Re: [Intel-gfx] [PATCH 6/8] drm/i915/gt: Fix memory leaks in per-gt sysfs

[Andrzej Hajda <andrzej.hajda@intel.com>]

On 10.05.2022 10:18, Tvrtko Ursulin wrote:
>
> On 10/05/2022 08:58, Andrzej Hajda wrote:
>> Hi Tvrtko,
>>
>> On 10.05.2022 09:28, Tvrtko Ursulin wrote:
>>>
>>> On 29/04/2022 20:56, Ashutosh Dixit wrote:
>>>> All kmalloc'd kobjects need a kobject_put() to free memory. For 
>>>> example in
>>>> previous code, kobj_gt_release() never gets called. The requirement of
>>>> kobject_put() now results in a slightly different code organization.
>>>>
>>>> v2: s/gtn/gt/ (Andi)
>>>>
>>>> Cc: Andi Shyti <andi.shyti@intel.com>
>>>> Cc: Andrzej Hajda <andrzej.hajda@intel.com>
>>>> Fixes: b770bcfae9ad ("drm/i915/gt: create per-tile sysfs interface")
>>>> Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
>>>> ---
>>>>   drivers/gpu/drm/i915/gt/intel_gt.c       |  1 +
>>>>   drivers/gpu/drm/i915/gt/intel_gt_sysfs.c | 29 
>>>> ++++++++++--------------
>>>>   drivers/gpu/drm/i915/gt/intel_gt_sysfs.h |  6 +----
>>>>   drivers/gpu/drm/i915/gt/intel_gt_types.h |  3 +++
>>>>   drivers/gpu/drm/i915/i915_sysfs.c        |  2 ++
>>>>   5 files changed, 19 insertions(+), 22 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c 
>>>> b/drivers/gpu/drm/i915/gt/intel_gt.c
>>>> index 92394f13b42f..9aede288eb86 100644
>>>> --- a/drivers/gpu/drm/i915/gt/intel_gt.c
>>>> +++ b/drivers/gpu/drm/i915/gt/intel_gt.c
>>>> @@ -785,6 +785,7 @@ void intel_gt_driver_unregister(struct intel_gt 
>>>> *gt)
>>>>   {
>>>>       intel_wakeref_t wakeref;
>>>>   +    intel_gt_sysfs_unregister(gt);
>>>>       intel_rps_driver_unregister(&gt->rps);
>>>>       intel_gsc_fini(&gt->gsc);
>>>>   diff --git a/drivers/gpu/drm/i915/gt/intel_gt_sysfs.c 
>>>> b/drivers/gpu/drm/i915/gt/intel_gt_sysfs.c
>>>> index 8ec8bc660c8c..9e4ebf53379b 100644
>>>> --- a/drivers/gpu/drm/i915/gt/intel_gt_sysfs.c
>>>> +++ b/drivers/gpu/drm/i915/gt/intel_gt_sysfs.c
>>>> @@ -24,7 +24,7 @@ bool is_object_gt(struct kobject *kobj)
>>>>     static struct intel_gt *kobj_to_gt(struct kobject *kobj)
>>>>   {
>>>> -    return container_of(kobj, struct kobj_gt, base)->gt;
>>>> +    return container_of(kobj, struct intel_gt, sysfs_gt);
>>>>   }
>>>>     struct intel_gt *intel_gt_sysfs_get_drvdata(struct device *dev,
>>>> @@ -72,9 +72,9 @@ static struct attribute *id_attrs[] = {
>>>>   };
>>>>   ATTRIBUTE_GROUPS(id);
>>>>   +/* A kobject needs a release() method even if it does nothing */
>>>>   static void kobj_gt_release(struct kobject *kobj)
>>>>   {
>>>> -    kfree(kobj);
>>>>   }
>>>>     static struct kobj_type kobj_gt_type = {
>>>> @@ -85,8 +85,6 @@ static struct kobj_type kobj_gt_type = {
>>>>     void intel_gt_sysfs_register(struct intel_gt *gt)
>>>>   {
>>>> -    struct kobj_gt *kg;
>>>> -
>>>>       /*
>>>>        * We need to make things right with the
>>>>        * ABI compatibility. The files were originally
>>>> @@ -98,25 +96,22 @@ void intel_gt_sysfs_register(struct intel_gt *gt)
>>>>       if (gt_is_root(gt))
>>>>           intel_gt_sysfs_pm_init(gt, gt_get_parent_obj(gt));
>>>>   -    kg = kzalloc(sizeof(*kg), GFP_KERNEL);
>>>> -    if (!kg)
>>>> +    /* init and xfer ownership to sysfs tree */
>>>> +    if (kobject_init_and_add(&gt->sysfs_gt, &kobj_gt_type,
>>>> +                 gt->i915->sysfs_gt, "gt%d", gt->info.id))
>>>
>>> Was there closure/agreement on the matter of whether or not there is 
>>> a potential race between "kfree(gt)" and sysfs access (last put from 
>>> sysfs that is)? I've noticed Andrzej and Ashutosh were discussing it 
>>> but did not read all the details.
>>>
>>
>> Not really :)
>> IMO docs are against this practice, Ashutosh shows examples of this 
>> practice in code and according to his analysis it is safe.
>> I gave up looking for contradictions :) Either it is OK, kobject is 
>> not fully shared object, docs are obsolete and needs update, either 
>> the patch is wrong.
>> Anyway finally I tend to accept this solution, I failed to prove it 
>> is wrong :)
>
> Like a question of whether hotunplug can be triggered while userspace 
> is sitting in a sysfs hook? Final kfree then has to be delayed until 
> userspace exists.
>
> Btw where is the "kfree(gt)" for the tiles on the PCI remove path? I 
> can't find it.. Do we have a leak?

intel_gt_tile_cleanup ?

>
> Regards,
>
> Tvrtko