xref: /rockchip-linux_mpp/CHANGELOG.md (revision 437bfbeb9567cca9cd9080e3f6954aa9d6a94f18)
1## 1.0.11 (2025-09-10)
2### Feature
3- [mpp_trie]: Add info name max length record
4- [mpp_enc_cfg]: Separate init function
5- [mpp]: Add jpeg roi function for RV1126B
6- [kmpp]: Add jpeg roi function for kmpp
7- [kmpp]: Set chan_fd to init cfg
8- [kmpp]: Replace frame_infos with kmpp_frame
9- [kmpp_frame]: Add self_meta in kmpp_frame
10- [kmpp_buffer]: Add ioctl to inc ref and flush
11- [mpp_meta]: Add more frame buffer key to meta
12- [base]: Add toml function
13- [base]: Use enc cfg obj
14- [smart_v3]: Add new frame qp interface
15- [kmpp]: Add KmppMeta module
16- [kmpp]: Add KmppBuffer module
17- [kmpp_obj]: Add priv prop support for objdef
18
19### Fix
20- [h265e]: Remove unused buffer
21- [mpp]: Add null check for sync pkt buffer
22- [mpp_meta]: Add user data deep copy support
23- [mpp_meta]: Add KEY_NPU_UOBJ_FLAG and KEY_NPU_SOBJ_FLAG
24- [kmpp_obj]: Fix obj ioctl typo
25- [mpp_trie]: Fix get err node issue
26- [vdpp] Fix building tests against musl libc
27- [script]: Prepend bash with /usr/bin/env
28- [kmpp_buffer]: Close fd when deinit
29- [mpp_thread]: Fix thread name is not set
30- Rename FFmpeg to FF for sdk release request
31- [kmpp_obj]: Fix kmpp obj get by sptr
32- [h265d]: Ensure the DTS is transmitted to the frame
33- [kmpp_obj]: Rename kmpp_obj_impl_put func
34- [kmpp_obj]: Fix kmpp frm/pkt self meta erro
35- [h264e_api_v2]: Fix bit_real calc in skip mode
36- [h264d]: Fix fast play mode not working in shell environment.
37- [kmpp_frame]: Remove unnecessary logs
38- [enc_test]: Set input block mode in init kcfg
39- [hal_h265e]: Fix nal type in tsvc mode
40- [h265d]: Fix log issue
41- [vepu511]: Add tune stat update
42- [kmpp_obj]: Update tbl after objdef registration
43- [mpp_cfg_io]: Add more mpp_cfg_io function
44- [kmpp_obj]: Fix grp_cfg and buf_cfg leak in kmpp_obj_test
45- [vproc]: Fix unit tests cannot be disabled
46
47### Docs
48- Update 1.0.11 CHANGELOG.md
49
50### Refactor
51- [sys_cfg]: Refactor C++ sys_cfg to C
52- [test]: Refactor C++ test file to C
53- [osal]: Refactor C++ osal file to C
54- [rc]: Refactor C++ rc/rc_base to C
55- [enc]: Use KmppShmPtr to represent osd buffer
56- [kmpp]: Fix kmpp obj compilation warning
57- [rc_api]: Refactor C++ rc_api to C
58
59### Test
60- [mpi_enc_test]: Add jpeg roi test
61
62### Chore
63- [dec_test]: Remove unused code
64- [mpp_singleton]: Update name print
65- [hal]: Organize the relevant processes for vepu fmt
66- Rename Dolby for sdk release requirement
67- [kmpp_meta]: Disable failure log
68- [mpp_enc_cfg]: Add base:smart_en option
69- [kmpp_obj]: Add is_kobj query function
70- [rc_smt]: Adjust code style for rc_smt
71
72## 1.0.10 (2025-06-23)
73### Feature
74- [mpp_log]: Add long log (llog) function
75- [mpp_buffer]: Add mpp_buffer discard function
76- [build]: add Android.bp support
77- [kmpp_packet]: Add kmpp_packet interface
78- [mpp_log]: Add external callback support
79- [kmpp_obj]: Refactor kmpp_obj helper
80- [kmpp_obj]: Add more kmpp_obj property
81- [kmpp_obj]: Add object update function
82- [kmpp_obj]: Add userspace objdef functions
83- [osal]: Add mpp_singleton module
84- [mpp_cfg_io]: Add mpp cfg io module
85- [kmpp]: Add kmpp_frame_test
86
87### Fix
88- [h265d]: Fix yuv400 decode error
89- [h265d]: Fix GDR stream decoding
90- [kmpp_obj]: Undef KMPP_OBJ_SGLN_ID macro
91- [osal]: Fix timeout expire too soon issue
92- [cmake]: Fix static build issue
93- [vp8e]: Remove unused vp8e_rc file
94- [h265d_rkv]: Fix dec err after cut streams
95- [mpp_singleton]: fix init order issue
96- [mpp_dec]: Fix compile warning
97- [h265d_parser]: Fix slice header parse
98- [mpp_sys_cfg]: afbc calc support yuv444sp_10bit
99- [kmpp_obj]: Update helper macro
100- [h263d]: Fix missing initializer for field problem
101- [enc_utils]: Remove duplicate option
102- [kmpp_obj]: Remove extra print in helper
103- [avsd_plus]: Fix page fault when filtering field data
104- [h265d_vdpu384a]: Fix CABAC error detection issue.
