Commit graph

119 commits

Author SHA1 Message Date
He Junyan
55769a16c7 libs: encoder: H265: Enable Main 12 profile support.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/375>
2020-08-06 00:21:50 +08:00
He Junyan
6c76e81cb6 libs: encoder: Add a helper function to check the tile support.
Encoding by tiles separation now is a very common feature for all
relative new codecs, such as HEVC, AV1, and VP9. Just make this
check as a common helper function of the encoder base class.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/294>
2020-06-05 19:27:11 +08:00
He Junyan
8486f82d98 libs: encoder: h265: Add support for MAIN 4:2:2 10 profile.
Using YUY2 as the input of the encoder can generate main 4:2:2 bit
streams and using Y210 as the input of the encoder can generate main
4:2:2 10 bit streams.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/-/merge_requests/334>
2020-06-05 08:42:35 +00:00
He Junyan
026c01875c libs: encoder: h265: Support MAIN 4:4:4 10 profile.
Using Y410 as the input of the encoder can generate main_444_10 bit
streams.
2020-04-02 23:01:46 +08:00
He Junyan
80b6e006bc libs: encoder: fix an inexact trace info in chroma type check. 2020-04-02 15:23:05 +08:00
Haihao Xiang
f680a8cba1 libs: encoder: make sure format array is not NULL when returning TRUE
This fixed segfault when running the pipeline below with iHD driver
(commit efe5e9a) on ICL

gst-launch-1.0 videotestsrc ! vaapivp9enc tune=low-power ! vaapivp9dec ! \
fakesink
2020-03-31 15:03:30 +08:00
U. Artie Eoff
861b4cc4b6 libs: encoder: set VA HRD param before RC param
This is a workaround for intel-media-driver bug
https://github.com/intel/media-driver/issues/865

The driver will force the RC method to CBR for HEVCe
when it parses the HRD param.  Thus, any RC method
param submitted "prior" to the HRD param will be lost.
Therefore, VBR, ICQ and QVBR for HEVCe can't be
effectively enabled if the RC method param "precedes"
the HRD param.

To work around this issue, set the HRD param before
the RC method param so the driver will parse the RC
method param "after" the HRD param.

Afaict, other codecs in the driver (and other drivers)
do not appear to be dependent on the order of HRD and
RC param submission.
2020-03-19 12:09:03 -07:00
Víctor Manuel Jáquez Leal
96cdebf185 libs: remove crumbs of libva < 0.39
All these guarded code seem like leftovers of commit 920b1ec7a. This
patch completes that missing clean up.
2020-02-11 12:32:33 +01:00
Víctor Manuel Jáquez Leal
eda173b3b0 libs: encoder: unref formats array if none
The formats array is always created, in order to keep the logic and
to avoid broken caps, if this formats array doesn't contain any
elements, it has to be unref and the function should return NULL.
2020-01-19 10:37:02 +00:00
He Junyan
25422b6766 plugin: encode: Add H265 main-444 profile.
Expose the main-444 profile to h265enc caps, when the upstream
chooses to use VUYA as input, we choose main 4:4:4 profile to encode
the frames.
2020-01-17 11:02:44 +00:00
Víctor Manuel Jáquez Leal
1db3ce56a0 vaapiencode: DMABuf only if PRIME is available
Add DMABuf capsfeature in encoders' allowed sinkcaps only if PRIME
memory type is available in the VA surface attributes of codec
context.
2020-01-17 11:31:57 +01:00
Víctor Manuel Jáquez Leal
9da5196b5a Remove all FEI related
FEI encoders are not actively mantained neither tested, and it is
using infrastructure that is changing and FEI is stopping this
effort.

Also it is required to rethink how FEI can be used in GStreamer.
2020-01-15 07:30:15 +00:00
He Junyan
36fd4d5d8a plugin: encode: List all possible profiles to detect input formats.
The current get_profile just return one possible profile for the encode,
which is not enough.  For example, if we want to support HEVC 4:4:4
profile, the input of encode should be VYUA rather than NV12 in HEVC
main profile. So the command line:

gst-launch-1.0 videotestsrc num-buffers=200 ! capsfilter \
caps=video/x-raw,format=VUYA,width=800,height=600 ! vaapih265enc \
tune=low-power init-qp=30 ! fakesink

can not work because vaapih265enc just report NV12 in sink caps, we need
to specify the profile obviously like:

gst-launch-1.0 videotestsrc num-buffers=200 ! capsfilter \
caps=video/x-raw,format=VUYA,width=800,height=600 ! vaapih265enc \
tune=low-power init-qp=30 ! capsfilter caps=video/x-h265, \
profile=main-444 ! fakesink

The encode should have the ability to choose the profile based on input
format automatically. If the input video format is VUYA, the main-444
profile should be auto choosed.

