Swap limit

[Mario Kleiner]

On Dec 13, 2010, at 11:21 AM, Michel Dänzer wrote:

> On Mon, 2010-12-13 at 11:35 +0200, Pauli Nieminen wrote:
>> On 12/12/10 19:49 -0500, ext Jerome Glisse wrote:
>>> On Fri, Dec 10, 2010 at 9:40 AM, Mario Kleiner
>>> <mario.kleiner at tuebingen.mpg.de> wrote:
>>>> On 12/08/2010 05:15 PM, Pauli Nieminen wrote:
>>>>>
>>>>> On 08/12/10 16:55 +0100, ext Alex Deucher wrote:
>>>>>>
>>>>>> One other thing that might be worth adding to DRI2 is a way  
>>>>>> for the
>>>>>> driver to access the swap interval.  If we could, then the driver
>>>>>> could dynamically disable things like vline waits for buffer  
>>>>>> blits or
>>>>>> do non-vsynced pageflipping more easily if the swap interval  
>>>>>> was 0.
>>>>>>
>>>>>
>>>>> Actually I tough DRI2 was already telling SwapInteval with
>>>>>
>>>>> *swap_target = pPriv->last_swap_target + pPriv->swap_interval;
>>>>>
>>>>> If swap_target doesn't advance then driver would know to flip  
>>>>> as fast as
>>>>> possible.
>>>>>
>>>>> But then I noticed a bug.
>>>>>
>>>>> First one is that we always blit with swapinterval is zero when  
>>>>> it should
>>>>> be
>>>>> possible to flip (even tear free).
>>>>>
>>>>
>>>> Let it tear :). I think swap_interval == 0 should cause  
>>>> immediate swaps with
>>>> tearing. All OpenGL implementations i know (Windows, MacOS  
>>>> classic, MacOS/X,
>>>> Linux with proprietary drivers, old SGI's) interpret a  
>>>> swapinterval of zero
>>>> as "swap as soon as rendering is finished, immediately, don't  
>>>> sync to
>>>> scanout cycle" and i think that makes the most sense. It's good for
>>>> benchmarking how fast your system can go if not throttled by the  
>>>> monitor
>>>> (useful), for crazy gamers that trade visual quality for fps and  
>>>> for special
>>>> applications that need to sometimes control swap timing  
>>>> themselves (like my
>>>> toolkit).
>>>>
>>>>> Code from DRI2ScheduleSwap:
>>>>> /* Old DDX or no swap interval, just blit */
>>>>> if (!ds->ScheduleSwap || !pPriv->swap_interval) {
>>>>>
>>>>
>>>> Not a "bug", just a half-done way of achieving the expected  
>>>> swap_interval
>>>> zero ;-) [ok, maybe that is a bug].
>>>>
>>>> It schedules an immediate copy-swap via blitting. Unfortunately  
>>>> the ddx
>>>> doesn't know about the swap_interval, so it still synchronizes  
>>>> the execution
>>>> of the blit to vsync via vline waits. That's tear-free, but it  
>>>> depends on
>>>> the location and size of the drawable and the current position  
>>>> of the
>>>> scanout if this will cause an immediate swap (if scanout is  
>>>> outside the
>>>> drawables area) or a vsync'ed swap. It's a bit undefined  
>>>> behaviour for
>>>> non-fullscreen drawables and it effectively enforces a minimum  
>>>> swap interval
>>>> of 1 for fullscreen drawables, which is not what we want.
>>>>
>>>> Both the intel ddx and (soon) the ati ddx have a xorg.conf  
>>>> parameter
>>>> "SwapBuffersWait" to disable vsync completely, then you'd get  
>>>> copy-swaps as
>>>> fast as possible for swap_interval zero, but for a non-zero  
>>>> swap_interval
>>>> you'd have some chance of tearing, so this is alos just a band-aid.
>>>>
>>>> If we wanted this for page-flipping we'd need a shortcut similar  
>>>> to this one
>>>> which would bypass the vblank scheduling and call the ddx  
>>>> pageflip routine
>>>> directly, plus an extension to the kernel's pageflip ioctl() to  
>>>> allow
>>>> non-vsynced flips.
>>>>
>>>> And we need a new interface to tell the ddx that a swap should be
>>>> non-vsync'ed.
>>>>
>>>
>>> Unless i miss read hw spec all pageflip are always vsync no matter
>>> what you do. IIRC the crtc sancout address is updated during vblank
>>> from the register, so what is scanned next is what ever address was
>>> written last.
>>>
>>
>> If I remember correctly RADEON_CRTC_OFFSET_FLIP_CNTL had some  
>> effect on that.
>
> Correct, if that bit is 1 the flip happens on the next horizontal  
> blank
> (the non-KMS radeon DRM always sets this bit for flipping). There's
> probably something similar for the AVIVO display engine.
>

On Radeons you can select if flip should happen at next vblank or  
hblank. Alex's initial pageflip patches for Radeon used this even for  
vblank sync'd flips, so we know it works, but we changed to a  
different method for higher overall robustness. I assume other gpu's  
aren't different, as on os/x and windows, all gpu's will swap  
immediately (with tearing) when you set the swap_interval to zero.

Adding this to the implementation wouldn't be difficult (at least for  
radeon's), it's just that the current pageflip ioctl() doesn't  
support an option to pass this to the kms-driver. Same with the ddx.  
Currently there isn't a specific interface for the x-server to  
communicate vsync vs. non-vsync to the ddx.

I have to read the other e-mails in this thread first, esp. Pauli's,  
i'm a bit slow at the moment.
-mario