From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3B426C433EF for ; Sun, 26 Sep 2021 01:20:58 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 0997861100 for ; Sun, 26 Sep 2021 01:20:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230191AbhIZBWc (ORCPT ); Sat, 25 Sep 2021 21:22:32 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46792 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229461AbhIZBWb (ORCPT ); Sat, 25 Sep 2021 21:22:31 -0400 Received: from mail-il1-x135.google.com (mail-il1-x135.google.com [IPv6:2607:f8b0:4864:20::135]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id B8369C061570 for ; Sat, 25 Sep 2021 18:20:55 -0700 (PDT) Received: by mail-il1-x135.google.com with SMTP id d11so14806505ilc.8 for ; Sat, 25 Sep 2021 18:20:55 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20210112.gappssmtp.com; s=20210112; h=subject:to:cc:references:from:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=h8rwZgo/X8fBt7Y1fQ9qY2zFB6VhIFdRa1FdI6tlUrk=; b=7+SCbgnOTiUL9TXsAxtrUkxuwbtL2rwU5E21LOFEyIQ+mQvQn1KR5LsWjEe8nZDexS oC5amrwhet1OISsckmJYby9FThJbzGzpaVbiXudE/o19PGFnma7VjcCgQ6GVVBytB5tD RbfBkuft9KA6G0caBop4NStrl3Yvq2Ks751xoJoZNI6FlyiFF7XrcGYkfJEfLHCutXSG ljLWjcj1GUyrRftRuRmPtUMDqpOAy7OB+XBC+BknY2UuOllgFJhY0aZXBSSZHuRN6fWI RNEt38p5Thc4VdnbZIP/4JoiVDObmNw3whnMnVAyhscBF1OsJtbiKdYCTI3I2P9RL1Se 59OQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:subject:to:cc:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=h8rwZgo/X8fBt7Y1fQ9qY2zFB6VhIFdRa1FdI6tlUrk=; b=dN1jCFM4VfiCR77CGc2t+hdBCxtBc70rXwzMgo5g8Pry9JeXlVTzve7SAGaskURnRn n/ZQHXx7MOR53GQ6gdgtFVBcqTDSGavGrnTyvilEE+GMUfJ/PjceXJdPsqP0KuCxSU2L QcsUSfYAGhW0kSQwPkQ34W+Lz5zVpBfZRZDi2Sl1AIz4XXFJq61U7ImhJa7z1pMA2Brr /Okfpk248hHkr3jmMWk4skb0gkPRFT1+nfONKajpzcfd1s4NMtw4s35PyA5iTUmOXwaG MGLA9wDs185+JDSDLeC7guIj2lOv5pFS9ZQoeqsY0hLxdfnyjC5eCiALx1Dy8ALcVee+ cQtw== X-Gm-Message-State: AOAM533wrO9iR1/txeXQTX6FRbt1WZEkzDQFcQOTXkJ/ncWbWZFq/4T9 NSnx4B7cuHzpRNrVkKnzZqqa6NWp7yOFJg== X-Google-Smtp-Source: ABdhPJwqHaUw5YmA24XByUvqlVoS7twFw04iPYBvHBNp6Pjy0VHUISsTbX6q+d8eaeGguTIt3ctl0g== X-Received: by 2002:a92:c56e:: with SMTP id b14mr3482303ilj.71.1632619254308; Sat, 25 Sep 2021 18:20:54 -0700 (PDT) Received: from [192.168.1.116] ([66.219.217.159]) by smtp.gmail.com with ESMTPSA id z5sm6628868ile.42.2021.09.25.18.20.53 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Sat, 25 Sep 2021 18:20:53 -0700 (PDT) Subject: Re: [GIT PULL] io_uring fixes for 5.15-rc3 To: Linus Torvalds , "Eric W. Biederman" , Oleg Nesterov , Al Viro Cc: io-uring References: From: Jens Axboe Message-ID: <0eeefd32-f322-1470-9bcf-0f415be517bd@kernel.dk> Date: Sat, 25 Sep 2021 19:20:52 -0600 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.10.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Precedence: bulk List-ID: X-Mailing-List: io-uring@vger.kernel.org On 9/25/21 5:05 PM, Linus Torvalds wrote: > On Sat, Sep 25, 2021 at 1:32 PM Jens Axboe wrote: >> >> - io-wq core dump exit fix (me) > > Hmm. > > That one strikes me as odd. > > I get the feeling that if the io_uring thread needs to have that > signal_group_exit() test, something is wrong in signal-land. > > It's basically a "fatal signal has been sent to another thread", and I > really get the feeling that "fatal_signal_pending()" should just be > modified to handle that case too. It did surprise me as well, which is why that previous change ended up being broken for the coredump case... You could argue that the io-wq thread should just exit on signal_pending(), which is what we did before, but that really ends up sucking for workloads that do use signals for communication purposes. postgres was the reporter here. -- Jens Axboe