We modify to let get_allowed_profiles of each encode sub class to return
an array of all supported profiles based on downstream's allowed caps, or
return NULL if no valid profiles specified by downstream.
If no allowed profiles found, all profiles which belong to the current
encoder's codec will be the candidates.
The function gst_vaapi_encoder_get_surface_attributes collects the surface's
attributes for that profile list we just get.

So for this case, both NV12 and VUYA should be returned.

TODO: some codec like VP9, need to implement the get_profile() function.
2020-01-14 11:36:18 +00:00
He Junyan
0a2a25981c libs: encoder: add a helper function to get all supported profiles 2020-01-14 11:36:17 +00:00
He Junyan
a6cf75e8c6 libs: encoder: get surfaces resolution the same time with formats.
We can get all the information about the video format at one shot
when we create the test context for getting the supported formats.

The current way to get the width and height ranges are inefficient,
since it calls the function gst_vaapi_profile_caps_append_encoder()
and it creates another temporal context to detect the resolution
information.

Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2020-01-02 18:01:45 +01:00
He Junyan
3398c64036 libs: encoder: set entrypoint based on tune automatically
Some profile, such as H265_MAIN_444 on new Intel platform, may only
support ENTRYPOINT_SLICE_ENCODE_LP entrypoint. This leads two
problems:

1. We need to specify the tune mode like `vaapih265enc tune=low-power`
   every time when we need to use this kind of profile. Or we can not
   create the encoder context successfully.

2. More seriously, we set the entrypoint to a fixed value in
   init_context_info() and so the create_test_context_config() can not
   create the test context for these profile and can not get the
   supported video formats, either.

We now change the entrypoint setting based on the tune option of the
encoder. If no tune property provided, we just choose the first
available entrypoint.
2019-12-29 17:58:36 +01:00
He Junyan
b50f06309a libs: encoder: set context info profile by encoder
Instead of init_context_info() setting the passed profile, it is
assumed that it has to be set by each encoder.

Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2019-12-29 17:58:36 +01:00
He Junyan
5ad67ea666 libs: encoder: Add NULL pointer check for context when finalize.
Context may be NULL if pipeline fail in early stage, and the
ensure_context will not be called. Need to add a pointer protection
for it.
2019-12-22 15:22:57 +08:00
He Junyan
a6289308ee libs: context: port to a plain C structure
The GstVaapiMiniObject is obsolete and we need to replace it. This
patch turns GstVaapiContext into a plain C structure with its own
reference counting mechanism.

Also this patch removes unused overlays attributes.

Signed-off-by: Víctor Manuel Jáquez Leal <vjaquez@igalia.com>
2019-12-20 05:01:07 +00:00
Víctor Manuel Jáquez Leal
7fb17b1a34 libs: encoder: remove gst_vaapi_encoder_{ref,unref}()
Since GstVaapiEncoder is a descendant of of GstObject, there is no
need to keep a custom ref()/unref() methods. This patch deletes them.
2019-09-23 19:52:20 +02:00
He Junyan
ea35de9914 libs: encoder: correct encoder's ref/unref function.
GstVaapiEncoder now is a standard gstobject and need to use
gst_object_ref/unref functions.
2019-09-24 01:03:02 +08:00
Víctor Manuel Jáquez Leal
f0d6263318 libs: encoders: use GST_PARAM_USER_SHIFT to define internal params
This patch makes use of GST_PARAM_USER_SHIFT to define the internal
param in encoders to decide which parameters to expose. Thus
gstreamer-vaapi will not interfere with any change in GStreamer in the
future.

