tech-toolchain archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: make timestamp granularity



On Sat, Feb 14, 2015 at 08:45:34PM +0100, Edgar Fuß wrote:
> Yes, I could read the code, but probably someone knows the answer off the top 
> of his head:
> It appears to me that make(1) compares timestamp only to seconds granularity.
> Is that true?
> Iff true, is this a feature?

Also, without looking at the code, only the man page: the operator
":" recreates if the "modification time" is newer.

_Portable_ stat/lstat/fstat(3) return structures contain a time_t,
which is counted in seconds; don't know if our make compares
the nanosecond field, too.

(Could be worse: OS-9's filesystem had modification times with a
granularity of err, two minutes? Mlelstv? So even pretty slow, by
today's standards, 68000 machines with OS-9/68k used to needlessly
recreate stuff on a make because the compiler became too fast.)

	-is
-- 
A medium apple... weighs 182 grams, yields 95 kcal, and contains no
caffeine, thus making it unsuitable for sysadmins. - Brian Kantor


Home | Main Index | Thread Index | Old Index