RE: [RFC][PATCH] fix csum_and_copy_..._user() idiocy. Re: AW: [PATCH] amd64: Fix csum_partial_copy_generic()

[David Laight]

From: Al Viro
> Sent: 22 October 2023 20:40
....
> We need a way for csum_and_copy_{from,to}_user() to report faults.
> The approach taken back in 2020 (avoid 0 as return value by starting
> summing from ~0U, use 0 to report faults) had been broken; it does
> yield the right value modulo 2^16-1, but the case when data is
> entirely zero-filled is not handled right.  It almost works, since
> for most of the codepaths we have a non-zero value added in
> and there 0 is not different from anything divisible by 0xffff.
> However, there are cases (ICMPv4 replies, for example) where we
> are not guaranteed that.
> 
> In other words, we really need to have those primitives return 0
> on filled-with-zeroes input.  So let's make them return a 64bit
> value instead; we can do that cheaply (all supported architectures
> do that via a couple of registers) and we can use that to report
> faults without disturbing the 32bit csum.

Does the ICMPv4 sum need to be zero if all zeros but 0xffff
if there are non-zero bytes in there?

IIRC the original buggy case was fixed by returning 0xffff
for the all-zero buffer.

Even if it does then it would seem more sensible to have the
checksum function never return zero, csum_and_copy() return
zero on fault and add extra code to the (unusual) ICMP reply
code to detect 0xffff and convert to zero if the buffer is
all zeros.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)