From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.org!.POSTED!not-for-mail From: Philipp Stephani Newsgroups: gmane.emacs.bugs Subject: bug#24756: 26.0.50; Emacs should not use pselect Date: Tue, 25 Oct 2016 11:45:22 +0000 Message-ID: References: NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: multipart/alternative; boundary=047d7bb03ddcc124ef053faf0b98 X-Trace: blaine.gmane.org 1477395994 29776 195.159.176.226 (25 Oct 2016 11:46:34 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Tue, 25 Oct 2016 11:46:34 +0000 (UTC) To: 24756@debbugs.gnu.org Original-X-From: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane.org@gnu.org Tue Oct 25 13:46:28 2016 Return-path: Envelope-to: geb-bug-gnu-emacs@m.gmane.org Original-Received: from lists.gnu.org ([208.118.235.17]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bz0BK-0005bM-JP for geb-bug-gnu-emacs@m.gmane.org; Tue, 25 Oct 2016 13:46:14 +0200 Original-Received: from localhost ([::1]:53596 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bz0BM-0001Ok-Vv for geb-bug-gnu-emacs@m.gmane.org; Tue, 25 Oct 2016 07:46:17 -0400 Original-Received: from eggs.gnu.org ([2001:4830:134:3::10]:60168) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bz0BE-0001Ms-QH for bug-gnu-emacs@gnu.org; Tue, 25 Oct 2016 07:46:09 -0400 Original-Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bz0B9-0001eZ-B3 for bug-gnu-emacs@gnu.org; Tue, 25 Oct 2016 07:46:08 -0400 Original-Received: from debbugs.gnu.org ([208.118.235.43]:40373) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1bz0B9-0001eV-77 for bug-gnu-emacs@gnu.org; Tue, 25 Oct 2016 07:46:03 -0400 Original-Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1bz0B7-0006nE-Kn for bug-gnu-emacs@gnu.org; Tue, 25 Oct 2016 07:46:03 -0400 X-Loop: help-debbugs@gnu.org Resent-From: Philipp Stephani Original-Sender: "Debbugs-submit" Resent-CC: bug-gnu-emacs@gnu.org Resent-Date: Tue, 25 Oct 2016 11:46:01 +0000 Resent-Message-ID: Resent-Sender: help-debbugs@gnu.org X-GNU-PR-Message: followup 24756 X-GNU-PR-Package: emacs X-GNU-PR-Keywords: Original-Received: via spool by 24756-submit@debbugs.gnu.org id=B24756.147739594026081 (code B ref 24756); Tue, 25 Oct 2016 11:46:01 +0000 Original-Received: (at 24756) by debbugs.gnu.org; 25 Oct 2016 11:45:40 +0000 Original-Received: from localhost ([127.0.0.1]:55772 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bz0Am-0006mb-5V for submit@debbugs.gnu.org; Tue, 25 Oct 2016 07:45:40 -0400 Original-Received: from mail-wm0-f54.google.com ([74.125.82.54]:37828) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1bz0Ak-0006mO-Bx for 24756@debbugs.gnu.org; Tue, 25 Oct 2016 07:45:38 -0400 Original-Received: by mail-wm0-f54.google.com with SMTP id c78so19392052wme.0 for <24756@debbugs.gnu.org>; Tue, 25 Oct 2016 04:45:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:references:in-reply-to:from:date:message-id:subject:to; bh=1sU67zIR5bVCL4/uyDQmFflfgtKWvLs12EnH3E2itu8=; b=RaYeJRlhrS9JXeZY/5JHR0G75gKKIh4QGD0Isai3q3KC0PVzAZrvKQvgkCKKQZX/Ru BELxB2MC+/gMthd3+E3WUFNQzacTVvsyIzsHMNqXEB8plpD/2uH1tHEa0VBDrdarrJJm 