summaryrefslogtreecommitdiff
path: root/libavcodec/x86
Commit message (Collapse)AuthorAge
* snowdsp: explicitily state instruction size.Ronald S. Bultje2012-05-02
| | | | Fixes a compile error with clang at -O0.
* dsputil x86: revert a test back to its previous valueChristophe GISQUET2012-04-28
| | | | | | Commit 356ee8d caused the initial inversion. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* rv34dsp x86: implement MMX2 inverse transformChristophe Gisquet2012-04-28
| | | | | | 141 cycles down to 51. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* h264: new assembly version of get_cabac for x86_64 with PICRoland Scheidegger2012-04-28
| | | | | | | | | | This adds a hand-optimized assembly version for get_cabac much like the existing one, but it works if the table offsets are RIP-relative. Compared to the non-RIP-relative version this adds 2 lea instructions and it needs one extra register. get_cabac() gets about 40% faster, for an overall speedup of about 5%. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* h264: use one table instead of several for cabac functionsRoland Scheidegger2012-04-28
| | | | | | | | | | | | | The reason is this is easier for PIC code (in particular on darwin...). Keep the old names as pointers (static in cabac_functions.h so gcc knows these are just immediate offsets) so the c code can nicely stay the same (alternatively could use offsets directly in the functions needing the tables). This should produce the same code as before with non-pic and better code (confirmed) with pic. The assembly uses the new table but still won't work for PIC case. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* h264: (trivial) remove unneeded macro argument in x86/cabac.hRoland Scheidegger2012-04-28
| | | | Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* Remove lowres video decodingMans Rullgard2012-04-21
| | | | | | | This feature is complex, of questionable utility, and slows down normal decoding. Signed-off-by: Mans Rullgard <mans@mansr.com>
* avcodec: remove AVCodecContext.dsp_maskMans Rullgard2012-04-21
| | | | | | | | This removes all references to AVCodecContext.dsp_mask and marks it for eviction at the next version bump. It has been superseded by av_set_cpu_flag_mask() which, unlike this field, works everywhere. Signed-off-by: Mans Rullgard <mans@mansr.com>
* h264: use proper PROLOGUE statement for a function using 8 registers.Ronald S. Bultje2012-04-16
| | | | Fixes crashes when using biweight on win64.
* dsputil: fix optimized emu_edge function on Win64.Ronald S. Bultje2012-04-13
| | | | | | | | Recent register allocation changes (x86inc.asm update) changed the register order and thus opcodes for the inner loops. One of them became >128bytes, which confuses other parts of this function where it jumps to fixed-offset positions to extend the edge by fixed amounts. A simple register change fixes this.
* ac3dsp: call femms/emms at the end of float_to_fixed24() for 3DNow and SSEJustin Ruggles2012-04-12
| | | | | | Fixes ac3-encode and eac3-encode FATE test failures with SSE2 disabled. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* h264: fix 10bit biweight functions after recent x86inc.asm fixes.Ronald S. Bultje2012-04-12
| | | | | This should have been updated in the x86inc.asm update, but was accidently forgotten.
* build: Consistently handle conditional compilation for all optimization OBJS.Diego Biurrun2012-04-12
|
* x86inc improvements for 64-bitHenrik Gramner2012-04-11
| | | | | | | | | | | | Add support for all x86-64 registers Prefer caller-saved register over callee-saved on WIN64 Support up to 15 function arguments Also (by Ronald S. Bultje) Fix up our asm to work with new x86inc.asm. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
* rv40dsp x86: use only one register, for both increment and loop counterChristophe GISQUET2012-04-10
| | | | | | Around 10 cycles faster for luma. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* rv40dsp: implement prescaled versions for biweight.Christophe GISQUET2012-04-10
| | | | | | | | | | Quite often, the original weights are multiple of 512. By prescaling them by 1/512 when they are computed (once per frame), no intermediate shifting is needed, and no prescaling on each call either. The x86 code already used that trick. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* dsputil x86: use SSE float instruction instead of SSE2 integer equivalentChristophe GISQUET2012-04-04
| | | | | | All the more required since the users are pure SSE functions. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* dsputil x86: remove deprecated parameter from scalarproduct_int16 prototypeChristophe GISQUET2012-04-04
| | | | Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* vp8dsp x86: perform rounding shift with a single instructionChristophe GISQUET2012-04-04
| | | | Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* cabac: add overread protection to BRANCHLESS_GET_CABAC().Ronald S. Bultje2012-03-28
| | | | Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
* cabac: increment jump locations by one in callers of BRANCHLESS_GET_CABAC().Ronald S. Bultje2012-03-28
|
* cabac: remove unused argument from BRANCHLESS_GET_CABAC_UPDATE().Ronald S. Bultje2012-03-28
|
* cabac: use struct+offset instead of memory operand in BRANCHLESS_GET_CABAC().Ronald S. Bultje2012-03-28
|
* h264: add overread protection to get_cabac_bypass_sign_x86().Ronald S. Bultje2012-03-28
|
* h264: reindent get_cabac_bypass_sign_x86().Ronald S. Bultje2012-03-28
|
* h264: use struct offsets in get_cabac_bypass_sign_x86().Ronald S. Bultje2012-03-28
|
* build: prettyprinting cosmeticsDiego Biurrun2012-03-26
|
* x86: dsputil: prettyprint gcc inline asmDiego Biurrun2012-03-25
|
* x86: K&R prettyprinting cosmetics for dsputil_mmx.cDiego Biurrun2012-03-25
|
* x86: conditionally compile H.264 QPEL optimizationsDiego Biurrun2012-03-25
|
* dsputil_mmx: Surround QPEL macros by "do { } while (0);" blocks.Diego Biurrun2012-03-25
| | | | This makes them safe to use in non-fully braced if-blocks and similar.
