From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <sourceware-bugzilla@sourceware.org>
Received: by sourceware.org (Postfix, from userid 48)
	id 198583858433; Wed, 21 Jun 2023 14:25:05 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 198583858433
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org;
	s=default; t=1687357506;
	bh=CPXZ0Eb9thxNBbXdXgqCx0r7c/3paXyKX0htrh+fMNs=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=IEdzV8WHEdFN/ShkHY7Unpul8ieqtPLM+/2FkG+stlLIWZmkR8et44HfHUn6/ATKZ
	 PMP6KfeC0P/wffI2XiEG6waDOTgCVRx1gwRUa40R30sjfFOJ/imfkVFZaD+lHI0n/c
	 u8Ug2A0NwMergJdyGKBaXIJ8N1KTJtgXPWh9iKis=
From: "adhemerval.zanella at linaro dot org"
 <sourceware-bugzilla@sourceware.org>
To: glibc-bugs@sourceware.org
Subject: [Bug nptl/29029] poll() spuriously returns EINTR during thread
 cancellation and with cancellation disabled
Date: Wed, 21 Jun 2023 14:25:04 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: glibc
X-Bugzilla-Component: nptl
X-Bugzilla-Version: 2.34
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: adhemerval.zanella at linaro dot org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Resolution: FIXED
X-Bugzilla-Priority: P2
X-Bugzilla-Assigned-To: adhemerval.zanella at linaro dot org
X-Bugzilla-Target-Milestone: 2.36
X-Bugzilla-Flags: security-
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-29029-131-fcViZsYyO6@http.sourceware.org/bugzilla/>
In-Reply-To: <bug-29029-131@http.sourceware.org/bugzilla/>
References: <bug-29029-131@http.sourceware.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://sourceware.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
List-Id: <glibc-bugs.sourceware.org>

https://sourceware.org/bugzilla/show_bug.cgi?id=3D29029

--- Comment #10 from Adhemerval Zanella <adhemerval.zanella at linaro dot o=
rg> ---
(In reply to Andres Freund from comment #9)
> FWIW, the fix / revert (404656009b) seems to have lead to a noticeable
> performance regression.

The 404656009b was a bugfix, meaning that the changes
8c1c0aae20/2b51742531/26cfbb7162 that possibly optimized the cancellation p=
ath
are not correct, and any performance implications should not be considered.

However, I have patch that refactor the cancellation handling to avoid most=
 of
atomic operations to enable/disable async on every cancellation entrypoint =
[1].
I am not sure about its performance implication though.

[1]
https://patchwork.sourceware.org/project/glibc/patch/20230524135126.3174670=
-1-adhemerval.zanella@linaro.org/

--=20
You are receiving this mail because:
You are on the CC list for the bug.=