Re: [bkpatch] add sys_sendfile64

[Benjamin LaHaise <bcrl@redhat.com>]

On Sun, Mar 03, 2002 at 07:10:23PM -0800, Chris Wedgwood wrote:
> We have a problem if off + count >= 2^32 here.
> 
> Ideally i think we need to check for 32-bit (31-bit?) overflow here
> and return -EOVERFLOW.  I made a similar patch last night for
> sendfile64 which included this check (although I was tired and the
> patch was slightly wrong).  Actually, I think wew are missing
> EOVERFLOW checks in a number of paths, ideally I'd like to make one
> function to check and have all other functions reference that if
> people agree that makes sense.

I was just following the semantics of the original code.  -EOVERFLOW 
checks are certainly doable; I'll post an update in a bit.

>     +	}
>     +	ret = common_sendfile(out_fd, in_fd, ppos, count);
>     +	if (offset)
>     +		put_user((off_t)pos, offset);
>     +	return ret;
> 
> What is another thread unmapped 'offset' during the system call?  Do
> we want to check the result of put_user here and return -EFAULT?
> (If so, there are other system calls to consider such as select).

Again, the original code didn't bother checking.  As far as how it 
should work, I'd rather send a segv to the app as otherwise it is 
impossible to determine how much data was actually transferred.

		-ben