From owner-freebsd-alpha Mon Oct 29 18:47: 2 2001 Delivered-To: freebsd-alpha@freebsd.org Received: from duke.cs.duke.edu (duke.cs.duke.edu [152.3.140.1]) by hub.freebsd.org (Postfix) with ESMTP id 746BE37B405 for ; Mon, 29 Oct 2001 18:46:53 -0800 (PST) Received: from grasshopper.cs.duke.edu (grasshopper.cs.duke.edu [152.3.145.30]) by duke.cs.duke.edu (8.9.3/8.9.3) with ESMTP id VAA15473; Mon, 29 Oct 2001 21:46:48 -0500 (EST) Received: (from gallatin@localhost) by grasshopper.cs.duke.edu (8.11.3/8.9.1) id f9U2kMN25302; Mon, 29 Oct 2001 21:46:22 -0500 (EST) (envelope-from gallatin@cs.duke.edu) From: Andrew Gallatin MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15326.5246.267516.981863@grasshopper.cs.duke.edu> Date: Mon, 29 Oct 2001 21:46:22 -0500 (EST) To: Marcel Moolenaar Cc: alpha@FreeBSD.ORG Subject: Re: Linuxulator MFC candidate for testing In-Reply-To: <20011026220239.A4254@dhcp01.pn.xcllnt.net> References: <20011026220239.A4254@dhcp01.pn.xcllnt.net> X-Mailer: VM 6.75 under 21.1 (patch 12) "Channel Islands" XEmacs Lucid Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Marcel Moolenaar writes: > Gang, > > Since I don't have a -stable Alpha, I can only do as much as > prepare an MFC and make it compile as much as possible. If > someone on this list can try out the patch and tell me if it > doesn't cause regressionsi (assuming it builds and loads), > I'd be grateful. > > The MFC candidate (patch) can be downloaded here: > > http://www.xcllnt.net/~marcel/linux.diff.gz > > Please remove /sys/alpha/linux/linux_ioctl.h before building. Two issues so far. DUMMY(mount) needs to be added to sys/alpha/linux/linux_dummy.c to make the module load. The second, much more serious problem, is that when I run the Compaq JRE/JDK for linux the system locks up. It responds to pings,etc. But all shells are dead, X locks up, etc. Its like its in a really tight loop in sigsuspend().. This is reproducible. The stack trace looks like this: db> tr siointr1() at siointr1+0x160 siointr() at siointr+0x14 isa_handle_intr() at isa_handle_intr+0x2c alpha_dispatch_intr() at alpha_dispatch_intr+0x8c interrupt() at interrupt+0x1d8 XentInt() at XentInt+0x28 --- interrupt (from ipl 0) --- tsleep() at tsleep+0x394 sigsuspend() at sigsuspend+0xd8 osf1_sigsuspend() at osf1_sigsuspend+0x84 syscall() at syscall+0x224 XentSys() at XentSys+0x5c --- syscall (111, Linux ELF, osf1_sigsuspend) --- --- user mode --- db> Yes, I realize that's an osf1_sigsuspend() getting called through the linux sysent. It works fine in the old, pre-patch code. I don't understand how you could have broken this with your patch. Any ideas? Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message