On 10/18/2016 2:27 AM, Petter Reinholdtsen wrote:
> [Petter Reinholdtsen]
>> This make me believe libeatmydata should change to behave according to the
>> POSIX specification.
>
> And just to make it more clear why I believe this. My understanding of
> eatmydata
> is that it should make the
[Petter Reinholdtsen]
> This make me believe libeatmydata should change to behave according to the
> POSIX specification.
And just to make it more clear why I believe this. My understanding of
eatmydata
is that it should make the computer behave as normal while ignoring
any calls to force file
Perhaps most interesting is the POSIX specification, which state
that both fsync() and fdatasync() should return EBADF when passed
an invalid file descriptor not open for writing.
See
http://pubs.opengroup.org/onlinepubs/009695399/functions/fsync.html > and
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
On 12/13/2014 09:12 AM, Mattia Rizzolo wrote:
On Fri, 15 Feb 2013 14:44:55 -0500 Phillip Susi ps...@ubuntu.com
wrote:
I think this is a bug in gnulib rather than eatmydata. fsync()
can't be expected to report errors that may have happened when
Control: reassign -1 libeatmydata1
On Fri, 15 Feb 2013 14:44:55 -0500 Phillip Susi ps...@ubuntu.com wrote:
I think this is a bug in gnulib rather than eatmydata. fsync() can't
be expected to report errors that may have happened when it doesn't
actually try to do anything. In other words,
5 matches
Mail list logo
