Epoll - revision history

Dank at 07:48, 14 May 2023

2023-05-14T07:48:41Z

← Older revision		Revision as of 07:48, 14 May 2023
Line 1:		Line 1:
	I don't like epoll so much as [[FreeBSD\|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_</code> and <code>signalfd_</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer. EPOLLRDHUP was not added until 2.6.17.		I don't like epoll so much as [[FreeBSD\|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_</code> and <code>signalfd_</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer. EPOLLRDHUP was not added until 2.6.17.

			<tt>EPOLLEXCLUSIVE</tt> was added in Linux 4.5, and wakes up only one thread for events when multiple threads are blocking on epoll.

	==System Limits==		==System Limits==
	* <tt>/proc/sys/fs/epoll/max_user_watches</tt>, since Linux 2.6.28, "specifies a limit on the total number of file descriptors that a user can register across all epoll instances on the system. The limit is per real user ID. Each registered file descriptor costs roughly 90 bytes on a 32-bit kernel, and roughly 160 bytes on a 64-bit kernel. Currently, the default value for max_user_watches is 1/25 (4%) of the available low memory, divided by the registration cost in bytes."		* <tt>/proc/sys/fs/epoll/max_user_watches</tt>, since Linux 2.6.28, "specifies a limit on the total number of file descriptors that a user can register across all epoll instances on the system. The limit is per real user ID. Each registered file descriptor costs roughly 90 bytes on a 32-bit kernel, and roughly 160 bytes on a 64-bit kernel. Currently, the default value for max_user_watches is 1/25 (4%) of the available low memory, divided by the registration cost in bytes."

			==External links==
			* [https://idea.popcount.org/2017-02-20-epoll-is-fundamentally-broken-12/ Epoll is fundamentally broken] by Marek 2017-02-20

Dank at 01:00, 7 January 2010

2010-01-07T01:00:08Z

← Older revision		Revision as of 01:00, 7 January 2010
Line 1:		Line 1:
	I don't like epoll so much as [[FreeBSD\|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_</code> and <code>signalfd_</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer.		I don't like epoll so much as [[FreeBSD\|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_</code> and <code>signalfd_</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer. EPOLLRDHUP was not added until 2.6.17.

	==System Limits==		==System Limits==
	* <tt>/proc/sys/fs/epoll/max_user_watches</tt>, since Linux 2.6.28, "specifies a limit on the total number of file descriptors that a user can register across all epoll instances on the system. The limit is per real user ID. Each registered file descriptor costs roughly 90 bytes on a 32-bit kernel, and roughly 160 bytes on a 64-bit kernel. Currently, the default value for max_user_watches is 1/25 (4%) of the available low memory, divided by the registration cost in bytes."		* <tt>/proc/sys/fs/epoll/max_user_watches</tt>, since Linux 2.6.28, "specifies a limit on the total number of file descriptors that a user can register across all epoll instances on the system. The limit is per real user ID. Each registered file descriptor costs roughly 90 bytes on a 32-bit kernel, and roughly 160 bytes on a 64-bit kernel. Currently, the default value for max_user_watches is 1/25 (4%) of the available low memory, divided by the registration cost in bytes."

Dank at 02:58, 25 June 2009

2009-06-25T02:58:07Z

← Older revision		Revision as of 02:58, 25 June 2009
Line 1:		Line 1:
	I don't like epoll so much as [[FreeBSD\|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_</code> and <code>signalfd_</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer.		I don't like epoll so much as [[FreeBSD\|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_</code> and <code>signalfd_</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer.

			==System Limits==
			* <tt>/proc/sys/fs/epoll/max_user_watches</tt>, since Linux 2.6.28, "specifies a limit on the total number of file descriptors that a user can register across all epoll instances on the system. The limit is per real user ID. Each registered file descriptor costs roughly 90 bytes on a 32-bit kernel, and roughly 160 bytes on a 64-bit kernel. Currently, the default value for max_user_watches is 1/25 (4%) of the available low memory, divided by the registration cost in bytes."

WikiSysop: 1 revision

2009-01-28T07:32:13Z

1 revision

← Older revision	Revision as of 07:32, 28 January 2009
(No difference)

Dank at 05:39, 29 June 2008

2008-06-29T05:39:39Z

New page

I don't like epoll so much as [[FreeBSD|FreeBSD's]] [[kqueue]], although many of its deficiencies have been overcome with the [http://kerneltrap.org/2.6.25 2.6.25 kernel release] and its addition of <code>timerfd_*</code> and <code>signalfd_*</code> functionality ([http://lwn.net LWN] had a good article regarding <code>timerfd</code> [http://lwn.net/Articles/245533/ here]). Note that these latter functions require [[glibc]] version 2.8 or later, or use of the <code>syscall()</code> multiplexer.