From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id D883343219; Fri, 27 Oct 2023 20:16:03 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 5A410402C2; Fri, 27 Oct 2023 20:16:03 +0200 (CEST) Received: from mail-pl1-f179.google.com (mail-pl1-f179.google.com [209.85.214.179]) by mails.dpdk.org (Postfix) with ESMTP id A630C40263 for ; Fri, 27 Oct 2023 20:16:01 +0200 (CEST) Received: by mail-pl1-f179.google.com with SMTP id d9443c01a7336-1bdf4752c3cso18672475ad.2 for ; Fri, 27 Oct 2023 11:16:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=networkplumber-org.20230601.gappssmtp.com; s=20230601; t=1698430560; x=1699035360; darn=dpdk.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date:from:to:cc:subject:date :message-id:reply-to; bh=NwL+eiqmDdyAzkjVeebThPjt6futkGBxeM3gynqzNBc=; b=C2y89aFOn0cPyG734AhiuMLtwFvHGQjZF0tP3KMcdH0lCmsP9x+XAPa/bR8xKIJtuA LWmPDC/DOxPdQOhwv2wsEFEVauv9ibUTnOgm8A4uk6x8SdA0LUXXni4Zy0iVrhkPsfMD ML3OEId2SNB0GqoEV7dHOcQdFGa5+OczW+Rx98vuiUd9cXyro1CJYIww3io8OaQ/SIOc E0fJwatolRCcMQYsEEMNJdRfj1f3bCeJI3U7CjSKVYqUbukfOD09mq8WfEFgLDrBg72T qDq845dsB02j+kvej7yT07mBM8XWZ3y101UzO/NosSGnli1qEHfjuOuGCfLkUt0y/18V 6NxQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1698430560; x=1699035360; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=NwL+eiqmDdyAzkjVeebThPjt6futkGBxeM3gynqzNBc=; b=RVJs27pL6mgHvHMBYXKn5nFfemuxbOlLrmXlSfIApqHbBRLRO8DK0Bb30aXjWuSIbE WaRF7KWUUNapxbQAgdiqVPw6qSeAypXufsdwPX1K9NCbfLVmefEgIj4qSeGG5NTdefqM so08S3XF60w9Ac3rIjp0nJRGPUGqhm18FSwr/VuECQu5K/I6UGm0/8BF1efe1AlzBwRm 4Zy4cUaj6tz1f/VyJytUugLqtzkwB0kkY3SRRtpADez/PtAQt0wIOPvUd+y32vMf2Qt2 7DXozcU8ofGEV2GtyaGKBr0ef8lyGP5D0ELoQioEko5GydriJbea8nCFWqEIowENtPwv 2h8g== X-Gm-Message-State: AOJu0YxcjTFupy7Eg1l/xmlmw9Y00sjhZTFFLOWYwTz7v63q+90SZQZR t9clzZV10GBOc40X5bThWopC9g== X-Google-Smtp-Source: AGHT+IGGtk9LYK9Ln7E1d60PgHUbKoTK/IDnkmPpSFs384wLgt2qtkTgpQQHwH0i4Y9Q6HfJYt8JWA== X-Received: by 2002:a17:902:e88f:b0:1c9:d46e:d523 with SMTP id w15-20020a170902e88f00b001c9d46ed523mr4191824plg.7.1698430560580; Fri, 27 Oct 2023 11:16:00 -0700 (PDT) Received: from hermes.local (204-195-123-141.wavecable.com. [204.195.123.141]) by smtp.gmail.com with ESMTPSA id m5-20020a170902db0500b001c5dea67c26sm1640141plx.233.2023.10.27.11.15.59 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 27 Oct 2023 11:16:00 -0700 (PDT) Date: Fri, 27 Oct 2023 11:15:58 -0700 From: Stephen Hemminger To: Morten =?UTF-8?B?QnLDuHJ1cA==?= Cc: "Thomas Monjalon" , , "David Marchand" , , "Anatoly Burakov" , "Tyler Retzlaff" , "Narcisa Vasile" , "Dmitry Kozlyuk" , "Konstantin Ananyev" , "Andrew Rybchenko" Subject: Re: [PATCH v6 1/1] eal/unix: allow creating thread with real-time priority Message-ID: <20231027111558.0e2688e2@hermes.local> In-Reply-To: <98CBD80474FA8B44BF855DF32C47DC35E9EFA3@smartserver.smartshare.dk> References: <20231024125416.798897-1-thomas@monjalon.net> <20231027081158.1358064-1-thomas@monjalon.net> <98CBD80474FA8B44BF855DF32C47DC35E9EFA3@smartserver.smartshare.dk> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On Fri, 27 Oct 2023 10:45:03 +0200 Morten Br=C3=B8rup wrote: > Is it 100 % certain that the system becomes unstable if not sleeping or u= sing blocking system calls from a real-time thread? > And technically, it's not the thread itself that becomes unstable. My experience is that the goal of real time threads is "do not let kernel h= ave higher priority than this thread". That means that if/when a kernel worker is needed on this core, instability= happens. It is very hard but possible to ensure that a kernel worker thread never ru= ns on that core. But if you are able to do configure that, then there is no point in using R= T threads. The best design is to avoid the kernel needing to run, and if it does then = let it run.