rsync-3.0.2 -- two build problems

Peter Breitenlohner peb at
Thu Jun 5 06:22:09 GMT 2008

On Wed, 4 Jun 2008, Wayne Davison wrote:

> On Mon, Jun 02, 2008 at 07:31:02PM +0200, Peter Breitenlohner wrote:
>> (1) With the Linux kernel >=2.6.20, "make check" occasionally fails, due to
>> subsecond timestamps sometimes being truncated and sometimes being rounded
>> upwards (both on i686 and x86_64).
> Rsync doesn't do any rounding, so I don't know where the inconsistency
> would be.  We just set the desination time based on the mtime value, and
> the tls program outputs an interpretation of the mtime value.

Yesterday I did some tests (on jfs with subsecond timestamps, ext2 doesn't
have them):

True, rsync just truncates, i.e. ignores the fractional seconds, as do
freerdist-0.92 and (GNU) tar.

However, "cp -p" and "touch -r" round to the nearest second. So it seems the
failures are caused by comparing a directory created by "cp" with one
created by "rsync".  Since spurious failures from "make check" defeat the
purpose of a test suite, it might be a good idea to avoid such comparisions.

Peter Breitenlohner <peb at>

More information about the rsync mailing list