diff mbox series

[3/3] media: staging: rkisp1: params: in 'stop_streaming' don't release the lock while returning buffers

Message ID 20200625174257.22216-4-dafna.hirschfeld@collabora.com (mailing list archive)
State New, archived
Headers show
Series media: staging: rkisp1: various bug fixes in params | expand

Commit Message

Dafna Hirschfeld June 25, 2020, 5:42 p.m. UTC
In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
there is no need to release the lock 'config_lock' and then acquire
it again at each iteration when returning all buffers.
This is because the stream is about to end and there is no need
to let the isr access a buffer.

Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
---
 drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

Comments

Robin Murphy June 26, 2020, 1:32 p.m. UTC | #1
Hi Dafna,

On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> there is no need to release the lock 'config_lock' and then acquire
> it again at each iteration when returning all buffers.
> This is because the stream is about to end and there is no need
> to let the isr access a buffer.
> 
> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> ---
>   drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
>   1 file changed, 1 insertion(+), 6 deletions(-)
> 
> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> index bf006dbeee2d..5169b02731f1 100644
> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
>   	/* stop params input firstly */
>   	spin_lock_irqsave(&params->config_lock, flags);
>   	params->is_streaming = false;
> -	spin_unlock_irqrestore(&params->config_lock, flags);
>   
>   	for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> -		spin_lock_irqsave(&params->config_lock, flags);
>   		if (!list_empty(&params->params)) {
>   			buf = list_first_entry(&params->params,
>   					       struct rkisp1_buffer, queue);
>   			list_del(&buf->queue);
> -			spin_unlock_irqrestore(&params->config_lock,
> -					       flags);
>   		} else {
> -			spin_unlock_irqrestore(&params->config_lock,
> -					       flags);
>   			break;
>   		}

Just skimming through out of idle curiosity I was going to comment that 
if you end up with this pattern:

	if (!x) {
		//do stuff
	} else {
		break;
	}

it would be better as:

	if (x)
		break;
	//do stuff

However I then went and looked at the whole function and frankly it's a 
bit of a WTF. As best I could decipher, this whole crazy loop appears to 
be a baroque reinvention of:

	list_for_each_entry_safe(&params->params, ..., buf) {
		list_del(&buf->queue);
		vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
	}

(assuming from context that the list should never contain more than 
RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)

Robin.

>   
> @@ -1508,6 +1502,7 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
>   			vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
>   		buf = NULL;
>   	}
> +	spin_unlock_irqrestore(&params->config_lock, flags);
>   }
>   
>   static int
>
Tomasz Figa June 26, 2020, 2:03 p.m. UTC | #2
On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
>
> Hi Dafna,
>
> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> > In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> > there is no need to release the lock 'config_lock' and then acquire
> > it again at each iteration when returning all buffers.
> > This is because the stream is about to end and there is no need
> > to let the isr access a buffer.
> >
> > Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> > ---
> >   drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> >   1 file changed, 1 insertion(+), 6 deletions(-)
> >
> > diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> > index bf006dbeee2d..5169b02731f1 100644
> > --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> > +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> > @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> >       /* stop params input firstly */
> >       spin_lock_irqsave(&params->config_lock, flags);
> >       params->is_streaming = false;
> > -     spin_unlock_irqrestore(&params->config_lock, flags);
> >
> >       for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> > -             spin_lock_irqsave(&params->config_lock, flags);
> >               if (!list_empty(&params->params)) {
> >                       buf = list_first_entry(&params->params,
> >                                              struct rkisp1_buffer, queue);
> >                       list_del(&buf->queue);
> > -                     spin_unlock_irqrestore(&params->config_lock,
> > -                                            flags);
> >               } else {
> > -                     spin_unlock_irqrestore(&params->config_lock,
> > -                                            flags);
> >                       break;
> >               }
>
> Just skimming through out of idle curiosity I was going to comment that
> if you end up with this pattern:
>
>         if (!x) {
>                 //do stuff
>         } else {
>                 break;
>         }
>
> it would be better as:
>
>         if (x)
>                 break;
>         //do stuff
>
> However I then went and looked at the whole function and frankly it's a
> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> be a baroque reinvention of:
>
>         list_for_each_entry_safe(&params->params, ..., buf) {
>                 list_del(&buf->queue);
>                 vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
>         }
>
> (assuming from context that the list should never contain more than
> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)

Or if we want to avoid disabling the interrupts for the whole
iteration, we could use list_splice() to move all the entries of
params->params to a local list_head under the spinlock, release it and
then loop over the local head. As a side effect, one could even drop
list_del() and switch to the non-safe variant of
list_for_each_entry().

Best regards,
Tomasz
Dafna Hirschfeld June 26, 2020, 3:48 p.m. UTC | #3
On 26.06.20 16:03, Tomasz Figa wrote:
> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
>>
>> Hi Dafna,
>>
>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
>>> there is no need to release the lock 'config_lock' and then acquire
>>> it again at each iteration when returning all buffers.
>>> This is because the stream is about to end and there is no need
>>> to let the isr access a buffer.
>>>
>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
>>> ---
>>>    drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
>>>    1 file changed, 1 insertion(+), 6 deletions(-)
>>>
>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
>>> index bf006dbeee2d..5169b02731f1 100644
>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
>>>        /* stop params input firstly */
>>>        spin_lock_irqsave(&params->config_lock, flags);
>>>        params->is_streaming = false;
>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
>>>
>>>        for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
>>> -             spin_lock_irqsave(&params->config_lock, flags);
>>>                if (!list_empty(&params->params)) {
>>>                        buf = list_first_entry(&params->params,
>>>                                               struct rkisp1_buffer, queue);
>>>                        list_del(&buf->queue);
>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>> -                                            flags);
>>>                } else {
>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>> -                                            flags);
>>>                        break;
>>>                }
>>
>> Just skimming through out of idle curiosity I was going to comment that
>> if you end up with this pattern:
>>
>>          if (!x) {
>>                  //do stuff
>>          } else {
>>                  break;
>>          }
>>
>> it would be better as:
>>
>>          if (x)
>>                  break;
>>          //do stuff
>>
>> However I then went and looked at the whole function and frankly it's a
>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
>> be a baroque reinvention of:
>>
>>          list_for_each_entry_safe(&params->params, ..., buf) {
>>                  list_del(&buf->queue);
>>                  vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
>>          }
Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
I see that many drivers implement it this way.
thanks!