105- [mpp_sys_cfg]: Fix stride issue on resolution change
106- [vepu_540c]: Reduce print hw_status when irq ret
107- [mpp_sys_cfg]: Fix ver_stride calc issue
108- [sys_cfg]: Fix ver stride calculation issue.
109- [vepu541]: Add warning for unsupport nv21/nv42
110- [avs2d]: fix vertical stride config
111- Revert "fix[mpp_enc_impl]: fix rc cfg for jpeg enc"
112- [h265d_ps]: Suppress YUV444 unsupported warning logs
113- [mpp_cfg]: Fix function define on C++ field
114- [h264e_dpb]: fix walk_len when refs_dryrun
115- [av1d_vdpu383]: fix segid page fault issue
116- [allocator]: Fix misc buffer group flag issue
117- [h265d_parser]: fix startcode finder for 00 00 00 xx case
118- [kmpp]: Fix eos frame with NULL buffer issue
119- [utils]: Remove duplicate assignments
120- [mpi_enc_test]: Sync mdc config of RV1126B
121- [sys_cfg]: Avoid frequent environment variable access.
122- [mpp_enc]: Add avc rc parameter set
123- [h265d_vdpu383]: Fix CABAC error detection issue.
124- [mpi]: Fix typo
125- [h264_vdpu384a]: Fix error proc issue
126- [h265e]: Correct tile syntax elements at PPS
127- [mpp]: Add atf set, atf value 0~3
128- [mpp_enc_cfg]: Add lambda_idx_i and lambda_idx_p
129- [mpp_enc]: Add encoder speed mode setup
130- [test]: Add qbias_arr and aq_rnge_arr init
131- [packet]: fix packet partition and eoi logic
132- [mpp]: add qpmap_en and enc_spd
133- [cmake]: Fix double object include issue
134- [sys_cfg]: Align to CTU64 to avoid info change.
135- [mpp]: Fix compile warning with ipc sdk toolchain
136
137### Docs
138- Update 1.0.10 CHANGELOG.md
139
140### Refactor
141- [base]: Refactor C++ mpp_enc_cfg to C
142- [base]: Refactor C++ mpp_meta to C
143- [base]: Refactor C++ mpp_packet to C
144- [base]: Refactor C++ mpp_frame to C
145- [base]: Refactor C++ mpp_buffer to C
146- [mpp_mem_pool]: Add exit leak pool print
147- [osal]: Refactor C++ mpp_server to C
148- [osal]: Refactor more module from C++ to C
149- [mpp_trace]: Refactor C++ mpp_trace to C
150- [mpp_runtime]: Refactor C++ mpp_runtime to C
151- [mpp_soc]: Refactor C++ mpp_soc to C
152- [mpp_platform]: Refactor C++ mpp_platform to C
153- [mem_pool]: Refactor C++ mem_pool to C
154- [mpp_mem]: Refactor C++ mpp_mem to C
155- [kmpp]: Replace venc_packet with KmppPacket
156- [osal/linux/os_log]: Use C constructor.
