Lines Matching +full:- +full:- +full:add +full:- +full:architecture
5 -----------------
6 If you plan to do non-x86 architecture specific optimizations (SIMD normally),
10 If you want to do x86 optimizations then you can either try to fine-tune the
16 --------------------------------------------
19 architecture-specific versions. It is recommended to look at older
22 Alternatively, look into the other architecture-specific versions in
28 (http://lists.ffmpeg.org/mailman/listinfo/ffmpeg-devel)
32 ----------------------------------
36 none of the following factors get worse due to an optimization -- speed,
37 binary code size, source size, source readability -- and at least one
45 -----------------------------------
53 Used in motion compensation of B-frames.
63 Used in motion estimation (encoding) with SAD of MPEG-4 4MV only.
72 Used in MPEG-4 qpel motion compensation (encoding & decoding).
74 the avg_* functions are used only for B-frames.
92 Used for MPEG-4 gmc.
97 Used for chroma blocks in MPEG-4 gmc with 1 warp point
115 Note, some optimized IDCTs have the add/put clamped code included and
131 Used in MPEG-1 en/decoding.
134 Used in MPEG-2 en/decoding.
137 Used in MPEG-4/H.263 en/decoding.
149 -------------
163 particularly important on Win64, where xmm6-15 are callee-save, and not
169 If gcc is not set to support sse (-msse) it will not accept xmm registers
181 - first of all, you're assuming that the compiler will not use xmm7 in
183 a poor assumption that will break at some point for some --cpu compiler flag
184 - secondly, you didn't mark xmm7 as clobbered. If you did, the compiler would
195 expansions, add DBG=1 to your make command-line: the input file will be
200 ---------------------------
205 - if your code is intended to be inlined in a C function, inline asm is always
207 - if your code calls external functions, external asm is always better
208 - if your code takes huge and complex structs as function arguments (e.g.
213 - in many cases, both can be used and it just depends on the preference of the
217 - if, for some reason, you believe that a particular chunk of existing external
219 way around), then please make the move from external asm <-> inline asm a
227 x86-specific:
228 -------------
231 The IA-32 Intel Architecture Software Developer's Manual, Volume 2:
238 http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/22007.pdf
241 ARM-specific:
242 -------------
243 ARM Architecture Reference Manual (up to ARMv5TE):
246 Procedure Call Standard for the ARM Architecture:
258 PowerPC-specific:
259 -----------------
267 http://www-01.ibm.com/chips/techlib/techlib.nsf/techdocs/30B3520C93F437AB87257060006FFE5E/$file/Lan…
268 http://www-01.ibm.com/chips/techlib/techlib.nsf/techdocs/9F820A5FFA3ECE8C8725716A0062585F/$file/CBE…
271 --------------
273 http://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html
275 a bit old (note "+" is valid for input-output, even though the next disagrees)
276 http://www.cs.virginia.edu/~clc5q/gcc-inline-asm.pdf