From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: [PATCH] poll: Avoid extra wakeups Date: Sun, 26 Apr 2009 12:46:39 +0200 Message-ID: <49F43B8F.2050907@cosmosbay.com> References: <49F3308B.1030507@cosmosbay.com> <20090426.020411.157511269.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: cl@linux.com, jesse.brandeburg@intel.com, netdev@vger.kernel.org, haoki@redhat.com, mchan@broadcom.com, davidel@xmailserver.org To: David Miller Return-path: Received: from gw1.cosmosbay.com ([212.99.114.194]:45450 "EHLO gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751012AbZDZKrp convert rfc822-to-8bit (ORCPT ); Sun, 26 Apr 2009 06:47:45 -0400 In-Reply-To: <20090426.020411.157511269.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: David Miller a =E9crit : >=20 > Great stuff Eric. >=20 > We've discussed splitting the wait queue up before, but shorter-term > your idea is pretty cool too :-) Well, I only got this idea because Davide did its previous work, he is the one who did the hard stuff :) About poll()/select() improvements, I believe following patch should be fine too. Note some lines on this patch are longer than 80 columns, I am aware of this but could not find an elegant/efficient way to avoid this. Thank you [PATCH] poll: Avoid extra wakeups in select/poll After introduction of keyed wakeups Davide Libenzi did on epoll, we are able to avoid spurious wakeups in poll()/select() code too. =46or example, typical use of poll()/select() is to wait for incoming network frames on many sockets. But TX completion for UDP/TCP=20 frames call sock_wfree() which in turn schedules thread. When scheduled, thread does a full scan of all polled fds and can sleep again, because nothing is really available. If number of fds is large, this cause significant load. This patch makes select()/poll() aware of keyed wakeups and useless wakeups are avoided. This reduces number of context switches by about 50% on some setups, and work performed by sofirq handlers. Signed-off-by: Eric Dumazet --- fs/select.c | 28 +++++++++++++++++++++++++--- include/linux/poll.h | 3 +++ 2 files changed, 28 insertions(+), 3 deletions(-) diff --git a/fs/select.c b/fs/select.c index 0fe0e14..2708187 100644 --- a/fs/select.c +++ b/fs/select.c @@ -168,7 +168,7 @@ static struct poll_table_entry *poll_get_entry(stru= ct poll_wqueues *p) return table->entry++; } =20 -static int pollwake(wait_queue_t *wait, unsigned mode, int sync, void = *key) +static int __pollwake(wait_queue_t *wait, unsigned mode, int sync, voi= d *key) { struct poll_wqueues *pwq =3D wait->private; DECLARE_WAITQUEUE(dummy_wait, pwq->polling_task); @@ -194,6 +194,16 @@ static int pollwake(wait_queue_t *wait, unsigned m= ode, int sync, void *key) return default_wake_function(&dummy_wait, mode, sync, key); } =20 +static int pollwake(wait_queue_t *wait, unsigned mode, int sync, void = *key) +{ + struct poll_table_entry *entry; + + entry =3D container_of(wait, struct poll_table_entry, wait); + if (key && !((unsigned long)key & entry->key)) + return 0; + return __pollwake(wait, mode, sync, key); +} + /* Add a new entry */ static void __pollwait(struct file *filp, wait_queue_head_t *wait_addr= ess, poll_table *p) @@ -205,6 +215,7 @@ static void __pollwait(struct file *filp, wait_queu= e_head_t *wait_address, get_file(filp); entry->filp =3D filp; entry->wait_address =3D wait_address; + entry->key =3D p->key; init_waitqueue_func_entry(&entry->wait, pollwake); entry->wait.private =3D pwq; add_wait_queue(wait_address, &entry->wait); @@ -418,8 +429,16 @@ int do_select(int n, fd_set_bits *fds, struct time= spec *end_time) if (file) { f_op =3D file->f_op; mask =3D DEFAULT_POLLMASK; - if (f_op && f_op->poll) + if (f_op && f_op->poll) { + if (wait) { + wait->key =3D POLLEX_SET; + if (in & bit) + wait->key |=3D POLLIN_SET; + if (out & bit) + wait->key |=3D POLLOUT_SET; + } mask =3D (*f_op->poll)(file, retval ? NULL : wait); + } fput_light(file, fput_needed); if ((mask & POLLIN_SET) && (in & bit)) { res_in |=3D bit; @@ -685,8 +704,11 @@ static inline unsigned int do_pollfd(struct pollfd= *pollfd, poll_table *pwait) mask =3D POLLNVAL; if (file !=3D NULL) { mask =3D DEFAULT_POLLMASK; - if (file->f_op && file->f_op->poll) + if (file->f_op && file->f_op->poll) { + if (pwait) + pwait->key =3D pollfd->events | POLLERR | POLLHUP; mask =3D file->f_op->poll(file, pwait); + } /* Mask out unneeded events. */ mask &=3D pollfd->events | POLLERR | POLLHUP; fput_light(file, fput_needed); diff --git a/include/linux/poll.h b/include/linux/poll.h index 8c24ef8..3327389 100644 --- a/include/linux/poll.h +++ b/include/linux/poll.h @@ -32,6 +32,7 @@ typedef void (*poll_queue_proc)(struct file *, wait_q= ueue_head_t *, struct poll_ =20 typedef struct poll_table_struct { poll_queue_proc qproc; + unsigned long key; } poll_table; =20 static inline void poll_wait(struct file * filp, wait_queue_head_t * w= ait_address, poll_table *p) @@ -43,10 +44,12 @@ static inline void poll_wait(struct file * filp, wa= it_queue_head_t * wait_addres static inline void init_poll_funcptr(poll_table *pt, poll_queue_proc q= proc) { pt->qproc =3D qproc; + pt->key =3D ~0UL; /* all events enabled */ } =20 struct poll_table_entry { struct file *filp; + unsigned long key; wait_queue_t wait; wait_queue_head_t *wait_address; };