Also, the internal symbol was change to
GST_VAAPI_PARAM_ENCODER_EXPOSURE to keep the namespacing.
2019-09-11 14:32:22 +02:00
He Junyan
adcb448276 libs: encoder: delete EncoderPropInfo related functions 2019-08-30 21:21:04 +08:00
He Junyan
25b6be1780 libs: encoder: delete encoder_set_property
We no longer need this obsolete set_property function now after
switch to standard gobject's property manner.
Also delete the old encoder's property enum in the header file.
2019-08-30 21:21:04 +08:00
He Junyan
69f7c95a29 libs: encoder: delete properties_get_default for base class 2019-08-30 21:21:04 +08:00
He Junyan
511a48da56 libs: encoder: delete the useless constructed func for encoder. 2019-08-30 21:21:04 +08:00
He Junyan
1256680486 gst: encode: enable new type of property mechanism. 2019-08-30 21:19:14 +08:00
He Junyan
bc2f8fd19e libs: encoder: add flags to all encoder properties.
G_PARAM_CONSTRUCT make all properties init correctly, we do not
need to init the properties manually.
G_PARAM_FLAG_VAAPI_ENCODER_EXPOSURE is a vaapi encoder specific
flag, means need to expose the property to according encode class.
2019-08-30 18:39:32 +08:00
He Junyan
beaf45cc6f libs: encoder: add properties and prop help functions
Add all common properties to encoder base class. rate-control and
tune are moved to sub class.
2019-08-30 13:30:04 +08:00
He Junyan
6978eae21a libs: encoder: delete useless gst_vaapi_encoder_new func.
GstVaapiEncoder is a abstract gobject and never be created directly.
2019-08-30 13:06:50 +08:00
He Junyan
c4a47f91ba lib: encoder: gstobjectfy all vaapi encoders.
Replace all gstvaapiobject in vaapi encoders with standard gstobject.
Let the gstobject common logic to handle all the init and finalize
works. But the property install/set/get still use the old way, need
to be improved later.
2019-08-30 13:05:50 +08:00
Víctor Manuel Jáquez Leal
3ba3966fc3 vaapiencode: enhance how the profile is defined
This code doesn't define the profile used by the internal encoder, but
it used to "predict" which is going to be used and to get the caps
restrictions.

Before the profile was predicted by checking the donwstream caps, but
sometimes they are not defined, setting an unknown profile. In order
to enhances this situation, the encoder asks to internal encoder if it
has one. If so, it is used.

To ask the internal encoder's profile a new accessor function was
added: gst_vaapi_encoder_get_profile()
2019-08-19 17:26:55 +00:00
U. Artie Eoff
ed7277a1b7 libs: encoder: increase bitrate prop max value
There are many profile levels that can support
more than 102400 kbps.  Thus, increase the max
allowed bitrate property value from 102400 kbps
to 2048000 kbps (same as msdk encoder plugins).
2019-06-05 08:30:10 +00:00
U. Artie Eoff
be496a66c5 libs: encoder: add target-percentage property
Allow users to set the target-percentage for
variable rate controls.  The default value is
70 (as hard-coded prior).

v2: minimum allowed value changed from 0 to 1
v3: target-percentage unchanged if CBR used

Resolves #129
2019-05-09 06:15:35 +00:00
He Junyan
a1ae75b500 libs: encoder: Enable trellis quantization method.
The advanced trellis algorithm is supported in VA driver. We add
its support as a property named "trellis" of encoder.
It only works for H264 now, should be more in future.
2019-05-09 05:54:09 +00:00
Víctor Manuel Jáquez Leal
2541a33d6c libs: encoder: continue if roi meta is NULL
Coverity scan bug:

If the function actually returns a null value, a null pointer
dereference will occur.

In gst_vaapi_encoder_ensure_param_roi_regions(): Return value of
function which returns null is dereferenced without checking
2019-05-03 08:39:03 +00:00
Víctor Manuel Jáquez Leal
f1877628bb libs: encoder: refactor to avoid code duplication
gst_vaapi_encoder_put_frame() and gst_vaapi_encoder_flush() duplicates
the same code segment where the coded buffer is created, the picture
encoded on it and pushed to the async queue.

The function gst_vaapi_encoder_encode_and_queue() refactor this.
2019-01-14 20:09:57 +01:00
Víctor Manuel Jáquez Leal
220016aa6c libs: encoder: h264/h265: flush pending ordered pictures
In order to flush the pending pictures, a new internal encoder vmethod
is used: get_pending_reordered()

This method follows an iterator pattern which will return the next
picture to encode and push.

The base encoder will call this function in a loop when flush() is called.

