ` [PATCH v8 5/4] arm64: Fix off-by-one in stack tracing when stepping off irq stack Jungseok Lee
2015-12-15 11:21 ` [PATCH v8 9/4] arm64: reduce stack use in irq_handler James Morse
2015-12-18 16:01 ` [PATCH v8 9/4] arm64: remove irq_count and do_softirq_own_stack() James Morse
2015-12-20 11:07   ` Jungseok Lee
2015-12-21 11:30     ` Will Deacon
2015-12-21 12:19       ` James Morse
2015-12-21 12:21         ` Will Deacon
2015-12-21 14:06           ` Jungseok Lee

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=20151209113852.GC9303@arm.com \
    --to=will.deacon@arm$(echo .)com \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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