From: "Timo Teräs" <timo.teras@iki•fi>
To: Herbert Xu <herbert@gondor•apana.org.au>
Cc: netdev@vger•kernel.org
Subject: Re: [PATCH 1/4] flow: virtualize flow cache entry methods
Date: Sun, 04 Apr 2010 08:50:41 +0300 [thread overview]
Message-ID: <4BB828B1.5090109@iki.fi> (raw)
In-Reply-To: <20100404020657.GA8520@gondor.apana.org.au>
Herbert Xu wrote:
> On Sat, Apr 03, 2010 at 11:19:04PM +0300, Timo Teräs wrote:
>> If someone is then removing a net driver, we still execute
>> flush on the 'device down' hook, and all stale bundles
>> get flushed.
>
> Not if the bundle belongs to a policy recently deleted.
>
>> But yes, this means that xfrm_policy struct can now be held
>> allocated up to ten extra minutes. But it's only memory that
>> it's holding, not any extra refs. And it's still reclaimable
>> by the GC.
>
> You also hold down the bundle xdst's along with it, which can
> hold netdev references preventing modules from being unloaded.
>
>> If this feels troublesome, we could add asynchronous flush
>> request that would be called on policy removal. Or even stick
>> to the synchronous one.
>
> How about change xfrm_flush_bundles to flush bundles from the
> cache instead of xfrm_policy?
For the common case:
1. Policy deleted; policy->walk.dead set, policy->genid incremented
2. NETDEV_DOWN hook called, calls flow_cache_flush()
3. flow_cache_flush enumerates all policy and bundle refs
in it's cache
4. for each bundle xfrm_bundle_check_fce() is called, which
calls stale_bundle()
5. all bundles using stale policy, fail that check because
xdst->policy_genid != xdst->pols[0]->genid
(checked in xfrm_bundle_ok)
6. flow cache calls entry's ->delete which is dst_free for bundles
7. flow_cache_flush() returns
flow_cache_flush really frees the bundles in it on flush.
But now that I look my code again. Your statement is true for
per-socket bundles. They would not get deleted in this case.
I'll change NETDEV_DOWN to call garbage collect instead of
flow cache flush which will then also free the per-socket bundles.
next prev parent reply other threads:[~2010-04-04 5:50 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-04-01 12:52 [PATCH 0/4] caching bundles, iteration 3 Timo Teras
2010-04-01 12:52 ` [PATCH 1/4] flow: virtualize flow cache entry methods Timo Teras
2010-04-01 13:05 ` Eric Dumazet
2010-04-01 13:07 ` Timo Teräs
2010-04-03 3:38 ` Herbert Xu
2010-04-03 8:36 ` Herbert Xu
2010-04-03 13:50 ` Timo Teräs
2010-04-03 14:17 ` Herbert Xu
2010-04-03 14:26 ` Timo Teräs
2010-04-03 15:53 ` Herbert Xu
2010-04-03 20:19 ` Timo Teräs
2010-04-04 2:06 ` Herbert Xu
2010-04-04 5:50 ` Timo Teräs [this message]
2010-04-04 5:58 ` Herbert Xu
2010-04-04 6:07 ` Timo Teräs
2010-04-04 6:19 ` Herbert Xu
2010-04-04 6:28 ` Timo Teräs
2010-04-04 8:35 ` Herbert Xu
2010-04-04 10:42 ` Herbert Xu
2010-04-04 10:50 ` Timo Teräs
2010-04-04 11:00 ` Herbert Xu
2010-04-04 11:06 ` Timo Teräs
2010-04-04 11:26 ` Herbert Xu
2010-04-04 11:31 ` Herbert Xu
2010-04-04 12:09 ` Timo Teräs
2010-04-01 12:52 ` [PATCH 2/4] xfrm: cache bundles instead of policies for outgoing flows Timo Teras
2010-04-01 12:52 ` [PATCH 3/4] xfrm: remove policy garbage collection Timo Teras
2010-04-01 12:52 ` [PATCH 4/4] flow: delayed deletion of flow cache entries Timo Teras
2010-04-02 3:00 ` [PATCH 0/4] caching bundles, iteration 3 David Miller
2010-04-02 13:12 ` Herbert Xu
-- strict thread matches above, loose matches on Subject: below --
2010-04-05 7:00 [PATCH 0/4] caching bundles, iteration 4 Timo Teras
2010-04-05 7:00 ` [PATCH 1/4] flow: virtualize flow cache entry methods Timo Teras
2010-04-05 8:33 ` Herbert Xu
2010-04-05 8:36 ` Timo Teräs
2010-04-05 8:44 ` Herbert Xu
2010-04-05 8:49 ` Herbert Xu
2010-04-05 8:53 ` Timo Teräs
2010-04-05 9:12 ` Herbert Xu
2010-04-05 17:01 ` Timo Teras
2010-04-06 12:34 ` Herbert Xu
2010-04-06 13:26 ` Timo Teräs
2010-04-07 9:15 ` David Miller
2010-04-07 10:30 [PATCH 0/4] caching bundles, iteration 5 Timo Teras
2010-04-07 10:30 ` [PATCH 1/4] flow: virtualize flow cache entry methods Timo Teras
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4BB828B1.5090109@iki.fi \
--to=timo.teras@iki$(echo .)fi \
--cc=herbert@gondor$(echo .)apana.org.au \
--cc=netdev@vger$(echo .)kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox