From: Julian Anastasov <ja@ssi.bg>
To: Simon Horman <horms@kernel.org>
Cc: lvs-devel@vger.kernel.org, netfilter-devel@vger.kernel.org,
netdev@vger.kernel.org, Dust Li <dust.li@linux.alibaba.com>,
Jiejian Wu <jiejian@linux.alibaba.com>,
Jiri Wiesner <jwiesner@suse.de>
Subject: Re: [PATCHv2 RFC net-next 08/14] ipvs: use resizable hash table for services
Date: Wed, 17 Jan 2024 15:09:55 +0200 (EET) [thread overview]
Message-ID: <31a84455-86e7-c99f-7011-5385bbe0179d@ssi.bg> (raw)
In-Reply-To: <20240117095414.GL588419@kernel.org>
Hello,
On Wed, 17 Jan 2024, Simon Horman wrote:
> On Tue, Dec 12, 2023 at 06:24:38PM +0200, Julian Anastasov wrote:
> > Make the hash table for services resizable in the bit range of 4-20.
> > Table is attached only while services are present. Resizing is done
> > by delayed work based on load (the number of hashed services).
> > Table grows when load increases 2+ times (above 12.5% with factor=3)
> > and shrinks 8+ times when load decreases 16+ times (below 0.78%).
> >
> > Switch to jhash hashing to reduce the collisions for multiple
> > services.
> >
> > Add a hash_key field into the service that includes table ID and
> > bucket ID. This helps the lookup and delete operations.
> >
> > Signed-off-by: Julian Anastasov <ja@ssi.bg>
>
> ...
>
> > @@ -391,18 +440,29 @@ static inline struct ip_vs_service *
> > __ip_vs_service_find(struct netns_ipvs *ipvs, int af, __u16 protocol,
> > const union nf_inet_addr *vaddr, __be16 vport)
> > {
> > - unsigned int hash;
> > + DECLARE_IP_VS_RHT_WALK_BUCKET_RCU();
> > + struct hlist_bl_head *head;
> > struct ip_vs_service *svc;
> > -
> > - /* Check for "full" addressed entries */
> > - hash = ip_vs_svc_hashkey(ipvs, af, protocol, vaddr, vport);
> > -
> > - hlist_for_each_entry_rcu(svc, &ipvs->svc_table[hash], s_list) {
> > - if (svc->af == af && ip_vs_addr_equal(af, &svc->addr, vaddr) &&
> > - svc->port == vport && svc->protocol == protocol &&
> > - !svc->fwmark) {
> > - /* HIT */
> > - return svc;
> > + struct ip_vs_rht *t, *p;
> > + struct hlist_bl_node *e;
> > + u32 hash, hash_key;
> > +
> > + ip_vs_rht_for_each_table_rcu(ipvs->svc_table, t, p) {
> > + /* Check for "full" addressed entries */
> > + hash = ip_vs_svc_hashval(t, af, protocol, vaddr, vport);
> > +
> > + hash_key = ip_vs_rht_build_hash_key(t, hash);
> > + ip_vs_rht_walk_bucket_rcu(t, hash_key, head) {
> > + hlist_bl_for_each_entry_rcu(svc, e, head, s_list) {
>
> Hi Julian,
>
> Smatch thinks that head is used uninitialised here.
> This does seem to be the case to me too.
It would crash and not find svc if that was true. May be it is
caused by the complex 'if (INIT then FAIL) {} else ANYTHING'
construct used in ip_vs_rht_walk_bucket_rcu() which allows
'if () {} else for () for () for () ...'.
Regards
--
Julian Anastasov <ja@ssi.bg>
next prev parent reply other threads:[~2024-01-17 13:10 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-12 16:24 [PATCHv2 RFC net-next 00/14] ipvs: per-net tables and optimizations Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 01/14] rculist_bl: add hlist_bl_for_each_entry_continue_rcu Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 02/14] ipvs: make ip_vs_svc_table and ip_vs_svc_fwm_table per netns Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 03/14] ipvs: some service readers can use RCU Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 04/14] ipvs: use single svc table Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 05/14] ipvs: do not keep dest_dst after dest is removed Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 06/14] ipvs: use more counters to avoid service lookups Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 07/14] ipvs: add resizable hash tables Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 08/14] ipvs: use resizable hash table for services Julian Anastasov
2024-01-17 9:54 ` Simon Horman
2024-01-17 13:09 ` Julian Anastasov [this message]
2023-12-12 16:24 ` [PATCHv2 RFC net-next 09/14] ipvs: switch to per-net connection table Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 10/14] ipvs: show the current conn_tab size to users Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 11/14] ipvs: no_cport and dropentry counters can be per-net Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 12/14] ipvs: use more keys for connection hashing Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 13/14] ipvs: add ip_vs_status info Julian Anastasov
2023-12-12 16:24 ` [PATCHv2 RFC net-next 14/14] ipvs: add conn_lfactor and svc_lfactor sysctl vars Julian Anastasov
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=31a84455-86e7-c99f-7011-5385bbe0179d@ssi.bg \
--to=ja@ssi.bg \
--cc=dust.li@linux.alibaba.com \
--cc=horms@kernel.org \
--cc=jiejian@linux.alibaba.com \
--cc=jwiesner@suse.de \
--cc=lvs-devel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=netfilter-devel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).