9OlSp3FFpAvizH7vDXF3z+jib8uYRevFEuoCB04W7ewl09/yaO3w5I8wxDSVXJDZkOX7 ok6TXQI/B0I0C8CE+aVFpEKBD16YvRT8NcxWilD4d5Q8OsOEBe7DmJ5NCi9to9hkQ6Ug EHP7llJ7KLAqpycTRiCdFbvUhbqbwRXWz87Ly7+xfocR5YwLrZUlQmO5TnjmcGYR5alZ 2OCw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to; bh=1sU67zIR5bVCL4/uyDQmFflfgtKWvLs12EnH3E2itu8=; b=B+eDoahHiNexgrnynVjNYR+Uz8v146Xy3cfoEXl3o3KTb82hAYCg9uzFs9Vec++Z9I k4zUnVJyhFQ4a1i5Hfyq3QyZyXbXX/hwNNnaaGBssIq9HQn42Ezt/ohHelgq+WfyoDjQ MK7KzkAkv5JDeS3x7J8QADHXlnV8ny+OJ4q3Ge7i2wvMAWvtGm2HkuTxa9HYb9J4Mfkr OgroUtFTvWOrsbWIQ1l6qJ4CmBfh4ToLlIUEclGj4eEv66d2lNcr/3loFriZRTvs0vVP VoIfAHkazJD5bshBFlB2YMHclUqAeavoqrQq/LxjWDL/8T+dQmOPq/SaJ2h5lV38vLG5 W4Cw== X-Gm-Message-State: ABUngveOdY4AsWCmWtsOlMb/CArhl/kdvxXwvvsT1aheemJ14OG3eb/feRY8wG3r8QvFm8ZksdT4amUMY/iEDw== X-Received: by 10.194.94.39 with SMTP id cz7mr18499234wjb.141.1477395932585; Tue, 25 Oct 2016 04:45:32 -0700 (PDT) In-Reply-To: X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 208.118.235.43 X-BeenThere: bug-gnu-emacs@gnu.org List-Id: "Bug reports for GNU Emacs, the Swiss army knife of text editors" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane.org@gnu.org Original-Sender: "bug-gnu-emacs" Xref: news.gmane.org gmane.emacs.bugs:124987 Archived-At: --047d7bb03ddcc124ef053faf0b98 Content-Type: text/plain; charset=UTF-8 Philipp Stephani schrieb am Fr., 21. Okt. 2016 um 17:42 Uhr: > > Emacs uses pselect for IO multiplexing, limiting the number of available > file descriptors to FD_SETSIZE. Rather, epoll/kqueue should be used, if > available. > > Some good counter-arguments were brought up: - pselect is in the POSIX standard, whereas the possible replacements (ppoll/epoll/kqueue) are not. - The file descriptor limit should only be hit in pathological cases, e.g. when leaking background processes, and is only a symptom of a bug somewhere else. A higher FD limit would only hide the symptom, but not fix the underlying root cause. Given that, I'm no longer convinced that switching away from pselect would be a good idea (but I'll leave this bug open for now to provide a space for discussion). --047d7bb03ddcc124ef053faf0b98 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


Philip= p Stephani <p.stephani2@gmail.c= om> schrieb am Fr., 21. Okt. 2016 um 17:42=C2=A0Uhr:

Emacs uses pselect for IO multiplexing, limiting the number of available file descriptors to FD_SETSIZE.=C2=A0 Rather, epoll/kqueue should be used, = if
available.


Some good counter-arguments were brought u= p:
- pselect is in the POSIX standard, whereas the possible repla= cements (ppoll/epoll/kqueue) are not.
- The file descriptor limit= should only be hit in pathological cases, e.g. when leaking background pro= cesses, and is only a symptom of a bug somewhere else. A higher FD limit wo= uld only hide the symptom, but not fix the underlying root cause.

Given that, I'm no longer convinced that switching away= from pselect would be a good idea (but I'll leave this bug open for no= w to provide a space for discussion).=C2=A0
--047d7bb03ddcc124ef053faf0b98--