From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14191 invoked by alias); 13 Nov 2011 23:25:35 -0000 Received: (qmail 14177 invoked by uid 22791); 13 Nov 2011 23:25:33 -0000 X-SWARE-Spam-Status: No, hits=4.7 required=5.0 tests=AWL,BAYES_40,BOTNET,RCVD_IN_DNSWL_LOW X-Spam-Check-By: sourceware.org Received: from outmail148108.authsmtp.net (HELO outmail148108.authsmtp.net) (62.13.148.108) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sun, 13 Nov 2011 23:25:20 +0000 Received: from mail-c194.authsmtp.com (mail-c194.authsmtp.com [62.13.128.121]) by punt5.authsmtp.com (8.14.2/8.14.2/Kp) with ESMTP id pADNPHAR065832; Sun, 13 Nov 2011 23:25:17 GMT Received: from asuka.2ndquadrant.com (157-157-43-108.dsl.dynamic.simnet.is [157.157.43.108]) (authenticated bits=0) by mail.authsmtp.com (8.14.2/8.14.2) with ESMTP id pADNPFbb055332; Sun, 13 Nov 2011 23:25:16 GMT To: GCC Java Cc: gc@linux.hpl.hp.com Subject: Re: Garbage collection segfault in OpenIndiana References: <4EBFF1C5.80806@redhat.com> From: "Johann 'Myrkraverk' Oskarsson" Date: Sun, 13 Nov 2011 23:25:00 -0000 In-Reply-To: <4EBFF1C5.80806@redhat.com> (Andrew Haley's message of "Sun, 13 Nov 2011 16:35:17 +0000") Message-ID: User-Agent: Gnus/5.1008 (Gnus v5.10.8) XEmacs/21.4.22 (usg-unix-v) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Server-Quench: bf1cf209-0e4e-11e1-80b9-0022640b883e X-AuthReport-Spam: If SPAM / abuse - report it at: http://www.authsmtp.com/abuse X-AuthRoute: OCdxZQATClZXSwQT DCUsVilMUBgsKgdF BwNFJgxfNl8UURhQ KkJXbgASJgRHAnde RXkJW1VVQF9yU2Fw YQtWIwRYZVRPWgJ0 VEtXQ1JaCwdtAxke BlBeI20MF3MUKn51 YEVhXD5YWEByOxQr Qx0CF20PeGc1bDUC UUENdh5ccFFDYx9F aFJ5UHNfaW0EMzQC El17EDkrNzxQJWxe RQUJIFkOQEENViUx SB4LADEzVUwDDwIr IhoHLFIdFUAYWgAA X-Authentic-SMTP: 61633235383639.1015:706 X-AuthFastPath: 0 (Was 255) X-AuthVirus-Status: No virus detected - but ensure you scan with your own anti-virus system. X-IsSubscribed: yes Mailing-List: contact java-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: java-owner@gcc.gnu.org X-SW-Source: 2011-11/txt/msg00002.txt.bz2 Andrew Haley writes: > On 11/12/2011 08:38 PM, Johann 'Myrkraverk' Oskarsson wrote: >> With GCC 4.6.2 on OpenIndiana 151a (Solaris) the boehm gc segfaults >> in os_dep.c:810 when JNI_CreateJavaVM() is called. > >> I thought of merging boehm gc 7.1 into the gcc source tree, but >> this seems non-trivial. In particular it's not possible just to >> make it a drop-in replacement. > > Do you have any reason to suspect that merging gc 7.1 might help? No concrete evidence. In the past I was unable to use 6.x on it while 7.x worked in my own projects. In 7.1 and 7.2alpha6 this function is now called GC_find_limit_with_bound() instead of just GC_find_limit(). The relevant line number in 7.2alpha6 is os_dep.c:994. I suppose the best way to be sure is to compare the output of gcc -E for os_dep.c. If you have any further ideas I can explore, please let me know. -- Johann Oskarsson http://www.2ndquadrant.com/ |[] PostgreSQL Development, 24x7 Support, Training and Services --+-- | Blog: http://my.opera.com/myrkraverk/blog/