Re: [PATCH] history locking with fcntl

[Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx>]

On Apr 18,  7:31pm, Wayne Davison wrote:
}
} On Fri, Apr 18, 2008 at 02:59:59AM +0200, Vincent Lefevre wrote:
} > If I do not set the option (which is equivalent to not using the patch),
} > then I get immediate history corruption when using two machines (well,
} > at least 64-bit ones), 100% reproducible.
} 
} Cool.  Then it will be easy for you to get the real error now

I don't think there is a "real error" in the sense of a -1 return from
a system call and/or an errno value.  I think everything appears to
succeed as seen from the C code, but the file is corrupted anyway.

I hope (in some sense) that I turn out to be wrong.