Talk:Call gate (Intel)

Talk
This was previously a transwikied dicdef; I have now fleshed it out into a proper stub: there is lots more that can go into this article regarding computer security, malware, and different implementations of the x86 architecture. -- The Anome 11:38, 21 July 2006 (UTC)

Modern Use
"With the introduction of SYSENTER/SYSEXIT and SYSCALL/SYSRET, a new faster mechanism was introduced for control transfers for x86 programs"

"Upon comparing call gates to interrupts, call gates are significantly faster."

2 methods... each is faster than the other. —Preceding unsigned comment added by 174.3.135.28 (talk) 14:03, 9 March 2011 (UTC)
 * Software interrupts are a third method of system call, although that last sentence should be investigated. Abubalay (talk) 18:14, 12 April 2011 (UTC)

Every generation believes that it invented sex
The concept of a call gate is not unique to the Intel x86 architecture and did not originate there; it goes back at least as far as Multics. The article should either be renamed, e.g., to Call gate (Intel), or rewritten to comply with Neutral point of view. Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:51, 2 May 2016 (UTC).

Reasons for call gates
The original Multics implementation of protection rings required that the Dynamic linker handle a segment fault for the first invocation of every routine. Call gates allowed eliminating that overhead. Shmuel (Seymour J.) Metz Username:Chatul (talk)
 * How does it do so on Multics? Or are you saying that this is a consequence of the x86 implementatino of call gates?
 * The primary reason for call gates is to disallow code running in an outer ring from jumping to arbitrary code that would run in an inner ring; see "Gate Address Validation" on page 144 in section B "Access Control to the Multics Virtual Memory" in http://www.bitsavers.org/pdf/honeywell/large_systems/multics/haley/AG95_part2_Jun72.pdf. Guy Harris (talk) 21:42, 30 November 2023 (UTC)

Need better hatnote
Call gate redirects to this article, which only refers to Intel's use. user:Widefox replaced with. However. protection ring does not discuss call gates, except to link back to Call gate (Intel). The summary of the edit cites WP:NOARTICLE, but there were no redlinks in the removed text.

I believe that either the redirect should be removed or the hatnote should specify that the term call gate has a wider meaning. Shmuel (Seymour J.) Metz Username:Chatul (talk) 02:29, 25 August 2020 (UTC)
 * User:Chatul Did you read WP:HATNOTE as asked? Their purpose is to help readers locate a different article if the one they are at is not the one they're looking for. (emphasis own). So I consider fixing the bad hatnote you added  is a good idea.
 * I see from the history you moved (and redirected) the broader scope title call gate to this arch-specific subtopic Call gate (Intel) . The obvious answer is undo that move and have x86 as the example. Much better than creating a main topic hole by sidelining the common example. It's only worth keeping as is if an article is about to be written at call gate. Widefox ; talk 18:03, 25 August 2020 (UTC)
 * If you've got someone willing to take on the task of writing a general article, that would certainly be better. In the meantime, having an article with a general title that only addresses a special case would seem to violate WP:NPOV. Shmuel (Seymour J.) Metz Username:Chatul (talk) 18:36, 25 August 2020 (UTC)
 * If it's not clear yet, I object to your move and will revert it. It was misguided like the hatnote without an article. To be considered is WP:OWN. The quickest way to get the general article is to not start from here, revert, and only when too big WP:SPINOUT the Intel example. The NPOV is more WP:RECENTISM / WP:WEIGHT which readers can at least get something, rather than the current nothing. Widefox ; talk 23:16, 26 August 2020 (UTC)