summaryrefslogtreecommitdiff
path: root/vpx_dsp
AgeCommit message (Collapse)Author
2016-12-27Add high bitdepth 8x8 idct NEON intrinsicsLinfeng Zhang
BUG=webm:1301 Change-Id: I56e3bc3aab9214e2debac93796389a7194991084
2016-12-21Clean idct 8x8 neon functionsLinfeng Zhang
BUG=webm:1301 Change-Id: I05f47dca1fddc155c8396e627cfccf6449677307
2016-12-19vpx_idct32x32_1024_add_neon: quiet uninitialized warningJames Zern
relocate the assignment to 'in' outside of the for loop. this quiets a spurious warning in visual studio builds since: 86e340c enable vpx_idct32x32_1024_add_neon in hbd builds + give the variable a more descriptive name BUG=webm:1294 Change-Id: I5c3da5c7939621477e0fc0ad3a1b2a3045c5bffd
2016-12-19Merge "Clean hbd idct 4x4 neon functions and other"Linfeng Zhang
2016-12-14postproc: neon down and across macroblock filterJohann
Implement vpx_post_proc_down_and_across_mb_row in NEON. Runs about 6-7x faster than C. BUG=webm:1320 Change-Id: Ic5c7d3552a88cfcf999ec5bf2bd46fee460642c2
2016-12-14Clean hbd idct 4x4 neon functions and otherLinfeng Zhang
BUG=webm:1301 Change-Id: I387b7eae716a7df15c691dc6f368b07602df7342
2016-12-12enable vpx_idct32x32_1024_add_neon in hbd buildsJames Zern
BUG=webm:1294 Change-Id: Ibdda54e6d1303b0f73bc7bc71417e4041d7618de
2016-12-12Cosmetics by unifying dest_stride to stride in idctLinfeng Zhang
Change-Id: Ie9336a808a3c3592bb4fd5d4ad3839028bfcafba
2016-12-09Move load_and_transpose to transpose_neon.hJohann
Allows for use outside the idcts without pulling in idct_neon.h Change-Id: I4a94c1af3dac3e1b5bc8296ec9eab0ddcc8cfecf
2016-12-08idct16x16_add_neon: fix arm visual studio buildsJames Zern
after: 2d3d95f enable vpx_idct16x16_256_add_neon in hbd builds reorder INCLUDEs and fix indent of IF/ENDIFs remove vpx_config.asm to avoid multiple symbol definitions in windows builds and shift idct_neon.asm.S to the top to allow use of CONFIG_VP9_HIGHBITDEPTH in the export list. Change-Id: I0dacfbae62a6ec8fe4a26940c1a52da2dfad2029
2016-12-07Merge "Update idct NEON optimization to not use narrowing saturating shift"Linfeng Zhang
2016-12-07Merge "enable vpx_idct16x16_256_add_neon in hbd builds"James Zern
2016-12-07Update idct NEON optimization to not use narrowing saturating shiftLinfeng Zhang
Change-Id: Iae517017217dbacd638d40fcfeeb0f4bba7b8b8b
2016-12-06enable vpx_idct16x16_256_add_neon in hbd buildsJames Zern
BUG=webm:1294 Change-Id: Ib421c150b0d29dee0a81390a612bf01a4a28cff1
2016-12-07Merge changes Ibad079f2,I7858a0a1James Zern
* changes: enable vpx_idct16x16_10_add_neon in hbd builds idct16x16,NEON: rm output_stride from pass1 fns
2016-12-06enable vpx_idct16x16_10_add_neon in hbd buildsJames Zern
BUG=webm:1294 Change-Id: Ibad079f25e673d4f5181961896a8a8333a51e825
2016-12-06idct16x16,NEON: rm output_stride from pass1 fnsJames Zern
vpx_idct16x16_256_add_neon_pass1, vpx_idct16x16_10_add_neon: this was a constant 8 in all cases meaning the results are stored contiguously, this allows the number of stores to be reduced. Change-Id: I7858a0a15a284883ef45c13dfd97c308df9ea09e
2016-12-05Refine 8-bit 8x8 idct NEON intrinsicsLinfeng Zhang
Change-Id: I4ec4ad1928ec2ed87f596f52f097bc52065278dd
2016-12-02Check in vpx_lpf_vertical_4_dual_neon() assemblyLinfeng Zhang
This replaces its C version. Change-Id: Ie39e9324305fdc0fff610ced608a037e44a85a1a
2016-12-02Merge changes I4afc130e,Iaa64d23fJames Zern
* changes: Add high bitdepth 4x4 idct NEON intrinsics Update idct x86 intrinsics to not use saturated add and sub
2016-11-30Add high bitdepth 4x4 idct NEON intrinsicsLinfeng Zhang
Change-Id: I4afc130effa05b8be2e9f982967216b1beb2ce4b
2016-11-29Update idct x86 intrinsics to not use saturated add and subLinfeng Zhang
Change-Id: Iaa64d23fdb45ca1f235b0ea57e614516e548eca4
2016-11-28idct16x16,NEON,cosmetics: normalize fn signaturesJames Zern
+ remove unused parameters from vpx_idct16x16_10_add_neon_pass2 Change-Id: Ie5912a4abdd308fab589380bca054a2e7234a2c4
2016-11-23enable vpx_idct32x32_135_add_neon in hbd buildsJames Zern