157- [base]: Remove MppDecCfgImpl
158- [base]: Refactor mpp_trie from C++ to C
159- [mpp_cfg_io]: Change cfg to trie interface
160
161### Test
162- [osal]: Add libc and OS compatibility checking
163- [resolution]: Add resolution test tool
164
165### Chore
166- [kmpp]: Modify kmpp_objs init / deinit order
167- [kmpp_obj]: Add from objs device macro
168- [kmpp_obj]: Add more obj function
169- [kmpp_obj]: Update flag calculation macro
170- [utils]: Add fbc frame data dump
171- A fix for company release requirement
172- [kmpp]: Remove get packet failed log
173
174## 1.0.9 (2025-04-03)
175### Feature
176- [kmpp_frame]: Add KmppFrame module
177- [vepu_511]: Add rv1126b 265e/264e/jpge support
178- [mpp_meta]: Add osd_data3 fmt for 1103b/1126b
179- [kmpp_obj]: Sync to new KmppEntry share object
180- [err_proc]: Add a new command: DIS_ERR_CLR_MARK
181- [mpi_enc_test]: Support enc for kmpp flow
182- [kmpp_obj]: Add more kmpp_obj functions
183- [vdpu384a]: Support RV1126B new features
184- [mpp_soc]: Support rv1126b soc
185- [kmpp_obj]: Sync to new kmpp_meta
186- [kmpp_obj]: Sync to loctbl without flag_type
187- [mpp_buf_slot]: buf_slot add coded width alignment config
188- [h265d]: Add vdpu383 hevc yuv444_10bit support
189- [vproc]: Add more log for debugging
190- [mpp]: Support kmpp access
191- [kmpp]: Add kmpp module
192- [rk_mpi_cmd]: Merge cmds from mpp_interface
193- [build]: Add --toolchain to config toolchain for linux
194- [mpp_meta]: Use trie to index the meta key
195- [mpp_packet]: Add realease callback info
196- [kmpp_obj]: Update to new objdef query mode
197- [mpp_trie]: Allow empty name trie for import
198- [enc]: Support setting temporal_id
199- [mpp_enc_cfg]: Merge enc cfgs from mpp_interface
200- [mpp_sys_cfg_st]: Provide packaging for use on products
201- [mpp_sys_cfg]: Add raster/tile/fbc buffer alignment
202- [mpp_sys_cfg]: Support sys_cfg buffer alignment
203- [kmpp_obj]: Add kmpp_obj_get_hnd func
204- [mpp_venc_kcfg]: Add mpp_venc_kcfg module
205
206### Fix
207- [sys_cfg]: Add debug info
208- [sys_cfg]: fix fbc ver stride calc issue
209- [sys_cfg]: Fix external configuration stride issue
210- [sys_cfg]: Support alignment for mpeg2/mpeg4/h263/vp8.
211- [sys_cfg]: AVC is aligned to ctu to avoid info change
212- [sys_cfg]: Fix RK3399 hor/ver stride calculation issue.
213- [sys_cfg]: Fix HAL layer buffer alignment issue
214- [h264d]: Recovery only takes effect when no IDR frames present
215- [hal_jpege_api]: Fix jpege api path judgment
216- [vdpp]: Fix vdpp blk_size calculation.
217- [mpp_venc_kcfg]: Revert to mpp interface
218- [cmake]: Fix kmpp_base symbol missing
219- [av1_syntax]: Fix array out-of-bounds issue.
220- [build]: fix build failure with CMake 4.0
221- [vepu_511]: Speed grade configuration of 0.67
222- [mpp_frame]: Add rk_fbc fmt for 1126b
223- [jpegd_rkv]: New JPEG IP supports tile 4x4 output by default.
224- [jpeg_rkv]: New JPEG IP defaults to no RGB support.
225- [hal_rcb]: Fix rcb buf size calc issue
226- [kmpp_obj]: Fix rockit compile error
227- [avsd]: Skip redundant zeros between fields inside one picture
228- [av1]: parameter is 16 bits
229- [base]: Fix strncpy compile warning
230- [hal_h265e_vepu580]: Fix overflow status check
231- [kmpp]: Fix channel dup issue
232- [os_log]: Modify default log option for linux
233- [kmpp_obj]: Fix warning on arm32
234- [kmpp]: Set KEY_OUTPUT_INTRA meta to packet
235- [sys_cfg]: Align rk3399 h_stride to an odd multiple of 265.
236- [mpp_sys]: Fix old IP vertical alignment to 16 issue
237- [kmpp_obj]: Disable /dev/kmpp_objs not found log
238- [mpp_soc]: Fix cap_fbc for rv1126b
239- [sys_cfg]: Optimize comparison information printing.
240- [sys_cfg]: Print comparison information only once.
241- [mpp_meta]: Fix compile error
242- [vepu510]: Mark frame first part when split slice out
243- [hdr_meta]: Fix hdr format for av1
244- [mpp_sys_cfg]: Fix align pixel stride on rk3576
245- [vproc]: fix height out of boundary problem
246- [mpp_sys_cfg]: Fix abnormal stride calculation.
247- [h264d]: disable ref erorr when decode recovery frame period
248- [jpege_vpu720]: Correct encoded size config
249- [buf_slot]: Correct coding mistakes.
250- [build]: Avoid exporting toolchain to system PATH
251- [mpp_enc]: Fix some exceptions when force pskip
252- [kmpp]: Fill pts/dts/flag to MppPacket
253- [vproc]: fix frame output disorder problem
254- [vproc]: Fix field disordered problem
255- [mpp_enc_cfg]: Remove a redundant atr_str
256- []: Fix abnormal FBC info issue in Info Change
257- [h264d]: Fix segment fault problem
258- [vproc]: Fix error info missed problem
259- [vproc]: Fix output blank buffer problem
260- [fbc]: Fix RK3588 av1 FBC usage issue
261- [sys_cfg/buf_slot]: support yuv422sp 10bit
262- [mpp_enc_cfg]: Add sao_bit_ratio from mpp_interface
263- [buf_slot]: Correct coding mistakes.
