[Toybox] dd cleanup notes

Rob Landley rob at landley.net
Sat Jul 16 21:57:33 PDT 2016


On 07/15/2016 01:02 PM, James McMechan wrote:
> On Tue, Jul 12, 2016 at 11:03 AM, Rob Landley <rob at landley.net> wrote:
> ...
>> longstanding issue I blogged about recently at
>> http://landley.net/notes-2016.html#10-07-2016 and the tl;dr of which is
>> "long being 32 bits on 32 bit systems is kinda awkward". I might switch
> Slight glitch, no http://landley.net/notes-2016.html#10-07-2016 it
> seems to stop at 2016-06-29

Remembering to copy the file to the webserver: a thing.

(Although the real problem is entries tend to trail off halfway through
and I need to go check them and finish the unfinished ones before
uploading, and sometimes I forget to write anything for a week or more
and then have to go check my email and twitter and git commits and such
to remind myself what I did.)

It's up now. :)

> And the wonderful hassle of standards continues http://xkcd.com/927/,
> recall that K&R which was I believe first and for a long time the
> definitive book stated that pointers could always fit in a "int", it
> seemed to take forever to get stdint.h where you could get a int8_t
> without your own header...

The "32/64 bit" section of http://landley.net/toybox/design.html links
to the LP64 standard, rationale, and context.

> of course that is only needed for files or protocols shared with other
> computers ;)

No, in this case storage limits affect specifying large file sizes as
command line arguments on 32 bit platforms.

> I am feeling old, I remember the joy of upgrading to ANSI C (C89) ~27 years ago

Which predated the first 64 bit microprocessor (apparently a mips r4k
variant used by SGI in Irix workstations) by 2 years, and DEC Alpha by 3
years.

> What will be next... long long being int128_t?

Possibly, LP64 only says that's "at least" 64 bits.

> Jim

Rob


More information about the Toybox mailing list