All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jerome Glisse <jglisse-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: "Michel Dänzer" <michel-otUistvHUpPR7s880joybQ@public.gmane.org>
Cc: Philip Yang <Philip.Yang-5C7GfCeVMHo@public.gmane.org>,
	amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org,
	linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org
Subject: Re: KASAN caught amdgpu / HMM use-after-free
Date: Wed, 27 Feb 2019 12:24:50 -0500	[thread overview]
Message-ID: <20190227172450.GB3296@redhat.com> (raw)
In-Reply-To: <e8466985-a66b-468b-5fff-6e743180da67-otUistvHUpPR7s880joybQ@public.gmane.org>

On Wed, Feb 27, 2019 at 06:02:49PM +0100, Michel Dänzer wrote:
> 
> See the attached dmesg excerpt. I've hit this a few times running piglit
> with amd-staging-drm-next, first on February 22nd.
> 
> The memory was freed after calling hmm_mirror_unregister in
> amdgpu_mn_destroy.

So that branch is not using the HMM changes queue up for 5.1 and thus
what you are doing is somewhat illegal. In 5.1 changes all is refcounted
and this bug should not be able to happen. So if you rebase your work
on top of 

https://cgit.freedesktop.org/~glisse/linux/log/?h=hmm-for-5.1

Or linux-next (i believe i saw those bits in linux-next) then this
error will vanish. Sorry if there was confusion between what is legal
now and what is legal tommorrow :)

Cheers,
Jérôme
_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

WARNING: multiple messages have this Message-ID (diff)
From: Jerome Glisse <jglisse@redhat.com>
To: "Michel Dänzer" <michel@daenzer.net>
Cc: Philip Yang <Philip.Yang@amd.com>,
	amd-gfx@lists.freedesktop.org, linux-mm@kvack.org
Subject: Re: KASAN caught amdgpu / HMM use-after-free
Date: Wed, 27 Feb 2019 12:24:50 -0500	[thread overview]
Message-ID: <20190227172450.GB3296@redhat.com> (raw)
In-Reply-To: <e8466985-a66b-468b-5fff-6e743180da67@daenzer.net>

On Wed, Feb 27, 2019 at 06:02:49PM +0100, Michel Dänzer wrote:
> 
> See the attached dmesg excerpt. I've hit this a few times running piglit
> with amd-staging-drm-next, first on February 22nd.
> 
> The memory was freed after calling hmm_mirror_unregister in
> amdgpu_mn_destroy.

So that branch is not using the HMM changes queue up for 5.1 and thus
what you are doing is somewhat illegal. In 5.1 changes all is refcounted
and this bug should not be able to happen. So if you rebase your work
on top of 

https://cgit.freedesktop.org/~glisse/linux/log/?h=hmm-for-5.1

Or linux-next (i believe i saw those bits in linux-next) then this
error will vanish. Sorry if there was confusion between what is legal
now and what is legal tommorrow :)

Cheers,
Jérôme


  parent reply	other threads:[~2019-02-27 17:24 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-27 17:02 KASAN caught amdgpu / HMM use-after-free Michel Dänzer
2019-02-27 17:02 ` Michel Dänzer
     [not found] ` <e8466985-a66b-468b-5fff-6e743180da67-otUistvHUpPR7s880joybQ@public.gmane.org>
2019-02-27 17:14   ` Yang, Philip
2019-02-27 17:14     ` Yang, Philip
     [not found]     ` <83fde7eb-abab-e770-efd5-89bc9c39fdff-5C7GfCeVMHo@public.gmane.org>
2019-02-27 17:33       ` Michel Dänzer
2019-02-27 17:33         ` Michel Dänzer
     [not found]         ` <c26fa310-38d1-acba-cf82-bc6dc2f782c0-otUistvHUpPR7s880joybQ@public.gmane.org>
2019-02-27 18:05           ` Yang, Philip
2019-02-27 18:05             ` Yang, Philip
     [not found]             ` <35d7e134-6eef-9732-8ebf-83256e40eb65-5C7GfCeVMHo@public.gmane.org>
2019-02-27 18:32               ` Deucher, Alexander
2019-02-27 18:32                 ` Deucher, Alexander
     [not found]                 ` <BN6PR12MB18090BDFE1DD800785C5ED76F7740-/b2+HYfkarSEx6ez0IUAagdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org>
2019-02-27 18:48                   ` Yang, Philip
2019-02-27 18:48                     ` Yang, Philip
     [not found]                     ` <b81bd33a-0041-392e-2c85-19036fc1c91d-5C7GfCeVMHo@public.gmane.org>
2019-02-28 11:51                       ` Michel Dänzer
     [not found]                         ` <f26e40c8-d981-03ca-839a-6f2340a29b97-otUistvHUpPR7s880joybQ@public.gmane.org>
2019-02-28 15:29                           ` Yang, Philip
2019-02-27 17:24   ` Jerome Glisse [this message]
2019-02-27 17:24     ` Jerome Glisse

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=20190227172450.GB3296@redhat.com \
    --to=jglisse-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
    --cc=Philip.Yang-5C7GfCeVMHo@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org \
    --cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
    --cc=michel-otUistvHUpPR7s880joybQ@public.gmane.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.