Use client input buffer queuing rate for power
calculations for non-realtime sessions
Change-Id: I498422ba1ae9ae96b782a2096dc0cd0aa7b05798
Signed-off-by: Deepa Guthyappa Madivalara <quic_dmadival@quicinc.com>
Add support for AV1D Intra-Block Copy (IBC) feature. This requires a
dedicated internal buffer (HFI_BUFFER_IBC_AV1D), which was previously
part of LINE buffer. A new internal buffer type called
HFI_BUFFER_PARTIAL_DATA is introduced, and this buffer type holds
HFI_BUFFER_IBC_AV1D for AV1D. HFI_BUFFER_PARTIAL_DATA is mapped to
NON_SECURE_PIXEL context bank for non-secure and to SECURE_PIXEL
context bank for secure use case.
Change-Id: If7ae22495f9981f275d54acc342c25ccdfc0c7b9
Signed-off-by: Mihir Ganu <quic_mganu@quicinc.com>
- add support for prepare_buf
- check if client has configured invalid fd for metadata
buffer fd control.
Change-Id: I576d02be05e4692a5ab45286895c34284f08651c
Signed-off-by: Darshana Patil <darshana@quicinc.com>
- Introduce V4L2_EVENT_VIDC_METADATA to post fence fd returned in
ETB metadata done buffer.
- V4L2_CID_MPEG_VIDC_SW_FENCE_FD g_ctrl support for client to get
fence fd for requested fence id.
- Addition of fence delivery and subscribtion support.
- Fence property packetization to send to firmware via FTB buffer.
Change-Id: Ifb289849e352af2c4729aa95040bd83753979970
Signed-off-by: Akshata Sahukar <quic_asahukar@quicinc.com>
Client can enable V4L2_CID_MPEG_VIDC_INPUT_METADATA_VIA_REQUEST_ENABLE
control and send input metadata buffer via request.
Change-Id: Icccdada8253d1d0291c01e2e539968141f7ddd2c
Signed-off-by: Darshana Patil <quic_darshana@quicinc.com>
[1] Remove CAP_FLAG_ROOT flag. Root willbe identified
with has_parents() api.
[2] Replace field(struct msm_vidc_inst_cap) cap -> cap_id.
[3] Rename below pool apis.
msm_memory_alloc -> msm_memory_pool_alloc
msm_memory_free -> msm_memory_pool_free
[4] Remove struct msm_vidc_inst_cap_entry usage in inst.
inst->firmware.list -> inst->firmware_list
inst->children.list -> inst->children_list
[5] Organize msm_vidc_inst_capability_type enums in such
a way that all parents placed first then children. It
will save CPU cycles while preparing caps_list.
Change-Id: I376f4e6ddec37ce326c393fba133e19ba70bab09
Signed-off-by: Govindaraj Rajagopal <quic_grajagop@quicinc.com>
Added changes to call do msm_vidc_core_init() inside
msm_vidc_probe_video_device() itself. It will uniform
core init and deinit as part of insmod and rmmod.
follow core_init and core_init_wait for all new session
open sequence.
Change-Id: I081d5ff3632296b5d2086992e6f3388daedfd1d7
Signed-off-by: Govindaraj Rajagopal <quic_grajagop@quicinc.com>
Added support for timestamp reorder. It is default
disabled.
Change-Id: I51a6fc785860081b651fc67f443132cb2675b9fd
Signed-off-by: Govindaraj Rajagopal <quic_grajagop@quicinc.com>
-Removed unwanted functions
-Removed unwanted part of code from function definitions
-Added proper format specifier according to the data type.
Change-Id: I700202f1cf588506b329d202ce3e2729c027dcae
Signed-off-by: Gaviraju Doddabettahalli Bettegowda <quic_gdoddabe@quicinc.com>
Signed-off-by: Govindaraj Rajagopal <quic_grajagop@quicinc.com>
AV1 bitstream frames can have different resolutions than
sequence resolutions. Subscribe to BITSTREAM_RESOLUTION and
CROP_OFFSETS metadatas and exopose them to clients so that
clients can receive per-frame resolution and crop.
Change-Id: I7c0503f25ddde201979d13ff18028681f685d649
Signed-off-by: Mihir Ganu <quic_mganu@quicinc.com>
check max mbpf as well while allowing
the real time session.