>>
>> (assuming from context that the list should never contain more than
>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> 
> Or if we want to avoid disabling the interrupts for the whole
> iteration, we could use list_splice() to move all the entries of

But this code runs when userspace asks to stop the streaming so I don't
think it is important at that stage to allow the interrupts.

Thanks,
Dafna

> params->params to a local list_head under the spinlock, release it and
> then loop over the local head. As a side effect, one could even drop
> list_del() and switch to the non-safe variant of
> list_for_each_entry().
> 
> Best regards,
> Tomasz
>
Tomasz Figa June 26, 2020, 3:59 p.m. UTC | #4
On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld
<dafna.hirschfeld@collabora.com> wrote:
>
>
>
> On 26.06.20 16:03, Tomasz Figa wrote:
> > On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
> >>
> >> Hi Dafna,
> >>
> >> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> >>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> >>> there is no need to release the lock 'config_lock' and then acquire
> >>> it again at each iteration when returning all buffers.
> >>> This is because the stream is about to end and there is no need
> >>> to let the isr access a buffer.
> >>>
> >>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> >>> ---
> >>>    drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> >>>    1 file changed, 1 insertion(+), 6 deletions(-)
> >>>
> >>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>> index bf006dbeee2d..5169b02731f1 100644
> >>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> >>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> >>>        /* stop params input firstly */
> >>>        spin_lock_irqsave(&params->config_lock, flags);
> >>>        params->is_streaming = false;
> >>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> >>>
> >>>        for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> >>> -             spin_lock_irqsave(&params->config_lock, flags);
> >>>                if (!list_empty(&params->params)) {
> >>>                        buf = list_first_entry(&params->params,
> >>>                                               struct rkisp1_buffer, queue);
> >>>                        list_del(&buf->queue);
> >>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>> -                                            flags);
> >>>                } else {
> >>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>> -                                            flags);
> >>>                        break;
> >>>                }
> >>
> >> Just skimming through out of idle curiosity I was going to comment that
> >> if you end up with this pattern:
> >>
> >>          if (!x) {
> >>                  //do stuff
> >>          } else {
> >>                  break;
> >>          }
> >>
> >> it would be better as:
> >>
> >>          if (x)
> >>                  break;
> >>          //do stuff
> >>
> >> However I then went and looked at the whole function and frankly it's a
> >> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> >> be a baroque reinvention of:
> >>
> >>          list_for_each_entry_safe(&params->params, ..., buf) {
> >>                  list_del(&buf->queue);
> >>                  vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> >>          }
> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
> I see that many drivers implement it this way.
> thanks!
>
> >>
> >> (assuming from context that the list should never contain more than
> >> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> >
> > Or if we want to avoid disabling the interrupts for the whole
> > iteration, we could use list_splice() to move all the entries of
>
> But this code runs when userspace asks to stop the streaming so I don't
> think it is important at that stage to allow the interrupts.

It's generally a good practice to reduce the time spent with
interrupts disabled. Disabling the interrupts prevents the system from
handling external events, including timer interrupts, and scheduling
higher priority tasks, including real time ones. How much the system
runs with interrupts disabled is one of the factors determining the
general system latency.

Best regards,
Tomasz
Robin Murphy June 26, 2020, 4:58 p.m. UTC | #5
On 2020-06-26 16:59, Tomasz Figa wrote:
> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld
> <dafna.hirschfeld@collabora.com> wrote:
>>
>>
>>
>> On 26.06.20 16:03, Tomasz Figa wrote:
>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
>>>>
>>>> Hi Dafna,
>>>>
>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
>>>>> there is no need to release the lock 'config_lock' and then acquire
>>>>> it again at each iteration when returning all buffers.
>>>>> This is because the stream is about to end and there is no need
>>>>> to let the isr access a buffer.
>>>>>
>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
>>>>> ---
>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
>>>>>
>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
>>>>> index bf006dbeee2d..5169b02731f1 100644
>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
>>>>>         /* stop params input firstly */
>>>>>         spin_lock_irqsave(&params->config_lock, flags);
>>>>>         params->is_streaming = false;
>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
>>>>>
>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
>>>>>                 if (!list_empty(&params->params)) {
>>>>>                         buf = list_first_entry(&params->params,
>>>>>                                                struct rkisp1_buffer, queue);
>>>>>                         list_del(&buf->queue);
>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>>>> -                                            flags);
>>>>>                 } else {
>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>>>> -                                            flags);
>>>>>                         break;
>>>>>                 }
>>>>
>>>> Just skimming through out of idle curiosity I was going to comment that
>>>> if you end up with this pattern:
>>>>
>>>>           if (!x) {
>>>>                   //do stuff
>>>>           } else {
>>>>                   break;
>>>>           }
>>>>
>>>> it would be better as:
>>>>
>>>>           if (x)
>>>>                   break;
>>>>           //do stuff
>>>>
>>>> However I then went and looked at the whole function and frankly it's a
>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
>>>> be a baroque reinvention of:
>>>>
>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
>>>>                   list_del(&buf->queue);
>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
>>>>           }
>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
>> I see that many drivers implement it this way.
>> thanks!
>>
>>>>
>>>> (assuming from context that the list should never contain more than
>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
>>>
>>> Or if we want to avoid disabling the interrupts for the whole
>>> iteration, we could use list_splice() to move all the entries of
>>
>> But this code runs when userspace asks to stop the streaming so I don't
>> think it is important at that stage to allow the interrupts.
> 
> It's generally a good practice to reduce the time spent with
> interrupts disabled. Disabling the interrupts prevents the system from
> handling external events, including timer interrupts, and scheduling
> higher priority tasks, including real time ones. How much the system
> runs with interrupts disabled is one of the factors determining the
> general system latency.

Right, with the way we handle interrupt affinity on Arm an IRQ can't 
target multiple CPUs in hardware, so any time spent with IRQs disabled 
might be preventing other devices' interrupts from being taken even if 
they're not explicitly affine to the current CPU.

Now that I've looked, it appears that vb2_buffer_done() might end up 
performing a DMA sync on the buffers, which, if it has to do 
order-of-megabytes worth of cache maintenance for large frames, is the 
kind of relatively slow operation that really doesn't want to be done 
with IRQs disabled (or under a lock at all, ideally) unless it 
absolutely *has* to be. If the lock is only needed here to protect 
modifications to the params list itself, then moving the whole list at 
once to do the cleanup "offline" sounds like a great idea to me.

