On Thu, Nov 30, 2017 at 4:38 PM, Herbert Xu [off-list ref] wrote:

On Thu, Nov 30, 2017 at 04:03:02PM -0800, Tom Herbert wrote:

quoted

This function is like rhashtable_walk_next except that it only returns
the current element in the inter and does not advance the iter.

This patch also creates __rhashtable_walk_find_next. It finds the next
element in the table when the entry cached in iter is NULL or at the end
of a slot. __rhashtable_walk_find_next is called from
rhashtable_walk_next and rhastable_walk_peek.

Signed-off-by: Tom Herbert <redacted>

Hi Tom:

Could you add some motivation for this feature into the patch
description? As it is it's difficult to deduce why we would want
to add something like this given that hashtable walks are always
unstable and there is no guarantee that two calls to peek or a
peek followed by a normal walk will see the same entry.

Hi Herbert,

We don't need a guarantee of stability, but what I am seeing is that
we're consisitently dropping entries on when doing a multi-part
netlink walk. We start iterating over the table filling in the netlink
info. But eventually the netlink info fills up and returns an error.
netlink dump gets called again but now the iter of the table returns
the object following the one that would have overflowed the netlink
buffer. So the result I was seeing is that we dropped one object in in
each pass.

This fixes the nldump for ila which will be in a follow on patch set.
In pseudo code it looks something like this:

rhashtable_walk_start(rhiter);

/* Get first entty */
ila = rhashtable_walk_peek(rhiter);

while (ila) {
     if (ila_dump_info(ila) < 0)
         break;
     ila = rhashtable_walk_next(rhiter);
}

rhashtable_walk_stop(rhiter);

return;

So peek is only called one and we only advance iter once the current
entry is successfully processed.

Tom

Thanks,
--
Email: Herbert Xu [off-list ref]
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt