From: Jeff Dike <jdike@ka...> - 2002-05-10 17:20:07
> I am attaching the log file which shows that now it runs all the way
> till it tries to mount root fs !!
> Looks like when the bread() eventually calls __get_request_wait(), it
> puts itself on the UBD filesystems queue and waits on the wait queue
> to read the super block.
> Seems like no one is actually doing the read, nor waking up this
What's supposed to be happening is that it's supposed to end up in
do_ubd_request, which writes the request to the io_thread. The io_thread
loops in io_thread(), reading requests from kernel_fd, doing the IO, and
writing the results back to kernel_fd.
That causes an IRQ which ends up in ubd_handler via ubd_intr, which calls
end_request, which causes the thread which wanted the data to be woken up.
From: Jeff Dike <jdike@ka...> - 2002-05-10 17:55:32
> Linux version 2.4.18-20um (chandan@...) (gcc version 2.95.3-5
> (cygwin special)) #107 SMP Thu May 9 21:15:06 2002
Is there any particular reason that you have CONFIG_SMP enabled? Not that
that's bad, but my inclination would be to leave it off just to avoid whatever
complications it might cause.
Hmm ... I had lots of problems trying to compile on Win32 without SMP
support. So I just turned it on. I will debug these eventually ...
On Fri, 10 May 2002, Jeff Dike wrote:
> chandan@... said:
> > Linux version 2.4.18-20um (chandan@...) (gcc version 2.95.3-5
> > (cygwin special)) #107 SMP Thu May 9 21:15:06 2002
> Is there any particular reason that you have CONFIG_SMP enabled? Not that
> that's bad, but my inclination would be to leave it off just to avoid whatever
> complications it might cause.
> Have big pipes? SourceForge.net is looking for download mirrors. We supply
> the hardware. You get the recognition. Email Us: bandwidth@...
> User-mode-linux-devel mailing list
On Fri, May 10, 2002 at 03:33:09PM -0400, Chandan Kudige wrote:
> Hmm ... I had lots of problems trying to compile on Win32 without SMP
> support. So I just turned it on. I will debug these eventually ...
Can you point out these problems? I rememeber I had no problems compiling
with UP when I worked on this port.
(hopefully I fixed my SMTP client this time)