Robin.
Dafna Hirschfeld Aug. 13, 2020, 10:44 a.m. UTC | #6
Am 26.06.20 um 18:58 schrieb Robin Murphy:
> On 2020-06-26 16:59, Tomasz Figa wrote:
>> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld
>> <dafna.hirschfeld@collabora.com> wrote:
>>>
>>>
>>>
>>> On 26.06.20 16:03, Tomasz Figa wrote:
>>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
>>>>>
>>>>> Hi Dafna,
>>>>>
>>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
>>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
>>>>>> there is no need to release the lock 'config_lock' and then acquire
>>>>>> it again at each iteration when returning all buffers.
>>>>>> This is because the stream is about to end and there is no need
>>>>>> to let the isr access a buffer.
>>>>>>
>>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
>>>>>> ---
>>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
>>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
>>>>>> index bf006dbeee2d..5169b02731f1 100644
>>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
>>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
>>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
>>>>>>         /* stop params input firstly */
>>>>>>         spin_lock_irqsave(&params->config_lock, flags);
>>>>>>         params->is_streaming = false;
>>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
>>>>>>
>>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
>>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
>>>>>>                 if (!list_empty(&params->params)) {
>>>>>>                         buf = list_first_entry(&params->params,
>>>>>>                                                struct rkisp1_buffer, queue);
>>>>>>                         list_del(&buf->queue);
>>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>>>>> -                                            flags);
>>>>>>                 } else {
>>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>>>>> -                                            flags);
>>>>>>                         break;
>>>>>>                 }
>>>>>
>>>>> Just skimming through out of idle curiosity I was going to comment that
>>>>> if you end up with this pattern:
>>>>>
>>>>>           if (!x) {
>>>>>                   //do stuff
>>>>>           } else {
>>>>>                   break;
>>>>>           }
>>>>>
>>>>> it would be better as:
>>>>>
>>>>>           if (x)
>>>>>                   break;
>>>>>           //do stuff
>>>>>
>>>>> However I then went and looked at the whole function and frankly it's a
>>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
>>>>> be a baroque reinvention of:
>>>>>
>>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
>>>>>                   list_del(&buf->queue);
>>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
>>>>>           }
>>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
>>> I see that many drivers implement it this way.
>>> thanks!
>>>
>>>>>
>>>>> (assuming from context that the list should never contain more than
>>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
>>>>
>>>> Or if we want to avoid disabling the interrupts for the whole
>>>> iteration, we could use list_splice() to move all the entries of
>>>
>>> But this code runs when userspace asks to stop the streaming so I don't
>>> think it is important at that stage to allow the interrupts.
>>
>> It's generally a good practice to reduce the time spent with
>> interrupts disabled. Disabling the interrupts prevents the system from
>> handling external events, including timer interrupts, and scheduling
>> higher priority tasks, including real time ones. How much the system
>> runs with interrupts disabled is one of the factors determining the
>> general system latency.
> 
> Right, with the way we handle interrupt affinity on Arm an IRQ can't target multiple CPUs in hardware, so any time spent with IRQs disabled might be preventing other devices' interrupts from being taken even if they're not explicitly affine to the current CPU.
> 
> Now that I've looked, it appears that vb2_buffer_done() might end up performing a DMA sync on the buffers, which, if it has to do order-of-megabytes worth of cache maintenance for large frames, is the kind of relatively slow operation that really doesn't want to be done with IRQs disabled (or under a lock at all, ideally) unless it absolutely *has* to be. If the lock is only needed here to protect modifications to the params list itself, then moving the whole list at once to do the cleanup "offline" sounds like a great idea to me.

ok, that might be a problem in v4l2 in general since vb2_buffer_done is actually often used inside an irq handler

> 
> Robin.
Laurent Pinchart Aug. 13, 2020, 10:53 a.m. UTC | #7
Hello,

On Thu, Aug 13, 2020 at 12:44:35PM +0200, Dafna Hirschfeld wrote:
> Am 26.06.20 um 18:58 schrieb Robin Murphy:
> > On 2020-06-26 16:59, Tomasz Figa wrote:
> >> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld wrote:
> >>> On 26.06.20 16:03, Tomasz Figa wrote:
> >>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy wrote:
> >>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> >>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> >>>>>> there is no need to release the lock 'config_lock' and then acquire
> >>>>>> it again at each iteration when returning all buffers.
> >>>>>> This is because the stream is about to end and there is no need
> >>>>>> to let the isr access a buffer.
> >>>>>>
> >>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> >>>>>> ---
> >>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> >>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
> >>>>>>
> >>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>>>>> index bf006dbeee2d..5169b02731f1 100644
> >>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> >>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> >>>>>>         /* stop params input firstly */
> >>>>>>         spin_lock_irqsave(&params->config_lock, flags);
> >>>>>>         params->is_streaming = false;
> >>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> >>>>>>
> >>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> >>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
> >>>>>>                 if (!list_empty(&params->params)) {
> >>>>>>                         buf = list_first_entry(&params->params,
> >>>>>>                                                struct rkisp1_buffer, queue);
> >>>>>>                         list_del(&buf->queue);
> >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>>>>> -                                            flags);
> >>>>>>                 } else {
> >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>>>>> -                                            flags);
> >>>>>>                         break;
> >>>>>>                 }
> >>>>>
> >>>>> Just skimming through out of idle curiosity I was going to comment that
> >>>>> if you end up with this pattern:
> >>>>>
> >>>>>           if (!x) {
> >>>>>                   //do stuff
> >>>>>           } else {
> >>>>>                   break;
> >>>>>           }
> >>>>>
> >>>>> it would be better as:
> >>>>>
> >>>>>           if (x)
> >>>>>                   break;
> >>>>>           //do stuff
> >>>>>
> >>>>> However I then went and looked at the whole function and frankly it's a
> >>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> >>>>> be a baroque reinvention of:
> >>>>>
> >>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
> >>>>>                   list_del(&buf->queue);
> >>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> >>>>>           }
> >>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
> >>> I see that many drivers implement it this way.
> >>> thanks!
> >>>
> >>>>>
> >>>>> (assuming from context that the list should never contain more than
> >>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> >>>>
> >>>> Or if we want to avoid disabling the interrupts for the whole
> >>>> iteration, we could use list_splice() to move all the entries of
> >>>
> >>> But this code runs when userspace asks to stop the streaming so I don't
> >>> think it is important at that stage to allow the interrupts.
> >>
> >> It's generally a good practice to reduce the time spent with
> >> interrupts disabled. Disabling the interrupts prevents the system from
> >> handling external events, including timer interrupts, and scheduling
> >> higher priority tasks, including real time ones. How much the system
> >> runs with interrupts disabled is one of the factors determining the
> >> general system latency.
> > 
> > Right, with the way we handle interrupt affinity on Arm an IRQ can't
> > target multiple CPUs in hardware, so any time spent with IRQs
> > disabled might be preventing other devices' interrupts from being
> > taken even if they're not explicitly affine to the current CPU.
> > 
> > Now that I've looked, it appears that vb2_buffer_done() might end up
> > performing a DMA sync on the buffers, which, if it has to do
> > order-of-megabytes worth of cache maintenance for large frames, is
> > the kind of relatively slow operation that really doesn't want to be
> > done with IRQs disabled (or under a lock at all, ideally) unless it
> > absolutely *has* to be. If the lock is only needed here to protect
> > modifications to the params list itself, then moving the whole list
> > at once to do the cleanup "offline" sounds like a great idea to me.

