Re: Linux 2.4.22-rc1

Mitch_at_0Bits.COM
Date: 08/07/03

  • Next message: Joel Becker: "Re: [PATCH][2.6-mm] Readahead issues and AIO read speedup"
    Date:	Thu, 7 Aug 2003 20:40:02 +0100 (BST)
    To: Fridtjof Busse <fbusse@gmx.de>
    
    

    > Well, I only see this with 2.4.22-pre/rc, so this is definitly not a
    > hardware-problem, it runs just fine with 2.4.21 at 9950 kB/s.

    Sounds exactly like the problem i was having. However the logical
    thought behind it was that it must be either 1) a hardware specific
    problem, or 2) a generic usb driver problem. Since only a handful
    of people seen the problem, then this logically leads one to the
    conclusion that our usb<->ide devices have some marginal timing
    issues that are being triggered with the new code.

    > And running my backup on USB 1.1 is not an option, way to slow.

    Well in my case i opted for the slow backup rather than the no backup
    if i wanted to move forward with the new kernels.

    Cheers
    Mitch

    -------- Original Message --------
    Subject: Re: Linux 2.4.22-rc1
    Date: Thu, 7 Aug 2003 17:45:06 +0200
    From: Fridtjof Busse <fbusse@gmx.de>
    To: linux-kernel@vger.kernel.org
    CC: Mitch@0Bits.COM
    References: <Pine.LNX.4.53.0308071119200.27424@mx.homelinux.com>
    <15050.1060270543@www56.gmx.net>

    * <Mitch@0Bits.COM>:
    > Not enough info.
    >
    > What usb controller do you have ? Which usb driver ?
    > ohci ? uhci ? ehci ? usb 2.0 ?

    nforce2 with the ehci-driver.

    > I reported this a long time ago on the usb lists, but
    > never got down to the bottom of the problem (my fault for
    > not following thru).

    I also reported this problem there, but didn't get a reply at all.

    > However if i disable the usb 2.0
    > driver (i.e. not loading the ehci driver) which my external
    > storage is connected to, then everything works fine - albeit
    > it much more slowly. Appears to be a timing issue on some
    > usb <-> ide controller chips since not everyone is seeing this.

    Well, I only see this with 2.4.22-pre/rc, so this is definitly not a
    hardware-problem, it runs just fine with 2.4.21 at 9950 kB/s.
    And running my backup on USB 1.1 is not an option, way to slow.

    -- 
    Fridtjof Busse
    panic("Lucy in the sky....");
    	2.2.16 /usr/src/linux/arch/sparc64/kernel/starfire.c
    -
    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: Joel Becker: "Re: [PATCH][2.6-mm] Readahead issues and AIO read speedup"

    Relevant Pages

    • Re: [PATCH][RFC][0/4] InfiniBand userspace verbs implementation
      ... > problem without any special hardware at all: ... # The driver did use get_user_pagesto elevated the refcount on all the ... # to by the driver at the time that the app accessed the page causing the ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: protecting source code in 2.6
      ... Why is there a reason to lock down the source of the actual driver? ... Can we assume the hardware is so cheap that all the smartness is in the ... - guarantee the users of your hardware a support period, say, three years ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [RFC] Small PCI core patch
      ... > On the hardware front, there are some signs of hope. ... > building X, DRI, and openGL from source isn't something I do casually. ... > radeon driver I mentioned above, it only seems to get touched every 2 months ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Wireless support
      ... > It doesn't actually say it works on Linux. ... when the project ends with the driver being merged into the kernel (like ... supported hardware leading users to think their newer hardware is ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Cant use SYSFS for "Proprietry" driver modules !!!.
      ... >> If the user has the source of the driver, he can port the driver or hire ... >> be an expensive piece of hardware). ... If you ignore the license, you will suffer the consequences of it, just ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)