optimization Tips (for libavcodec):
+===================================
What to optimize:
-if u plan to do non-x86 architecture specific optimiztions (SIMD normally) then
-take a look in the i386/ directory, as most important functions are allready
-optimized for MMX
+-----------------
+If you plan to do non-x86 architecture specific optimizations (SIMD normally),
+then take a look in the x86/ directory, as most important functions are
+already optimized for MMX.
-if u want to do x86 optimizations then u can either try to finetune the stuff in the
-i386 directory or find some other functions in the c source to optimize, but there
-arent many left
-
-Understanding these overoptimized functions:
-as many functions, like the c ones tend to be a bit unreadable currently becouse
-of optimizations it is difficult to understand them (and write arichtecture
-specific versions, or optimize the c functions further) it is recommanded to look
-at older CVS versions of the interresting files (just use CVSWEB at
-(http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/ffmpeg/ffmpeg/libavcodec/))
-or perhaps look into the other architecture specific versions in i386/, ppc/,
-alpha/, ...; even if u dont understand the instructions exactly it could help
-understanding the functions & how they can be optimized
-
-NOTE:!!! if u still dont understand some function then ask at our mailing list!!!
-(http://lists.sourceforge.net/lists/listinfo/ffmpeg-devel)
+If you want to do x86 optimizations then you can either try to finetune the
+stuff in the x86 directory or find some other functions in the C source to
+optimize, but there aren't many left.
-
-wtf is that function good for ....:
-the primary purpose of that list is to avoid wasting time to optimize functions
-which are rarely used
+Understanding these overoptimized functions:
+--------------------------------------------
+As many functions tend to be a bit difficult to understand because
+of optimizations, it can be hard to optimize them further, or write
+architecture-specific versions. It is recommended to look at older
+revisions of the interesting files (web frontends for the various FFmpeg
+branches are listed at http://ffmpeg.org/download.html).
+Alternatively, look into the other architecture-specific versions in
+the x86/, ppc/, alpha/ subdirectories. Even if you don't exactly
+comprehend the instructions, it could help understanding the functions
+and how they can be optimized.
+
+NOTE: If you still don't understand some function, ask at our mailing list!!!
+(http://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel)
+
+
+When is an optimization justified?
+----------------------------------
+Normally, clean and simple optimizations for widely used codecs are
+justified even if they only achieve an overall speedup of 0.1%. These
+speedups accumulate and can make a big difference after awhile. Also, if
+none of the following factors get worse due to an optimization -- speed,
+binary code size, source size, source readability -- and at least one
+factor improves, then an optimization is always a good idea even if the
+overall gain is less than 0.1%. For obscure codecs that are not often
+used, the goal is more toward keeping the code clean, small, and
+readable instead of making it 1% faster.
+
+
+WTF is that function good for ....:
+-----------------------------------
+The primary purpose of this list is to avoid wasting time optimizing functions
+which are rarely used.
put(_no_rnd)_pixels{,_x2,_y2,_xy2}
- used in motion compensation (en/decoding)
+ Used in motion compensation (en/decoding).
avg_pixels{,_x2,_y2,_xy2}
- used in motion compensation of B Frames
- these are less important then the put*pixels functions
+ Used in motion compensation of B-frames.
+ These are less important than the put*pixels functions.
avg_no_rnd_pixels*
- unused
+ unused
pix_abs16x16{,_x2,_y2,_xy2}
- used in motion estimation (encoding) with SAD
+ Used in motion estimation (encoding) with SAD.
pix_abs8x8{,_x2,_y2,_xy2}
- used in motion estimation (encoding) with SAD of MPEG4 4MV only
- these are less important then the pix_abs16x16* functions
+ Used in motion estimation (encoding) with SAD of MPEG-4 4MV only.
+ These are less important than the pix_abs16x16* functions.
put_mspel8_mc* / wmv2_mspel8*
- used only in WMV2
- it is not recommanded that u waste ur time with these, as WMV2 is a
- ugly and relativly useless codec
+ Used only in WMV2.
+ it is not recommended that you waste your time with these, as WMV2
+ is an ugly and relatively useless codec.
mpeg4_qpel* / *qpel_mc*
- use in MPEG4 qpel Motion compensation (encoding & decoding)
- the qpel8 functions are used only for 4mv
- the avg_* functions are used only for b frames
- optimizing them should have a significant impact on qpel encoding & decoding
-
+ Used in MPEG-4 qpel motion compensation (encoding & decoding).
+ The qpel8 functions are used only for 4mv,
+ the avg_* functions are used only for B-frames.
+ Optimizing them should have a significant impact on qpel
+ encoding & decoding.
+
qpel{8,16}_mc??_old_c / *pixels{8,16}_l4
- just used to workaround a bug in old libavcodec encoder
- dont optimze them
+ Just used to work around a bug in an old libavcodec encoder version.
