From owner-freebsd-current Tue Jul 31 20:50:47 2001 Delivered-To: freebsd-current@freebsd.org Received: from ises00.j.dendai.ac.jp (ises00.j.dendai.ac.jp [133.14.49.10]) by hub.freebsd.org (Postfix) with ESMTP id 1C64D37B40A for ; Tue, 31 Jul 2001 20:50:40 -0700 (PDT) (envelope-from fujimoto@j.dendai.ac.jp) Received: from j.dendai.ac.jp (yebisu [133.14.49.224]) by ises00.j.dendai.ac.jp (8.9.3+Sun/8.9.3) with ESMTP id MAA09350 for ; Wed, 1 Aug 2001 12:53:22 +0900 (JST) Message-ID: <3B677C89.D72AA4C8@j.dendai.ac.jp> Date: Wed, 01 Aug 2001 12:50:33 +0900 From: FUJIMOTO Kou Organization: Tokyo Denki University X-Mailer: Mozilla 4.75 [ja] (Windows NT 5.0; U) X-Accept-Language: ja,en MIME-Version: 1.0 To: freebsd-current@freebsd.org Subject: ohci_process_done hangs when uhub detached Content-Type: text/plain; charset=iso-2022-jp Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Hi, Last weekend I replaced my socket7 MVP3 board (w/ K6-2+) to ASUS CUSI-FX socket370 SiS630E board (w/ VIA C3 733MHz), then I found kernel panics when USB HUBs are detached. My USB setup is as follows: FreeBSD PC----------+ | CPU_switcher----HUB1----HUB2-----Mouse | | Windows PC----------+ +----USB-PS/2---Keyboard Mouse and USB-PS/2 translator are detached safely. When switching FreeBSD to Windows, kernel hangs with flood of "db>" prompts. This is also the case when I pull out USB cable between switcher and HUB1. When pulling cable between HUB1-HUB2 out, the following error messages are indicated and I can check trace result: ==================================================================== uhub3: detached ohci_device_ctrl_close: pipe=0xc122e580 ohci_process_done: err cc=5 (DEVICE_NOT_RESPONDING), xfer=0xc103b500 Fatal trap 12: page fault while in kernel mode fault virtual address = 0xdeadc0e6 fault code = supervisor write, page not present instruction pointer = 0x8:0xc01daa5d stack pointer = 0x10:0xc79a5f20 frame pointer = 0x10:0xc79a5f34 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def31 1, gran 1 processor eflags = interrupt enabled, resume, IOPL=0 current process = 18 (irq 9: ohci0++) kernel: type 12 trap, code=0 Stopped at ohci_process_done+0x21d: movl %eax,0x8(%edx) db> trace ohci_process_done(c080e000,580620,c1024340,c1016a80,4)at ohci_process_done+0x21d ohci_intr1(c080e000,c79a5f7c,c0214157,c080e000,c0213e98) at ohci_intr1+0x139 ohci_intr(c080e000) at ohci_intr+0x15 ithread_loop(c1016a80,c79af5fa8) at ithread_loop+0x2bf fork_exit(c0213e98,c1016a80,c79a5fa8) at fork_exit+0xb4 fork_trampoline() at fork_trampoline+0x8 db> ====================================================================== With MVP3 chipset (UHCI) the setup above have worked peacefully, so I guess it's due to OHCI code and/or SiS 630E chipset. -- FUJIMOTO Kou, Dept. of Information Sciences, Tokyo Denki Univ. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message