264- [mpp_venc_kcfg]: Get objdef at runtime
265- [jpegd]: Avoid buffer overrun
266- [sys_cfg/buf_slot]: fix fbc yuv444sp buf calculation issue
267- [kmpp_obj]: Add extern C
268
269### Docs
270- Update 1.0.9 CHANGELOG.md
271
272### Refactor
273- [kmpp]: Move kmpp to seperate directory
274- [mpp_trie]: Replace root import
275- [mpp_enc_cfg]: Adjust cu_qp_delta_depth
276
277### Chore
278- [mpp_buf_slot]: Modify sys_cfg mismatch print
279
280## 1.0.8 (2024-12-30)
281### Feature
282- [enc]: Add switch for disable IDR encoding when FPS changed.
283- [test]: Add PSNR info for video encoder
284- [mpp_buf_slots]: Add coding attribute to buf slots
285- [mpp_sys_cfg]: Add mpp_sys_cfg function
286- [dec_nt_test]: Support jpeg decoding on decode
287- [mpp_dec]: Add jpeg put/get decode support
288- [mpp_obj]: Add mpp_obj for kernel object
289- [mpp_trie]: Add functions for import / export
290- [rk_type.h]: Add kernel driver compat define
291- [mpp_dec]: add control for select codec device
292- [mpp_dec]: support hdr10plus dynamic metadata parse
293- [hal_avsd]: enable hw dec timeout
294- [vpu_api]: Support configuration to disable decoding errors
295- [enc]: Support use frame meta to cfg pskip
296- [vepu510]: Add scaling list regs setup
297
298### Fix
299- [enc]: Fix CPB size not enough problem
300- [m4v_parser]: Fix split_parse setting failure issue
301- [mpp_trie]: Remove a redundant variables from log
302- [mpp_enc]: Set frm type in pkt meta
303- [mpp_sys_cfg]: Fix compile warning
304- [rc_smt]: Fix the variable overflow issue
305- [h264e_sps]: fix constraint_set3_flag flag issue
306- [vpu_legacy]: Fix vpu fbc configuration issue
307- [mpp_buffer]: Fix buffer put log
308- [mpp_mem_pool]: Record pool buffer allocator caller
309- [mpp]: Fix input_task_count for async enc
310- [av1d]: Fix uninitialized fbc_hdr_stride issue
311- [cfg]: fix cfg test segment fault problem
312- [drm]: Call drop master by default
313- [vepu580]: fix is_yuv/is_fbc typo
314- [misc]: Fix compile on 32bit platform
315- [jpegd]: replace packet size with stream length
316- [av1_vdpu383]: Fix the CDF issue between GOPs
317- [mpp_enc_impl]: fix rc cfg for jpeg enc
318- [av1_vdpu383]: fix cdf usage issue
319- [hal_h265d]: Avoid reg offset duplicate setting issue
320- [vepu580]: fix incorrect color range problem
321- [buf_slots]: Fix the issue of fmt conv during info change
322- [h264d]: force reset matrix coefficients when parse unknown value
323- [h264d]: Parse hdr parameters on enable_hdr_meta enabled
324- [h264d_parser]: Fix pps parsing issue
325- [hal_vdpu383]: fix fbc hor_stride mismatch issue
326- [hal_vepu580]: re-get roi buf when resolution switch
327- [hal_vepu541]: re-get roi buf when resolution switch
328- [iep2]: Remove unnessary log on init failed
329- [h264_dpb]: Add env variables to force fast play mode
330- [h265e_slice]: fix compilation warning
331- [hal_avs2d_vdpu383]: handle scene reference frame
332- [debain]: fix typo in compat version
333- [debian]: Update debian control
334- [debain]: Update debian/control
335- [debain]: Update compat to 10
336- [h264e_pps]: add pic_scaling_matrix_present check
337- [h2645d_sei]: fix read byte overflow error
338- [m2vd]: Fix refer frame error on beginning
339- [vdpu383]: fix err detection mask issue
340- [test]: Fix AQ table error
341- [vepu580]: Add md info internal buffer
342- [vepu580]: Add ATF weight adjust switch for H.265
343- [tune]: Replace qpmap_en with deblur_en
344- [vepu580]: Adjust frame-level QP for VI frame
345- [hal_jpegd]: fix huffman table selection
346- [h265]: fix pskip when enable tile mode
347- [smt_rc]: Fix first frame QP error
348- [h264d]: fix no output for mvc stream
349- [vepu580]: Fix motion level assignment error
350- [avsd]: Fix attach dev error issue
351- [h265d]: Fix conformance window offsets for chroma formats
352- [test]: Fix mdinfo size according to soc type
353- [h265d_vdpu383]: fix dec err when ps_update_flag=0
354- [vepu510]: Sync code from enc_tune branch
355- [mpp_cfg]: Fix compile warning
356- [h265d]: fix output err causeby refs cleard
357- [h264d]: remove error check for B frame has only one ref
358- [test]: Fix test demo stuck issue
359
360### Docs
361- Update 1.0.8 CHANGELOG.md
362- update doc for fast play
363
364### Refactor
365- [hal]: Update the reg offset setting method.
