Re: [PATCH net-next] net: tipc: fix FB_MTU eat two pages

On Wed, Jun 09, 2021 at 03:34:33AM -0400, Jon Maloy wrote:

[...]

It seems like I have been misleading you. It turns out that these messages
*will* be sent out over the nework in some cases, i.e. at
multicast/broadcast over an UDP bearer.
So, what we need is two macros, one with the conditional crypto
head/tailroom defined as you first suggested, and one that only use the
non-crypto head/tailroom as we have been discussing now.
The first one can be defined inside bcast.c, the latter  inside msg.c.
It might also be a good idea to give the macros more descriptive names, such
as ONEPAGE_MTU in the broadcast version, and ONEPAGE_SKB in the node local
version.

Does that make sense?

I think it's another point which can be optimized. However, with
CONFIG_TIPC_CRYPTO=y, the BUF_HEADROOM used in tipc_buf_acquire() is
always the crypto version, so it donen't work to define FB_MTU with
non-crypto BUF_HEADROOM.

So the point is to make a non-crypto version tipc_buf_acquire(), which
is used to alloc data for non-crypto message with non-crypto
BUF_HEADROOM. And that require us to distinguish the message that don't
crypto. Is there simple way to achieve this goal?

(Btw, I resended the patches for the 'two pages' problems.)

Thanks!
Menglong Dong

quoted