|Subject:||Re: [PATCH] dd: sparse conv flag|
|Date:||Tue, 26 Jul 2011 12:13:17 +0400|
|User-agent:||Thunderbird 220.127.116.11 (X11/20100331)|
Pádraig Brady wrote:
Thank you for appreciations.On 25/07/11 21:45, Eric Blake wrote:On 07/23/2011 10:08 AM, Roman Rybalko wrote:--- doc/coreutils.texi | 4 +++ src/dd.c | 22 +++++++++++++++- tests/Makefile.am | 1 + tests/dd/sparse | 70 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 96 insertions(+), 1 deletions(-) create mode 100755 tests/dd/sparseThanks for taking the time to do this. I agree that since BSD dd understands a sparse option, that it makes sense to teach coreutils dd this option as well. However, the size of your patch is non-trivial; would you be willing to start the copyright assignment process to give rights to your patch to the FSF?I agree that this would fit well. I wonder should be try to use the efficient sparse copy code done for cp? I guess that could be done after this in any case.
Sadly that that guy http://lists.gnu.org/archive/html/bug-coreutils/2007-12/msg00200.html didn't manage the work done.
I'm not against copyright assignment or whatever to be done to merge my patch to the master branch. Please, mail private emails to romanr.info
BTW, I don't think that my sparse code is efficient, I doubt it will stuck on large zero-filled files (like rsync does). But at least it just works.
PS: there is a bug, I've fixed that and do heavy testing, I'll post a fixed patch in a few days when I'm sure (actually I've already done for preview, to the http://debbugs.gnu.org/cgi/bugreport.cgi?bug=9157 that was automatically(?) opened for me)
-- WBR, Roman Rybalko
|[Prev in Thread]||Current Thread||[Next in Thread]|