366- [mpi]: Add ops name when assign for reading friendly
367- [av1d_vdpu383]: Regs definition sync with other protocols.
368- [vproc]: Refactor iep2 progress
369- [h265]: unify calculation tile width
370
371### Chore
372- [hal_jpegd]: Remove reset / flush functions
373- [test]: Use put/get in mpi_dec_test for jpeg
374- [MppPacket]: Add caller log on check failure
375
376## 1.0.7 (2024-09-04)
377### Feature
378- [rc_smt]: Add rc container for smart mode
379- [vepu580]: Optimization to improve VMAF
380- [vepu580]: Optimize hal processing for smart encoding
381- [vepu580]: Add qpmap and rc container interface
382- [vepu510]: Add anti-smear regs setup for H.264
383- [vepu510]: Add H.264 tuning setup
384- [vepu510]: Sync code from enc_tune branch
385- [vepu510]: Sync code from enc_tune branch
386- [vepu510]: Sync code from enc_tune branch
387- [mpp_trie]: Add trie context filling feature
388- [mpp_trie]: Add trie tag and shrink feature
389- [h264d]: support hdr meta parse
390- [h265e]: Support force mark & use ltr
391- [vpu_api]: support yuv444sp decode ouput pixel format
392
393### Fix
394- [h265d]: fix infochange loss when two sps continuous
395- [hal_h264e]: Fix CAVLC encode smartP stream err
396- [mpi_enc_test]: Remove redundant code about smart encoding
397- [h264e_sps]: fix the default value of max mv length
398- [enc_roi]: Fix cu_map init in vepu_54x_roi
399- [hal_vp9]: Optimize prob memory usage
400- [hal_h265d]: Allow reference missing for GDR
401- [osal]: Fix mpp_mem single instance issue
402- [hal_vp9d_com]: Fixed memory leak issue
403- [hal_h265d]: Avoid risk of segment fault
404- [hal_h265d]: fix error slot index marking
405- [h265d]: Adjust condition of scan type judgement
406- [mpp_hdr]: Fix buffer overflow issue
407- [mpp_buffer]: Synchronous log addition point
408- [hal_vepu]: fix split regs assignment
409- [vepu580]: poll max set to 1 on split out lowdelay mode
410- [mpp_common]: fix compile err on F_DUPFD_CLOEXEC not defined
411- [h265d]: return error on sps/pps read failure
412- [build]: The first toolchains is selected by default
413- [265e]:Fix the st refernce frame err in tsvc
414- [av1d]: when MetaData found then it is new frame
415- [m2vd]: Fix seq_head check error
416- [h265e_vepu510]: Fix a memory leak
417- [h265d]: auto output frame in dpb when ready
418- [m2vd]: Remove ref frame when info changed
419- [mpp_meta]: Missing data in the instance
420- [mpp_bitread]: Fix negative shift error
421- [osal]: fix 128 odd plus 64 bytes alignment
422- [h265d_parser]: Fix fmt configuration issue
423- [hal_av1d_vdpu383]: modify av1 segid wr/rd base config
424- [h265d_parser]: Fix fmt configuration issue
425- [hal_av1d_vdpu383]: add segid reg base config
426
427### Docs
428- Update 1.0.7 CHANGELOG.md
429- [readme]: Add more repo info
430
431### Refactor
432- [mpp_cfg]: Refactor MppTrie and string cfg
433
434### Chore
435- [mpp_mem]: Add mpp_realloc_size
436- [mpp_cfg]: Remove some unused code
437- fix compile warning
438
439## 1.0.6 (2024-06-12)
440### Feature
441- [vdpu383]: refine rcb info setup
442- [enc_265]: Support get Largest Code Unit size
443- [mpp_dec_cfg]: Add disable dpb check config
444- [vdpu383]: support 8K downscale mode
445
446### Fix
447- [drm]: Fix permission check issue on GKI kernel
448- [hal_h265e]: Amend 510 tid and sync cache
449- [hal_h265e]: Fix nalu type avoid stream warning
450- [h265e]: Fix vps/sps max temparal layers val
451- [hal_jpeg_vdpu1]: fix dec failed on RK3036 problem
452- [osal]: rv1109/rv1126 vcodec_type mismatch problem
453- [h264e_vepu2]: Adjust inter favor table
454- [h264d]: fix drop packets after reset when err stream
455- [h265d]: Allow filtering of consecutive start code
456- [hal_h264d_vdpu383]: fix spspps update issue
457- [mpp]: fix mpp frame leak when async enc
458- [enc]: Add use_lt_idx to output packet meta
459- [hal_h265e]: fix sse_sum get err
460- [mpp_enc_async]: fix mpp packet leak when thread quit
