ms-20240504.cu
Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| ms-20240504.cu [2024/06/19 22:12] – mino | ms-20240504.cu [2024/10/16 15:05] (current) – mino | ||
|---|---|---|---|
| Line 20: | Line 20: | ||
| sys | sys | ||
| - | == Note added on 2024.06.19 == | + | == Note added on 2024.06.20 == |
| - | The function void LastCols( ) in the code contains the following | + | The function void LastCols( ) in the code contains the following |
| psingl = psingl | psingl = psingl | ||
| or rev_max32_c == v2 | or rev_max32_c == v2 | ||
| - | | + | |
| - | It should be replaced by | + | which should be replaced by |
| psingl = psingl | psingl = psingl | ||
| or rev_max32_c == unsigned( v2 ) | or rev_max32_c == unsigned( v2 ) | ||
| - | because the upper 32 bits of v2 should be ignored | + | because the upper 32 bits of v2 should be ignored |
| - | However, this flaw is irrelevant to the validity of the result because this comparison | + | However, this flaw is irrelevant to the validity of the result because this logical operation |
| - | For odd N, this flaw can be relevant, but for N < 6, upper 32 bits of v2 are always 0s and the cast has no effect in any way. You have to be careful if you want to use this code for N=7. | + | |
| + | This part of the code and other parts related | ||
| + | We will gain 3% improvement of the performance by this optimization. | ||
| <file cpp ms-20240504.cu> | <file cpp ms-20240504.cu> | ||
ms-20240504.cu.1718802778.txt.gz · Last modified: 2024/06/19 22:12 by mino