Ouch.

> ok, that might be a problem in v4l2 in general since vb2_buffer_done
> is actually often used inside an irq handler

Correct. The DMA sync should be moved to DQBUF time, there shouldn't be
any reason to do it in the IRQ handler. I thought this had already been
fixed :-(
Tomasz Figa Aug. 13, 2020, 12:50 p.m. UTC | #8
On Thu, Aug 13, 2020 at 12:53 PM Laurent Pinchart
<laurent.pinchart@ideasonboard.com> wrote:
>
> Hello,
>
> On Thu, Aug 13, 2020 at 12:44:35PM +0200, Dafna Hirschfeld wrote:
> > Am 26.06.20 um 18:58 schrieb Robin Murphy:
> > > On 2020-06-26 16:59, Tomasz Figa wrote:
> > >> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld wrote:
> > >>> On 26.06.20 16:03, Tomasz Figa wrote:
> > >>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy wrote:
> > >>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> > >>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> > >>>>>> there is no need to release the lock 'config_lock' and then acquire
> > >>>>>> it again at each iteration when returning all buffers.
> > >>>>>> This is because the stream is about to end and there is no need
> > >>>>>> to let the isr access a buffer.
> > >>>>>>
> > >>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> > >>>>>> ---
> > >>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> > >>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
> > >>>>>>
> > >>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> > >>>>>> index bf006dbeee2d..5169b02731f1 100644
> > >>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> > >>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> > >>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> > >>>>>>         /* stop params input firstly */
> > >>>>>>         spin_lock_irqsave(&params->config_lock, flags);
> > >>>>>>         params->is_streaming = false;
> > >>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> > >>>>>>
> > >>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> > >>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
> > >>>>>>                 if (!list_empty(&params->params)) {
> > >>>>>>                         buf = list_first_entry(&params->params,
> > >>>>>>                                                struct rkisp1_buffer, queue);
> > >>>>>>                         list_del(&buf->queue);
> > >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> > >>>>>> -                                            flags);
> > >>>>>>                 } else {
> > >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> > >>>>>> -                                            flags);
> > >>>>>>                         break;
> > >>>>>>                 }
> > >>>>>
> > >>>>> Just skimming through out of idle curiosity I was going to comment that
> > >>>>> if you end up with this pattern:
> > >>>>>
> > >>>>>           if (!x) {
> > >>>>>                   //do stuff
> > >>>>>           } else {
> > >>>>>                   break;
> > >>>>>           }
> > >>>>>
> > >>>>> it would be better as:
> > >>>>>
> > >>>>>           if (x)
> > >>>>>                   break;
> > >>>>>           //do stuff
> > >>>>>
> > >>>>> However I then went and looked at the whole function and frankly it's a
> > >>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> > >>>>> be a baroque reinvention of:
> > >>>>>
> > >>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
> > >>>>>                   list_del(&buf->queue);
> > >>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> > >>>>>           }
> > >>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
> > >>> I see that many drivers implement it this way.
> > >>> thanks!
> > >>>
> > >>>>>
> > >>>>> (assuming from context that the list should never contain more than
> > >>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> > >>>>
> > >>>> Or if we want to avoid disabling the interrupts for the whole
> > >>>> iteration, we could use list_splice() to move all the entries of
> > >>>
> > >>> But this code runs when userspace asks to stop the streaming so I don't
> > >>> think it is important at that stage to allow the interrupts.
> > >>
> > >> It's generally a good practice to reduce the time spent with
> > >> interrupts disabled. Disabling the interrupts prevents the system from
> > >> handling external events, including timer interrupts, and scheduling
> > >> higher priority tasks, including real time ones. How much the system
> > >> runs with interrupts disabled is one of the factors determining the
> > >> general system latency.
> > >
> > > Right, with the way we handle interrupt affinity on Arm an IRQ can't
> > > target multiple CPUs in hardware, so any time spent with IRQs
> > > disabled might be preventing other devices' interrupts from being
> > > taken even if they're not explicitly affine to the current CPU.
> > >
> > > Now that I've looked, it appears that vb2_buffer_done() might end up
> > > performing a DMA sync on the buffers, which, if it has to do
> > > order-of-megabytes worth of cache maintenance for large frames, is
> > > the kind of relatively slow operation that really doesn't want to be
> > > done with IRQs disabled (or under a lock at all, ideally) unless it
> > > absolutely *has* to be. If the lock is only needed here to protect
> > > modifications to the params list itself, then moving the whole list
> > > at once to do the cleanup "offline" sounds like a great idea to me.
>
> Ouch.
>
> > ok, that might be a problem in v4l2 in general since vb2_buffer_done
> > is actually often used inside an irq handler
>
> Correct. The DMA sync should be moved to DQBUF time, there shouldn't be
> any reason to do it in the IRQ handler. I thought this had already been
> fixed :-(

For reference, there was a patch [1] proposed, but it moved the
synchronization to a wrong place in the sequence, already after the
.buf_finish queue callback, ending up breaking the drivers which need
to access the buffer contents there.

