[Linux/Unix编程] nv30_fragtex

说明：handle base_level when not using a mip filter, min max level is unfortunately ignored by the hardware otherwise.this is a tad stupid of the hardware, but there s no non-rcomp z16 z24 texture formats to be had, we have to
<touwongyie> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] find_first_of_pred

说明：Also test lp_build_assert.
<mlpiuyw> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] tie_ostream

说明：Elliptic Cauer filters designed with 0.1 dB passband ripple, 80 dB minimum stopband attenuation, and [0.95 : 0.15 : 0.35] normalized cut off frequencies.
<yuelaolao> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] SkStrokeRec

说明：Specify the strokewidth, and optionally if you want stroke + fill.
<mzyiuqin> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] draw_pipe_twoside

说明：copy back attribs to front attribs.
<mzyiuqin> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] LPC_analysis_filter

说明：LPC analysis filter.Allowing wrap around so that two wraps can cancel each other. The rare cases where the result wraps around can only be triggered by invalid streams.
<juelvcp> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] algorithm_header_test

说明：12th order LPC analysis filter, does not write first 12 samples.
<juelvcp> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] BlurRectBench

说明：Use of this source code is governed by a BSD-style license that can be found in the LICENSE file.disable Gaussian benchmarks the algorithm works well enough and serves as a baseline for ground truth. -Use of this sourc
<dinzplz> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] SkColorTable

说明：SkColorTable holds an array SkPMColors (premultiplied 32-bit colors) used by 8-bit bitmaps, where the bitmap bytes are interpreted as indices into the colortable.
<litrgon> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] fpu_control

说明：FPU control word bits. x86-64 version.Note that this file sets on x86-64 only the x87 FPU, it does not touch the SSE unit.
<hnnjla> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] quant_levels

说明：For all x in [0.284], x155 100 is bitwise equal to (x101581) >> 16. The smallest precision for that is (x6349) >> 12 but 16 is a good word size.Maximum number of convergence steps.Start with uniformly spread
<byzingyang> 在 2025-02-22 上传 | 大小：2kb | 下载：0

[Linux/Unix编程] GrBlend

说明：vp8_quantize_mbX functions here differs corresponding ones in quantize.c only by using quantize_b_pair function pointer instead of the regular quantize_b function pointer. -vp8_quantize_mbX functions here differs corr
<belagx> 在 2025-02-22 上传 | 大小：2kb | 下载：0