[x265] [PATCH] testbench: Added stress test cases for chroma_pp, chroma_ps and chroma_hps filter functions

[chen]

At 2014-02-03 19:07:31,nabajit at multicorewareinc.com wrote:
># HG changeset patch
># User Nabajit Deka
># Date 1391425643 -19800
>#      Mon Feb 03 16:37:23 2014 +0530
># Node ID 89b7060e631754de11577dbd1cab735d0df6df7e
># Parent  ae56333a326830d07ee2f7a25c2a5939154888bf
>testbench: Added stress test cases for chroma_pp, chroma_ps and chroma_hps filter functions
>
>diff -r ae56333a3268 -r 89b7060e6317 source/test/ipfilterharness.cpp
>--- a/source/test/ipfilterharness.cpp Mon Feb 03 16:06:24 2014 +0530
>+++ b/source/test/ipfilterharness.cpp Mon Feb 03 16:37:23 2014 +0530
>@@ -151,28 +151,31 @@
> 
> bool IPFilterHarness::check_IPFilterChroma_primitive(filter_pp_t ref, filter_pp_t opt)
> {
>-    int rand_srcStride, rand_dstStride, rand_coeffIdx;
>+    int rand_srcStride, rand_dstStride;
> 
>-    for (int i = 0; i <= 100; i++)
>+    for (int i = 0; i < ITERS; i++)
>     {
>-        rand_coeffIdx = rand() % 8;                // Random coeffIdex in the filter
>+        int index = i % TEST_CASES;
+        for (int coeffIdx = 0; coeffIdx < 8; coeffIdx++)


why combo TEST_CASES and coeffIdx, I think every specially case have exact coeffIdx value.
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20140203/d37dc7dd/attachment.html>