[RFC] Copy as much as you can during a 32-bit stat before returning EOVERFLOW?

[Carlos O'Donell]

On 02/26/2013 04:00 PM, Roland McGrath wrote:
> The argument in favor of this API change seems quite thin.  An old
> program will have to be modified to accept EOVERFLOW failures, so why
> not modify it to use *64 interfaces or -D_FILE_OFFSET_BITS=64 instead?
> It may seem at first blush that the change would be simpler in complex
> programs.  But, in fact, to be robust when using older libcs a program
> would have to do something very special to distinguish a library call
> (new-style) that delivered some truncated values from one (old-style)
> that delivered some or all uninitialized fields.

A given users needs are far more focused. In practice they want to move
to a newer distribution or filesystem and keep down the cost of the 
upgrade while incrementally fixing applications.

> I don't see any defensible rationale for putting such a change into
> libc.

I agree. I didn't want to colour the conversation with my initial opinion,
but it seems like everyone is pretty well agreed that the change would
complicate the API without enough gain.

Cheers,
Carlos.