From: Vladimir Davydov <vdavydov.dev@gmail.com> To: Serge Petrenko <sergepetrenko@tarantool.org> Cc: tarantool-patches@freelists.org Subject: Re: [tarantool-patches] [PATCH v2] iproto: introduce a proxy module. Date: Mon, 8 Oct 2018 22:45:15 +0300 [thread overview] Message-ID: <20181008194515.h5tgx7erp5ryjywp@esperanza> (raw) In-Reply-To: <20181002180554.1142-1-sergepetrenko@tarantool.org> On Tue, Oct 02, 2018 at 09:05:54PM +0300, Serge Petrenko wrote: > +static void > +proxy_forward_request(struct iproto_connection *con, struct iproto_msg *msg, > + struct proxy_connection *pcon) > +{ > + struct sync_hash_entry *cs = (struct sync_hash_entry *)malloc(sizeof(*cs)); > + cs->sync = msg->header.sync; > + cs->con = con; > + msg->header.sync = ++pcon->sync_ctr; > + struct mh_i64ptr_node_t node = {pcon->sync_ctr, cs}; > + if (mh_i64ptr_put(pcon->sync_hash, &node, NULL, NULL) == > + mh_end(pcon->sync_hash)) { > + diag_set(OutOfMemory, sizeof(node), "malloc", "sync_hash"); > + diag_raise(); > + } > + coio_write_xrow(&pcon->connection.io, &msg->header); > + > + /* > + * After forwarding the request, mark it as read and > + * delete the msg. > + */ > + msg->p_ibuf->rpos += msg->len; > + iproto_msg_delete(msg); > +} > +static int > +proxy_replier_f(va_list ap) > +{ > + struct proxy_connection *pcon = va_arg(ap, struct proxy_connection *); > + > + struct ev_io io; > + coio_create(&io, pcon->connection.io.fd); > + > + while (!fiber_is_cancelled()) { > + struct xrow_header row; > + coio_read_xrow(&io, &pcon->connection.ibuf, &row); > + uint64_t key = row.sync; > + mh_int_t i = mh_i64ptr_find(pcon->sync_hash, key, NULL); > + if (i == mh_end(pcon->sync_hash)) { > + /* > + * Some error. We recieved a reply with sync > + * not corresponding to any connection > + */ > + say_warn("sync recieved from remote instance is not in sync_hash"); > + continue; > + } > + struct mh_i64ptr_node_t *node = mh_i64ptr_node(pcon->sync_hash, i); > + if (row.type != IPROTO_CHUNK) { > + mh_i64ptr_remove(pcon->sync_hash, node, NULL); > + } > + struct sync_hash_entry * cs = (struct sync_hash_entry *)node->val; > + struct iproto_connection *con = cs->con; > + uint64_t sync = cs->sync; > + row.sync = sync; > + free(cs); I may be wrong, but I think that using a hash table for mapping client connection syncs to server connection syncs, like you do here, may hurt performance (think of thousands requests in flight). If we really want to multiplex, I'd rather consider extending xrow_header with something like connection_id. This might need some benchmarking though. > + coio_write_iproto_response(&con->output, &row); > + > + fiber_gc(); > + } > + return 0; > +}
next prev parent reply other threads:[~2018-10-08 19:45 UTC|newest] Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-10-02 18:05 Serge Petrenko 2018-10-03 8:49 ` Vladimir Davydov 2018-10-04 11:54 ` [tarantool-patches] " Georgy Kirichenko 2018-10-08 10:44 ` [tarantool-patches] " Vladimir Davydov 2018-10-16 18:35 ` [tarantool-patches] " Konstantin Osipov 2018-10-08 16:48 ` [tarantool-patches] " Vladimir Davydov 2018-10-16 18:39 ` [tarantool-patches] " Konstantin Osipov 2018-10-17 8:35 ` Vladimir Davydov 2018-10-17 15:31 ` Konstantin Osipov 2018-10-08 19:45 ` Vladimir Davydov [this message] 2018-10-16 18:42 ` Konstantin Osipov 2018-10-23 17:26 ` Konstantin Osipov
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=20181008194515.h5tgx7erp5ryjywp@esperanza \ --to=vdavydov.dev@gmail.com \ --cc=sergepetrenko@tarantool.org \ --cc=tarantool-patches@freelists.org \ --subject='Re: [tarantool-patches] [PATCH v2] iproto: introduce a proxy module.' \ /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
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox