James Carlson wrote:
Rao Shoaib writes:
This is one of the reasons NFS does not use this approach. Infact
desballoca was implemented specifcially for NFS so that the buffer is
freed right away as opposed to esballoca which delays the freeing of the
buffer. However we ran into a file corruption issue. It turns out some
NIC drivers (do not remember which) wrote on the mblks.
That doesn't sound complete to me.
This implies that those drivers were writing to a dblk with db_ref >
1, as the minimum ref count for the *alloca() functions is 2.
Wouldn't that just be a fatal driver bug? If that were a problem, why
wouldn't we fix that bug?
Yes it's a driver bug. I am not sure why this was not pursued. This must
not be the case anymore or else sendfile would not work.
Rao.
_______________________________________________
networking-discuss mailing list
[email protected]