[1] https://patchwork.linuxtv.org/project/linux-media/patch/1494255810-12672-4-git-send-email-sakari.ailus@linux.intel.com/

Best regards,
Tomasz
Laurent Pinchart Aug. 13, 2020, 1:02 p.m. UTC | #9
Hi Tomasz,

On Thu, Aug 13, 2020 at 02:50:17PM +0200, Tomasz Figa wrote:
> On Thu, Aug 13, 2020 at 12:53 PM Laurent Pinchart wrote:
> > On Thu, Aug 13, 2020 at 12:44:35PM +0200, Dafna Hirschfeld wrote:
> > > Am 26.06.20 um 18:58 schrieb Robin Murphy:
> > > > On 2020-06-26 16:59, Tomasz Figa wrote:
> > > >> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld wrote:
> > > >>> On 26.06.20 16:03, Tomasz Figa wrote:
> > > >>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy wrote:
> > > >>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> > > >>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> > > >>>>>> there is no need to release the lock 'config_lock' and then acquire
> > > >>>>>> it again at each iteration when returning all buffers.
> > > >>>>>> This is because the stream is about to end and there is no need
> > > >>>>>> to let the isr access a buffer.
> > > >>>>>>
> > > >>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> > > >>>>>> ---
> > > >>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> > > >>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
> > > >>>>>>
> > > >>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> > > >>>>>> index bf006dbeee2d..5169b02731f1 100644
> > > >>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> > > >>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> > > >>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> > > >>>>>>         /* stop params input firstly */
> > > >>>>>>         spin_lock_irqsave(&params->config_lock, flags);
> > > >>>>>>         params->is_streaming = false;
> > > >>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> > > >>>>>>
> > > >>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> > > >>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
> > > >>>>>>                 if (!list_empty(&params->params)) {
> > > >>>>>>                         buf = list_first_entry(&params->params,
> > > >>>>>>                                                struct rkisp1_buffer, queue);
> > > >>>>>>                         list_del(&buf->queue);
> > > >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> > > >>>>>> -                                            flags);
> > > >>>>>>                 } else {
> > > >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> > > >>>>>> -                                            flags);
> > > >>>>>>                         break;
> > > >>>>>>                 }
> > > >>>>>
> > > >>>>> Just skimming through out of idle curiosity I was going to comment that
> > > >>>>> if you end up with this pattern:
> > > >>>>>
> > > >>>>>           if (!x) {
> > > >>>>>                   //do stuff
> > > >>>>>           } else {
> > > >>>>>                   break;
> > > >>>>>           }
> > > >>>>>
> > > >>>>> it would be better as:
> > > >>>>>
> > > >>>>>           if (x)
> > > >>>>>                   break;
> > > >>>>>           //do stuff
> > > >>>>>
> > > >>>>> However I then went and looked at the whole function and frankly it's a
> > > >>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> > > >>>>> be a baroque reinvention of:
> > > >>>>>
> > > >>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
> > > >>>>>                   list_del(&buf->queue);
> > > >>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> > > >>>>>           }
> > > >>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
> > > >>> I see that many drivers implement it this way.
> > > >>> thanks!
> > > >>>
> > > >>>>>
> > > >>>>> (assuming from context that the list should never contain more than
> > > >>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> > > >>>>
> > > >>>> Or if we want to avoid disabling the interrupts for the whole
> > > >>>> iteration, we could use list_splice() to move all the entries of
> > > >>>
> > > >>> But this code runs when userspace asks to stop the streaming so I don't
> > > >>> think it is important at that stage to allow the interrupts.
> > > >>
> > > >> It's generally a good practice to reduce the time spent with
> > > >> interrupts disabled. Disabling the interrupts prevents the system from
> > > >> handling external events, including timer interrupts, and scheduling
> > > >> higher priority tasks, including real time ones. How much the system
> > > >> runs with interrupts disabled is one of the factors determining the
> > > >> general system latency.
> > > >
> > > > Right, with the way we handle interrupt affinity on Arm an IRQ can't
> > > > target multiple CPUs in hardware, so any time spent with IRQs
> > > > disabled might be preventing other devices' interrupts from being
> > > > taken even if they're not explicitly affine to the current CPU.
> > > >
> > > > Now that I've looked, it appears that vb2_buffer_done() might end up
> > > > performing a DMA sync on the buffers, which, if it has to do
> > > > order-of-megabytes worth of cache maintenance for large frames, is
> > > > the kind of relatively slow operation that really doesn't want to be
> > > > done with IRQs disabled (or under a lock at all, ideally) unless it
> > > > absolutely *has* to be. If the lock is only needed here to protect
> > > > modifications to the params list itself, then moving the whole list
> > > > at once to do the cleanup "offline" sounds like a great idea to me.
> >
> > Ouch.
> >
> > > ok, that might be a problem in v4l2 in general since vb2_buffer_done
> > > is actually often used inside an irq handler
> >
> > Correct. The DMA sync should be moved to DQBUF time, there shouldn't be
> > any reason to do it in the IRQ handler. I thought this had already been
> > fixed :-(
> 
> For reference, there was a patch [1] proposed, but it moved the
> synchronization to a wrong place in the sequence, already after the
> .buf_finish queue callback, ending up breaking the drivers which need
> to access the buffer contents there.
> 
> [1] https://patchwork.linuxtv.org/project/linux-media/patch/1494255810-12672-4-git-send-email-sakari.ailus@linux.intel.com/