* aacsbr: handle m_max values smaller than 4.Ronald S. Bultje2012-03-23
| | | | | | | | Prevents a signflip in the counter, and a subsequent crash because of overreads/overwrites. Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind CC: libav-stable@libav.org
* vp8: convert mbedge loopfilter x86 assembly to use named arguments.Ronald S. Bultje2012-03-10
|
* vp8: convert inner loopfilter x86 assembly to use named arguments.Ronald S. Bultje2012-03-10
|
* sbrdsp.asm: convert all instructions to float/SSE ones.Reimar Döffinger2012-03-07
| | | | | | | | | | | Since the values are floats, using the float operations makes sense, improves performance on some CPUs and makes the code SSE compatible instead of needing SSE2. Based on suggestion by Jason. Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de> Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* dsputil: remove shift parameter from scalarproduct_int16Christophe GISQUET2012-03-07
| | | | | | | | | There is only one caller, which does not need the shifting. Other use cases are situations where different roundings would be needed. The x86 and neon versions are modified accordingly. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* x86: Remove duplicated AVG_3DNOW_OP / AVG_MMX2_OP macros from h264_qpel_mmx.c.Diego Biurrun2012-03-07
|
* SBR DSP: fix SSE code to not use SSE2 instructions.Reimar Döffinger2012-03-06
| | | | | | | | movq from SSE register _to_ memory is an SSE2 instruction. Use the SSE movlps function instead that does the same thing. Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de> Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* x86: clean up ff_dsputil_init_mmx()Mans Rullgard2012-03-05
| | | | | | | | This splits ff_dsputil_init_mmx() into multiple functions, one for each MMX/SSE level, somewhat simplifying the nested conditions. Signed-off-by: Mans Rullgard <mans@mansr.com> Signed-off-by: Diego Biurrun <diego@biurrun.de>
* vp8: convert simple loopfilter x86 assembly to use named arguments.Ronald S. Bultje2012-03-03
|
* vp8: convert idct x86 assembly to use named arguments.Ronald S. Bultje2012-03-03
|
* vp8: convert mc x86 assembly to use named arguments.Ronald S. Bultje2012-03-03
|
* vp8: convert loopfilter x86 assembly to use cpuflags().Ronald S. Bultje2012-03-03
|
* vp8: convert idct/mc x86 assembly to use cpuflags().Ronald S. Bultje2012-03-03
|
* h264: change underread for 10bit QPEL to overread.Ronald S. Bultje2012-03-02
| | | | | This prevents us from reading before the start of the buffer, and thus prevents crashes resulting from this behaviour. Fixes bug 237.
* vp8: disable mmx functions with sse/sse2 counterparts on x86-64.Ronald S. Bultje2012-03-02
| | | | | x86-64 is guaranteed to have at least SSE2, therefore the MMX/MMX2 functions will never be used in practice.
* vp8: change int stride to ptrdiff_t stride.Ronald S. Bultje2012-03-02
| | | | | On 64bit platforms with 32bit int, this means we won't have to sign- extend the integer anymore.
* h264: fix mmxext chroma deblock to use correct TC values.Ronald S. Bultje2012-02-27
|
* SBR DSP x86: implement SSE sbr_hf_g_filtChristophe GISQUET2012-02-23
| | | | | | | | | | | | | | | Unrolling the main loop to process, instead of 4 elements: - 8: minor gain of 2 cycles (not worth the extra object size) - 2: loss of 8 cycles. Assigning STEP to a register is a loss. Output address (Y) is almost always unaligned. Timings: - C (32/64 bits): 117/109 cycles - SSE: 57 cycles Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
* SBR DSP x86: implement SSE sbr_sum_square_sseChristophe GISQUET2012-02-23
| | | | | | | | | | | | | The 32bits targets have been compiled with -mfpmath=sse for proper reference. sbr_sum_square C /32bits: 82c (unrolled)/102c C /64bits: 69c (unrolled)/82c SSE/32bits: 42c SSE/64bits: 31c Use of SSE4.1 dpps to perform the final sum is slower. Not unrolling to perform 8 operations in a loop yields 10 more cycles. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-30 11:13:36 +0000