Change-Id: I1091582a02e01a4daa9c58ec5b50400b7d856610
Signed-off-by: Manikanta Kanamarlapudi <kmanikan@codeaurora.org>
Currently MSM_VIDC_BUF_DPB buffer allocated and queued to firmware
as part of output port streamon. Sometimes firmware raises b2b ipsc
with only change in fw_min_count. So client will not follow port
reconfig sequence and calls directly HFI_CMD_RESUME on output port.
So firmware holds only prev set of DPB buffers, which is not sufficient
and it continues to wait for further DPB buffers and that is leading
to stall and testcase fails. So added change to requeue new set of
DPB buffers, incase of b2b ipsc(with only change in count).
Change-Id: I213c752be4ca9dc16f838ab2990c996c4d78e95f
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
There are 2 possible race issues with msm_vidc_stats_handler.
[1] msm_vidc_close acquired inst->lock and called cancel_delayed_work,
by that time stats_handler already fired and incremented inst->kref
via get_inst_ref. Close sequence releases the lock and called
put_inst().
So now inst strong ref is held by stats_handler, which will acquire
inst->lock and schedules new stats_work and does put_inst. inst->kref
count reaches zero and it will free inst struct(using close_helper).
So that will lead to use-after-free crash at core kernel side.
[2] msm_vidc_close acquired inst->lock and called cancel_delayed_work,
by that time stats_handler is scheduled. So process_one_work() from
workqueue is ready to call stats_handler api. But before it invokes
stats handler(context switch), msm_vidc_close sequence continued to
run and completed the close sequence and called put_inst. So inst
struct got freed up(because inst->kref count reached to zero). Now if
core kernel(workqueue) attempts to invoke stats_handler by calling
worker->current_func(work), will lead to again use-after-free crash.
Added changes to avoid above mentioned issues.
Change-Id: I55bc33a753f4dbae4a8cdc6373b5465d183da3bc
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
add checks to limit max number of sessions supported based on
resolution.
max number of 720P sessions: 16
max number of 1080P sessions: 8
max number of 4K sessions: 4
max number of 8K sessions: 2
Change-Id: I2a784a0d28ba050923bdc7f42a21ee9801d01d86
Signed-off-by: Darshana Patil <darshana@codeaurora.org>
Query final values of session params like stage,
pipe and quality mode from FW and check if they
are the same intended values as configured by
the driver.
Change-Id: I1a504401a599860ab971e34e1e184e67631d6120
Signed-off-by: Chandrakant I Viraktamath <civirakt@codeaurora.org>
add support for HFI_PROP_MAX_NUM_REORDER_FRAMES.
add ctrl for this metadata for avc and hevc.
Change-Id: I9a7b95f130e58f810255aff23e9b605304841072
Signed-off-by: Darshana Patil <darshana@codeaurora.org>
- Added change to print buffer statistics like total and avg
etb, ebd, ftb, fbd counts, client set fps, actual fps,
bitrate values in 1 sec window and these stats currently
configured to print at every 1sec boundary.
- Convert meta buffer prints to low level logs.
Change-Id: I1c93eea90fbd5460eb51bbf133c513157003d51e
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
Ignore input port or output port streamoff if corresponding port is already
streamed off.
Change-Id: I708126b24758ee0f76bc2646527d1baee59e5178
Signed-off-by: Akshata Sahukar <asahukar@codeaurora.org>
1. Calculate framerate based on buffer timestamp
2. If framerate changed and stable for 2 frames,
update it to firmware and in driver internally.
Change-Id: I7feda86dec8fcfc1dff6defac8c15a97c1b8393e
Signed-off-by: Priyanka Gujjula <pgujjula@codeaurora.org>
- Maintain free & busy pool to track buffers separately
- track double-free cases during msm_memory_free()
- track memory leak, by checking busy pool buffer count.
it is expected to be zero at the end of a session.
- maintain common api's to init, alloc, free, destroy
for all below buffer types.
- struct msm_vidc_buffer
- struct msm_vidc_alloc
- struct msm_vidc_map
- struct msm_vidc_timestamp.
Change-Id: Ib8655047c3883ea6601344661058d7a629ffd5bd
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
Enable HFI_PROP_SECURE when secure session is enabled
in encoder. Same behaviour as Decoder.
Change-Id: I5edfe3d735fda8edea9e491ae17546945b4431a1
Signed-off-by: Chinmay Sawarkar <chinmays@codeaurora.org>
- To determine if session is admissible, check only the
decoder input width/height and encoder crop width/height.
- Update session MBPF Calculations.
