What this patch actually fixes is the interrupt handling in the ISA controller.
Since both IDE and NE2K go through ISA on Prep...

Date: Fri, 19 Oct 2007 21:39:25 +0200
From: Aurelien Jarno <aurelien@aurel32.net>
To: Jocelyn Mayer <l_indien@magic.fr>
Subject: Re: [Qemu-devel] I got a kernel booted under qemu-system-ppc !
Message-ID: <20071019193925.GA30604@hall.aurel32.net>

On Fri, Oct 19, 2007 at 07:39:47PM +0200, Jocelyn Mayer wrote:
> > The small patch below fixes the IDE problem, but not the NE2000 ISA one.
> > Please apply.
> 
> Interesting, thanks. I'll test this and apply or check for more fixes if
> needed... I'll also try to check what's happening with the NE2000. Could
> it be the ne2000_irq table in ppc_prep.c would not be correct ? It may
> also be usefull to be able to use PCI network devices, if the target
> runs properly, isn't it ?

Well the problem is actually in the kernel, MIPS got the exact same
problem a few times ago. The same fix apply for PowerPC:

Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>

diff --git a/arch/powerpc/sysdev/i8259.c b/arch/powerpc/sysdev/i8259.c
index ad87adc..9134b2e 100644
--- a/arch/powerpc/sysdev/i8259.c
+++ b/arch/powerpc/sysdev/i8259.c
@@ -138,6 +138,7 @@ static void i8259_unmask_irq(unsigned int irq_nr)
 static struct irq_chip i8259_pic = {
 	.typename	= " i8259    ",
 	.mask		= i8259_mask_irq,
+	.disable	= i8259_mask_irq,
 	.unmask		= i8259_unmask_irq,
 	.mask_ack	= i8259_mask_and_ack_irq,
 };


I will submit it later this week-end.


> If the proposed ROM image is best suitable to make PreP target run (and
> as I don't spend a lot of time hacking OHW those days), it may also be
> interesting to add a ppc_prep_rom.bin to the repository... In fact, it
> was maybe not a good idea to try to use the same BIOS image on all
> PowerPC targets...

I would say that the main avantage of this proposed ROM is that it is
plainly functional with a recent kernel. I think that it is something
really important, as it will bring more users who will improve qemu ppc
by reporting bugs or even submitting patches. 

And if a better ROM image (e.g. OpenBIOS) that matches real hardware
comes later it would be possible to change that again.


I have used QEMU CVS with a Debian Sid image. It basically works, I am
even able to login via SSH, but I have noticed two problems:

- Some process hang, stay into "D" state and become unkillable. It seems
  it can happen to all processes, but it is always reproducible with
  uptime or top. I still don't know if it is a problem of the kernel or
  if it comes from the emulation.
- The target CPU never gets into idle loop, so the host CPU is always
  used at 100%

-- 
  .''`.  Aurelien Jarno	            | GPG: 1024D/F1BCDB73
 : :' :  Debian developer           | Electrical Engineer
 `. `'   aurel32@debian.org         | aurelien@aurel32.net
   `-    people.debian.org/~aurel32 | www.aurel32.net