BUG=webm:1294 Change-Id: Ide6d3994fe01c4320c9d143e6d059b49568048e4
2016-11-23idct_neon: rename load_tran_low_to_s16 -> ...s16qJames Zern
BUG=webm:1294 Change-Id: I164cfcbe9bc4511d1d04af9206cf351a0ec2957b
2016-11-24Merge changes Icc4ead05,Ib019964b,I3b5fd3b3,Ieedadee2James Zern
* changes: Update vpx_idct4x4_16_add_neon() to pass SingleExtremeCoeff test Refine 8-bit 4x4 idct NEON intrinsics Add idct speed test. Update partial_idct_test.cc to support high bitdepth
2016-11-23Change C/MSA post proc to match SSE2.Jerome Jiang
BUG=webm:1321 Change-Id: I719023375dc48cf7d8ed72188853f0f1ccc4ad7f
2016-11-22Merge "Add 32x32 d45 and 8x8, 16x16, 32x32 d135 NEON intra prediction"Linfeng Zhang
2016-11-22Update vpx_idct4x4_16_add_neon() to pass SingleExtremeCoeff testLinfeng Zhang
Change-Id: Icc4ead05506797d12bf134e8790443676fef5c10
2016-11-22Refine 8-bit 4x4 idct NEON intrinsicsLinfeng Zhang
Change-Id: Ib019964bfcbce7aec57d8c3583127f9354d3c11f
2016-11-22Fix mips dspr2 build warningKaustubh Raste
Change-Id: Ia8fb3ed124f01384e7896e309c9ff22c05b40719
2016-11-19Merge "Fix SingleLargeCoeff idct test"Kaustubh Raste
2016-11-19Merge "Clean horizontal intra prediction NEON optimization"James Zern
2016-11-17Change *_xmm to *_sse2 in deblocker assembly functions.Jerome Jiang
Some cosmetic changes because xmm is an anachronism. Change-Id: I436a5b78a3c52776c20d6640939311f2a84a9bc7
2016-11-17Fix SingleLargeCoeff idct testKaustubh Raste
Updated idct code to handle single large coefficient (-32768) Change-Id: Ia13ab1ab434a9a1b9954a5914088977a88841cc7
2016-11-17Merge "Change C and msa to match results from sse2."Jerome Jiang
2016-11-16Change C and msa to match results from sse2.Jerome Jiang
Re-enable the tests to check CvsAssembly. BUG=webm:1321 Change-Id: Id7f7d74b06c469fb6c8f5d04e91359e9cd9097a6
2016-11-15Add high bitdepth intra prediction NEON optimization (mode tm)Linfeng Zhang
BUG=webm:1316 Change-Id: Ib014de06836ac12726f4a2c9f0833ec4eb4d233b
2016-11-12Add high bitdepth intra prediction NEON optimization (h and v)Linfeng Zhang
BUG=webm:1316 Change-Id: I47eeac698a98a31d1af5f72441052302e9fa4f46
2016-11-10Merge changes I339088b2,Iaade219e,If142afb1,I4257c4b3James Zern
* changes: fdct8x8_test: add vpx_idct8x8_64_add_neon in hbd fdct4x4_test: add vpx_idct4x4_16_add_neon in hbd partial_idct_test,NEON: add missing idct variants enable vpx_idct32x32_34_add_neon in hbd builds
2016-11-09Add high bitdepth intra prediction NEON optimization (mode d45 and d135)Linfeng Zhang
BUG=webm:1316 Change-Id: I6a330874348df04df24a6d9efdc06f567e04bf8e
2016-11-08enable vpx_idct32x32_34_add_neon in hbd buildsJames Zern
replace load_and_transpose_s16_8x8() in idct32_6_neon() with a separate load_tran_low_to_s16() and transpose_s16_8x8(). the combined function is used in idct32_8_neon() where the input is the correctly sized output from the earlier stage. BUG=webm:1294 Change-Id: I4257c4b3a421b2cf5d13651f966eee0680ef98a9
2016-11-08Optimize idct32x32_135_add for NEONJohann
BUG=webm:1295 Change-Id: I7f80ef4d29813fcb401fc6075babf19e3c195462
2016-11-08Merge "Add high bitdepth intra prediction NEON optimization (mode dc)"Linfeng Zhang
2016-11-07Rename vpx_highbd_idct8x8_10{*}() to vpx_highbd_idct8x8_12{*}()Linfeng Zhang
Also update its trigger threshold from 10 to 12. Change-Id: Ib8dddd87a5a22a12ca66e7084d342fbb027b0a2f
2016-11-07Merge "Replace highbd_dct_const_round_shift with dct_const_round_shift"Linfeng Zhang
2016-11-05Update vp9_fdct8x8_quant_ssse3 for highbitdepthJohann
Borrow transition functions from fdct.h nee vpx_quantize_b_sse2 BUG=webm:1304 Change-Id: I9c88c3eec3ff8bb461411d98c26c3c236ea28ef1
2016-11-04Replace highbd_dct_const_round_shift with dct_const_round_shiftLinfeng Zhang
They are identical. Change-Id: I1ccaf03c81c3cbf88e82d77ffeb8204f5b063c61
2016-11-04Merge "Cosmetics of inv_txfm.c"Linfeng Zhang
2016-11-04Merge "Extract high bit depth helper functions"Johann Koenig
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-10 01:15:54 +0000