461- [enc_roi]: Support ROI cfg under CQP mode
462- [hal]: Fix the lib interdependence issue
463- [vepu_510]: fix same log type when enc feedback
464- [mpp_buffer]: fix dec/inc ref_count in multi threads
465- [mpp_enc_async]: fix debreath not work on async flow
466- [base]: fix AV1 and AVS2 string info missing problem
467- [mpp]: Add encoder input/output pts log
468- [hal_vepu580/510]: fix split out err when pass1 frame
469- [hal]: Fix target link issue
470- [hal_enc]: Fix lib dependency issue
471- [hal_h265d_vdpu383]: fix ref_err mark for special poc
472- [rc2_test]: fix pkt buffer overflow error
473- [enc_utils]: Support read odd resolution image
474- [allocator]: fix on invalid DMA heap allocator
475- [hal_h265e_vepu580]: fix reg config err for 2pass
476- [jpegd_vdpu]: Adjust file dump path
477- [mpp_common]: fix 128 odd plus 64 alignment
478- [cmake]: fix static build
479- [vdpu383]: Update vdpu383 error detection
480
481### Docs
482- Update 1.0.6 CHANGELOG.md
483
484### Refactor
485- [hal_jpegd]: init devices at hal_jpegd_api
486- [dec]: get deocder capability via common routine
487- [hal_av1d]: Migrate av1d from vpu to rkdec
488
489### Chore
490- [h265d]: Reduce malloc/free frequency of vps
491- [mpp_service]: fix typo err
492- [hal_h265d]: use INT_MAX for poc distance initiation
493- [cmake]: remove duplicate code
494
495## 1.0.5 (2024-04-19)
496### Feature
497- [vdpu383]: align hor stride to 128 odds + 64 byte
498- [vdpu383]: support 2x2 scale down
499- [mpp_buffer]: Add MppBuffer attach/detach func
500- [mpp_dev]: Add fd attach/detach operation
501- [vdpp]: Add libvdpp for hwpq
502- [vdpp]: Add capacity check function
503- [cmake]: Add building static library
504- [vdpp_test]: Add vdpp slt testcase
505- [av1d]: Add tile4x4 frame format support
506- [mpp_enc_cfg]: Add H.265 tier config
507- [jpeg]: Add VPU720 JPEG support
508- [enc]: Add config entry for output chroma format
509- [vdpu383]: Add vdpu383 av1 module
510- [vdpu383]: Add vdpu383 vp9 module
511- [vdpu383]: Add vdpu383 avs2 module
512- [vdpu383]: Add vdpu383 H.264 module
513- [vdpu383]: Add vdpu383 H.265 module
514- [vdpu383]: Add vdpu383 common module
515- [vdpp]: Add vdpp2 for rk3576
516- [vdpp]: Add vdpp module and vdpp_test
517- [vepu_510]: Add vepu510 h265e support
518- [vepu_510]: Add vepu510 h264e support
519- [mpp_frame]: Add tile format flag
520- [vepu_510] add vepu 510 common for H264 & h265
521- [mpp_soc]: support rk3576 soc
522
523### Fix
524- [avs2d_vdpu383]: Optimise dec result
525- [vdpu383]: Fix compiler warning
526- [vdpp]: Fix dmsr reg size imcompat error
527- [vdpu383]: hor stride alignment fix for vdpu383
528- [h265d_ref]: fix set fbc output fmt not effect issue
529- [vdpu383]: Fix memory out of bounds issue
530- [h264d_vdpu383]: Fix global parameter config issue
531- [avs2_parse]: add colorspace config to mpp_frame
532- [hal_h264e]:fix crash after init vepu buffer failure
533- [vpu_api]: Fix frame format and eos cfg
534- [av1d_vdpu383]: fix fbc hor_stride error
535- [av1d_parser]: fix fmt error for 10bit HDR source
536- [avs2d]: fix stuck when seq_end shows at pkt tail
537- [av1d_vdpu]: Fix forced 8bit output failure issue
538- [enc_async]: Invalidate cache for output buffer
539- [hal_av1d_vdpu383]: memleak for cdf_bufs
540- [av1d_vdpu383]: fix rcb buffer size
541- [vp9d_vdpu383]: Fix segid config issue
542- [vepu510]: Add split low delay output mode support
543- [avs2d_vdpu383]: Fix declaring shadow local variables issue
544- [av1]: Fix global config issue
545- [hal_av1d]: Delte cdf unused value
546- [av1]: Fix av1 display abnormality issue
547- [avs2d]: Remove a unnecessary log
548- [vepu510]: Adjust regs assignment
549- [hal_jpegd]: Add stream buffer flush
550- [265e_api]: Support cons_intra_pred_flag cfg
551- [mpp_enc]: Add device attach/detach on enc flow
552- [mpp_dec]: Add device attach/detach on dec flow
553- [vdpp]: Add error detection
