One last thing, there is no need to change your kernel modules from linked
to builtin. They are stored in /boot/modules and if you see the four
penquins, this means grub is setup right and your system has acces to that
path.
Op 23 aug. 2014 14:44 schreef "Ronnie van Aarle" <[email protected]>
het volgende:

> If your udev and devtempfs are funtioning correcty, you might want to
> check the mountpoints in /etc/fstab. You can boot debian and enter fdisk -l
> to see how it is configured there to get a clue but still you need to sort
> out the right path descriptors yourself, because when you run the kernel
> from the external drive your in a different chrooted environment. Chroot
> determines the kernels mountpoint and the rest of the storage media
> mountpoints are relative to the root mountpoint. However, this doesnt apply
> to virtual filesystems and devices physically connected to another hardware
> controller.
> Op 23 aug. 2014 14:27 schreef "Ronnie van Aarle" <
> [email protected]> het volgende:
>
>> In /var/log on your usb drive you can find several logs. Those also
>> include error messages. Most common boot issues are allready solved and you
>> can find solutions on the forums.
>> Op 23 aug. 2014 07:56 schreef "Patrick Kennedy" <[email protected]>
>> het volgende:
>>
>>> Speaking of not compiling in things as modules, when I do the following
>>> after "make defconfig" -
>>>
>>> cat .config | grep =m
>>>
>>> I see quite a few modules.  Should I replace all =m with =y ?  I mean,
>>> obviously, I tried that too, but it doesn't fix my boot problem.  When my
>>> LFS boots, it says: Error: unknown filesytem.
>>>
>>> I have the generic device driver support for devtempfs; I have ext4
>>> support.
>>>
>>> I'm still trying to confirm the proper driver for this external USB
>>> Seagate 500MB drive, but no cigar.
>>>
>>> I did "lsusb" and can determine the vendor and product id numbers with
>>> other info (0bc2:5021 Seagate RSS LLC FreeAgent GoFlex USB 2.0) for it, but
>>> I can't find the precise device name per se.  I could start over with LFS
>>> on a spare computer, and do everything on the internal hard drive.  I'm
>>> pretty sure it would work, but to be this close, it just seems that I
>>> should be able to deduce the device_name for this Seagate USB drive
>>> somehow.  Ideas from Linux Kernel in a Nutshell are helpful but, but I
>>> can't close the gap.
>>>
>>> Anyways, sure I replace all =m with =y ?   Any other ideas to kick
>>> around for a couple of days before I try something completely new and
>>> drastic?
>>>
>>> Thanks.
>>>
>>> ~PK
>>>
>>>
>>>
>>> On Thu, Aug 21, 2014 at 10:28 PM, Patrick Kennedy <[email protected]>
>>> wrote:
>>>
>>>> Oh!
>>>>
>>>>
>>>> On Thu, Aug 21, 2014 at 10:24 PM, Hazel Russman <
>>>> [email protected]> wrote:
>>>>
>>>>> On Thu, 21 Aug 2014 21:59:08 +0800
>>>>> Patrick Kennedy <[email protected]> wrote:
>>>>>
>>>>> > Okay, it crashed and burned gloriously, and I botched the grub to
>>>>> > boot.  ;-)
>>>>> >
>>>>> > After some studies on grub, I can now boot my distro Debian from the
>>>>> > primary hard drive via grub commands...very cool.  I can also attempt
>>>>> > to boot my LFS from auxiliary hard drive, and it starts to boot up,
>>>>> > and I see four penguins, a screen's worth of boot messages, and then
>>>>> > it freezes up.
>>>>> >
>>>>> > Questions:
>>>>> >
>>>>> > Firstly, there is /dev, but it only has console and null in it, which
>>>>> > was created earlier per the book.  Is that sufficient?  Is /dev
>>>>> > populated more as the boot progresses?  Seems deficient.
>>>>> No, that should be sufficient. udev should make the other devices as
>>>>> the kernel detects the hardware. But you need the kernel's DEVTMPFS to
>>>>> be set to "y".
>>>>> >
>>>>> > Secondly, I don't see any initrd.img file.  Do I need one?  Here's
>>>>> > what I see in /boot:
>>>>> >
>>>>> > config-3.13.3
>>>>> > grub
>>>>> > System-map-3.13.3
>>>>> > vmlinuz-3.13.3-lfs-7.5
>>>>> >
>>>>> Normally you only need an initrd for a stock kernel. When you build
>>>>> your own, you should compile in the necessary disk drivers rather than
>>>>> building them as modules, so that nothing needs to be loaded at boot.
>>>>>
>>>>> > I just did "make defconfig" when building the kernel.  I figure that
>>>>> > would be the easiest way to test...and just to see if it works.
>>>>> > Maybe I need to load a better driver for auxiliary USB hard drives,
>>>>> > etc., and somehow make that apart of the kernel building, but I would
>>>>> > need to study more on that aspect.
>>>>>
>>>>> The defconfig kernel should contain all the drivers you need but I
>>>>> don't know which ones it compiles in. You can check by looking in
>>>>> the config file in /boot. The SATA and ext4 drivers should definitely
>>>>> be
>>>>> compiled in, not as modules.
>>>>>
>>>>> --
>>>>> http://lists.linuxfromscratch.org/listinfo/lfs-support
>>>>> FAQ: http://www.linuxfromscratch.org/blfs/faq.html
>>>>> Unsubscribe: See the above information page
>>>>>
>>>>
>>>>
>>>
>>> --
>>> http://lists.linuxfromscratch.org/listinfo/lfs-support
>>> FAQ: http://www.linuxfromscratch.org/blfs/faq.html
>>> Unsubscribe: See the above information page
>>>
>>>
-- 
http://lists.linuxfromscratch.org/listinfo/lfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to