For now, only H.264 and H.265 encoders implement this flushing mechanism.
2019-01-14 20:09:57 +01:00
Wangfei
f50fb8748f libs: encoder: h264/h265: fix encode lose frame issue.
Instead of dropping all remain frames in reorder_frame_list during
flush, keep encoding.

https://gitlab.freedesktop.org/gstreamer/gstreamer-vaapi/issues/97
2019-01-14 19:08:29 +00:00
Víctor Manuel Jáquez Leal
6ccd5d6fb2 libs: display: remove gst_vaapi_display_ref()
Replace it with gst_object_ref()

https://bugzilla.gnome.org/show_bug.cgi?id=796470
2018-06-14 16:41:00 +02:00
Víctor Manuel Jáquez Leal
c49a17ed15 libs: encoder: reimplement ROI using meta
Check input buffers for ROI metas and pass them to VA. Also added a
new "default-roi-delta-qp" property in order to tell the encoder what
delta QP should be applied to ROI by default.

Enabled it for H264 and H265 encoders.

https://bugzilla.gnome.org/show_bug.cgi?id=768248
2018-02-27 17:51:02 -05:00
Víctor Manuel Jáquez Leal
d3110713b8 Revert "libs: encoder: add api gst_vaapi_encoder_add/del_roi"
This reverts commit 7a6f690340.

https://bugzilla.gnome.org/show_bug.cgi?id=768248
2018-02-27 17:51:02 -05:00
Matteo Valdina
02c7c1bc17 libs: encoder: add zero as valid value for periodic keyframe.
Enabled zero as valid value for keyframe-period property.

https://bugzilla.gnome.org/show_bug.cgi?id=793829
2018-02-26 10:04:29 -06:00
Víctor Manuel Jáquez Leal
d18d6a8bd0 libs: encoder: code-style fix 2018-02-22 08:24:12 -06:00
Matteo Valdina
83886ced66 libs: encoder: h264,h265: extend max periodic keyframe.
Increased max values of periodic key frame for h26x codecs.
This allow more fine tunning of encoder that in certian scenario
want higher periodic key frame.

For example: it doesn't want a key frame each 10 seconds but
each 120 seconds.

https://bugzilla.gnome.org/show_bug.cgi?id=786320
2018-02-19 22:34:53 +01:00
Hyunjun Ko
acfabb3561 libs: encoder: allow to set bitrate on runtime
In case of streaming, controlling bitrate dynamically for encoder might be
important to manage quality of the streaming.

This patch is to support such a scenario.

https://bugzilla.gnome.org/show_bug.cgi?id=786321
2017-10-20 13:22:51 +02:00
Víctor Manuel Jáquez Leal
e308b452d6 libs: encoder: change mbbrc from uint to enum
Instead of handling the macroblock bitrate control as a integer, this
patch changes it as a enum, which is more self documented in the
GStreamer elements.

https://bugzilla.gnome.org/show_bug.cgi?id=787855
2017-09-18 20:19:20 +02:00
Víctor Manuel Jáquez Leal
29cf49d56a libs: encoder: don't unref properties
This patch fixes a regression introduced in commit 148f867c, since the
props variable is set to object's member variable
encoder->properties. And it is set in the instance initialization,
thus it will not be leaked.

https://bugzilla.gnome.org/show_bug.cgi?id=787733
2017-09-15 18:34:49 +02:00
Hyunjun Ko
148f867c12 vaapiencode/libs: encoder: fix leaks of properties
https://bugzilla.gnome.org/show_bug.cgi?id=786321
2017-09-15 11:07:55 +02:00
Sreerenj Balachandran
94483de40f FEI: libs: Add virtual method for secondary context creation.
Add a new vitrual method ensure_secondary_context to the
base encoder which is only required for the FEI entrypoint, that too
only when user configures the ENC+PAK mode. ENC+PAK mode is not something
supported directly by libva or driver, but this can be enabled
from the middleware.

Original Author of this idea: Leilei Shang <leilei.shang@intel.com>

Signed-off-by: Leilei Shang <leilei.shang@intel.com>
Signed-off-by: xiaominc <xiaomin.chen@intel.com>
Signed-off-by: Sreerenj Balachandran <sreerenj.balachandran@intel.com>

https://bugzilla.gnome.org/show_bug.cgi?id=785712
https://bugzilla.gnome.org/show_bug.cgi?id=784667
2017-09-01 11:32:24 +02:00