Am 22.08.19 um 22:02 schrieb Dave Airlie:
On Fri, 23 Aug 2019 at 03:25, Thomas Zimmermann
> I was traveling and could reply earlier. Sorry for taking so long.
> Am 13.08.19 um 11:36 schrieb Feng Tang:
>> Hi Thomas,
>> On Mon, Aug 12, 2019 at 03:25:45PM +0800, Feng Tang wrote:
>>> Hi Thomas,
>>> On Fri, Aug 09, 2019 at 04:12:29PM +0800, Rong Chen wrote:
>>>>>> Actually we run the benchmark as a background process, do we need
>>>>>> disable the cursor and test again?
>>>>> There's a worker thread that updates the display from the shadow
>>>>> The blinking cursor periodically triggers the worker thread, but the
>>>>> actual update is just the size of one character.
>>>>> The point of the test without output is to see if the regression
>>>> >from the buffer update (i.e., the memcpy from shadow buffer to VRAM),
>>>> >from the worker thread. If the regression goes away after disabling
>>>>> blinking cursor, then the worker thread is the problem. If it
>>>>> goes away if there's simply no output from the test, the screen
>>>>> is the problem. On my machine I have to disable the blinking cursor,
>>>>> I think the worker causes the performance drop.
>>>> We disabled redirecting stdout/stderr to /dev/kmsg, and the regression
>>>> f1f8555dfb9 drm/bochs: Use shadow buffer for bochs framebuffer console
>>>> 90f479ae51a drm/mgag200: Replace struct mga_fbdev with generic
>>>> f1f8555dfb9a70a2 90f479ae51afa45efab97afdde testcase/testparams/testbox
>>>> ---------------- -------------------------- ---------------------------
>>>> %stddev change %stddev
>>>> \ | \
>>>> 43785 44481
>>>> 43785 44481 GEO-MEAN
>>> Till now, from Rong's tests:
>>> 1. Disabling cursor blinking doesn't cure the regression.
>>> 2. Disabling printint test results to console can workaround the
>>> Also if we set the perfer_shadown to 0, the regression is also
>> We also did some further break down for the time consumed by the
>> new code.
>> The drm_fb_helper_dirty_work() calls sequentially
>> 1. drm_client_buffer_vmap (290 us)
>> 2. drm_fb_helper_dirty_blit_real (19240 us)
>> 3. helper->fb->funcs->dirty() ---> NULL for mgag200 driver
>> 4. drm_client_buffer_vunmap (215 us)
> It's somewhat different to what I observed, but maybe I just couldn't
> reproduce the problem correctly.
>> The average run time is listed after the function names.
>> From it, we can see drm_fb_helper_dirty_blit_real() takes too long
>> time (about 20ms for each run). I guess this is the root cause
>> of this regression, as the original code doesn't use this dirty worker.
> True, the original code uses a temporary buffer, but updates the display
> My guess is that this could be a caching problem. The worker runs on a
> different CPU, which doesn't have the shadow buffer in cache.
>> As said in last email, setting the prefer_shadow to 0 can avoid
>> the regrssion. Could it be an option?
> Unfortunately not. Without the shadow buffer, the console's display
> buffer permanently resides in video memory. It consumes significant
> amount of that memory (say 8 MiB out of 16 MiB). That doesn't leave
> enough room for anything else.
> The best option is to not print to the console.
Wait a second, I thought the driver did an eviction on modeset of the
scanned out object, this was a deliberate design decision made when
writing those drivers, has this been removed in favour of gem and
generic code paths?
Yes. We added back this feature for testing in . It was only an
improvement of ~1% compared to the original report. I wouldn't mind
landing this patch set, but it probably doesn't make a difference either.
Graphics Driver Developer
SUSE Linux GmbH, Maxfeldstrasse 5, 90409 Nuernberg, Germany
GF: Felix Imendörffer, Mary Higgins, Sri Rasiah
HRB 21284 (AG Nürnberg)