- Amend work mode setting to consider decoder input width,
height and GOP size.
Change-Id: If4053c222de2a55ad5b85a8ab96c12c47b145aa3
Signed-off-by: Mihir Ganu <mganu@codeaurora.org>
Currently driver prints hex values for codec type, pixel
format and port number. So added change to print strings
instead of hex values.
Removed unnecessary logs to avoid overlogging timeout
issues.
Change-Id: Ia44401990797dc20eae0a35631e9ed6a0b094c20
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
qbuf will be deferred on all ports before streamon,
So explicitly queue pending buffers at the time of
streamon.
Change-Id: I673a14f0d43563184aac887f5f23b63ed13d59df
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
Encoder supports downscale with scaling ratio upto 1/8 of width
and 1/8 of height. So added necessary scaling checks at streamon.
Change-Id: I3a29b43c79cf4e693ba2c0d9f98ec24410d50fbd
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
Add strict check in msm_vidc_check_mbpf_supported() to
reject image session, if total image load exceeds 16K.
Added msm_vidc_print_insts_info() support to print all
running instance info, if new session is rejected due
to mbpf/mbps checks.
Change-Id: I98bab985fcf4f4d20c6f263c48e0c701ec86ab6f
Signed-off-by: Govindaraj Rajagopal <grajagop@codeaurora.org>
Occationally when multiple sessions are waiting for firmware
response and at the same time if there was a system error
received which will deinitalize the core and remove all sessions
from core list. If client opened new session which will initialize
the core and start the new session. As there will be no response
for old sessions, eventually, they will timeout and try to deinitialize
the core which will reset new session also which is not correct.
So check if timed out session is valid (present in core list) before
deintializing the core from timeout session.
This change also addresses below race condition issue -
If client is opening multiple new sessions and multiple old sessions
are also timedout at the same time and all are waiting for core lock
to be acquired. The new session does init_completion(&core->init_done);
and release core lock before waiting for completion event. As core lock
was released the timed out session will acquire the core lock and deinit
the core and release core lock. As core was deinited, another new session
will do init_completion(&core->init_done); on which the previous session
was waiting for completion event. doing init_completion again on an
already waiting completion event will result in list corruption issues.
Change-Id: I94766ead8b3ecad7d0b87f4b999608f435d0cee2
Signed-off-by: Maheshwar Ajja <majja@codeaurora.org>
adjust MAX_MAP_OUTPUT_COUNT based on resolution
to enhance performance and reduce memory pressure.
- For 8K session: count = 20
- For 4K session: count = 32
- For 1080p session: count = 48
- For all remaining sessions: count = 64
reduce MAX_DPB_COUNT to 32.
reduce DEFAULT_MAX_HOST_BUF_COUNT to 64.
Change-Id: I0e6d25121947524b843e9cce96b75871aba174af
Signed-off-by: Darshana Patil <darshana@codeaurora.org>
To avoid memory leakage, clean children and firmware lists
at session close, error cases, and at the end of each usage.
Change-Id: I1eadbf81da207bd987c50428c0c7cf4967a8885f
Signed-off-by: Akshata Sahukar <asahukar@codeaurora.org>
Due to delayed unmap feature on decoder output buffers,
refcount will become one when no one uses the buffer
in driver. In release done functonality, remove the
buffer from maps list only if refcount equal to one.
If refcount is more than one indicates the buffer is
being used, so do not remove from maps list. There are
chances that same decoder output buffer might be queued
by the client in which case we are not supposed to
unmap it from maps list in release done.
Change-Id: I69250c3efa89d77055e89770fd441af1fb50b755
Signed-off-by: Maheshwar Ajja <majja@codeaurora.org>
add extra size for input meta buffer
when ROI metadata is enabled.
add dependency of rate control and
color format.
Change-Id: Icefd3f9e319d596f88090cee331417c0b2946180
Signed-off-by: Darshana Patil <darshana@codeaurora.org>
- subscribe for dpb list property in combined
mode on output port.
- copy dpb list property payload from fw to
static array.
- using this array, mark read_only list buffers
as non-reference if not part of array. if count
of such buffers is greater than output min count,
send to fw for release.
- once fw returns them, destroy these buffers.
- unmap stale output mappings due to lazy unmap feature.
- unify release internal functions into common function.
Change-Id: Id32f04efb19eecaff453cc4383ee8296a0246263
Signed-off-by: Darshana Patil <darshana@codeaurora.org>