554- [hal_265e_510]: modify srgn_max & rime_lvl val
555- [vdpu383]: spspps data not need copy all range ppsid
556- [vpu_api_legacy]: fix frame CodingType err
557- [av1]: Fix 10bit source display issue
558- [mpp_enc]: Expand the hdr_buf size
559- [av1]: Fix delta_q value read issue
560- [vdpu383]: Enable error detection
561- [ext_dma]: fix mmap permission error
562- [jpege_vpu720]: sync cache before return task
563- [mpp_buffer]: fix buffer type assigning
564- [vepu510]: Configure reg of Subjective param
565- [vepu510]: Checkout and optimize 510 reg.h
566- [mpp_dec]: Optimize HDR meta process
567- [av1d]: Fix scanlist calc issue
568- [h265e]: fix the profile tier cfg
569- [av1d]: Fix av1d ref stride error
570- [hal_h265e_vepu510]: Add cudecis reg cfg
571- [av1d]: Only rk3588 support 10bit translate to 8bit
572- [vp9d]: Fix vp9 hor stride issue
573- [rc]: Add i quality delta cfg on fixqp mode
574- [hal_h265d]: Fix filter col rcb buffer size calc
575- [av1d]: Fix compiler warning
576- [h264d]: Fix error mvc stream crash issue
577- [hal_h264e]: Fix qp err when fixqp mode
578- [h264d]: Fix H.264 error chroma_format_idc
579- [vdpu383]: Fix av1 rkfbc output error
580- [av1d]: Fix compatibility issues
581- [hal_h264d_vdpu383]: Reduce mpp_put_bits calls
582- Fix clerical error
583- [avs2d]: Fix get ref_frm slot idx error
584- [vdpu383]: Fix av1 global params bit pos issue
585- [vdpp]: fix sharp config error
586- [hal_av1d]: fix av1 dec err for rk3588
587- [vdpu383]: Fix av1 global params issue
588- [vepu510]: Fix camera record stuck issue
589- [utils]: fix read and write some YUV format
590- [mpp_bitput]: fix put bits overflow
591- [mpp_service]: fix rcb info env config
592- [vepu510]: Fix compile warning
593- [hal_vp9d]: fix colmv size calculator err
594- [avsd]: Fix the ref_frm slot idx erro in fast mode.
595
596### Docs
597- Update 1.0.5 CHANGELOG.md
598- [mpp_frame]: Add MppFrameFormat description
599
600### Refactor
601- [hal_av2sd]: refactor hal_api assign flow
602- [hal_h264d]: refactor hal_api assign flow
603- Using soc_type for compare instead of soc_name
604
605### Chore
606- [hal_h264e]: clean some unused code
607
608## 1.0.4 (2024-02-07)
609### Feature
610- [vpu_api_legacy]: Support RGB24 setup
611- [avsd]: keep codec type if not avs+
612- [mpi_enc_test]: add YUV400 fmt support
613- [mpp_enc]: Add YUV400 support for vepu580/540
614
615### Fix
616- [h265e]: fix hw stream size check error
617- [hal_vdpu]: unify colmv buffer size calculation
618- [vproc]: Fix deadlock in vproc thread
619- [h265e]: disable tmvp by default
620- [h265e]: Amend temporal_id to stream
621- [mpp_dump]: add YUV420SP_10BIT format dump
622- [hal_h265d]: Fix register length for rk3328/rk3328H
623- [hal_avsd]: Fix crash on no buffer decoded
624- [mpp_enc]: allow vp8 to cfg force idr frame
625- [m2vd]: fix unindentical of input and output pts list
626- [h265e_vepu580]: fix SIGSEGV when reencoding
627- [mpp_dmabuf]: fix align cache line size calculate err
628- [h265e_vepu580]: flush cache for the first tile
629- [dmabuf]: Disable dmabuf partial sync function
630- [iep_test]: use internal buffer group
631- [common]: Add mpp_dup function
632- [h265e]: Adapter RK3528 when encoding P frame skip
633- [h265e]: fix missing end_of_slice_segment_flag problem
634- [hal_av1d_vdpu]: change rkv_hor_align to 16 align
635- [av1d_parser]: set color info per frame
636- [jpegd]: add sof marker check when parser done
637
638### Docs
639- Update 1.0.4 CHANGELOG.md
640
641### Chore
642- [script]: add rebuild and clean for build
643- [mpp_enc_roi_utils]: change file format dos to unix
644
645## 1.0.3 (2023-12-08)
646### Feature
647- [dec_test]: Add buffer mode option
648- [mpp_dmabuf]: Add dmabuf sync operation
649- [jpege]: Allow rk3588 jpege 4 tasks async
650- [rc_v2]: Support flex fps rate control
651
652### Fix
653- [av1d_api]: fix loss last frame when empty eos
654- [h265e_dpb]: do not check frm status when pass1
