Gentoo Archives: gentoo-council

From: Zac Medico <zmedico@g.o>
To: Ciaran McCreesh <ciaran.mccreesh@××××××××××.com>
Cc: Denis Dupeyron <calchan@g.o>, Ulrich Mueller <ulm@g.o>, gentoo-council@l.g.o
Subject: Re: [gentoo-council] mtime preservation
Date: Sat, 21 Nov 2009 02:01:44
Message-Id: 4B07362D.2010108@gentoo.org
In Reply to: Re: [gentoo-council] mtime preservation by Ciaran McCreesh
1 Ciaran McCreesh wrote:
2 > On Thu, 19 Nov 2009 17:14:20 -0700
3 > Denis Dupeyron <calchan@g.o> wrote:
4 >> On Thu, Nov 19, 2009 at 3:12 PM, Ciaran McCreesh
5 >> <ciaran.mccreesh@××××××××××.com> wrote:
6 >>> I honestly don't think we should be going anywhere with this until
7 >>> Portage is using a mechanism that guarantees correct sub-second
8 >>> resolution preservation in all cases.
9 >> Any idea, or even the beginning of one, of how to make that happen?
10 >
11 > My understanding is that someone with Python knowledge would have to
12 > write an external C library that provides a replacement for stat and a
13 > wrapper for either utimensat or futimens.
14
15 Alternatively, we could simply make portage spawn the mv binary
16 whenever rename fails (it fails when the source and destination are
17 on different devices). Although it's relatively slow, it should
18 solve the problem.
19 --
20 Thanks,
21 Zac