From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [216.205.24.124]) by sourceware.org (Postfix) with ESMTP id 838693854835 for ; Thu, 18 Mar 2021 17:57:31 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org 838693854835 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-565-c945IrEwORGyRPg5M1ePEQ-1; Thu, 18 Mar 2021 13:57:29 -0400 X-MC-Unique: c945IrEwORGyRPg5M1ePEQ-1 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id AB8BB8143FE for ; Thu, 18 Mar 2021 17:57:28 +0000 (UTC) Received: from oldenburg.str.redhat.com (ovpn-115-121.ams2.redhat.com [10.36.115.121]) by smtp.corp.redhat.com (Postfix) with ESMTPS id CB6825D6A8; Thu, 18 Mar 2021 17:57:27 +0000 (UTC) From: Florian Weimer To: DJ Delorie Cc: libc-alpha@sourceware.org Subject: Re: tunables vs osxsave vs checkpointing vs _dl_runtime_resolve References: Date: Thu, 18 Mar 2021 18:57:36 +0100 In-Reply-To: (DJ Delorie's message of "Thu, 18 Mar 2021 13:47:43 -0400") Message-ID: <87mtv0qr8v.fsf@oldenburg.str.redhat.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux) MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.15 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain X-Spam-Status: No, score=-6.6 required=5.0 tests=BAYES_00, DKIMWL_WL_HIGH, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, RCVD_IN_DNSWL_LOW, RCVD_IN_MSPIKE_H3, RCVD_IN_MSPIKE_WL, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 18 Mar 2021 17:57:32 -0000 * DJ Delorie: > Florian Weimer writes: >> What happens if you migrate to a machine that has XSAVE/XSAVEC, but >> requires more storage? > > We'd need a new tunable to always use the maximum size, or some > specified size, I suppose. But there are limits to the miracles we can > perform, and "size" is not the root problem - the layout changes too! The layout matters only if the process is captured at a very unlucky time. So hopefully they can just avoid that. I think the main issue with the size change is that it prevents restore consistently, even in the non-race case. For XSAVE, it would be possible to query the size each time using CPUID. >> I expect that this is the case that people actually care about these >> days. 8-) > > In this case, they just want to disable xsave completely, so it's as > portable as possible. But I don't think it actually is. Disabling XSAVE won't play nicely with the AVX2 and later calling conventions. Thanks, Florian