l...@gnu.org (Ludovic Courtès) writes:
> Mark H Weaver <m...@netris.org> skribis:
>
>> Since the code below will behave badly if 'c_count' does not fit in an
>> 'ssize_t', we should validate here that it _does_ fit.
>
> Oops, indeed. (Note that sendfile(2) and write(2) have that problem:
> they take a size_t and return a ssize_t...)
>
> There was also the other issue of making sure we use the right function,
> depending on _FILE_OFFSET_BITS & co.
>
> Here are the changes compared to the previous patch:
[...]
> WDYT?
Looks good to me! :)
Mark
