On 2018-01-22 09:57 AM, Andrey Grodzovsky wrote:
> You guys just need to revert it on dal-dev branch and don't promote to amd-staging-drm-next.
>
K. I'll leave it out of amd-staging-drm-next and we'll revert on our internal branch.
Thanks for following up on this and good to hear we found a better solution.
Harry
>
> So NAK for the change.
>
>
>
> Thanks,
>
> Andrey
>
>
> On 01/22/2018 09:38 AM, Lipski, Mikita wrote:
>>
>> Hi Andrey,
>>
>>
>> I've checked to revert this change and use Alex's change on switching irq destruction sequence and it worked no problem.Â
>> You can reject this change if Alex's change is pulled in.
>>
>>
>> Thanks,
>>
>> Nick
>>
>> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>> *From:* Grodzovsky, Andrey
>> *Sent:* Friday, January 19, 2018 1:33:51 PM
>> *To:* Wentland, Harry; amd-gfx at lists.freedesktop.org
>> *Cc:* Lipski, Mikita
>> *Subject:* Re: [PATCH 15/24] drm/amd/display: Fix deadlock when flushing irq
>> Â
>> What this spin lock is protecting here ? Seems to me it's just a read of
>> an array element which is always there.
>>
>> Regarding subsequent remove_timer_handler and timer queue destruction it
>> seems to me to be obsolete code, I don't think DAL is still using the
>> timer queue,
>>
>> so seems to me everything related to it should be removed.
>>
>> Thanks,
>>
>> Andrey
>>
>>
>> On 01/18/2018 04:03 PM, Harry Wentland wrote:
>> > From: Mikita Lipski <mikita.lipski at amd.com>
>> >
>> > Lock irq table when reading a work in queue,
>> > unlock to flush the work, lock again till all tasks
>> > are cleared
>> >
>> > Signed-off-by: Mikita Lipski <mikita.lipski at amd.com>
>> > Reviewed-by: Harry Wentland <Harry.Wentland at amd.com>
>> > ---
>> >Â Â drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c | 5 +++--
>> >Â Â 1 file changed, 3 insertions(+), 2 deletions(-)
>> >
>> > diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
>> > index 1874b6cee6af..fb60c91a1bfe 100644
>> > --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
>> > +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
>> > @@ -400,14 +400,15 @@ void amdgpu_dm_irq_fini(struct amdgpu_device *adev)
>> >Â Â {
>> >Â Â Â Â Â Â Â int src;
>> >Â Â Â Â Â Â Â struct irq_list_head *lh;
>> > +Â Â Â Â unsigned long irq_table_flags;
>> >Â Â Â Â Â Â Â DRM_DEBUG_KMS("DM_IRQ: releasing resources.\n");
>> > -
>> >Â Â Â Â Â Â Â for (src = 0; src < DAL_IRQ_SOURCES_NUMBER; src++) {
>> > -
>> > +Â Â Â Â Â Â Â Â Â Â Â Â DM_IRQ_TABLE_LOCK(adev, irq_table_flags);
>> >Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â /* The handler was removed from the table,
>> >Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â * it means it is safe to flush all the 'work'
>> >Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â * (because no code can schedule a new one). */
>> >Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â lh = &adev->dm.irq_handler_list_low_tab[src];
>> > +Â Â Â Â Â Â Â Â Â Â Â Â DM_IRQ_TABLE_UNLOCK(adev, irq_table_flags);
>> >Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â flush_work(&lh->work);
>> >Â Â Â Â Â Â Â }
>> >Â Â
>>
>