I think we need to fix the drivers. We just can't do cache sync in IRQ
context by default because a few drivers need to access the buffer
contents. Those drivers should instead deffer access to a work queue,
and sync explicitly. We could possibly provide helpers for that, making
it transparent if a queue flag is set.
Tomasz Figa Aug. 13, 2020, 1:05 p.m. UTC | #10
On Thu, Aug 13, 2020 at 3:02 PM Laurent Pinchart
<laurent.pinchart@ideasonboard.com> wrote:
>
> Hi Tomasz,
>
> On Thu, Aug 13, 2020 at 02:50:17PM +0200, Tomasz Figa wrote:
> > On Thu, Aug 13, 2020 at 12:53 PM Laurent Pinchart wrote:
> > > On Thu, Aug 13, 2020 at 12:44:35PM +0200, Dafna Hirschfeld wrote:
> > > > Am 26.06.20 um 18:58 schrieb Robin Murphy:
> > > > > On 2020-06-26 16:59, Tomasz Figa wrote:
> > > > >> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld wrote:
> > > > >>> On 26.06.20 16:03, Tomasz Figa wrote:
> > > > >>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy wrote:
> > > > >>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> > > > >>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> > > > >>>>>> there is no need to release the lock 'config_lock' and then acquire
> > > > >>>>>> it again at each iteration when returning all buffers.
> > > > >>>>>> This is because the stream is about to end and there is no need
> > > > >>>>>> to let the isr access a buffer.
> > > > >>>>>>
> > > > >>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> > > > >>>>>> ---
> > > > >>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> > > > >>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
> > > > >>>>>>
> > > > >>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> > > > >>>>>> index bf006dbeee2d..5169b02731f1 100644
> > > > >>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> > > > >>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> > > > >>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> > > > >>>>>>         /* stop params input firstly */
> > > > >>>>>>         spin_lock_irqsave(&params->config_lock, flags);
> > > > >>>>>>         params->is_streaming = false;
> > > > >>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> > > > >>>>>>
> > > > >>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> > > > >>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
> > > > >>>>>>                 if (!list_empty(&params->params)) {
> > > > >>>>>>                         buf = list_first_entry(&params->params,
> > > > >>>>>>                                                struct rkisp1_buffer, queue);
> > > > >>>>>>                         list_del(&buf->queue);
> > > > >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> > > > >>>>>> -                                            flags);
> > > > >>>>>>                 } else {
> > > > >>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> > > > >>>>>> -                                            flags);
> > > > >>>>>>                         break;
> > > > >>>>>>                 }
> > > > >>>>>
> > > > >>>>> Just skimming through out of idle curiosity I was going to comment that
> > > > >>>>> if you end up with this pattern:
> > > > >>>>>
> > > > >>>>>           if (!x) {
> > > > >>>>>                   //do stuff
> > > > >>>>>           } else {
> > > > >>>>>                   break;
> > > > >>>>>           }
> > > > >>>>>
> > > > >>>>> it would be better as:
> > > > >>>>>
> > > > >>>>>           if (x)
> > > > >>>>>                   break;
> > > > >>>>>           //do stuff
> > > > >>>>>
> > > > >>>>> However I then went and looked at the whole function and frankly it's a
> > > > >>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> > > > >>>>> be a baroque reinvention of:
> > > > >>>>>
> > > > >>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
> > > > >>>>>                   list_del(&buf->queue);
> > > > >>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> > > > >>>>>           }
> > > > >>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
> > > > >>> I see that many drivers implement it this way.
> > > > >>> thanks!
> > > > >>>
> > > > >>>>>
> > > > >>>>> (assuming from context that the list should never contain more than
> > > > >>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> > > > >>>>
> > > > >>>> Or if we want to avoid disabling the interrupts for the whole
> > > > >>>> iteration, we could use list_splice() to move all the entries of
> > > > >>>
> > > > >>> But this code runs when userspace asks to stop the streaming so I don't
> > > > >>> think it is important at that stage to allow the interrupts.
> > > > >>
> > > > >> It's generally a good practice to reduce the time spent with
> > > > >> interrupts disabled. Disabling the interrupts prevents the system from
> > > > >> handling external events, including timer interrupts, and scheduling
> > > > >> higher priority tasks, including real time ones. How much the system
> > > > >> runs with interrupts disabled is one of the factors determining the
> > > > >> general system latency.
> > > > >
> > > > > Right, with the way we handle interrupt affinity on Arm an IRQ can't
> > > > > target multiple CPUs in hardware, so any time spent with IRQs
> > > > > disabled might be preventing other devices' interrupts from being
> > > > > taken even if they're not explicitly affine to the current CPU.
> > > > >
> > > > > Now that I've looked, it appears that vb2_buffer_done() might end up
> > > > > performing a DMA sync on the buffers, which, if it has to do
> > > > > order-of-megabytes worth of cache maintenance for large frames, is
> > > > > the kind of relatively slow operation that really doesn't want to be
> > > > > done with IRQs disabled (or under a lock at all, ideally) unless it
> > > > > absolutely *has* to be. If the lock is only needed here to protect
> > > > > modifications to the params list itself, then moving the whole list
> > > > > at once to do the cleanup "offline" sounds like a great idea to me.
> > >
> > > Ouch.
> > >
> > > > ok, that might be a problem in v4l2 in general since vb2_buffer_done
> > > > is actually often used inside an irq handler
> > >
> > > Correct. The DMA sync should be moved to DQBUF time, there shouldn't be
> > > any reason to do it in the IRQ handler. I thought this had already been
> > > fixed :-(
> >
> > For reference, there was a patch [1] proposed, but it moved the
> > synchronization to a wrong place in the sequence, already after the
> > .buf_finish queue callback, ending up breaking the drivers which need
> > to access the buffer contents there.
> >
> > [1] https://patchwork.linuxtv.org/project/linux-media/patch/1494255810-12672-4-git-send-email-sakari.ailus@linux.intel.com/
>
> I think we need to fix the drivers. We just can't do cache sync in IRQ
> context by default because a few drivers need to access the buffer
> contents. Those drivers should instead deffer access to a work queue,
> and sync explicitly. We could possibly provide helpers for that, making
> it transparent if a queue flag is set.

The drivers don't access the buffers explicitly from the IRQ. The vb2
queue .buf_finish callback is called at DQBUF time. It was just the
patch mentioned that moved it to a part of DQBUF executed too late.

Best regards,
Tomasz
Laurent Pinchart Aug. 13, 2020, 1:09 p.m. UTC | #11
Hi Tomasz,