+ Don't optimize them.
tpel_mc_func {put,avg}_tpel_pixels_tab
- used only for SVQ3, so only optimze them if u need fast SVQ3 decoding
+ Used only for SVQ3, so only optimize them if you need fast SVQ3 decoding.
add_bytes/diff_bytes
- for huffyuv only, optimize if u want a faster ff-huffyuv codec
+ For huffyuv only, optimize if you want a faster ffhuffyuv codec.
get_pixels / diff_pixels
- used for encoding, easy
-
+ Used for encoding, easy.
+
clear_blocks
- easiest, to optimize
-
+ easiest to optimize
+
gmc
- used for mpeg4 gmc
- optimizing this should have a significant effect on the gmc decoding speed but
- its very likely impossible to write in SIMD
+ Used for MPEG-4 gmc.
+ Optimizing this should have a significant effect on the gmc decoding
+ speed.
gmc1
- used for chroma blocks in mpeg4 gmc with 1 warp point
- (there are 4 luma & 2 chroma blocks per macrobock, so
- only 1/3 of the gmc blocks use this, the other 2/3
- use the normal put_pixel* code, but only if there is
- just 1 warp point)
- Note: Divx5 gmc always uses just 1 warp point
+ Used for chroma blocks in MPEG-4 gmc with 1 warp point
+ (there are 4 luma & 2 chroma blocks per macroblock, so
+ only 1/3 of the gmc blocks use this, the other 2/3
+ use the normal put_pixel* code, but only if there is
+ just 1 warp point).
+ Note: DivX5 gmc always uses just 1 warp point.
pix_sum
- used for encoding
-
+ Used for encoding.
+
hadamard8_diff / sse / sad == pix_norm1 / dct_sad / quant_psnr / rd / bit
- specific compare functions used in encoding, it depends upon the command line
- switches which of these are used
- dont waste ur time with dct_sad & quant_psnr they arent really usefull
+ Specific compare functions used in encoding, it depends upon the
+ command line switches which of these are used.
+ Don't waste your time with dct_sad & quant_psnr, they aren't
+ really useful.
put_pixels_clamped / add_pixels_clamped
- used for en/decoding in the IDCT, easy
- Note, some optimized IDCTs have the add/put clamped code included and then
- put_pixels_clamped / add_pixels_clamped will be unused
+ Used for en/decoding in the IDCT, easy.
+ Note, some optimized IDCTs have the add/put clamped code included and
+ then put_pixels_clamped / add_pixels_clamped will be unused.
idct/fdct
- idct (encoding & decoding)
- fdct (encoding)
- difficult to optimize
-
+ idct (encoding & decoding)
+ fdct (encoding)
+ difficult to optimize
+
dct_quantize_trellis
- used for encoding with trellis quantization
- difficult to optimize
+ Used for encoding with trellis quantization.
+ difficult to optimize
dct_quantize
- used for encoding
-
+ Used for encoding.
+
dct_unquantize_mpeg1
- used in mpeg1 en/decoding
+ Used in MPEG-1 en/decoding.
dct_unquantize_mpeg2
- used in mpeg2 en/decoding
+ Used in MPEG-2 en/decoding.
dct_unquantize_h263
- used in mpeg4/h263 en/decoding
+ Used in MPEG-4/H.263 en/decoding.
FIXME remaining functions?
-btw, most of these are in dsputil.c/.h some are in mpegvideo.c/.h
+BTW, most of these functions are in dsputil.c/.h, some are in mpegvideo.c/.h.
+
-
Alignment:
-some instructions on some architectures have strict alignment restrictions,
-for example most SSE/SSE2 inctructios on X86
-the minimum guranteed alignment is writen in the .h files
-for example:
+Some instructions on some architectures have strict alignment restrictions,
+for example most SSE/SSE2 instructions on x86.
+The minimum guaranteed alignment is written in the .h files, for example:
void (*put_pixels_clamped)(const DCTELEM *block/*align 16*/, UINT8 *pixels/*align 8*/, int line_size);
+General Tips:
+-------------
+Use asm loops like:
+__asm__(
+ "1: ....
+ ...
+ "jump_instruction ....
+Do not use C loops:
+do{
+ __asm__(
+ ...
+}while()
+
+For x86, mark registers that are clobbered in your asm. This means both
+general x86 registers (e.g. eax) as well as XMM registers. This last one is
+particularly important on Win64, where xmm6-15 are callee-save, and not
+restoring their contents leads to undefined results. In external asm (e.g.
+yasm), you do this by using:
+cglobal functon_name, num_args, num_regs, num_xmm_regs
+In inline asm, you specify clobbered registers at the end of your asm:
+__asm__(".." ::: "%eax").
+If gcc is not set to support sse (-msse) it will not accept xmm registers
+in the clobber list. For that we use two macros to declare the clobbers.
