From owner-freebsd-hackers Tue Apr 15 11:40:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA28724 for hackers-outgoing; Tue, 15 Apr 1997 11:40:11 -0700 (PDT) Received: from athena.veritas.com (athena.veritas.com [192.203.46.191]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA28716 for ; Tue, 15 Apr 1997 11:40:08 -0700 (PDT) Received: from megami.veritas.com by athena.veritas.com with smtp (Smail3.1.29.0 #9) id m0wHD93-000iRaC; Tue, 15 Apr 97 11:39 PDT Received: from sigma by megami.veritas.com with smtp (Smail3.1.29.0 #7) id m0wHD92-000iVIC; Tue, 15 Apr 97 11:39 PDT Message-Id: From: Aaron Smith To: John Polstra cc: hackers@freebsd.org Subject: Re: mmap vs. disk errors In-reply-to: Your message of "Tue, 15 Apr 1997 09:51:15 PDT." <199704151651.JAA04914@austin.polstra.com> Date: Tue, 15 Apr 1997 11:39:54 -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 15 Apr 1997 09:51:15 PDT, John Polstra writes: >What happens if an application has a file mapped into memory and tries >to access part of it, but the kernel can't bring the pages into memory >because of a disk error (bad block in the file)? The kernel can't satisfy the request, so you'd have to return an error, but the only way to do that is to fault the access. The app would have to be handed a segv. -- Aaron Smith aaron@veritas.com unless explicitly stated, i do not speak for my employer. support free speech.