[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Problem using patched loader.c

On 2/14/07, Bill Nottingham <notting redhat com> wrote:
James Olin Oden (james oden gmail com) said:
> One the developers where I work has attempted to patch loader.c with a
> slight sleep (5 seconds) to occur before scanning for devices but now
> he is getting a segv.  The original problem was that when we booted
> from a USB device, anaconda would load, but then be unable to find the
> device from which it was booted.  This appeared to be a race condition
> where the udev files are not being created before anaconda searches
> the bootable device list. This is one of the problems addressed by the
> patch sent by Uday Prakash to this list on Jan 25.  He grabbed the
> relevant portion of that patch and just added the logMessage and 5
> second delay to loader.c.  The build completed successfully, and then
> I copied the newly built loader executable into the initrd at
> /sbin/loader.

Make sure your kudzu and new loader were both built against the
same pciutils - that can cause crashes if the ABI changes.

I'm missing something Bill.  Since loader is statically linked how can
there be ABI differences?


Anaconda-devel-list mailing list
Anaconda-devel-list redhat com

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]