dri-devel.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
To: Thomas Zimmermann <tzimmermann@suse.de>,
	daniel@ffwll.ch, airlied@linux.ie, mripard@kernel.org,
	maarten.lankhorst@linux.intel.com
Cc: dri-devel@lists.freedesktop.org
Subject: Re: [PATCH] drm: Set vm_ops to GEM object's values during mmap
Date: Fri, 15 Jan 2021 09:32:10 +0000	[thread overview]
Message-ID: <318a8d3a-d99f-2db6-6429-20754c93288b@ideasonboard.com> (raw)
In-Reply-To: <11206023-8595-6b68-5e74-d4b7edcc528e@suse.de>

On 15/01/2021 09:28, Thomas Zimmermann wrote:
> Hi
> 
> Am 15.01.21 um 10:17 schrieb Kieran Bingham:
>> Hi Thomas,
>>
>> On 15/01/2021 08:39, Thomas Zimmermann wrote:
>>> The GEM mmap code relies on the GEM object's mmap callback to set the
>>> VMA's vm_ops field. This is easily forgotten and lead to a memory leak
>>
>>   s/lead/can lead/
>>
>>> in the CMA helpers. Instead set the vm_ops field in the DRM core code
>>> to the GEM object's value. Drivers with different needs can override
>>> this in their mmap callback.
>>>
>>> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
>>> Fixes: f5ca8eb6f9bd ("drm/cma-helper: Implement mmap as GEM CMA
>>> object functions")
>>> Reported-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
>>
>> This also works here.
>>   - https://paste.ubuntu.com/p/S2Q586rgwT/
>>
>> Tested-by: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
> 
> Great! I have to send out another iteration of the patch, but I think
> the change is such that I can keep the Tested-by.

I have this test automated now, so I'll kick off a validation run on the
new patch as well.

--
Kieran


> 
> Best regards
> Thomas
> 
>>
>>> Cc: Maxime Ripard <mripard@kernel.org>
>>> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
>>> Cc: David Airlie <airlied@linux.ie>
>>> Cc: Daniel Vetter <daniel@ffwll.ch>
>>> Cc: Eric Anholt <eric@anholt.net>
>>> Cc: dri-devel@lists.freedesktop.org
>>> ---
>>>   drivers/gpu/drm/drm_gem.c   | 23 ++++++++++++-----------
>>>   drivers/gpu/drm/drm_prime.c |  4 ++++
>>>   2 files changed, 16 insertions(+), 11 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
>>> index 34b2f111c01c..54d95621fcbb 100644
>>> --- a/drivers/gpu/drm/drm_gem.c
>>> +++ b/drivers/gpu/drm/drm_gem.c
>>> @@ -1069,27 +1069,28 @@ int drm_gem_mmap_obj(struct drm_gem_object
>>> *obj, unsigned long obj_size,
>>>         vma->vm_private_data = obj;
>>>   +    if (!obj->funcs->vm_ops) {
>>> +        ret = -EINVAL;
>>> +        goto err_drm_gem_object_put;
>>> +    }
>>> +    vma->vm_ops = obj->funcs->vm_ops;
>>> +
>>>       if (obj->funcs->mmap) {
>>>           ret = obj->funcs->mmap(obj, vma);
>>> -        if (ret) {
>>> -            drm_gem_object_put(obj);
>>> -            return ret;
>>> -        }
>>> +        if (ret)
>>> +            goto err_drm_gem_object_put;
>>>           WARN_ON(!(vma->vm_flags & VM_DONTEXPAND));
>>>       } else {
>>> -        if (obj->funcs->vm_ops)
>>> -            vma->vm_ops = obj->funcs->vm_ops;
>>> -        else {
>>> -            drm_gem_object_put(obj);
>>> -            return -EINVAL;
>>> -        }
>>> -
>>>           vma->vm_flags |= VM_IO | VM_PFNMAP | VM_DONTEXPAND |
>>> VM_DONTDUMP;
>>>           vma->vm_page_prot =
>>> pgprot_writecombine(vm_get_page_prot(vma->vm_flags));
>>>           vma->vm_page_prot = pgprot_decrypted(vma->vm_page_prot);
>>>       }
>>>         return 0;
>>> +
>>> +err_drm_gem_object_put:
>>> +    drm_gem_object_put(obj);
>>> +    return ret;
>>>   }
>>>   EXPORT_SYMBOL(drm_gem_mmap_obj);
>>>   diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
>>> index 683aa29ecd3b..0549ec190583 100644
>>> --- a/drivers/gpu/drm/drm_prime.c
>>> +++ b/drivers/gpu/drm/drm_prime.c
>>> @@ -717,6 +717,10 @@ int drm_gem_prime_mmap(struct drm_gem_object
>>> *obj, struct vm_area_struct *vma)
>>>       vma->vm_pgoff += drm_vma_node_start(&obj->vma_node);
>>>         if (obj->funcs && obj->funcs->mmap) {
>>> +        if (!obj->funcs->vm_ops)
>>> +            return -EINVAL;
>>> +        vma->vm_ops = obj->funcs->vm_ops;
>>> +
>>>           ret = obj->funcs->mmap(obj, vma);
>>>           if (ret)
>>>               return ret;
>>>
>>
>> _______________________________________________
>> dri-devel mailing list
>> dri-devel@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/dri-devel
>>
> 

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

      reply	other threads:[~2021-01-15  9:32 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-15  8:39 [PATCH] drm: Set vm_ops to GEM object's values during mmap Thomas Zimmermann
2021-01-15  9:17 ` Kieran Bingham
2021-01-15  9:28   ` Thomas Zimmermann
2021-01-15  9:32     ` Kieran Bingham [this message]

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=318a8d3a-d99f-2db6-6429-20754c93288b@ideasonboard.com \
    --to=kieran.bingham+renesas@ideasonboard.com \
    --cc=airlied@linux.ie \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=mripard@kernel.org \
    --cc=tzimmermann@suse.de \
    /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).