Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 30 Jan 2009 03:38:00 -0800
From:      Maxim Sobolev <sobomax@FreeBSD.org>
To:        Peter Jeremy <peterjeremy@optushome.com.au>
Cc:        Tim Kientzle <kientzle@FreeBSD.org>, "'current@FreeBSD.org'" <current@FreeBSD.org>
Subject:   Re: RFC: Change mtree nsec handling?
Message-ID:  <4982E698.1090204@FreeBSD.org>
In-Reply-To: <20090130112307.GJ1755@server.vk2pj.dyndns.org>
References:  <49829D49.10306@freebsd.org> <4982CBC7.5050102@FreeBSD.org> <20090130112307.GJ1755@server.vk2pj.dyndns.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Peter Jeremy wrote:
> On 2009-Jan-30 01:43:35 -0800, Maxim Sobolev <sobomax@freebsd.org> wrote:
>> Tim Kientzle wrote:
>>> For example, a timestamp of 1233295862.000001
>>> (1233295682 seconds and 1000 nanoseconds)
>>> will be printed like this by mtree:
>>>    time=1233295862.1000
>>> Unsurprisingly, the mtree parsing works the same
>>> way in reverse.
>> Given the age of mtree(8) I guess there are lot of existing mtree specs 
>> out there who rely on this behavior.
> 
> The existing code to read nanoseconds will handle either the old
> format or a %09d format (the for() loop that Tim added is unnecessary)
> so existing specs won't have a problem.  I think adding leading zeroes
> is the correct way to proceed.

My point is that it would not restore correct timestamp, not that it 
would not read it. The 1233295862.000001 before change would become 
1233295862.1000 after. I don't know how important is it, but I can 
imagine some applications where it could be an issue (e.g. incremental 
backup).

-Maxim



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4982E698.1090204>