RE: Re: Kernel bug: mm/rmap.c:483

From: man_josewanadoo.es (man_jose_at_wanadoo.es)
Date: 01/17/05

  • Next message: Roland McGrath: "[PATCH] i386/x86_64 fpu: fix x87 tag word simulation using fxsave"
    To: linux-kernel@vger.kernel.org
    Date:	Mon, 17 Jan 2005 02:26:24 +0100
    
    

    I'm trying to send the next post to:
    "Frank Denis \(Jedi/Sector One\)" j@pureftpd.org

    But my email provider doesn't let me to send this message to: j@pureftpd.org
    It's about a bug.
    -------------------
    Hello Frank,
    I supose that I should write to you directly, or am I wrong? Besides, I'm Spanish, so please sorry my english.
    The 2.6.11-rc1-mm1-jedi1 keeps on failing. Disclaimer : I'm a newbie. This is the first time I patch something. And this friday was my first contact with Gentoo, ebuilds, emerge and all those things (I came from Mandrake 10.1 because I feel that there was something working bad).

    What I have done (is it OK?):
    tar -xvjpf linux-2.6.11-rc1
    bzip2 -dc 2.6.11-rc1-mm1.bz2 | patch -p1
    bzip2 -dc 2.6.11-rc1-mm1-jedi1.batch.z2 | patch -p1

    Everything works fine, but when I do "emerge eagle-usb":
    ------------[ cut here ]------------
    kernel BUG at mm/rmap.c:482!
    invalid operand: 0000 [#2]
    Modules linked in: rtc
    CPU: 0
    EIP: 0060:[<c0146aa9>] Not tainted VLI
    EFLAGS: 00010296 (2.6.11-rc1-mm1-jedi1)
    EIP is at page_remove_rmap+0x29/0x40
    eax: fffffff0 ebx: 0000d000 ecx: c1205020 edx: c1205020
    esi: cee7834c edi: c1205020 ebp: 00025000 esp: d1306e6c
    ds: 007b es: 007b ss: 0068
    Process doman (pid: 12367, threadinfo=d1306000 task=cf46c0c0)
    Stack: c0140218 c1205020 00000000 0f014067 40417000 10281067 084c6000 d0c3c084
           080eb000 00000000 c0140383 c04f9028 d0c3c080 080c6000 00025000 00000000
           080c6000 d0c3c084 080eb000 00000000 c01403da c04f9028 d0c3c080 080c6000
    Call Trace:
     [<c0140218>] zap_pte_range+0x128/0x240
     [<c0140383>] zap_pmd_range+0x53/0x70
     [<c01403da>] zap_pud_range+0x3a/0x60
     [<c0140470>] unmap_page_range+0x70/0x90
     [<c0140586>] unmap_vmas+0xf6/0x1f0
     [<c0144d88>] exit_mmap+0x78/0x140
     [<c0113da7>] mmput+0x37/0xd0
     [<c0118047>] do_exit+0xa7/0x330
     [<c0118344>] do_group_exit+0x34/0x70
     [<c010304f>] syscall_call+0x7/0xb
    Code: 26 00 8b 54 24 04 8b 02 f6 c4 08 75 28 83 42 08 ff 0f 98 c0 84 c0 74 11 8b 42 08 40 78 0d 9c 58 fa ff 0d 50 61 50 c0 50 9d 90 c3 <0f> 0b e2 01 a0 ae 3d c0 eb e9 0f 0b df 01 a0 ae 3d c0 eb ce 8d
     BUG: atomic counter underflow at:
     [<c0118258>] do_exit+0x2b8/0x330
     [<c0103cf0>] do_invalid_op+0x0/0xd0
     [<c0103cf0>] do_invalid_op+0x0/0xd0
     [<c0103930>] do_trap+0x0/0x110
     [<c0103d9e>] do_invalid_op+0xae/0xd0
     [<c0146aa9>] page_remove_rmap+0x29/0x40
     [<c0136ea2>] prep_new_page+0x52/0x60
     [<c0137594>] buffered_rmqueue+0xe4/0x200
     [<c0136ea2>] prep_new_page+0x52/0x60
     [<c01031f7>] error_code+0x2b/0x30
     [<c0146aa9>] page_remove_rmap+0x29/0x40
     [<c0140218>] zap_pte_range+0x128/0x240
     [<c0140383>] zap_pmd_range+0x53/0x70
     [<c01403da>] zap_pud_range+0x3a/0x60
     [<c0140470>] unmap_page_range+0x70/0x90
     [<c0140586>] unmap_vmas+0xf6/0x1f0
     [<c0144d88>] exit_mmap+0x78/0x140
     [<c0113da7>] mmput+0x37/0xd0
     [<c0118047>] do_exit+0xa7/0x330
     [<c0118344>] do_group_exit+0x34/0x70
     [<c010304f>] syscall_call+0x7/0xb

    ----- Mensaje Original -----
    Remitente: "Frank Denis \(Jedi/Sector One\)" j@pureftpd.org
    Destinatario: "man_josewanadoo.es" man_jose@wanadoo.es
    Fecha: Domingo, Enero 16, 2005 10:23pm
    Asunto: Re: Kernel bug: mm/rmap.c:483

    >On Sun, Jan 16, 2005 at 10:12:21PM +0100, man_josewanadoo.es wrote:
    >> What is next is with the kernel-2.6.10-r4
    >> Would you post me to say if the bug is closed or if is my system
    >who is broken?
    >> kernel BUG at mm/rmap.c:483!
    > Your system is not broken, this is a known bug.
    >
    > Can you check whether 2.6.11-rc1-mm1-jedi1 fixes it?
    >
    > 2.6.11-rc1 : ftp://ftp.kernel.org:/pub/linux/kernel/v2.6/testing/
    > -mm1 patch : ftp://ftp.kernel.org:/pub/linux/kernel/people/akpm/
    >patches/2.6/
    > -jedi1 : ftp://ftp.c9x.org/linux-kernel/
    > Best regards,
    >--
    >Frank - my st
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/


  • Next message: Roland McGrath: "[PATCH] i386/x86_64 fpu: fix x87 tag word simulation using fxsave"

    Relevant Pages

    • Re: SATA_SIL works with 2.6.7-bk8 seagate drive, but oops
      ... So that's a kernel bug, ... that those architectures expose the _internal_ HZ rather than some ... it's a bug, and it helps make excuses for _other_ bugs. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: 2.6.13-rc3-mm3
      ... to kernel/softirq.c in that patch... ... kernel BUG at kernel/sched.c:2988! ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Alpha/2.6.4: Kernel bug at mm/page_alloc.c:752
      ... I get a BUG right after mounting the root filesystem, ... Compiled using gcc-3.3.3, binutils from cvs. ... Kernel bug at mm/page_alloc.c:752 ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: filesystem corruption (ReiserFS, 2.6.6): regions replaced by 00 bytes
      ... if you slept in the right spot you could hit it on UP. ... Any idea about when the bug was introduced? ... who then figured out that it was a kernel bug, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Kernel BUG at fs/mpage.c:489
      ... I found a kernel bug. ... Internal Connector Type: On Board IDE ... Internal Reference Designator: SECONDARY IDE/HDD ...
      (Linux-Kernel)