On Thu, Aug 13, 2020 at 03:05:09PM +0200, Tomasz Figa wrote:
> On Thu, Aug 13, 2020 at 3:02 PM Laurent Pinchart wrote:
> > On Thu, Aug 13, 2020 at 02:50:17PM +0200, Tomasz Figa wrote:
> >> On Thu, Aug 13, 2020 at 12:53 PM Laurent Pinchart wrote:
> >>> On Thu, Aug 13, 2020 at 12:44:35PM +0200, Dafna Hirschfeld wrote:
> >>>> Am 26.06.20 um 18:58 schrieb Robin Murphy:
> >>>>> On 2020-06-26 16:59, Tomasz Figa wrote:
> >>>>>> On Fri, Jun 26, 2020 at 5:48 PM Dafna Hirschfeld wrote:
> >>>>>>> On 26.06.20 16:03, Tomasz Figa wrote:
> >>>>>>>> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy wrote:
> >>>>>>>>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> >>>>>>>>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> >>>>>>>>>> there is no need to release the lock 'config_lock' and then acquire
> >>>>>>>>>> it again at each iteration when returning all buffers.
> >>>>>>>>>> This is because the stream is about to end and there is no need
> >>>>>>>>>> to let the isr access a buffer.
> >>>>>>>>>>
> >>>>>>>>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> >>>>>>>>>> ---
> >>>>>>>>>>     drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> >>>>>>>>>>     1 file changed, 1 insertion(+), 6 deletions(-)
> >>>>>>>>>>
> >>>>>>>>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>>>>>>>>> index bf006dbeee2d..5169b02731f1 100644
> >>>>>>>>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> >>>>>>>>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>>>>>>>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> >>>>>>>>>>         /* stop params input firstly */
> >>>>>>>>>>         spin_lock_irqsave(&params->config_lock, flags);
> >>>>>>>>>>         params->is_streaming = false;
> >>>>>>>>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> >>>>>>>>>>
> >>>>>>>>>>         for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> >>>>>>>>>> -             spin_lock_irqsave(&params->config_lock, flags);
> >>>>>>>>>>                 if (!list_empty(&params->params)) {
> >>>>>>>>>>                         buf = list_first_entry(&params->params,
> >>>>>>>>>>                                                struct rkisp1_buffer, queue);
> >>>>>>>>>>                         list_del(&buf->queue);
> >>>>>>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>>>>>>>>> -                                            flags);
> >>>>>>>>>>                 } else {
> >>>>>>>>>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>>>>>>>>> -                                            flags);
> >>>>>>>>>>                         break;
> >>>>>>>>>>                 }
> >>>>>>>>>
> >>>>>>>>> Just skimming through out of idle curiosity I was going to comment that
> >>>>>>>>> if you end up with this pattern:
> >>>>>>>>>
> >>>>>>>>>           if (!x) {
> >>>>>>>>>                   //do stuff
> >>>>>>>>>           } else {
> >>>>>>>>>                   break;
> >>>>>>>>>           }
> >>>>>>>>>
> >>>>>>>>> it would be better as:
> >>>>>>>>>
> >>>>>>>>>           if (x)
> >>>>>>>>>                   break;
> >>>>>>>>>           //do stuff
> >>>>>>>>>
> >>>>>>>>> However I then went and looked at the whole function and frankly it's a
> >>>>>>>>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> >>>>>>>>> be a baroque reinvention of:
> >>>>>>>>>
> >>>>>>>>>           list_for_each_entry_safe(&params->params, ..., buf) {
> >>>>>>>>>                   list_del(&buf->queue);
> >>>>>>>>>                   vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> >>>>>>>>>           }
> >>>>>>> Hi, indeed this is a much simpler implementation, greping 'return_all_buffers'
> >>>>>>> I see that many drivers implement it this way.
> >>>>>>> thanks!
> >>>>>>>
> >>>>>>>>>
> >>>>>>>>> (assuming from context that the list should never contain more than
> >>>>>>>>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> >>>>>>>>
> >>>>>>>> Or if we want to avoid disabling the interrupts for the whole
> >>>>>>>> iteration, we could use list_splice() to move all the entries of
> >>>>>>>
> >>>>>>> But this code runs when userspace asks to stop the streaming so I don't
> >>>>>>> think it is important at that stage to allow the interrupts.
> >>>>>>
> >>>>>> It's generally a good practice to reduce the time spent with
> >>>>>> interrupts disabled. Disabling the interrupts prevents the system from
> >>>>>> handling external events, including timer interrupts, and scheduling
> >>>>>> higher priority tasks, including real time ones. How much the system
> >>>>>> runs with interrupts disabled is one of the factors determining the
> >>>>>> general system latency.
> >>>>>
> >>>>> Right, with the way we handle interrupt affinity on Arm an IRQ can't
> >>>>> target multiple CPUs in hardware, so any time spent with IRQs
> >>>>> disabled might be preventing other devices' interrupts from being
> >>>>> taken even if they're not explicitly affine to the current CPU.
> >>>>>
> >>>>> Now that I've looked, it appears that vb2_buffer_done() might end up
> >>>>> performing a DMA sync on the buffers, which, if it has to do
> >>>>> order-of-megabytes worth of cache maintenance for large frames, is
> >>>>> the kind of relatively slow operation that really doesn't want to be
> >>>>> done with IRQs disabled (or under a lock at all, ideally) unless it
> >>>>> absolutely *has* to be. If the lock is only needed here to protect
> >>>>> modifications to the params list itself, then moving the whole list
> >>>>> at once to do the cleanup "offline" sounds like a great idea to me.
> >>>
> >>> Ouch.
> >>>
> >>>> ok, that might be a problem in v4l2 in general since vb2_buffer_done
> >>>> is actually often used inside an irq handler
> >>>
> >>> Correct. The DMA sync should be moved to DQBUF time, there shouldn't be
> >>> any reason to do it in the IRQ handler. I thought this had already been
> >>> fixed :-(
> >>
> >> For reference, there was a patch [1] proposed, but it moved the
> >> synchronization to a wrong place in the sequence, already after the
> >> .buf_finish queue callback, ending up breaking the drivers which need
> >> to access the buffer contents there.
> >>
> >> [1] https://patchwork.linuxtv.org/project/linux-media/patch/1494255810-12672-4-git-send-email-sakari.ailus@linux.intel.com/
> >
> > I think we need to fix the drivers. We just can't do cache sync in IRQ
> > context by default because a few drivers need to access the buffer
> > contents. Those drivers should instead deffer access to a work queue,
> > and sync explicitly. We could possibly provide helpers for that, making
> > it transparent if a queue flag is set.
> 
> The drivers don't access the buffers explicitly from the IRQ. The vb2
> queue .buf_finish callback is called at DQBUF time. It was just the
> patch mentioned that moved it to a part of DQBUF executed too late.