655- [hal_bufs]: clear buffer when hal_bufs get failed
656- [dma-buf]: Add dma-buf.h for old ndk compiling
657- [enc]: Fix sw enc path segment_info issue
658- [cmake]: Remove HAVE_DRM option
659- [m2vd]: update frame period on frame rate code change
660- [test]: Fix mpi_enc_mt_test error
661- [dma_heap]: add dma heap uncached node checking
662- [mpp_mem]: Fix MEM_ALIGNED macro error
663- [mpeg4_api]: fix drop frame when two stream switch
664- [script]: fix shift clear input parameter error
665- [hal_h265e_vepu541]: fix roi buffer variables incorrect use
666
667### Docs
668- Update 1.0.3 CHANGELOG.md
669
670### Refactor
671- [allocator]: Refactor allocator flow
672
673### Chore
674- [vp8d]: optimize vp8d debug
675- [mpp_enc]: Encoder changes to cacheable buffer
676- [mpp_dec]: Decoder changes to cacheable buffer
677- [mpp_dmabuf]: Add dmabuf ioctl unit test
678
679## 1.0.2 (2023-11-01)
680### Feature
681- [mpp_lock]: Add spinlock timing statistic
682- [mpp_thread]: Add simple thread
683- add more enc info to meta
684
685### Fix
686- [vepu540c]: fix h265 config
687- [h264d]: Optimize sps check error
688- [utils]: adjust format range constraint
689- [h264d]: fix mpp split eos process err
690- [h264d]: add errinfo for 4:4:4 lossless mode
691- [h264d]: fix eos not updated err
692- [camera_source]: Fix memory double-free issue
693- [mpp_dec]:fix mpp_destroy crash
694- [mpp_enc]: Fix async multi-thread case error
695- [jpeg_dec]: Add parse & gen_reg err check for jpeg dec
696- [h265e_vepu580]: fix tile mode cfg
697- [vp9d]: Fix AFBC to non-FBC mode switch issue
698- [h264e_dpb]: fix modification_of_pic_nums_idc error issue
699- [allocator]: dma_heap allocator has the highest priority
700- [camera_source]: enumerate device and format
701- [utils]: fix hor_stride 24 aligned error
702
703### Docs
704- Update 1.0.2 CHANGELOG.md
705- Add mpp developer guide markdown
706
707### Chore
708- [scipt]: Update changelog.sh
709
710## 1.0.1 (2023-09-28)
711### Feature
712- [venc]: Modify fqp init value to invalid.
713- [vepu580]: Add frm min/max qp and scene_mode cmd param
714- [venc]: Add qbias for rkvenc encoder
715- Support fbc mode change on info change stage
716- [hal_vepu5xx]: Expand color transform for 540c
717- Add rk3528a support
718
719### Fix
720- [mpp_enc_impl]: fix some error values without return an error
721- [av1d_cbs]: fix read 32bit data err
722- [Venc]: Fix jpeg and vpx fqp param error.
723- [h265e_vepu580]: dual cores concurrency problem
724- [hal_h264e_vepu]: terminate task if not support
725- [vdpu_34x]: disable cabac err check for rk3588
726- [enc]: fix duplicate idr frame
727- [h264e_amend]: fix slice read overread issue
728- [hal_jpegd]: add pp feature check
729- [enc]: fix duplicate sps/pps information
730- [h264e_slice]: fix pic_order_cnt_lsb wrap issue
731- [hal_h264e_vepu540c]: fix reg configuration
732- [hal_h264e_vepu540c]: Amend slice header
733- [h264d]: fix crash on check reflist
734- [hal_vp9d]: not support fast mode for rk3588
735- [h264d]: fix frame output order when dpb full
736- [mpp_frame]: setup color default UNSPECIFIED instead 0
737- [hal_h264d]: adjust position of env_get
738- [h264e_slice]: fix pic_order_cnt_lsb wrap issue
739- [hal_avs2d]: fix some issue
740- fix redundant prefix NALU amended problem
741- [hal_jpegd]: fix rgb out_fmt mismatch error
742- [utils]: fix convert format error
743- [h265e]: check input profile_idc
744- [hal_h264e_vepu580]: fix SEGV on GDR setting
745- [h264d]: fix TSVC decode assert error.
746- [hal_vepu580]: fix comiple issue
747- [h264d]: fix MVC DPB allocation
748- [h264d]: fix SEI packet parsing
749- [hal_vp8e]: fix entropy init
750- [mpp_soc]: fix rk356x vepu2 capability
751
752### Docs
753- Add 1.0.1 CHANGELOG.md
754- update readme.txt
755
756### Refactor
757- move same tables to common module
758
759## 1.0.0 (2023-07-26)
760### Docs
761- Add 1.0.0 CHANGELOG.md
762