- If OUTPUT_BLOCK_SIZE is negative, use a format like "127k" if
- possible, using powers of -OUTPUT_BLOCK_SIZE; otherwise, use
- ordinary decimal format. Normally -OUTPUT_BLOCK_SIZE is either
- 1000 or 1024; it must be at least 2. Most people visually process
- strings of 3-4 digits effectively, but longer strings of digits are
- more prone to misinterpretation. Hence, converting to an
- abbreviated form usually improves readability. Use a suffix
- indicating which power is being used. For example, assuming
- -OUTPUT_BLOCK_SIZE is 1024, 8500 would be converted to 8.3k,
- 133456345 to 127M, 56990456345 to 53G, and so on. Numbers smaller
- than -OUTPUT_BLOCK_SIZE aren't modified. */
+ If (OPTS & human_autoscale), deduce the output block size
+ automatically; TO_BLOCK_SIZE must be 1 but it has no effect on the
+ output. Use powers of 1024 if (OPTS & human_base_1024), and powers
+ of 1000 otherwise. For example, assuming powers of 1024, 8500
+ would be converted to 8.3, 133456345 to 127, 56990456345 to 53, and
+ so on. Numbers smaller than the power aren't modified.
+ human_autoscale is normally used together with human_SI.
+
+ If (OPTS & human_SI), append an SI prefix indicating which power is
+ being used. If in addition (OPTS & human_B), append "B" (if base
+ 1000) or "iB" (if base 1024) to the SI prefix. When ((OPTS &
+ human_SI) && ! (OPTS & human_autoscale)), TO_BLOCK_SIZE must be a
+ power of 1024 or of 1000, depending on (OPTS &
+ human_base_1024). */