Aahhh it's better than I thought :-) Shouldn't be too hard to fix then.
Thanks for refreshing my memory.
Dafna Hirschfeld Aug. 14, 2020, 11:04 a.m. UTC | #12
Am 26.06.20 um 16:03 schrieb Tomasz Figa:
> On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
>>
>> Hi Dafna,
>>
>> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
>>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
>>> there is no need to release the lock 'config_lock' and then acquire
>>> it again at each iteration when returning all buffers.
>>> This is because the stream is about to end and there is no need
>>> to let the isr access a buffer.
>>>
>>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
>>> ---
>>>    drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
>>>    1 file changed, 1 insertion(+), 6 deletions(-)
>>>
>>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
>>> index bf006dbeee2d..5169b02731f1 100644
>>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
>>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
>>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
>>>        /* stop params input firstly */
>>>        spin_lock_irqsave(&params->config_lock, flags);
>>>        params->is_streaming = false;
>>> -     spin_unlock_irqrestore(&params->config_lock, flags);
>>>
>>>        for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
>>> -             spin_lock_irqsave(&params->config_lock, flags);
>>>                if (!list_empty(&params->params)) {
>>>                        buf = list_first_entry(&params->params,
>>>                                               struct rkisp1_buffer, queue);
>>>                        list_del(&buf->queue);
>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>> -                                            flags);
>>>                } else {
>>> -                     spin_unlock_irqrestore(&params->config_lock,
>>> -                                            flags);
>>>                        break;
>>>                }
>>
>> Just skimming through out of idle curiosity I was going to comment that
>> if you end up with this pattern:
>>
>>          if (!x) {
>>                  //do stuff
>>          } else {
>>                  break;
>>          }
>>
>> it would be better as:
>>
>>          if (x)
>>                  break;
>>          //do stuff
>>
>> However I then went and looked at the whole function and frankly it's a
>> bit of a WTF. As best I could decipher, this whole crazy loop appears to
>> be a baroque reinvention of:
>>
>>          list_for_each_entry_safe(&params->params, ..., buf) {
>>                  list_del(&buf->queue);
>>                  vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
>>          }
>>
>> (assuming from context that the list should never contain more than
>> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> 
> Or if we want to avoid disabling the interrupts for the whole
> iteration, we could use list_splice() to move all the entries of

Hi, list_splice combines two lists together, I guess you meant list_cut_position
which cut a list into two

Thanks,
Dafna

> params->params to a local list_head under the spinlock, release it and
> then loop over the local head. As a side effect, one could even drop
> list_del() and switch to the non-safe variant of
> list_for_each_entry().
> 
> Best regards,
> Tomasz
>
Tomasz Figa Aug. 14, 2020, 11:23 a.m. UTC | #13
On Fri, Aug 14, 2020 at 1:04 PM Dafna Hirschfeld
<dafna.hirschfeld@collabora.com> wrote:
>
>
>
> Am 26.06.20 um 16:03 schrieb Tomasz Figa:
> > On Fri, Jun 26, 2020 at 3:32 PM Robin Murphy <robin.murphy@arm.com> wrote:
> >>
> >> Hi Dafna,
> >>
> >> On 2020-06-25 18:42, Dafna Hirschfeld wrote:
> >>> In the stop_streaming callback 'rkisp1_params_vb2_stop_streaming'
> >>> there is no need to release the lock 'config_lock' and then acquire
> >>> it again at each iteration when returning all buffers.
> >>> This is because the stream is about to end and there is no need
> >>> to let the isr access a buffer.
> >>>
> >>> Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
> >>> ---
> >>>    drivers/staging/media/rkisp1/rkisp1-params.c | 7 +------
> >>>    1 file changed, 1 insertion(+), 6 deletions(-)
> >>>
> >>> diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>> index bf006dbeee2d..5169b02731f1 100644
> >>> --- a/drivers/staging/media/rkisp1/rkisp1-params.c
> >>> +++ b/drivers/staging/media/rkisp1/rkisp1-params.c
> >>> @@ -1488,19 +1488,13 @@ static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
> >>>        /* stop params input firstly */
> >>>        spin_lock_irqsave(&params->config_lock, flags);
> >>>        params->is_streaming = false;
> >>> -     spin_unlock_irqrestore(&params->config_lock, flags);
> >>>
> >>>        for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
> >>> -             spin_lock_irqsave(&params->config_lock, flags);
> >>>                if (!list_empty(&params->params)) {
> >>>                        buf = list_first_entry(&params->params,
> >>>                                               struct rkisp1_buffer, queue);
> >>>                        list_del(&buf->queue);
> >>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>> -                                            flags);
> >>>                } else {
> >>> -                     spin_unlock_irqrestore(&params->config_lock,
> >>> -                                            flags);
> >>>                        break;
> >>>                }
> >>
> >> Just skimming through out of idle curiosity I was going to comment that
> >> if you end up with this pattern:
> >>
> >>          if (!x) {
> >>                  //do stuff
> >>          } else {
> >>                  break;
> >>          }
> >>
> >> it would be better as:
> >>
> >>          if (x)
> >>                  break;
> >>          //do stuff
> >>
> >> However I then went and looked at the whole function and frankly it's a
> >> bit of a WTF. As best I could decipher, this whole crazy loop appears to
> >> be a baroque reinvention of:
> >>
> >>          list_for_each_entry_safe(&params->params, ..., buf) {
> >>                  list_del(&buf->queue);
> >>                  vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> >>          }
> >>
> >> (assuming from context that the list should never contain more than
> >> RKISP1_ISP_PARAMS_REQ_BUFS_MAX entries in the first place)
> >
> > Or if we want to avoid disabling the interrupts for the whole
> > iteration, we could use list_splice() to move all the entries of
>
> Hi, list_splice combines two lists together, I guess you meant list_cut_position
> which cut a list into two

No, I meant list_splice() and it does the expected thing here. It adds
all the elements from one list (the list argument) to another list
(the head argument). Since an element can't be on two lists at the
same time, the first list is rendered invalid.

Best regards,
Tomasz
diff mbox series

Patch

diff --git a/drivers/staging/media/rkisp1/rkisp1-params.c b/drivers/staging/media/rkisp1/rkisp1-params.c
index bf006dbeee2d..5169b02731f1 100644
--- a/drivers/staging/media/rkisp1/rkisp1-params.c
+++ b/drivers/staging/media/rkisp1/rkisp1-params.c
@@ -1488,19 +1488,13 @@  static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
 	/* stop params input firstly */
 	spin_lock_irqsave(&params->config_lock, flags);
 	params->is_streaming = false;
-	spin_unlock_irqrestore(&params->config_lock, flags);
 
 	for (i = 0; i < RKISP1_ISP_PARAMS_REQ_BUFS_MAX; i++) {
-		spin_lock_irqsave(&params->config_lock, flags);
 		if (!list_empty(&params->params)) {
 			buf = list_first_entry(&params->params,
 					       struct rkisp1_buffer, queue);
 			list_del(&buf->queue);
-			spin_unlock_irqrestore(&params->config_lock,
-					       flags);
 		} else {
-			spin_unlock_irqrestore(&params->config_lock,
-					       flags);
 			break;
 		}
 
@@ -1508,6 +1502,7 @@  static void rkisp1_params_vb2_stop_streaming(struct vb2_queue *vq)
 			vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
 		buf = NULL;
 	}
+	spin_unlock_irqrestore(&params->config_lock, flags);
 }
 
 static int