+XMM_CLOBBERS should be used when there are other clobbers, for example:
+__asm__(".." ::: XMM_CLOBBERS("xmm0",) "eax");
+and XMM_CLOBBERS_ONLY should be used when the only clobbers are xmm registers:
+__asm__(".." :: XMM_CLOBBERS_ONLY("xmm0"));
+
+Do not expect a compiler to maintain values in your registers between separate
+(inline) asm code blocks. It is not required to. For example, this is bad:
+__asm__("movdqa %0, %%xmm7" : src);
+/* do something */
+__asm__("movdqa %%xmm7, %1" : dst);
+- first of all, you're assuming that the compiler will not use xmm7 in
+ between the two asm blocks. It probably won't when you test it, but it's
+ a poor assumption that will break at some point for some --cpu compiler flag
+- secondly, you didn't mark xmm7 as clobbered. If you did, the compiler would
+ have restored the original value of xmm7 after the first asm block, thus
+ rendering the combination of the two blocks of code invalid
+Code that depends on data in registries being untouched, should be written as
+a single __asm__() statement. Ideally, a single function contains only one
+__asm__() block.
+
+Use external asm (nasm/yasm) or inline asm (__asm__()), do not use intrinsics.
+The latter requires a good optimizing compiler which gcc is not.
+
+Inline asm vs. external asm
+---------------------------
+Both inline asm (__asm__("..") in a .c file, handled by a compiler such as gcc)
+and external asm (.s or .asm files, handled by an assembler such as yasm/nasm)
+are accepted in FFmpeg. Which one to use differs per specific case.
+
+- if your code is intended to be inlined in a C function, inline asm is always
+ better, because external asm cannot be inlined
+- if your code calls external functions, yasm is always better
+- if your code takes huge and complex structs as function arguments (e.g.
+ MpegEncContext; note that this is not ideal and is discouraged if there
+ are alternatives), then inline asm is always better, because predicting
+ member offsets in complex structs is almost impossible. It's safest to let
+ the compiler take care of that
+- in many cases, both can be used and it just depends on the preference of the
+ person writing the asm. For new asm, the choice is up to you. For existing
+ asm, you'll likely want to maintain whatever form it is currently in unless
+ there is a good reason to change it.
+- if, for some reason, you believe that a particular chunk of existing external
+ asm could be improved upon further if written in inline asm (or the other
+ way around), then please make the move from external asm <-> inline asm a
+ separate patch before your patches that actually improve the asm.
+
Links:
+======
http://www.aggregate.org/MAGIC/
-X86 specific:
+x86-specific:
+-------------
http://developer.intel.com/design/pentium4/manuals/248966.htm
-The IA-32 Intel Architecture Software Developer's Manual, Volume 2:
+The IA-32 Intel Architecture Software Developer's Manual, Volume 2:
Instruction Set Reference
http://developer.intel.com/design/pentium4/manuals/245471.htm
AMD Athlon Processor x86 Code Optimization Guide:
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/22007.pdf
+
+ARM-specific:
+-------------
+ARM Architecture Reference Manual (up to ARMv5TE):
+http://www.arm.com/community/university/eulaarmarm.html
+
+Procedure Call Standard for the ARM Architecture:
+http://www.arm.com/pdfs/aapcs.pdf
+
+Optimization guide for ARM9E (used in Nokia 770 Internet Tablet):
+http://infocenter.arm.com/help/topic/com.arm.doc.ddi0240b/DDI0240A.pdf
+Optimization guide for ARM11 (used in Nokia N800 Internet Tablet):
+http://infocenter.arm.com/help/topic/com.arm.doc.ddi0211j/DDI0211J_arm1136_r1p5_trm.pdf
+Optimization guide for Intel XScale (used in Sharp Zaurus PDA):
+http://download.intel.com/design/intelxscale/27347302.pdf
+Intel Wireless MMX2 Coprocessor: Programmers Reference Manual
+http://download.intel.com/design/intelxscale/31451001.pdf
+
+PowerPC-specific:
+-----------------
+PowerPC32/AltiVec PIM:
+www.freescale.com/files/32bit/doc/ref_manual/ALTIVECPEM.pdf
+
+PowerPC32/AltiVec PEM:
+www.freescale.com/files/32bit/doc/ref_manual/ALTIVECPIM.pdf
+
+CELL/SPU:
+http://www-01.ibm.com/chips/techlib/techlib.nsf/techdocs/30B3520C93F437AB87257060006FFE5E/$file/Language_Extensions_for_CBEA_2.4.pdf
+http://www-01.ibm.com/chips/techlib/techlib.nsf/techdocs/9F820A5FFA3ECE8C8725716A0062585F/$file/CBE_Handbook_v1.1_24APR2007_pub.pdf
+
+SPARC-specific:
+---------------
+SPARC Joint Programming Specification (JPS1): Commonality
+http://www.fujitsu.com/downloads/PRMPWR/JPS1-R1.0.4-Common-pub.pdf
+
+UltraSPARC III Processor User's Manual (contains instruction timings)
+http://www.sun.com/processors/manuals/USIIIv2.pdf
+
+VIS Whitepaper (contains optimization guidelines)
+http://www.sun.com/processors/vis/download/vis/vis_whitepaper.pdf
+
GCC asm links:
+--------------
official doc but quite ugly
http://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html
-a bit old (note "+" is valid for input-output, even though the next says its not)
+a bit old (note "+" is valid for input-output, even though the next disagrees)
http://www.cs.virginia.edu/~clc5q/gcc-inline-asm.pdf
-