2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
|
}
/*
* ----------------------------------------------------------------------
*
* NOTES --
*
* Some measurements show that it is faster to use a table to to perform
* uuencode and base64 value encoding than to calculate the output (at
* least on intel P4 arch).
*
* Conversely using a lookup table for the decoding is slower than just
* calculating the values. We therefore use the fastest of each method.
*
* Presumably this has to do with the size of the tables. The base64
|
|
|
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
|
}
/*
* ----------------------------------------------------------------------
*
* NOTES --
*
* Some measurements show that it is faster to use a table to perform
* uuencode and base64 value encoding than to calculate the output (at
* least on intel P4 arch).
*
* Conversely using a lookup table for the decoding is slower than just
* calculating the values. We therefore use the fastest of each method.
*
* Presumably this has to do with the size of the tables. The base64
|